Gibbs EmbodimentAndCognitiveScience

 

This page intentionally left blank 

 

Embodiment and Cognitive Science

This book explores how people’s subjective, felt experiences of their  bodies in action provide providepart part of the fundamental grounding groundingfor for human cognition and language. Cognition is what occurs when the body engage ga gess th thee ph phys ysic ical al an and d cu cult ltur ural al wo worl rld, d, an and d it mu must st be st stud udie ied d in te term rmss of the dynamical interactions between people and the environment. Huma Hu man n la lang ngua uage ge an and d th thou ough ghtt em emer erge ge fr from om re recu curr rrin ing g pa patt tter erns ns of em em- bodied activity that constrain ongoing intelligent behavior. behavior. We We must not assume cognition to be purely internal, symbolic, computational, computational, and disembodied, but seek out the gross and detailed ways in which language and thought are inextricably shaped by embodied action. Embodiment and Cognitive Science describes Science  describes the abundance of empirical evidence from many disciplines, including work on perception, concep con cepts, ts, ima imager gery y and re reaso asonin ning, g, lan langua guage ge and com commun munica icatio tion, n, cog cog-nitive development, and emotions and consciousness, that support the idea that the mind is embodied. Raymond W. Gibbs, Jr. is Professor of Psychology at the University   The Poetics of Mind and Mind  and of California, Santa Cruz. He is the author of  The Intentions in the Experience of Meaning. Meaning . He is coeditor (with G. Steen)  Metaphor in Cognitive Linguistics and Linguistics and editor of the interdisciplinary of  Metaphor interdisciplinary  journal Metaphor  journal  Metaphor and Symbol. Symbol.

 

Embodiment and Cognitive Science

RAYMOND W. GIBBS, JR. University of California, Santa Cruz

 

     

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge   , UK  Published in the United States of America by Cambridge University Press, New York 

 www.c ambrid ridge. ge.org  org   www .camb Information on this title: www.cambridge.org/9780521811743 © Cambridge University Press 2005 This publication is in copyright. Subject to statutory exception and to the provision of  relevant collective licensing agreements, no reproduction of any part may take place  without the written permission of Cambridge University Press. First published in print format 2005

- 

eBook (EBL) --- eBook (EBL)

- 

----

- 

---

- 

---- paperback 

- 

--- paperback 

- 

----

hardback  hardback 

Cambridge University Press has no responsibility for the persistence or accuracy of s for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

 

Contents

 Acknowledgments

 

 vii page vii page

1 2   3   4   5   6   7  

Introduction Bodies and Persons Perception and Action Concepts Imagery,, Memory Imagery Memory,, and Reasonin Reasoning g Language and Communication Cognitive Development

1    

42

 

79      

8   Emotion and Consciousness 9   Conclusion

References Index

14

   

158 208

   

123

239 275 283 325

v

 

Acknowledgments

I thank Greg Bryant, Christin Izett, Melissa Newman, and Nicole Wilson for their important comments on earlier versions of some of the chapters in this book. Ben Bergen and Alan Cienki and his students at Emory University also offered extremely extremely helpful comments on parts of this book. Many conversations with Herb Colston Col ston and Guy Van Orden were critical in sharpening some of the ideas discussed here. Many Ma ny th than anks ks to Ph Phil il La Laug ughl hlin in an and d th thee en enti tirre st staf afff at Ca Camb mbri ridg dgee Un Univ iver er-sity Press for their wonderful support and expertise while this book was  being written and produced produced for publication. This book is dedicated to Christin Izett in appreciation of her love and

support throughout the writing of this book.

vii

 

1 Introduction

Embodiment in the field of cognitive science refers to understanding the roleofanagent’sownbodyinitseveryday,situatedcognition.Forexample, how do our bodies influence the ways we think and speak? Consider the following narrative written by a   23-year-old woman, Sandra, who was aske as ked d to de desc scri ribe be a rec ecen ent, t, im impo port rtan antt li life fe ev even ent. t. Sa Sand ndra ra be bega gan n he herr na narr rrat ativ ivee  by noting that she was engaged to be married to an older man who worked in the computer industry in northern California. Quite recently, Sandra’s fianc´e asked her to sign a prenuptial agreement and this request evoked many feelings that Sandra struggled to deal with. wit h. I know that I shouldn’t be so naive about a bout this sort of thing, but when he presented me with a draft of the agreement, it was so formal and legal and felt so cold to me that th at I ju just st br brok okee do down wn cr cryi ying ng.. I si simp mply ly co coul uldn dn’t ’t st stan and d to se seee ou ourr fu futu ture re rel elat atio ions nshi hip p  be reduced to questions que stions of money. money. It seemed like li ke Barry didn’t trust me, or that he lack la cked ed fa fait ith h in ou ourr fu futu ture re.. I ha had d al alwa ways ys th thou ough ghtt th that at we we werre in th this is to toge geth ther er,, go goin ing g forw fo rwar ard d as pa part rtne ners rs as we st star arte ted d da dati ting ng,, go gott se seri riou ous, s, th then en en enga gage ged, d, an and d ho hope pefu full lly y soon married. Now my parents want me to consult with a lawyer to insure I don’t get screwed by the pre-nup.

I m trying hard to find the right balance between understanding Barry s needs to prot pr otec ectt hi hims msel elff an and d my ow own n ne need edss fo forr em emot otio iona nall se secu curi rity ty . . . I’ I’m m tr tryi ying ng to be fle flexi xibl blee about the the whole thing thing . . . I love Barry Barry and I know know he loves loves me and I wish that that the feeling of love would be enough to sustain us through anything. But the idea of  getting divorced, even before we have been married, makes me ill. Everyone tells me that I’ll get over this and that doing the pre-nup is probably the right thing to do. That may be so. The wedding is in August. Hopefully by that time, I’ll be mellow about what we’re going through right now now..

This narrative is not particularly remarkable in terms of how Sandra described her recent experience. However, a closer look at what is said reveals how various embodied experiences help structure the narrative. For example, Sandra commented early on that “I couldn’t stand to see 1

 

2

 

Embodiment and Cognitive Science

ourr fu futu turre rel elat atio ions nshi hip p be red educ uced ed to qu ques esti tion onss of mo mone ney y,” ref efer erri ring ng to th thee ou physical experience of standing, or failing to remain standing, to describe how ho w sh shee fe felt lt ab abou outt he herr re rela latio tions nshi hip p be beco comin ming g so ce cent nter ered ed on mo mone ney y is issu sues es.. Late La terr on on,, Sa San ndr draa sa said id,, “I had al alwa ways ys th thou oug ght th that at we we werre in th this is to toge geth ther er,, goin go ing g fo forw rwar ard d as pa part rtne ners rs as we st star arte ted d da dati ting ng,, go gott se seri riou ous, s, th then en en enga gage ged, d, and hopefully soon married.” At this moment, Sandra clearly talked of  her relationship in terms of being physically together with her boyfriend as they started out on a journey, beginning when they first began dating, soon traveling traveling to the point of gettin getting g serio serious, us, and then moving forward forward along a path toward the eventual destination of marriage. Sand Sa ndra ra al also so no note ted d he herr st stru rugg ggle le “t “to o fin find d th thee ri righ ghtt ba bala lanc ncee be betw twee een n un unde derrstan st andi ding ng Ba Barr rry’ y’ss ne need edss to pr prot otec ectt hi hims msel elff an and d my ow own n ne need edss fo forr em emot otio iona nall security.” This emotional experience is referred to metaphorically, as if  Sandra were physically balancing two opposing weights while trying to remain upright. As she worked to come to terms with her fianc´ fiance’s e´ ’s request for a prenuptial agreement, Sandra was “trying to be flexible about the whole thing,” showing she to is conceptualizing her emotional experience as ifagain her body mustthat adjust remain flexible in order not to  be injured when confronted confronted with physical burdens. Finally Finally,, Sandra hoped hoped for her wedding that “by that time, I’ll be mellow about what we’re going through right now,” now,” referring to the physical obstacle that she and her fianc´e were struggling to overcome along the path of their relationship  journey..  journey Sandra’s narrative illustrates how the ways we think about our experiences may be shaped by embodiment. She specifically talked of her mental/emotional experiences in terms of recurring patterns of embodied action (e.g., standing, being flexible, movement along paths toward goals, remaining balanced). Sandra was likely not conscious of the embodied character of her words, and readers probably do not think of her emotional experiences as specific embodied actions. Yet Sandra’s description of he herr em emot otio iona nall ex expe peri rien ence cess in te term rmss of em embo bodi died ed ac acti tion on is no nott a li ling ngui uist stic ic accident, but demonstrates how embodiment provides the foundation for how people interpret their lives and the world around them. What must a body be like for it to support cognition, language, and consciousness? Did Sandra’s embodied experience shape the way she thought about particular topics, or did she merely talk that way? One of 

the traditional beliefs in the cognitive sciences is that intelligent behavior, behavior, including the ability to perceive, think, and use language, need not arise from fr om an any y sp spec ecifi ificc bo bodi dily ly fo form rm.. Th Ther ermo most stat ats, s, co comp mput uter ers, s, rob obot ots, s, an and d br brai ains ns in vats may all, under the right circumstances, exhibit sophisticated cognitive skills. Under this view, cognitive systems are best characterized in terms of their functional states (i.e., their logical and computational process ce sses es)) wi with thou outt co conc ncer ern n fo forr ho how w th thes esee st stat ates es ar aree ph phys ysic ical ally ly rea ealiz lized ed (i (i.e .e., ., as human brains, silicon chips, or robots). The building materials that shape

 

Introduction

 

3

the contents of mental life simply do not matter. Minds may be realized in flesh, silicon, or even cream cheese (Putnam,  1975). To be in a specific mental state is simply to be in a physical device of whatever type satisfies a specific formal/functional description. This Th is tr trad aditi ition onal al co conc ncep epti tion on of mi mind nd an and d bo body dy ha hass imp impos osed ed se seri riou ouss lim limiitati ta tion onss on th thee sc scho hola larl rly y st stud udy y of me ment ntal al li life fe in co cogn gnit itiv ivee sc scie ienc nce. e. Al Alth thou ough gh psychologists and others readily admit that much knowledge is derived from sensory perception, few scholars, until recently, have emphasized the importance of kinesthetic action in theoretical t heoretical accounts of how people percei per ceive, ve, lea learn, rn, thi think, nk, exp experi erienc encee emo emotio tions ns and con consci scious ousnes ness, s, and use lan lan-guage. This book advances the idea that the traditional disembodied view of mind is mistaken, because human cognition is fundamentally shaped  by embodied experience. My aim is to describe the way in which many aspects of cognition are grounded in embodiment, especially in terms of the phenomenological experience of our bodies in action. Embodiment may nott provi no vid de th thee sin ingl glee fo foun unda dati tion on fo forr al alll th thou oug ght an and d la lang ngua uage ge,, bu butt it is an esse es sent ntia iall pa part rt of th thee pe perrce cept ptua uall an and d co cogn gnit itiv ivee pr proc oces esse sess by wh whic ich h we ma make ke sense of has our cognitive experiences in the been world. Why science so neglectful of embodiment in constructing theories of perception, cognition, and language? The denial of  the body in consideration of human thought has been part of the Western Western intellectual tradition since the time of the ancient Greeks. Perhaps Perhaps the best voice for this earlier view was Plato, as shown in the following dialogue from the “Phaedo”: All these considerations, said Socrates, must surely prompt serious philosophers to review review the the positio position n in some such such way as this this . . . So long long as we keep the body body and our soul contaminated contaminated with this imperfection, imperfection, there is no chance of our ever attaining attaini ng satisfactor satisfactorily ily to our object, object, which which we assert assert to be the the truth. . . . The body fills us with loves and desires and fears and all sorts of nonsense, with the result thatt we literally tha literally never never get an opp opport ortunit unity y to think at all about anythin anything g . . . That is why, why, on all these accounts, accounts, we have have so little time time for philosophy philosophy.. . . . It seems, seems, to  judge from the argument, that the wisdom which we desire and an d upon u pon which wh ich we profess profe ss to have have set our our hearts will be be attainable attainable only only when we we are dead dead . . . It seems seems that so long as we are alive, we shall continue closest to knowledge if we avoid as much as we can all contact and association with the body, except when they are absolutely necessary, and instead of allowing ourselves to become infected with its nature, purify ourselves from it until God himself gives us deliverance. (Hamilton & Cairns, 1961: 49)

Plato viewed the body as a source of distraction in intellectual life that must be eradicated in the practice of philosophy. Separation of the mind

andbodyandthehierarchicalorderingofmindoverbodyhauntthehistory of Wes este tern rn ph phil ilos osop ophi hica call ac acco coun unts ts of kn know owle ledg dgee fr from om Pla Plato to,, Ar Aris isto totl tle, e, an and d Augustine through to Descartes and Kant. Forrivaled example, in early Christian writings, bodily sensations and desires were in contests against a

 

4

 

Embodiment and Cognitive Science

higher form of Truth, or closeness to God. As St. Augustine wrote in the fifth fif th ce cent ntur ury y, “M “Mor oree an and d mo morre, O Lo Lorrd, yo you u wi will ll in incr crea ease se yo your ur gi gift ft in me me,, so that my soul may follow me to you, freed from the concupiscence which  binds it, and rebel no more against itself” (Augustine, 1961: 234). Aug August ustine ine fixed fix ed th thee bo body dy as a so sour urce ce of si sin, n, we weak akne ness ss,, an and d th thee me meas asur uree ag agai ains nstt wh whic ich h the strength of his will toward God is knowable. Inthe 17th cen centur tury y, Ren Renee Des Descar cartes tes’’ str strugg uggle le wit with h a pur purely ely mat materi erial al bod body y and a perfectly insubstantial mind led him to propose that the body is, in fact, an idea in the mind (Descartes,  1984,  1985). The body’s materiality, along with other objects that are impressed upon body substances, is a literalization of this idea in our experience. When we pay attention to it, the body materializes, and we become aware of the body as an object. However,, as our attention centers on other things, or on thought itself, the However  body disappears. Mental phenomena, according to Descartes, have no place in the quantifiable world of physics, but have a completely autonomous status: “I am a sub ubssta tanc ncee th thee wh whol olee na natu turre or es esse sen nce of wh whic ich h is to th thin ink, k, an and d wh whic ich h fo forr its existence does not need any place or depend on any material thing” (Descartes, Discourse, Part IV). Descartes distinguished, then, physical substances (“res extensa”), which can be measured and between divided, and thinking substances (“res cogitans”), which are unextended and indivisi vi sibl ble. e. Th Thee hu huma man n bo body dy,, in incl clud udin ing g th thee br brai ain n an and d ne nerv rvou ouss sy syst stem em,, be belo long ngss to the first group, whereas the mind, including all thoughts, desires, and volitions, belongs to the second. Cartesian dualism arose from Descartes’ claim that he could doubt the exis ex iste tenc ncee of ph phys ysic ical al ob obje ject cts, s, in incl clud udin ing g hi hiss ow own n bo body dy,, bu butt no nott th thee ex exis iste tenc ncee of his thoughts or thinking. Although Descartes worried about possible interactions of mind and body, Cartesian dualism evolved into an epistemological tradition that separated the mind as rational, thinking, immaterial, and private from the body as an irrational, corrupt, and physical substance that merely provided public, physical exertion on the material worl wo rld. d. Th This is bi bifu furrca cati tion on of th thee pe pers rson on in into to mi mind nd an and d bo body dy ha hass su subs bseq eque uent ntly ly given rise to many other dualisms, including subjective as opposed to ob jective, knowledge as opposed to experience, reason as opposed to feeling, theory as opposed to practice, and verbal as opposed to nonverbal. Cartesianism has also led to the romantic view of the body as the last bastion of what is natural, unspoiled, preconceptual, and primitive in experience. Bodily movement is viewed as behavior behavior,, with little relevance to language, thought, or consciousness, and not as meaningful action. The Western tradition since Descartes has generally assumed that the  body is a solid object and the self, in particular the mind, is an ethereal subject mysteriously infused into i nto the body object. Throughout history, history, the mind has been modeled as a series of different material objects (e.g., a hydraulic dra ulic mac machin hine, e, a tel teleph ephone one swi switch tchboa board rd,, a hol hologr ogram, am, a dig digital italcom comput puter) er)..

 

Introduction

 

5

Cog Cognit nitive ive sci scienc ence, e, as an int inter erdis discip ciplin linary ary re resea searc rch h ent enterp erpris rise, e, cam camee int into o being in the  1950s with the rise of the MIND IS A COMPUTER metaphor, which whi ch re resul sulted ted fr from om tec techno hnolog logica icall adv advanc ances es in com comput puting ing mac machin hinery ery.. Ala Alan n Turi uring ng (1950) ou outl tlin ined ed a me meth thod od fo forr as asse sess ssin ing g th thee qu ques estio tion n “C “Can an ma mach chin ines es think?” Following in Descartes’ footsteps, Turing emphasized the importance of drawing a “fairly sharp line” (p.  434) between a person’s physical capacities and his/her intellectual abilities. Turing asked us to consider a scenario that included three people – a man (A), a woman (B), and an interrogator of unspecified sex (C). The interrogator was in a separate room from the man and the woman, and the interrogator’s task was to determine which of the two was a man and which was a woman on the basis of  thei th eirr wr writ itte ten n an answ swer erss to ce cert rtai ain n qu ques esti tion onss (e (e.g .g., ., “W “Wha hatt is th thee le leng ngth th of yo your ur hair?”). It is A’s task to confuse the interrogator and B’s task to help. The test proper comes into play by swapping the man (A) with a machine. If  thee in th inte terr rrog ogat ator or ma make kess th thee sa same me se sett of ju judg dgme ment nts, s, de dedu duct ctio ions ns,, an and d gu gues esse sess after the swap as before (i.e., the interrogator is unable to distinguish the mach ma chin ine’ e’ss an answ swer erss fr from om th thee ma man’ n’ss an answ swer ers) s),, th then en th thee ma mach chin inee ha hass pa pass ssed ed the “Turing test.” The machine whose behavior is indistinguishable from the intellect of the man is the machine that thinks. Cognitive science models of intelligent human activity have mostly continued to assume, like Turing, that cognition is autonomous, logical, and disembodied. In his history of cognitive science, Gardner ( 1985) claimed that the exclusion of the body was, in fact, a benign methodological decision: “Though mainstream cognitive scientists do not necessarily  bear any animus against the affective realm, against the context that surrounds any actor or thought, or against historical or cultural analyses, in practice they attempt to factor out these elements to the maximum extent possible possi ble . . . This may may be a question question of of practicali practicality: ty: if one one were were to take into accoun acc ountt the these se ind individ ividual ualizin izing g and phe phenom nomena enalis listic tic ele elemen ments, ts, cog cognit nitive ive sci sci-ence might well become impossible” (p.  41). Some cognitive scientists question whether the exclusion of the phenomenological body, body, along with other aspects of experience such as emotion and consciousness, is merely a methodological issue, and not really consti con stitut tutive ive of wha whatt cog cognit nitive ive sci scient entist istss bel believ ievee is ess essent ential ial abo about ut cog cognit nition ion.. Of co cour urse se,, ma many ny sc scho hola lars rs no now w tr try y to av avoi oid d th thee st stri rict ct se sepa para rati tion on of mi mind nd an and d  body assumed by Cartesian dualism. The most popular strategy strategy,, especially in rec ecen entt de deca cade des, s, ha hass be been en to red educ ucee me ment ntal al ev even ents ts to br brai ain n pr proc oces esse sess an and d replace internal explanations with instrumental ones. In some cases, the reduction of mind to brain carries with it the reduction of body to brain. Neuroscientists, for instance, seldom acknowledge the role played by the  body as a whole in the cognitive operation of the brain. The body is reduced to its representation in the somatosensory cortex and is considered impo im port rtan antt on only ly to th thee ex exte tent nt th that at it pr prov ovid ides es th thee ra raw w se sens nsor ory y in inpu putt re requ quir ired ed for cognitive computations. In other cases, the body is first reduced to the

 

6

 

Embodiment and Cognitive Science

psychology, mind, and then reduced to the brain. This is especially true in psychology, where the body is first treated as an intentional object (i.e., an image, a mental representation) and then reduced to neural computations. Cont Co ntem empo pora rary ry ph phil ilos osop ophe hers rs ar argu guee ov over er wh whet ethe herr a ph phys ysica icall bo body dy is ne neccessa es sary ry fo forr kn know owle ledg dgee an and d co cogn gnit itio ion, n, of ofte ten n by co cons nsid ider erin ing g th thee imp implic licat atio ions ns of di difffe ferren entt th thou ough ghtt ex expe peri rime ment ntss in wh whic ich h th thee mi mind nd ma may y be di divo vorrce ced d fr from om critical aspects of bodily experience. For instance, consider the following scenario: Imagine a brilliant neuroscientist named Mary, who has lived her entire life in a room that is rigorously controlled to display only various shades of black, white, and grey. She learns about the outside world by means of a black/white television monitor, and being brilliant, she manages to transcend these obstacles. She  becomes the world’s greatest neuroscientist, all from within this room. In particular, she comes to know everything there is to know about the physical structure and activity of the brain and its visual system, of its actual and possible states. (Churchland, 1985: 22)

Philosophers argue, based on examples like the above, over whether qual qu alia ia (i (i.e .e., ., th thee ph phen enom omen enal al ch char arac acte terr of ou ourr ex expe peri rien ence ce), ), su such ch as on one’ e’ss su subb jective sensations of color color,, must be mental states that are causally related to the neurophysiology of the brain (see Churchland, 1984; Jackson, 1982, 1986). Th Thes esee sc scen enar ario ios, s, ho howe weve verr, dr dram amat atic ical ally ly fa fail il to re reco cogn gniz izee th thee ne need ed fo forr a real living body in knowing about the world. There is no acknowledgment me nt of Ma Mary ry as a li livi ving ng pe pers rson on,, ma made de of fle flesh sh,, bl bloo ood, d, an and d bo bone ne,, wh who o mo move vess and has awareness of the felt qualities of her own actions. Mary’s firstpers pe rson on ex expe peri rien ence cess of he herr ow own n bo body dy in re rela latio tion n to th thee en envi virron onme ment nt pr prov ovid idee knowledge that is “qualitatively incommensurate” with whatever may  be happening in her own brain, or anyone else’s (Sheets-Johnstone, 1999: 167). Mary learns about qualia because she subjectively experiences them through her own bodily actions. Knowledge of a set of abstract propositions, tio ns, suc such h as Mar Mary’s y’s und unders erstan tandin ding g of the neu neuro rophy physio siolog logy y of col color or vis vision ion,, means nothing unless a person experiences in some embodied sense the physical world to which these propositions refer (Sheets-Johnstone, 1999). Cognitive psychologists, like many philosophers, often fail to recognize the significance of embodied action in the study of human mental life. Most experimental investigations of perception and cognition occur in laboratory situations where a person passively observes stimuli and then responds in some specified manner to what has been presented. In some so me in inst stan ance ces, s, th thee pe pers rson on is ph phys ysic ical ally ly res estr tric icte ted d in hi hiss or he herr mo move veme ment ntss (e.g., head rests are used in psychophysical experiments). In cases where the participant must move to respond to stimuli, such as having to push a  button or speak aloud, psychologists work hard to eliminate the movement from their theoretical understanding of the processes involved in perception and cognition. Cognitive processes, especially, are viewed as strictly

 

Introduction

 

7

mental phenomena that have little to do with embodied experience. The  body is the vessel for the mind and brain, but has negligible importance in characterizing the essence of mental life. However, the situation is now changing. Consider just three examples of how psychologists now pay attention to embodied action when studying different cognitive phenomena. First, the classic empirical work on mental imagery investigates possible correspondences between mental ima imager gery y and vis visual ual per percep ceptio tion. n. For exa exampl mple, e, par partic ticipa ipants nts in one classic study were presented with two-dimensional drawings of pairs of three dimensional objects. The participants’ task was to determine whether the two represented objects were identical except for orientation (Shepard & Metzler,  1971). Some of the figures required rotation solely within the picture plane, whereas others required rotation in depth (“into” the page). The general result was that, whether for two- or threedimens dim ension ional al ro rotat tation ions, s, par partic ticipa ipants nts see seemed med to ro rotat tatee the obj object ectss men mental tally ly at a fix fixed ed ra rate te of ap appr prox oxim imat atel ely y 60 deg degre rees/ es/sec second ond.. For man many y yea years, rs, psy psycho chollogists assumed that cognitive abilities, such as those observed in mental rota ro tatio tion n st stud udie ies, s, de demo mons nstr trat atee th thee ti tigh ghtt li link nk be betw twee een n vi visu sual al pe perrce cept ptio ion n an and d mental men tal ima imager gery y. Alt Althou hough gh num numer erous ous stu studie diess exa examin minee peo people ple’s ’s kin kinest esthet hetic ic and motor imagery, scholars traditionally have not searched for explicit relations between kinesthetic activity and mental imagery imagery.. However,, recent work suggests that many aspects of visual However vi sual and motor imagery share a common representational, and possibly neuropsychological, substrate. Various studies demonstrate that the ability to transform mental images is linked to motor processes, so that rotating one’s hands in the direction opposite to the required mental rotation slows down the speed of mental rotation (Wexler, Kosslyn, & Berthoz, 1998). Researchers now claim that “visuomotor anticipation is the engine that drives mental rotation” (Wexler et al.,  1998). Under this view, similar mechanisms drive both visual image transformation and the production of embodied movements. The ability to plan movements as simulated actions, and not as actual motor plans, may be the common element underlying embodied action act ion and men mental tal ima imager gery y per perfor forman mance ce (Jo (Johns hnson, on, 2000). Th Thes esee ne new w de deve vellopments in cognitive psychology illustrate how correcting for a previous neglect of embodied experience in experimental studies leads to a richer picture of the importance of embodiment in human cognition. Psyc Ps ycho holi ling ngui uist stss ha have ve al also so sl slow owly ly be begu gun n to se seek ek ou outt th thee em embo bodi died ed fo foun un-dation of linguistic structure and meaning. Recall Sandra’s earlier comment in response to her fianc´e’s e’s request for a prenuptial agreement that “I couldn’t stand to see our future relationship be reduced to questions of  mone mo ney y.” Wh Why y is it th that at Sa Sand ndra ra us used ed th thee wo worrd “s “sta tand nd”” to ref efer er to an ab abst stra ract ct,, mental experience of her adjusting to her fianc´e’s e’s demand? Traditional studies on how people process ambiguous, or polysemous, words such as “stand” generally assume that each sense of a word is listed as part of 

 

8

 

Embodiment and Cognitive Science

its entry in the mental lexicon. For example, do people immediately access ce ss al alll th thee po poss ssib ible le se sens nses es fo forr th thee wo worrd “s “sta tand nd,” ,” wi with th co cont ntex extt de dete term rmin inin ing g which meaning is appropriate afterward? Or does context constrain lexical access so that only the correct meaning of “stand” is accessed during immediate utterance interpretation? These empirical questions have been

studied extensively (Gorfein, 2001). Psycholinguists rarely ask whether people have intuitions about why “stand,” or any polysemous word, has the variety of meanings it does. Recent studies, however, demonstrate that people’s intuitions about the me mean anin ings of “sta “s tand nd” ” ar are e sh shap ed by thei th r em embo died ed, pe expe ex peri rien ces s ofre stan st andi ding ng (Gibbs (Gi bbs, , gs Beitel Bei tel, , Har Harrin ringto gton, n, aped & San Sander ders, s,eir 1994 )bodi . Th Thus us, peop ople leence taci ta citly tly reco cogn gnize ize that Sandra’s use of “stand” has a metaphorical meaning that is related to their embodied experiences of struggling to remain physically upright when some physical force acts against them. People’s understandings of  lingu lin guis isti ticc me mean anin ings gs ar aree no nott di divo vorrce ced d fr from om th thei eirr em embo bodi died ed ex expe peri rien ence ces, s, bu butt rather are fundamentally constrained by them in predictable ways. Following Piaget’s early writings, developmental psychology has also started to meaningfully explore how embodied action may underlie children’s acquisition of perceptual/conceptual knowledge. For example, infants’ interest in things that move assists them in understanding some causecau se-ef effec fectt re relat lation ionss in the phy physic sical al wor world. ld. Sop Sophis histic ticate ated d stu studie diess ind indica icated ted that infants 12 months old and younger are capable, in the right setting, of  making causal attributions to the behavior of objects they see in the world (Gerg (Ge rgely ely,, Nad Nadasd asdy y, Csi Csiba, ba, & Bir Biro, o, 1995; Spe Spelke lke,, Phi Philip lip,, & Wood oodwar ward, d, 1995). Thee in Th infa fant nt’s ’s de deve velo lopi ping ng se sens nsiti itivi vity ty to ca caus usal al rel elat atio ions ns ma may y un unde derl rlie ie th thee ac ac-quis qu isiti ition on of a co conc ncep eptt fo forr ag agen ency cy (i (i.e .e., ., th thin ings gs mo move ve be beca caus usee of in inte tern rnal al fo forc rces es or human intentions). These studies, however, despite their brilliance, situate the child as a passive observer who learns to reason about the physical world by visual inspection of real-world events. Several experiments now demonstrate the importance of the child’s bodily exploration of the physical world in learning about objects and their behaviors (Adolph,  1997, 2000; Ber Berten tentha thal, l, Cam Campos pos,, & Ker Kermoia moian, n, 1994; Hertenstein, 2002; Needham, Barrett, & Peterman, 2002). This empirical work suggests that many basic conc co ncep epts ts ma may y ar aris isee fr from om ru rudi dime ment ntar ary y bo bodi dily ly ac acti tion onss an and d yo youn ung g ch child ildrren en’s ’s felt fe lt ex expe peri rien ence cess of th them em.. Ca Caus usat atio ion n an and d ag agen ency cy,, fo forr ex exam ampl ple, e, ma may y be ro root oted ed in infants’ phenomenological sense of their own bodies’ interactions with objects and other people. Even before infants possess any ability to physicall ca lly y ma mani nipu pula late te ob obje ject ctss wi with th th thei eirr ha hand ndss an and d fe feet et,, th they ey di dire rect ctly ly ex expe peri rien ence ce caus ca usee an and d ef effe fect ct fr from om th thee mo move veme ment nt of th thei eirr lip lips, s, to tong ngue ues, s, an and d mo mout uths hs du durring in g br brea east stfe feed edin ing, g, or fr from om ch chew ewin ing g fo food od,, wh whic ich h tr tran ansf sfor orms ms it to so some meth thin ing g that th at ca can n be sw swal allo lowe wed d ea easi sily ly.. An en enco cour urag agin ing g tr tren end d in de deve velo lopm pmen enta tall ps psyycholog cho logy y is gr great eater er att attent ention ion giv given en to inf infant ants’ s’ phe phenom nomeno enolog logica icall exp experi erienc encee in relation to cognitive growth.

 

Introduction

 

9

These brief examples illustrate how looking for embodied action in thought and language may provide a different picture of human cognition than has traditionally been assumed within cognitive science. Much recent work in cognitive science views embodiment as a matter of brain states and neural activity. We have indeed learned a great deal from these neuroscientific studies. However, as Roger Sperry noted over sixty-five years ago, “An objective psychologist, hoping to get at the physiological side of behavior, is apt to plunge immediately into neurology trying to corr co rrel elat atee br brai ain n ac acti tivi vity ty wi with th mo mode dess of ex expe peri rien ence ce.. Th Thee res esul ultt in ma many ny ca case sess

only accentuates the gap between the total experience as studied by the ps psyc ycho holo logis gist t an and d ne neur uron onal al ac activ ity as an anal alyz yzed ed by th thee and neur ne urol olog ogis ists ts. . Bu Butt th thee experience of the organism istivity integrated, organized, has its meaning in terms of coordinated movement” (1939: 295). The psychologist Scott Kelso more recently suggested, “It is important to ke keep ep in mi mind nd . . . th that at th thee br brai ain n di did d no nott ev evol olve ve me merrel ely y to reg egis iste terr rep eprres esen en-tations of the world; rather, it evolved for adaptive actions and behaviors. Muscul Mus culosk oskele eletal tal str struct uctur ures es coe coevol volved ved wit with h app appro ropri priate ate bra brain in str struct uctur ures es so that th at th thee en enti tirre un unit it mu must st fu func ncti tion on to toge geth ther er in an ad adap apti tive ve fa fash shio ion n . . . it is th thee entire system of muscles, joints, and proprioceptive and kinesthetic functions and appropriate parts of the brain that evolve and function together in a unitary way” (1995: 268). The brain is certainly part of an integrated dynamic system devoted to the mom moment ent-by -by-mo -momen mentt emb embodi odied ed dyn dynami amics cs of eve everyd ryday ay lif life. e. View iewing ing the  brain simply as an information-process information-processing ing or computational device, as the center of cognition, ignores the centrality of animate form in human thought (Sheets-Johnstone, 1999). Thiss boo Thi book k des descri cribes bes the way wayss tha thatt per percep ception tion,, con concep cepts, ts, men mental tal ima imager gery y, memory,, reasoning, cognitive development, language, emotion, and conmemory scio sc ious usne ness ss ha have ve,, to va vary ryin ing g ex exte tent nts, s, gr grou ound ndin ings gs in em embo bodi dime ment nt.. My st stra rattegy eg y in ex expl plor orin ing g th thee si sign gnifi ifica canc ncee of em embo bodi dime ment nt in th thee st stud udy y of th thes esee to topi pics cs adopts what may be called the “embodiment premise”: People’s subjective, felt experiences of their bodies in action provide part of the fundam fun dament ental al gr grou ound nding ing fo forr lan langua guage ge and th thoug ought ht.. Co Cogni gniti tion on is wh what at oc occur curss wh when en the body engages the physical, cultural world and must be studied in terms of the dynamical interactions between people and the environment. Human language and thought emerge from recurring patterns of embodied activity that constrain ongoing intelligent behavior beha vior.. We We must not assume cognition to be purely internal, symbolic, computational, and disembodied, but seek out the gross and detailed ways that language and thought are inextricably shaped by embodied action.

Thekeyfeatureofthispremiseistheideathatunderstandingtheembodied nature of human cognition demands that researchers specifically look for possible mind-body and language-body connections. Understanding embodied experience is not simply a matter of physiology or kinesiology

 

10

 

Embodiment and Cognitive Science

(i.e., the body as object), but demands recognition of how people dynamically move in the physical/ physical/cultural cultural world (i.e., the body experienced from fr om a fir first st-p -per erso son, n, ph phen enom omen enol olog ogic ical al pe pers rspe pect ctiv ive) e).. Th Thee mi mind nd (i (its ts im imag ages es,, thoughts, representations) representations) is created from ideas that are closely related to  brain representations representations of the body body and to the body’s continued activities in the real world. Fort Fo rtun unat atel ely y, th ther eree is an ac accu cumu mula lati ting ng bo body dy of em empi piri rica call ev evid iden ence ce sh show ow-ing how embodied activities shape human cognition. In the spirit of  cognitive science, this “empirical” evidence includes data collected from controlled contr olled labor laboratory atory studie studies, s, natur naturalisti alisticc field obser observation vations, s, neur neuropsyc opsychohological case studies, linguistic resear research, ch, artificial intelligence (and artificial life) andscholars variouswhose phenomenological studies and reports. Toenbe sure,modeling, many of the studies are described here may not

tirely tirel y ag agrree wi with th my in inte terp rpre reta tati tion on of th thei eirr wo work rk as su supp ppor ortt fo forr “e “emb mbod odie ied” d” cogn co gnit ition ion.. So Some me of th thes esee di disa sagr gree eeme ment ntss ce cent nter er ar arou ound nd wh what at is me mean antt by th thee term te rmss “e “emb mbod odie ied” d” an and d “e “emb mbod odim imen ent. t.”” I ar argu guee th that at “e “emb mbod odim imen ent” t” ma may y re re-ferr to fe to,, at le leas ast, t, th thrree le leve vels ls of pe pers rson onho hood od (s (see ee La Lako kofff & Jo John hnso son, n, 1999): ne neuural events, the cognitive unconscious, and phenomenological experience. Although amazing advances have been made in understanding neural processes, insignificant attention has been given to people’s phenomenologica log icall exp experi erienc encee in exp explai lainin ning g man many y asp aspect ectss of per percep ceptio tion, n, cog cognit nition ion,, and language. I address this problem in the pages that follow follow.. At the same time, special emphasis will be given in the following chapters to two important development developmentss in cognitive science. science. The first is the appr ap proa oach ch to co cogn gnit ition ion kn know own n as dy dyna namic mical al sy syst stem emss th theo eory ry.. Dyn Dynam amic ical al ap ap-proa pr oach ches es em emph phas asiz izee th thee te temp mpor oral al di dime mens nsio ions ns of co cogn gniti ition on an and d th thee wa ways ys in which an individual’s behavior emerges from interactions of brain, body, body, and environment. Simple and complex behavioral patterns are higherorder ord er pro products ductsof of selfself-org organiza anization tion pro processe cesses. s. Virtual irtually ly all living livingorg organism anismss self-assemble, or are self-organizing systems, “as emergent consequences of nonlinear interaction among active components” (Kelso, 1995: 67). Selforga or gani nize zed d pa patt tter erns ns of be beha havi vior or em emer erge ge as st stab able le st stat ates es fr from om th thee in inte tera ract ctio ion n of many subsystems. Yet Yet the emerging higher-order behavior is also capa ble of “enslaving” lower-level components in such a way that behavioral patterns can often be described by relatively few dimensions. Much of the emphasis, then, in dynamical systems theory is on the structure of spaces of poss po ssib ible lebetween beha be havi vior oral al tr traj ajec ecto tori ries and an d th thee in inte tern rnal al an and d these exte ex tern rnal al fo forrce cess (i (i.e .e., couplings brain, body, body ,es and world) that shape trajectories as., they unfold. More specifically, specifically, dynamical systems theory is a set of mathematical tools that can be applied to characterize different states of the system as these evolve in time. In this way, a dynamical view aims to describe how the body’s continuous interactions with the world provide for coordinated patterns of adaptive behavior, rather than focusing on how the external world become repr represented esented in the inner mind.

 

Introduction

 

11

dyna nami mica call ap appr proa oach ch re reje ject ctss th thee id idea ea th that at co cogn gniti ition on is be best st un unde ders rsto tood od A dy in terms of representational content (either for neurons or for parts of the mind), or that cognitive systems can be decomposed into inner functional subsystems or modules. Linear decomposition of cognitive performance into functional subsystems (i.e., “boxology”) is inadequate to understand the dynamical systems that cut across brain-body-world divisions. Most researchers working within a dynamical framework adopt the conservative ti ve st stra rate tegy gy of se seei eing ng ho how w fa farr on onee ca can n go in ex expl plai aini ning ng va vari riou ouss be beha havi vior oral al data without invoking repre representational sentational explanations. Dynamical systems theeor th ory y ha hass ha had d it itss mo most st prof ofou ound nd ef effe fect ct in co cogn gnit itiv ivee sci cien encce in th thee stu tudy dy of  perception/action relations, relations, or couplings, and in the development of situated at ed,, em embo bodi died ed ag agen ents ts,, or ro robo bots ts,, ca capa pabl blee of mi mini nima mally lly co cogn gnit itiv ivee be beha havi vior or.. Although there is debate over whether dynamical approaches can “scale up” to explain higher-order aspects of cognition, including language use and consciousness, I am enthusiastic about this perspective because it direct re ctly ly ac ackn know owle ledg dges es th thee in inte tera ract ctio ion n of an ag agen ent’ t’ss ph phys ysic ical al bo body dy (i (inc nclu ludi ding ng its brain and nervous system), its experience of its body body,, and the structure

of the environment and social context to produce meaningful adaptive  behavior..  behavior As will become evident in what follows, a dynamical perspective is not given to all of the topics discussed in this book, mostly because such applications are still in their infancy, and because my purpose here is not to argue for a single metatheory that explains all cognition. Nonetheless, understanding cognition as an embodied activity demands recognition of  the situated dynamics that serve to generate meaningful behavior in a complex world. A second area of special emphasis in this book will be the important work in cognitive linguistics on the embodied nature of mind and language. Cognitive linguistics does not view language as arising from an autonomous part of the mind/brain, but seeks to discover the ways in which linguistic structures are related to and motivated by human conceptual knowledge, bodily experience, and the communicative functions of discourse (Croft & Cruse,  2004;  Lakoff & Johnson,  1999). One of the many notable findings from cognitive linguistic research is that the body serves as a significant resource for people’s understanding of many abstract concepts. Metaphor is especially important in mapping experiences of the body to help structure abstract ideas that are fundamental to how people speak and think. theasis research from cognitive linguistics is co cont ntro rove vers rsia ial, l, gi give ven n itsAlthough heav he avy y em emph phas is on la lang ngua uage ge, , an and d th thee in indi divi vidu dual al intu in tuit itio ions ns of li ling ngui uist sts, s, I ai aim m to gi give ve co cogn gnit itiv ivee li ling ngui uist stic ic ev evid iden ence ce it itss pr prop oper er posi po siti tion on in co cogn gnit itiv ivee sc scie ienc ncee as a le lead adin ing g em empi piri rica call an and d th theo eorret etic ical al fo forrce in establishing the importance of embodiment for human cognition. Over Ov eral all, l, th ther eree is cu curr rren entl tly y en enou ough gh em empi piri rica call ev evid iden ence ce to pr pres esen entt a fu fulle llerr, more embodied picture of human cognition than has typically been the

 

12

 

Embodiment and Cognitive Science

generally, experiences of the body are case in the cognitive sciences. Most generally, repr re pres esen ente ted d as id idea eass in th thee mi mind nd,, an and d th thee bo body dy pr prov ovid ides es va valua luabl blee re reso sour urce cess for off-loading cognition such that mind is distributed across brain, body, body, and world interactions. This new view of the embodied mind has the followi lo wing ng sp spec ecifi ificc ch char arac acte teri rist stic ics, s, wh whic ich h wi will ll be ex expl plor ored ed in mo morre de deta tail il in th thee following chapters: –   Conce Concepts pts of the self, and who we are as persons, persons, are tightly tightly linked to tactile-kinesthetic activity. –   Embodiment is more than physiological and/or brain activity, activity, and is constituted by recurring patterns of kinesthetic, proprioceptive action that provide much of people’s felt, subjective experience. –   Perception is not something that only only occurs through specific sensory sensory appa ap para ratu tuss (e (e.g .g., ., ey eyeb eball allss an and d th thee vi visu sual al sy syst stem em)) in co conj njun unct ctio ion n wi with th pa parrticular brain areas, but is a kinesthetic activity that includes all aspects of th thee bo body dy in ac acti tion on.. Pe Perrce cept ptio ion n is ti tigh ghtl tly y li link nked ed to su subj bjun unct ctiv ivee th thou ough ghtt processes whereby objects are perceived by imagining how they may  be physically manipulated. –   Many abstract concepts are partly partly embodied, because they arise from embodied experience and continue to remain rooted in systematic patterns of bodily action. –   Hu Huma man n mi mind ndss ev evol olve ved d wi with th ne neur ural al res esou ourc rces es th that at ar aree pr prim imar arily ily de devo vote ted d

to pe perrce cept ptua uall an and d mo moto tori ricc pr proc oces essi sing ng,, an and d wh whos osee co cogn gnit itive ive ac acti tivi vity ty co connsists largely of on-line interaction with the environment. –   Cognitive processes are not located exclusively inside a person’s skin as computations upon mental representations (e.g., propositions, productions, mental images, connectionist networks). Cognitive C ognitive processes are partly constituted by physical and bodily movements and manipulations of objects in real-world environments. Cognitive mechanisms have evolved to operate in conjunction with environmental structures. Thus, cognitive processes are composed of both internal processes and  bodily manipulation of external objects objects outside the skin. –   Language reflects important important aspects of human conceptualization and thus is not independent from mind (i.e., as a separate module). Systematic at ic pa patt tter erns ns of li ling ngui uist stic ic st stru ruct ctur uree an and d be beha havi vior or ar aree no nott ar arbi bitr trar ary y, or du duee to con conven ventio tions ns or pur purely ely lin lingui guisti sticc gen genera eraliza lization tions, s, but ar aree mot motiva ivated ted by rec ecur urri ring ng pa patt tter erns ns of em embo bodi died ed ex expe peri rien ence ce (i (i.e .e., ., im imag agee sc sche hema mas) s),, wh which ich are often metaphorically extended. –   Memory Memory,, mental imagery, imagery, and problem solving do not arise from internal, computational, and disembodied processes but are closely linked to sensorimotor simulations. –   Children’s developing perception and cognition begins with and is rooted in embodied action.

 

Introduction

 

13

Emo motio tion, n, co cons nsci ciou ousn snes ess, s, an and d la lang ngua uage ge ev evol olve ved, d, an and d co cont ntin inue ue to ex exis istt in – E many ways, as extensions of animate motion. –   Bod Bodies ies ar aree not cul cultur ture-f e-fre reee obj object ects, s, bec becaus ausee all asp aspect ectss of emb embodi odied ed exp expeerience are shaped by cultural processes. Theories of human conceptual systems should be inherently cultural in that the cognition that occurs when the body meets the world is inextricably culturally based. Some of these ideas are not entirely new, but descend from a variety of scholarly works, in many academic disciplines, that have explored the  body in mind. These developments, developments, described in the chapters that that follow, follow, have decisively influenced the ways cognitive scientists now think about, and an d em empi piri rica call lly y ex exam amin ine, e, co cogn gnit itiv ivee be beha havi vior or.. My ar argu gume ment ntss in fa favo vorr of th thee embodied mind, again, do not imply i mply that embodied experience is the sole underlying factor driving human cognition. Many of the above characteristics of the embodied mind do not tell the whole story of why concepts, language, development, emotions, and consciousness take the particular forms they do in human life. However, I attempt to make a strong case for the importance of embodiment in providing a better blueprint for how cognitive scientists study and describe minds. The time is right for this kind ki nd of rea eapp ppra rais isal al of th thee bo body dy’s ’s rol olee in hu huma man n co cogn gnit itio ion. n. Ou Ourr bo bodi dies es,, an and d our felt experiences of our bodies in action, finally take center stage in the empirical study of perception, cognition, and language and in cognitive science’s theoretical accounts of human behavior. behavior.

 

2 Bodies and Persons

Each of us feels some intimate connection between who we are and our  bodies. When someone punches my nose, I, Raymond W. Gibbs, Jr Jr.,., and not someone else, experience pain. When I wonder if I feel happy, I consider this in terms of my own embodied being, and not someone else’s. When I expe ex peri rien ence ce th thee pl plea easu surres of se sex, x, th thee dis disco comf mfor ortt of fe feel elin ing g co cold ld,, or th thee fa fati tigu guee from running five miles, I clearly know that my body is the sole source of  these sensations. Yet when I think about the existence of God, or try to solve a complex math problem, I have little awareness that my body has a place in my thoughts. Cognitive processes seemingly occur with little input from our  bodies. The very act of perception focuses on perceptual objects/events out in the world in such a way that the body recedes into the background and feels almost unnecessary (Leder,  1990). This “corporeal disappearance” allows us, for better or worse, to objectify the body. We see the body as a material object, whereas the self and the mind are ethereal entities

that somehow mysteriously invade or permeate the body body.. The traditional t raditional maxim of “mind over matter” captures the common belief that the immaterial mind rules supreme over the corporeal body. What is the relationship relationship between between pers persons ons and their bodies? bodies? One may argu ar guee th that at on ones esel elf, f, on one’ e’ss pe pers rson on,, is a pu purre th thin inki king ng be bein ing, g, si simi mila larr to th that at en en-visa vi sage ged d by De Desc scar arte tes. s. Af Afte terr al all, l, th ther eree ar aree ma many ny ti time mess wh when en my tr true ue es esse senc ncee as a pe pers rson on se seem emss ut utte terl rly y im imma mate teri rial al,, as wh when en my bo body dy be beco come mess fa fati tigu gued ed,, ill, il l, or di disfi sfigu gurred ed,, ye yett I st stil illl be beli liev evee th that at my “s “sel elf” f” is un unch chan ange ged. d. Th Thee pe pers rson on I am, my self-conception, feels unrelated to my body, with my body only  being the vehicle for my thoughts. But this kind of quick introspectionist analysis may be misleading, and due as much to one’s cultural “folk  beliefs” as it is to veridical phenomenological insight. Systematic examination of one’s experience of the self and its relation to having a particular kind of body suggests that personhood may be deeply connected to bodies. A body is not just j ust something that we own, it is i s something that we are. 14

 

Bodies and Persons

 

15

I claim that the regularities in people’s kinesthetic-tactile experience not only constitute the core of their self-conceptions as persons, but form the foundation for higher-order cognition. chapter begins to shape the outline for this claim, setting the stage forThis more detailed consideration of various perceptual, cognitive, and linguistic phenomena that follow in later chapters. I begin by considering traditional conceptions about personhood, the ways we notice our bodies, the difference between body schema and body image, the importance of  move mo veme ment nt in ou ourr ex expe peri rien ence cess as “p “per erso sons ns,” ,” di diso sord rder ered ed bo bodi dies es,, cu cult ltur uree an and d embodiment, and the three levels of embodiment that make up the totality of who and what we are as persons. My specific aim here is to begin closing the mind-body gap, so persistent in the Western view of mind, by establishing a tight connection between our sense of ourselves as unique pers pe rson onss an and d ou ourr bo bodi dies es.. Re Rece cent nt at atte temp mpts ts wi with thin in co cogn gnit itiv ivee sc scie ienc ncee to ta talk lk of  thee “e th “emb mbod odie ied d mi mind nd”” to too o of ofte ten n do so in th thee co cont ntex extt of sp spec ecifi ificc pr prop oper erti ties es of  the brain. But there is a great need to understand embodiment as a aspect of wh whol olee pe pers rson onss in inte tera ract ctin ing g wi with th on onee an anot othe herr an and d th thee wo worl rld d ar arou ound nd th them em.. The First-Person Perspective

The intimate connection between who we are as persons and our bodies has recently been explored by the philosopher Lynne Baker (2000). She argued that a “person” is constituted by a human body body,, but a “person” is nott id no iden enti tica call to hi hiss or her bo body dy.. A hu huma man n or orga gani nissm is a per erso son n by vi virt rtue ue of  having a capacity to adopt a “first-person perspective.” Under this view, pers pe rson onss ar aree no nott di dist stin ingu guis ishe hed d fr from om ot othe herr th thin ings gs by vir virtu tuee of ha havi ving ng ce cert rtai ain n mental states, conscious or otherwise. Many mammals have mental states of belief and desire, and many mammals also have conscious states. Instead, the distinguishing mark of human persons is their capacity have a complex mental property – a first-person perspective that allows to me, for instance, to conceive of my body and mental states as my own, to have various intentional states such as believing, desiring, hoping, fearing, and so on on,, an and d to be se self lf-c -con onsc scio ious us ab abou outt th thee pl plan anss an and d go goal alss I de deci cide de to pu purs rsue ue.. From a first-person perspective, I can think about myself as myself (e.g.,

“I wonder whether I’ll be happy in ten years”), which demonstrates that I have some concept of myself as myself (Baker,  2000: 92). Even if totall totally y paralyzed, an individual has a first-person relation to his or her own body if the thought “I wonder if I’ll ever be able to move my legs again” can be entertained (Baker, 2000: 94). Two hu huma man n pe pers rson onss ma may y be di dist stin ingu guis ishe hed d fr from om on onee an anot othe herr by th thee fa fact ct that th at th they ey ar aree co cons nstit titut uted ed by di diff ffer eren entt bo bodi dies es,, ea each ch of wh whic ich h su supp ppor orts ts di diff ffer er-ent fir ent first st-p -per erso son n in inte tent ntio iona nall st stat ates es.. Th Thus us,, an any y re repl plic icaa of me ha hass a fir first st-p -per erso son n relation to his body, not mine. Moreover, a single body cannot constitute two persons at the same time. A single person may sometimes feel as if 

 

16

 

Embodiment and Cognitive Science

he or she is different persons at different times (e.g., The Three Faces of  Eve), and a commissurotomy patient may be manipulated in an experimental situation into simultaneously trying to put on his pants with one hand and trying to take them off with the other. But these are examples of a disordered, single first-person and not of :two first-person perspectives within the perspective, same body (Baker,  2000  108,different but also see the discussion below of conjoined twins). Although a human body starts out as entirely organic, it can acquire nonorganic nonorganic parts. An artificial leg that I think of as my own, and that I can move merely by intending to move it, becomes a part of my (still human) body. How much of the human body may be replaced and still remain a human body? Technology will surely have an increasing role in supplementing and extending the body. The philosopher Andy Clark (2003) claims that the melding of  flesh fle sh an and d ma mach chin inee is a na natu tura rall pr prog ogre ress ssio ion n in ou ourr lo long ng-d -dev evel elop opin ing g ca capa paci city ty to incorporate tools into our living environments to reduce the demands placed on brains and minds (see Chapter  5). But as long as we continue to be sustained by organic processes, processes, to some significant extent, each of us should be considered to have a genuine human body. These observations illustrate the central importance of how we define who we are as persons in terms of our first-person bodily perspective. Moreover, nothing can be your body if there is no you. At the very least, linking first-person bodily perspective with some concept of the self goes agai ag ains nstt ar argu gume ment ntss th that at we ar aree me mere re bu bund ndle less of re resp spir irin ing g ce cell lls, s, or co comp mput uter er programs, or that our existence as human persons is a metaphysical illusion. Each of these ideas has been seriously debated within the history of  philosophy.. Some contemporary philosophers, for example, maintain that philosophy the phrases “human body” and “one’s body” introduce i ntroduce much philosophical confusion and should be avoided in discussions of personal identity (Olson, 2003). Even if one’s personhood may be more than the body, there is no self without a body. Bodies and World

One tr One trad aditi ition onal al be belie lieff in Wes este tern rn cu cult ltur ures es is th that at hu huma man n bo bodi dies es ar aree se sepa para rate te from fro m the external world. Many cogni cognitive tive scientists scientists embra embrace ce this idea by assuming that individuals learn to know the world by re-presenting it to thei th eirr min minds ds.. Hu Huma man n bo bodie dies, s, th thro roug ugh h th thee fiv fivee ma majo jorr se sens nses es,, ar aree co cond ndui uits ts fo forr this re-presentation of the world. Yet bodies are independent of the world

as defined by the boundaries of skin (i.e., metaphysical or person-world dualism). But man many y phi philos losoph ophers ers and cog cognit nitive ive sci scient entist istss now re rejec jectt per person son-wo -world rld dualism and advocate that persons be understood, and scientifically studied, in terms of organism-environment mutuality and reciprocity. For example, Merleau-Ponty (1962)  claimed that the body exists primordially,

 

Bodies and Persons

 

17

 before there is thought or a reflected world, and the world exists for us only in an and d th thro roug ugh h th thee bo body dy.. Ph Phen enom omen enol olog ogy y sh show owss ho how w th thee en envi viro ronm nmen entt an and d people’s perceptions of it are interrelated. As Merleau-Ponty wrote: “My  body is the fabric into which all objects are woven, and it is, at least in rela re latio tion n to th thee pe perc rcei eive ved d wo worl rld, d, th thee ge gene nera rall in inst stru rume ment nt of my ‘c ‘com ompr preh ehen en-sion’” (1962: 235). Contrary to Descartes, who saw self-knowledge as the foundation of one’s knowledge of the world and others persons, MerleauPonty suggested that the full explanation of our knowledge of self arises from fr om par partic ticipa ipator tory y int intera eracti ction on wit with h our emb embodi odied ed exi existe stence nce.. Whe When n we con con-sider the concept of time, for instance, Merleau-Ponty argued that it is not helpful to think of time as a river that flows through our lives, independent of and precedent to our relation to it. We do not “observe” time as it goes by. Instead, time comes into being as a function of our embodied interaction with the world. A more recent perspective on person-environment mutuality is the enactive view of personhood and cognition (Varela, Thompson, & Rosch, 1991). The enactive view has the explicit goal to “negotiate a middle path  between the Scylla of cognition as recovery of a pregiven outer world (realism), and the Charybdis of cognition as the projection of a pregiven world (idealism)” (p. 172 ). Cognition is understood as enaction, or a history of  structural couplings that “brings forth a world” either by taking part in an existing world, as happens during development and maturation, or by shaping a new one, as happens over the history of a species. Because enacti ac tion on co cons nsis ists ts pa part rtia iall lly y in co coup upli ling ng,, th thee ag agen entt an and d th thee wo worl rld d ar aree no nott rea eall lly y separate, because they are “mutually specifying.” A person’s world is determined by the agent’s behavior and the sensorimotor capabilities that allow the individual to cope with a local situation. What people perceive depends upon what they are able to do, and what they do, in time, alters what they perceive. “Perception and action, sensorium and motorium are linked together as successfully emergent and mutually selecting pattern”  163). When a person enacts or brings forth a world, the person and the (p. worl wo rld d ar aree co coup uple led. d. Th This is po poss ssib ibil ilit ity y do does es no nott im impl ply y th that at th thee bo body dy an and d mi mind nd aree on ar onee an and d th thee sa same me.. Bu Butt ou ourr bo bodi dies es ar aree cl clos osel ely y de defin fined ed,, an and d ex expe peri rien ence ced, d, in terms of the specific actions we engage in as we move about the world. Ourr bo Ou bodi dies es an and d th thee wo worl rld d ar aree di difffe ferren ent, t, al alth thou ough gh th they ey ca can n be se seem emin ingl gly y absorbed absor bed into one another on many occasions. occasions. Philos Philosopher opher Drew Leder describes this embodied coupling of self and the world in the following personal pers onal examp example le (Led (Leder er,, 1990: 165): “I am wa walk lkin ing g do down wn a fo forres estt pa path th.. Yet et,, I am not attendin ing g to my world in a bodily or mindful way. I am caught up in my own worries – a paper that needs completion, a financial problem. My thoughts are running their private race, unrelated to the landscape. I am dimly aware of the sights and sound of nature, but it is a surface

awareness. The landscape neither penetrates into me, nor I into it. We are two bodies.

 

18

 

Embodiment and Cognitive Science

“Y “Yet, et, once again, it is possible to imagine an existential shift. Over time, through the rhythm of my walking, the calmness of the scene, my mind  begins to quiet. Something catches my ear – the trilling of a bird. I glance up in time to see the bird hopping from branch to branch, its bright colors shining in the sunlight. I gradually become aware of other birds, other songs, and, as if awakening from a dream, realize that I stand in the midst ofawildchorus.Iambeginningtoabsorbtheworldaroundmeandbecome absorbed in it.” Does Do es th this is ex expe peri rien ence ce so soun und d fa famil milia iar? r? As Le Lede derr co comm mmen ents ts,, “T “The he bo boun unddarie ar iess be betw twee een n th thee in inne nerr an and d out uter er th thus us be beco come me po porrou ous. s. As I cl clos osee my ey eyes es,, I feel the sun and hear the bird songs both within-me and without-me. They Th ey ar aree no nott se sens nsee da data ta in inte tern rnal al to co cons nsci ciou ousn snes ess, s, bu butt ne neit ithe herr ar aree th they ey ‘o ‘out ut there’ somewhere. They are part of a rich body-world chasm that eludes dualistic characterization” (Leder, (Leder, 1990: 165–6). The world becomes alive for us from being incorporated into our bodies, while, at the same time, we experience ourselves being absorbed into the body of the world. This fusion of body and world makes it difficult, at times, to strictly distinguish between the two. Gregory Bateson’s (1972) famous example of the  blind person who experiences the tip of his walking stick as part of his  bodily being illustrates this problem. The clothes we wear wear,, our eyeglass, hearing aids, artificial hearts, and other prosthetic devices are not natural parts of our bodies at birth, but eventually becomes so to some individuals. Postmodern philosophers, and science fiction enthusiasts, explore the consequences of blurring the self/other dichotomy with advances in technological devices that extend the body and integrate nonorganic material with human flesh. These developments work to dissolve any clear  boundary between bodies bodies and world. Bodies and Selves

The cultural anthropologist Clifford Geertz once suggested that the WestWestern view of the self is: “A bounded, unique, more or less integrated motivational and cognitive universe, a dynamic center of awareness, emotion,  judgment, and action organized into a distinctive whole and set contrastively both against other such wholes and against a social and natural  background” (Geertz, 1979: 229). Many Westerners agree that this definition captures a great deal of the essence of how people conceive of themselves. But this definition does not explicitly acknowledge the body’s role in the creation of a self-concept. There has historically been a strong tendency to conceive of the self  as indivisible, and separate from any bodily incarnation. Following in Descartes’ footsteps, the  18th-century Scottish philosopher Thomas Reid famously wrote, “A part of a person is a manifest absurdity absurdity.. When a man loses his estate, his health, his strength, he is still the same person, and

 

Bodies and Persons

 

19

has lost nothing of his personality. If he has a leg or an arm cut off, he is the same person he was before. The amputated member is no part of  his person, otherwise it would have a right to his estate, and be liable for parrt of hi pa hiss eng ngaage geme ment nts. s. It wo woul uld d be ent ntit itle led d to a sha harre of his me merrit an and d de de-meri me rit, t, wh whic ich h is ma mani nife fest stly ly ab absu surd rd.. A pe pers rson on is so some meth thin ing g in indi divi vidu dual al . . . My thoughts, and actions, and feelings change every moment; they have no cont co ntin inue ued, d, bu butt a su succ cces essi sive ve ex exis iste tenc nce; e; bu butt th thee se self lf or I, to wh whic ich h th they ey be belo long ng is pe perm rman anen ent, t, an and d ha hass th thee sa same me rel elat atio ion n to al alll th thee su succ ccee eedi ding ng th thou ough ghts ts an and d actions which I call mine” (from Flanagan, 2002: 173). Reid Re id’s ’s at atte temp mptt to lo loca cate te th thee se self lf in th thee im imma mate terria ial, l, an ech cho o to th thee “I th thin ink k ther th eref efor oree I am am”” ma maxi xim, m, do does es no nott ne nece cess ssar aril ily y ho hold ld up to ph phen enom omen enol olog ogic ical al examination. Many people perceive their selves, the coherence of what we think of as the “self,” as being founded on the perceived unity and  boundedness of the body. body. As William William James observed over 100 years ago, “The nucleus of the ‘me’ is always the bodily existence felt to be present at the time” (James, 1890; 194). Our identity through time consists in the identity of our bodies (Ayer, 1936). Part of our felt sense of ourselves as persons comes from how sensory information is correlated in experience. I know who I am, and that I am, in part, because I see my body (e.g., hands, legs, arms, stomach, feet) as I move mo ve an and d ex expe peri rien ence ce sp spec ecifi ificc se sens nsat atio ions ns as a re resu sult lt of ac actio tion. n. Re Rese sear arch ch wi with th adults adu lts sug sugges gests ts tha thatt sel self-k f-know nowled ledge ge par partly tly eme emerg rges es fr from om vis visual ual,, tac tactile tile,, and propr pr oprioc iocept eptive ive inf inform ormati ation on fr from om our bod bodies ies.. For exa exampl mple, e, peo people ple ar aree mor moree accu ac cura rate te at pr pred edic icti ting ng a da dart rt’s ’s la land ndin ing g po posi siti tion on wh when en wa watc tchi hing ng vi vide deo o cl clip ipss of themselves throwing the dart than when watching a clip of someone else el se to toss ssin ing g th thee da dart rt (K (Kno nobl blic ich h & Fl Flac ach, h, 2001). Pe Peop ople le ev even en be bett tter er rec ecog ogni nize ze light-displays of their own movements than they do of those of other people, despite rarely seeing their entire bodies in i n movement (Beardworth & Buckner, 1981). Furthermore, people find it more difficult to identify their own hands, and their actions, in situations in which their hands are perceived as incongruent with their bodily orientation (van den Bos &  Jeannerod, 2002). This research suggests that people are normally quite awaree of th awar thei eirr bo bodi dies es an and d th thei eirr pa past st an and d po pote tent ntia iall fu futu ture re ac acti tion ons. s. We po poss sses esss fair fa irly ly de deta taile iled d se self lf-s -sch chem emas as th that at ar aree roo oote ted d in ou ourr ex expe peri rien ence cess of em embo bodi died ed possibilities. One unusual experiment suggests how people’s identification of their  bodies depends crucially on intermodal correlation. Botvinik and Cohen (1998) had participants “seated with the left arm resting on a small table. A study screen was positioned beside the arm to hide it from the subject’s view and a life-sized rubber model of a left hand and arm was placed on thee ta th tabl blee di dirrec ectl tly y in fr fron ontt of th thee su subj bjec ect. t. Th Thee pa part rtic icip ipan ants ts sa satt wi with th ey eyes es fix fixed ed on the artificial hand while we used two small paintbrushes to stroke the rubber hand and the subject’s hidden hand, synchronizing the timing of  the brushing as closely as possible” (Botvinik & Cohen, 1998: 766). After a

 

20

 

Embodiment and Cognitive Science

short interval, participants had the distinct and unmistakable feeling that they sensed the stroking and tapping in the visible rubber hand, and not on the hand which, in fact, was being touched. Further tests revealed that if the experimenters asked participants, with eyes closed, to point to the left le ft ha hand nd wi with th th thee hi hidd dden en ha hand nd,, th thei eirr po poin inti ting ng,, af afte terr ex expe peri rien ence ce of th thee il illu lu-sion, were displaced toward the rubber hand. Botvinik and Cohen argued that these results support the idea that our sense of our bodies as our own depe de pend ndss le less ss on th thei eirr di difffe ferren enti tiat atio ion n fr from om ot othe herr ob obje ject ctss an and d bo bodi dies es th than an on their participation in specific forms of intermodal correlation. Thus, our tactile-kinesthetic sensations, and how they are correlated across modalities, provide a strong foundation for our sense of self. The link between self and body need not imply that there must be a single self in the same way that each of us may possess, or is, a single  body.. Many scholars today acknowledge that there is not a single “self.”  body Self is fragmented and, at best, provides the center of our “narrative gravity” (Dennett,  1992;  Gergen,  1991). In fact, the complexity of our bodily expe ex peri rien ence cess pr prom omot otes es an eq equa uall lly y co comp mple lex x se sett of se self lf-i -ide dent ntiti ities es.. Pe Peop ople le ta talk lk abou ab outt th thei eirr in inne nerr se selv lves es in di diff ffer eren entt wa ways ys at di diff ffer eren entt ti time mes, s, us usin ing g a ra rang ngee of  metaph met aphori orical cal con concep cepts ts tha thatt ari arise se fr from om the their ir var varied ied bod bodily ily exp experi erienc ences es in the phys ph ysic ical al an and d so soci cial al wo worl rld d (L (Lak akof offf & Jo John hnso son, n, 1999). Met Metaph aphori orical cal con concep cepts ts expr ex pres esss fu fund ndam amen enta tall me ment ntal al ma mapp ppin ings gs by wh which ich kn know owle ledg dgee fr from om on onee do do-main (i.e., the target) is structured and understood by information from a dissimilar domain (i.e., the source). In many cases, these concepts reflect differentt kinds of embodied correlations, including (1) the correlation bedifferen tween body control and control of physical objects (e.g., SELF CONTROL IS OBJECT CONTROL – “After being knocked down, the boxer picked himself up off the canvas”), (2) the correlation between being in a certain norm no rmal al lo loca cati tion on an and d ex expe peri rien enci cing ng a se sens nsee of co cont ntro roll (e (e.g .g., ., SE SELF LF CO CONT NTRO ROL L IS BE BEIN ING G IN ON ONE’ E’S S NO NORM RMAL AL LO LOCA CATI TION ON – “I “I’m ’m ju just st be besi side de my myse self lf wi with th anger,” “Peter is out of his mind”), (3) the correlation between self action and the movement of objects (e.g., CAUSING THE SELF TO ACT IS THE FORCED MOVEMENT OF AN OBJECT – “You’re pushing yourself too hard,” “I can’t seem to get myself going”), ( 4) the correlation between our sense of self control and our control of unified containers (e.g., SELF CONTROL IS HAVING THE SELF TOGETHER AS A CONTAINER – “She is falling to pieces,” “Pull yourself together”), and (5) the correlation  between our sense of self and the search for things at particular locations (e.g., SELF AS AN ESSENCE THAT THAT IS A FOUND OBJECT – “He is trying to find himself,” “She went to India to look for her true self,” “He found himself in writing”). These metaphorical ways of understanding the self are not consistent,  because there is no single, monolithic self-concept. Yet these metaphors appear to be found in a variety of cultures, and capture important qualities of how we conceive of our inner lives, partly based on varying bodily

 

Bodies and Persons

 

21

experiences (Lakoff & Johnson,  1999). Cognitive scientists have, in fact, prop pr opos osed ed th that at th ther eree ar aree di diff ffer eren entt le leve vels ls of th thee se self lf,, ea each ch on onee ro root oted ed in di diff ffer er--

ent kinds of embodiments. For example, cognitive neuroscientist Antonio Damasio (1999) distinguishes between three kinds of self. The “proto self” is un unco cons nsci ciou ouss an and d co cons nsti titu tute ted d by “i “int nter erco conn nnec ecte ted d an and d te temp mpor orar arily ily co cohe herrent collection of neural patterns which represent the state of the organism” (Damasio, 1999: 154). Lower animals, even lobsters, possess a proto self. The “core self” is the nonverbal, or preverbal, subject of conscious experience. Dogs, cats, and human infants have core selves. The “autobiographical self,” or “extended consciousness,” is the record of one’s life experiences and is usually thought to require a conceptual structure and maybe language. All three kinds of self are closely rooted in the brain and body, with each level being built upon its predecessor (Damasio, 1999). The proto self  req equi uirres a bo body dy th that at se send ndss si sign gnal alss (v (via ia ch chem emic ical alss in th thee bl bloo oods dstr trea eam) m) to th thee  basal forebrain, hypothalamus, and brain stem, which causes the release of  certai cer tain n neu neuro rotra transm nsmitte itters rs in the cen centra trall cor cortex tex,, tha thalam lamus, us, and bas basal al gan ganglia glia.. These pathways provide the proper material to unconsciously represent different bodily states, including the body’s relation to the environment. The core self requires a functioning proto self as well as the cingulate cortex, the thalamus, parts of the prefro prefrontal ntal cortex, and the superior colliculi. On Once ce th thes ese difffe di ferren entt br brai ain n ar area eass co come me in into to pl play ay,, pe peop ople le ca can n ha have ve a ge genu nuin inee subject ofeexperience. The organism feels things, and feels itself feeling things. Autobiographical memories are stored in various sensory cortices and activated by convergence zones in the temporal and frontal higher cortices, as well as subcortical areas such as the amygdala, which together are important in experiencing and remembering how certain experiences feel and felt. Most generally, the emergence of self requires an organism with a particular kind of brain and to live in a world with other similarly embodied creatur creatures. es. Our sense of ourselves as persons that endure through physical (aging) and an d me ment ntal al (b (bel elie iefs fs,, at atti titu tude des) s) ch chan ange gess is pr prim imar aril ily y ba base sed d on ou ourr bo bodi dily ly in in-teractions with the physical/cultural physical/cultural world. As James Gibson (1966, 1979) long argued, our perception of the sensory world is given to us directly  by “affor “affordances.” dances.” An affordan affordance ce is a resour resource ce that the environment offers an animal, such as surfaces that provide support, objects that can be manipulated, and substances that can be eaten, each of which is a property ert y spe specifi cified ed as sti stimul mulus us inf inform ormati ation on in ani animal mal-en -envir vironm onment ent int intera eracti ctions ons.. Each person/animal has a vast set of possibilities for action, based on the perception of affordances (e.g., chairs that can be sat on, streetcars that can be caught if running) that implicitly define who we are (White, 1999). Gibso Gib son n de defin fined ed th thee ec ecol olog ogic ical al se self lf as fo follo llows ws:: “A “Awa warren enes esss of th thee pe pers rsis isti ting ng and changing environment (perception) is concurrent with the persisting and changing self (proprioception in any extended use of the term). This

 

22

 

Embodiment and Cognitive Science

includes the body and its parts and all its activities from locomotion to thou th ough ght, t, wi with thou outt an any y di dist stin inct ctio ion n be betw twee een n th thee ac activ tiviti ities es ca calle lled d ‘m ‘men enta tal’ l’ an and d those called ‘physical.’ Oneself and one’s body exist along with the environm ro nmen ent, t, th they ey ar aree co co-p -per erce ceiv ived ed”” (1987: 418). Ou Ourr se self lf-c -con once cept pt is im impl plic icit it in ourr pe ou perc rcep eptu tual al,, em embo bodi died ed in inte tera ract ctio ions ns wi with th th thee wo worl rld, d, an and d ou ourr ki kine nest sthe heti ticc experiences of our own bodies (i.e., proprioception; see Chapter 3).

Our self-concepts depend, then, on the patterns of bodily actions we engage in on a daily basis. The sense of agency, as the causal basis for action, is perhaps the most convincing evidence for the “I” we experience as pe pers rson ons. s. Fo Forr in inst stan ance ce,, I ma make ke a co cons nscio cious us de deci cisi sion on to ra rais isee my ri righ ghtt ha hand nd,, and an d my bo body dy so some meho how w re resp spon onds ds ac acco cord rdin ingly gly.. Mu Much ch of th thee pe pers rsis iste tent nt be belie lief  f  that th at we ar aree th thee “a “aut utho hors rs”” of ou ourr ac acti tion onss is roo oote ted d in th thee sy syst stem emat atic ic pa patt tter erns ns of actions that appear to follow foll ow from our willful intentions. But this belief about the link between agency and action may be based on a misinterpretation of brain and unconscious cognitive processes. Several er al li line ness of res esea earrch su sugg gges estt th that at th thee fe feel elin ing g th that at ou ourr co cons nsci ciou ouss wi will ll se serv rves es as the causal basis for our actions may be illusory. For example, in an infamous study, Libet (1985) asked students to move their hands whenever they wished, while noting on a fast-moving analog clock when they made their the ir dec decisio isions. ns. The par partic ticipa ipants nts’’ EEG EEGss wer weree als also o con concur curre rentl ntly y mea measur sured. ed. If  will wi llfu full bo bodi dily ly ac acti tion on is ca caus used ed by a co cons nsci ciou ouss de deci cisi sion on,, or wi will ll,, th then en pa part rtic ic-ipants should indicate that they made their decisions to act prior to when  brain processes executed the hand movement. In –fact, the opposite after was observed: the decision to move occurred about 350 400 milliseconds relevant cerebral cortex activity began. One interpretation of this finding suggests that conscious decisions (i.e., sense of agency) do not cause human behaviors, at least for simple motor actions. Some scholars claim that Libet’s findings, along with others, demonstrate that the concept of “free will” is an illusion (Wegner, 2002). Many students, as well as cognitive scientists, are disturbed by the implic pl icat atio ions ns of Li Libe bet’ t’ss wo work rk.. Mo Most st pe peop ople le st stil illl ma main inta tain in a be beli lief ef in a co conc ncep eptt of  self-bodydualism,atleasttotheextentthateachpersonpossessesaselfthat is th thee “p “pil ilot ot of on one’ e’ss sh ship ip.” .” Yet pe peop ople le ca can n be fo fool oled ed in into to be beli liev evin ing g th that at th thei eirr conscious wills are the causes of their actions, even when the true cause resides outside their brains. One study aimed to show this by applying tran tr ansc scra rani nial al ma magn gnet etic ic st stim imula ulatio tion n (T (TMS MS)) to on onee si side de of pa part rtic icip ipan ants ts’’ br brai ains ns  by the motor area (W (Wegner egner & Wheatley Wheatley,, 1999). TM TMS S in indu duce cess mo moto torr ne neur uron onss to fir firee so th that at pe peop ople le au auto toma matic tical ally ly mo move ve th thei eirr lim limbs bs.. Th Thes esee mo move veme ment ntss ar aree involuntary, similar to the “knee jerk” that occurs when one’s knee is tapped with a hammer. Participants in this study were subjected to TMS on one side of the brain and asked to make spontaneous limb movements. Interestingly,, the participants believed that their conscious decisions were Interestingly the cause of their “knee-jerk” reactions just as much when the true cause was the TMS as when they moved their limbs l imbs voluntarily. voluntarily. Results such as

 

Bodies and Persons

 

23

thes call ll in into to qu ques esti tion on th thee si simp mple le id idea ea th that at th thee co cons nsci ciou ouss se self lf is al alwa ways ys th thee thesee ca author of one’s bodily action. Chapter 8 ex expl plor ores es so some me of th thee co comp mple lex x re rela lati tion onss be betw twee een n ne neur ural al ac activ tiv-ity, conscious will, and bodily action. For now, it appears that our feeling of ownership for our actions may be explained in terms of a matching of  motor activity with sensory feedback, instead of any thought-then-action caus ca usal al rel elat atio ions nshi hip. p. Bo Both th th thee th thou ough ghtt an and d th thee ac acti tion on ar aree du duee to a si sing ngle le,, un un-cons co nsci ciou ouss br brain ain pr proc oces esss wi with th th thee th thou ough ghtt of wh what at is ab abou outt to ha happ ppen en ar aris isin ing g into consciousness a bit before the overt action is completed. Only when we reflect upon what has happened do we invoke a concept of agency to

explain the reasons for our actions., even if we are often incorrect in our explanations as to why we behaved in a certain way (Wilson, 2002). Individuals may claim ownership for their actions in the sense that actions arise from a complex interplay of brain processes, fast-acting cognitive mechanisms, and feelings of conscious awareness, all of which are experienced within the body. body. A central, even defining, aspect of our sense of self is our ability to predict our future actions. This predictive power is mistakenly attributed to some thought-to-action causal link, but is really an emergent property of brain, body body,, and world couplings. Our ability to predict our future actions explains, among other things, why it is difficult to tickle oneself. When trying to tickle ourselves, we can easily predict the dire di rect ctio ion n of ou ourr mo move veme ment nts, s, wh whic ich h gr grea eatl tly y red educ uces es th thee se sens nsit itivi ivity ty to ta tact ctile ile stimuli compared to when they these are randomly applied by someone else. Many schizophrenics, thebetween cause oftickling their behaviors to outside sources, do not feel who muchattribute difference themselves and an d be bein ing g ti tick ckle led d by so some meon onee el else se (B (Bla lake kemo morre, Wol olpe pert rt,, D. D.,, & Fi Firt rth, h, 2000). Characterizing the self as an emergent property of brain, body, and worl wo rld d in inte tera ract ctio ions ns mo most stly ly as assu sume mess th that at ea each ch pe pers rson on ha hass on onee br brai ain n an and d on onee  body.. But conjoined, or Siamese, twins offer an interesting challenge to this  body idea. Depending on how they are conjoined, these twins may share body space, various organs, and limbs, and experience areas of joint sensation and an d mo move veme ment nt.. Ver ery y fe few w st stud udie iess ha have ve ex exam amin ined ed tw twin ins’ s’ ex expe peri rien ence ce of bo body dy  boundaries as this relates to the self/ self/other other distinction. For example, the first well we ll-k -kno nown wn pa pair ir,, Ch Chan ang g an and d En Eng, g, bo born rn in 1871, we werre jo join ined ed to tog get ethe herr at th thee  base of their chests by a band band of cartilaginous cartilaginous tendon 5–6 inches long. The  brothers had both regions of common sensitivity and areas of individual sensation (Murray,  2001). When touched on the middle of their band of  union, both Chang and Eng felt the stimulus. But if a stimulus was moved even 1.2 in inch ches es to towa warrd on onee si side de,, on only ly on onee br brot othe herr co coul uld d st stil illl fe feel el it it.. An Anot othe herr famous set of twins were the Tocci brothers, born in  1875. Each twin had a usable pair of arms. However, they were completely joined below the sixth rib, sharing a common abdomen, anus, and penis and one pair of  legs le gs.. Ea Eacch brot oth her ha had d co con ntr trol ol ov over er th thee le leg g on hi hiss si side de of th thee bod ody y, bu butt th they ey were unable to walk.

 

24

 

Embodiment and Cognitive Science

One set of four-month-old conjoined twins were connected on the ventral surface between the umbilicus and sternum so that they always faced one ano anothe therr (St (Stern ern,, 1985). On Onee tw twin in wo woul uld d of ofte ten n su suck ck on th thee ot othe her’ r’ss fin finge gers rs,, and vice versa, but they had no confusion as to which fingers belonged to whom wh om.. Th This is su sugg gges ests ts th that at ea each ch tw twin in “k “kne new” w” th that at on one’ e’ss ow own n mo mout uth h su suck ckin ing g a finger and one’s own fingers being sucked did not make a coherent self. Even with body fusion, the individual twins seemed to make distinctions  between parts of their own own bodies and those of the other twin. However, many conjoined twins who were surgically separated as young children or adolescents report confusion as to whether they are the same sa me pe pers rson on af afte terw rwar ard d (e (e.g .g., ., “I “Iss it rea eall lly y me me?” ?” “A “Am m I rea eall lly y my myse self lf?” ?”)) (S (Sep ep-arating conjoined twins is obviously a tremendous medical challenge. Yet surgeons actually construct, rather than merely separate, bodies, because there are no natural ways of making two bodies out of one that preserve

each twin s self identity and self body relationship [Murray, [Murray, 2001]). It seems impossible for an individual to phenomenally feel from the inside part of another person’s body. body. But cases of autositic-parasitic twins provide an exception to this idea (Murray,  2001). With autositic-parasitic twin tw ins, s, on onee tw twin in di dies es at an ea earl rly y st stag agee of em embr bryo yoni nicc de deve velo lopm pmen ent, t, bu butt va vari ri-ousportionsofitsbody(parasitic)becomeattachedto,andaresustainedby, the surviving twin (autositic). Most twins with upper-body parasites give their “companion” a name and treat the parasite as a person (one parasite wasevenbaptized).InthecaseofLaloo,bornin1874, th thee pa para rasi siti ticc tw twin in wa wass atta at tach ched ed to th thee lo lowe werr pa parrt of his wa wais istt and had tw two o ar arms ms an and d tw two o le leg gs an and da penis, peni s, wh whic ich h ha had d er erec ecti tion onss an and d di disc scha harg rged ed ur urin inee in inde depe pend nden ently tly of La Lalo loo’ o’ss cont co ntro rol. l. Th Thee pa para rasi site te’s ’s “u “unr nrul uly y be beha havio vior” r” su sugg gges ests ts th that at an ag agen ency cy se sepa para rate te from that of the autositic twin was attributed to it. Laloo could, however, feel sensations whenever any part of the parasitic twin was touched. These case studies raise more questions than they answer about how Siam Si ames esee tw twin inss co conc ncei eive ve of se selflf-bo body dy rel elat atio ions nshi hips ps.. On Onee re reas ason onab able le co conc nclulusion, though, is that the t he ways individuals know their body boundaries are actually contingencies that, although reliable for most of us, can lead to ambiguity between body and self for conjoined twins (Murray (Murray,, 2001). Most generally, understanding how brain, body, and world function to produce a sense of self requires that we view this interaction as part of  a self-organized dynamic system. Brains operate at different levels, from the microscopic level of single neurons, to populations of neurons or cell assemblies, to levels at which mental functions are experienced from a first fir st-p -per erso son n pe pers rspe pect ctiv ive. e. Hi High gher er an and d lo lowe werr le leve vels ls in inte tera ract ct,, an and d ha have ve ca caus usal al influence, in both directions. Yet despite our feelings that this is the case, there is no single control center (i.e., the self) that oversees the operation of the different levels or their interaction. Self-organizing dynamic systems have a kind of permanency whose existence can only be understood

 

Bodies and Persons

 

25

from the inside (Flanagan,   2002). Thus, brain, body, and environment interaction gives rise to the sense of self that, again, has a feeling of some permanency. What Do We Notice about Our Bodies?

What do adults ordinarily notice about their bodies? When I run uphill, I cert ce rtai ainl nly y fe feel el th thee mu musc scle le te tens nsio ion n in my le legs gs,, an and d th thee ex expa pans nsio ion n of my lu lung ngss as I struggle to breathe. But I have little awareness of the hair on my head, thee mo th move veme ment nt of my ha han nds ds,, th thee se sens nsaati tion onss in th thee pit of my st sto oma mach ch.. Th Thus us,, thee bo th body dy do does es no nott ap appe pear ar to co cons nsci ciou ousn snes esss as a no norm rmal al ob obje ject ct of aw awar aren enes esss as we actively engage with our surroundings. Phenom Phe nomeno enolog logica icall phi philos losoph ophers ers hav havee lon long g str strugg uggled led to cha charac racter terize ize the felt sensations in mental life and bodily action (Husserl,  1977; Heiddeger, 1962; Merleau-Ponty, 1962; Sartre, 1956; Sheets-Johnstone, 1999). Contempora po rary ry Wes este tern rn cu cultu lturres al also so wr wres estle tle wi with th ho how w to th thin ink k ab abou outt an and d de desc scri ribe be  bodily experience. We certainly have become obsessed with the care and appe ap pear aran ance ce of th thee bo body dy.. Mu Much ch of th this is at atte tent ntio ion n is wi with th th thee bo body dy-a -ass-ob obje ject ct,, or some so meth thin ing g th that at ca can n lo look oked ed at an and d de desi sirred ed,, an and d no nott wi with th th thee bo body dy-a -ass-su subj bjec ect, t,

or th thee fir first st-p -per erso son n ex expe peri rien ence ce of on one’ e’ss bo body dy in ac acti tion on.. Of co cour urse se,, mo most st of us acknowledge that looking good on the outside makes us feel good on the insi in side de!! Th Ther eree is a si sign gnifi ifica cant nt tr tren end, d, wh whic ich h re rema main inss qu quit itee ac acti tive ve in Ca Calif lifor orni niaa as a remnant of the “human potential movement,” to develop practices that enhance one’s felt understanding of one’s body (e.g., massage, polarity therapy, progressive relaxation, rolfing, meditation, Vipassana meditation, yoga, biofeedback, the Feldenkrais method, autogenic training, the Alexander technique) (Marrone, 1990). Many of these techniques emphasize how each individual’s sense of self is composed of a constellation of  physical, emotional, and intellectual habits that can become limiting and restr re stricti ictive. ve. Vari arious ous mov moveme ement nt pra practi ctices ces ena enable ble peo people ple to “ge “gett in tou touch ch wit with h thei th eirr bo bodi dies es”” in wa ways ys th that at mo most st of us si simp mply ly do no nott se sens nsee wi with th ap appr prop opri riat atee attention atten tion in ever everyday yday life. This version of “emb “embodied odied psychology” psychology” aims to develop a reciprocity of body and mind as part of psychotherapeutic healing processes. Few systematic studies in cognitive science, however, have explored what people ordinarily notice about their bodies. An exception to this is one study that directly examined adults’ intuitions of their daily em bodied experiences (Pollio, Henley, Henley, & Thompson, 1997). Men and women responded at length to two questions: ( 1) “Could you tell me some times when wh en yo you u ar aree awa warre of you ourr bod ody? y?”” an and d (2) “C “Cou ould ld yo you u te tell ll me of wh what at you are aware of in that situation?” The first question was aimed at revealing the “whens” of bodily experience, and the second question the “whats” of   bodily experience.

 

26

 

Embodiment and Cognitive Science

Participants’ responses indicated that bodily experience tended to be focused on eight general situations: i. i. Us  Usin ing g th thee bo body dy (i (i.e .e., ., an aw awaaren enes esss of th thee bo bod dy wh when en en enga gage ged d in an activity or project); ii. ii.   Sens Sensing ing the body (i.e., an awareness awareness of the body such as that felt when whe n exp experi erienc encing ing ach aches es and pai pains, ns, illn illness ess and fat fatigu igue, e, and var variou iouss pleasures); iii.   Presenting the body (i.e., an awareness of the body as presented to iii. other people, such as posture and mode of dress); iv iv..  Pregnancy and sexuality (i.e., the feelings associated with being pregnant, including nursing, and sexual intimacy and arousal); v.   Chang Changes es over time (i.e., an awar awarenes enesss of comparisons comparisons between a present experience of the body and past ones); vi. Identity vi.  Identity (i.e., an awareness of the meaning of some event, such as when a person is aware of his/her body “as a Christian”); vii.   Awareness of others (i.e., an awareness of the presence or absence vii. of others); viii.   Awa viii. ware rene ness ss of af affe fect ct (i (i.e .e., ., an aw awar aren enes esss of st strron ong g em emot otio ions ns as a ma majo jorr aspect of some situation). The Pollio et al. interviews also revealed three major themes that reflect fle ct ea each ch pe pers rson on’s ’s un uniqu iquee mo mode de of ex expe peri rien enci cing ng hi his/ s/he herr bo body dy.. Ea Each ch th them emee

included two specific subthemes: 1.   Experiences of engagement

a.  Body in vitality  b.  Body in activity; 2.   Experiences of corporeality a.  Body as instrument  b.  Body as object; 3.   Experiences of interpersonal interpersonal meaning a.  Body as appearance  b.   Body as expression expression of self. Experiences of engagement occur when a person experiences his or her  body as fully engaged in some project out in the world. Vitality refers to expe ex peri rien ence cess in whi hich ch th thee pe perrso son n is fu full lly y en enga gage ged d in th thee wo worrld wi with th li litt ttle le or no sense of the body-as-physical. Instead, there is a feeling of well-being and absorption in the world (e.g., “ . . . the feeling of well-being, you go outside and take a deep breath and feel good all over”). Activity refers directly to the concrete movements a person is engaged in, in which there is a general awareness of the body as central to the experience (e.g., “In runn ru nnin ing g I en enjo joy y th thee fe feel elin ing g of th thee mu musc scle less bu burn rnin ing, g, th thee ti tigh ghtn tnes esss of th thee sk skin in with the wind on it”).

 

Bodies and Persons

 

27

Experi erienc ences es of cor corpor poreal eality ity ari arise se as an awa aware renes nesss of the bod body-a y-assExp physical becomes present as an object in a world of objects or as a means for achieving goals. These experiences include both acting upon things and an d be bein ing g ac acte ted d up upon on by th them em.. In Inst stru rume ment nt ref efer erss to ex expe peri rien enci cing ng th thee bo body dy as a tool for accomplishing things and as something that can be brought up in learning skillful performance (e.g., “In the early stages of learning how to dance, you have to think about the process of all these pieces”). Object refers to bodily limits, similar to the ways that objects have limits. Thus, the body can be impaired through illness, or may be seen in terms of possession and ownership. Furthermore, Furthermore, the body may call attention to itself and reorient a person back to the world (e.g., “I’m most aware of my  body after I eat, aware aware and self-conscious of my belly”). belly”). Experi Exp erienc ences es of int interp erpers ersona onall mea meanin ning g ar aree tho those se whe where re the thebod body y is und under er-stood in terms of its social and symbolic meaning. People feel their bodies as they th part pa ake in th the th e sh shar ared edance mean me ings gs the th ever ev eryd yday ay worl d.ghli Expr Ex pres essi sion of th the eeyse self lf rtak ref efer ere to thos ose e in inst stan cessanin wher wh ereeofth the e ebo body dy-a -ass-se self lfwo isrld. high hi ligh ghte ted, d,ons in-sin cluding aspects of lifestyle, character character,, and interpersonal stance (e.g., “I’ve always been hard on myself, seeing room for improvement. I guess I’ve fall fa llen en in into to th thee tr trap ap of th thee mo mode dern rn Am Amer eric ican an ma male le al alwa ways ys tr tryi ying ng to lo look ok li like ke the model in Esquire in Esquire”). ”). Appearance refers to concrete ways that a person’s  body looks in the eyes of the self and to other people (e.g., “I try to dress in a way that calls attention away from my body and toward my face. Being fat means you have to try harder to just look okay”). The six subthemes identified above often blend together in differe different nt aspects of bodily experience. For example, “(when running) I’m often aware of my pace, the pace of my feet, and heartbeat and my breath. When I’m feel fe elin ing g at my pe peak ak,, th that at’s ’s wh when en I’ I’m m mo most st aw awar aree of th thin ings gs wo work rkin ing g to toge geth ther er.” .”

Another blend is when object, experience of self, vitality, and activity are combined: “When I’m doing what I want to be doing, there isn’t a separation (between my body and me) . . . I perceive separatedness when I’m critical of my body or in pain. If I’m doing something well, I’m aware of   being me . . . To be aware of my body . . . me mean anss th that at so some meth thin ing g is isn’ n’tt ri righ ght. t.”” This analysis of people’s attention to their own bodies illustrates, once again, the reciprocity that exists between bodies and environment. People mostly notice their bodies in relational situations involving the environment and interacting with other persons. Movement, Body Schemas, and Body Images

Moveme Move ment nt is ce cent ntra rall to ho how w we co conc ncei eive ve of th thee rel elat atio ion n be betw twee een n ou ours rsel elve vess and our bodies. We do not feel subjective experiences to be specific brain stat st ates es,, bu butt se sens nsat atio ions ns of ou ourr bo bodi dies es in ac acti tion on.. Ba Babi bies es be begi gin n li life fe by wi wigg ggli ling ng,, stretc str etchin hing, g, ope openin ning g and clo closin sing g the their ir mou mouths ths,, swa swallo llowin wing, g, kic kickin king, g, cry crying ing,, reaching out to touch people and objects, and so on. In a very literal sense

 

28

 

Embodiment and Cognitive Science

we kinesthetically grow into our bodies. Infants come to realize that different parts of their bodies are capable of specific movement (e.g., arms that extend, fingers that touch, spines that bend, knees that flex, mouths that open and shut, and so on). As infants make kinesthetic sense of their physical experiences, they progressively build more complex mental understandings having to do with containment, balance, weight, physical effort, and the consequences of their embodied actions upon the world. As the philosopher Edmund Husserl once observed, primal movement is “the mother of all cognition” (Husserl, 1980: 69). What underlies people’s abilities to move as they do and have any awareness of their bodies? People have a specific body sense that yields specific knowledge about their bodies. Various Various information systems yield information about the state and performance of the body. Among these systems are the following (Bermudez, Marcel, & Eilaan, 1995: 13): (a)   Infor (a) Information mationabout about pre pressur ssure, e, tempe temperatur rature, e, and fricti friction on fro from m rec receptor eptorss on the skin and beneath the surface. (b) Information (b)  Information about the relative state of body signals from receptors in the joints, some sensitive to static position, some to dynamic information. (c)   In (c) Info form rmat ation ion ab abou outt ba bala lanc ncee an and d po post stur uree fr from om th thee ve vest stib ibul ular ar sy syst stem em in the inner ear and the head/trunk dispositional system and information from pressure pressure on any parts of the body that might be in contact with gravity-resisting surfaces. (d)   Information from skin stretch (d) stretch and bodily disposition and volume. (e) Information (e)  Information from receptors in the internal organs about nutritional and other states relevant to homeostasis and well-being. (f)   Information about effort and muscular fatigue from (f) from muscles. (g)   blood (g) Info In form rmat ation ion ab abou outt ge gene nera rall fa fatig tigue ue fr from om ce cere rebr bral al sy syst stem emss se sens nsiti itive ve to composition. These information systems work together in complex ways to produce

 both a “body schema” and a “body image.” Unfortunately these terms have sometimes been used indiscriminately to refer to quite different types of   bodily repr representations. esentations. Some of the ways the terms “body image” and “body schema” have been used include the following (Bermudez et al., 1995: 15): (a)   One’s conscious experience of the body at a particular time. (a)  (b) A (b)  A ch chan angin ging g no nonc ncon onsc scio ious us rec ecor ord d of th thee mo mome ment ntar ary y re rela lativ tivee di disp spos osiition of, and space occupied by, one’s body parts. (c)   A no (c) nonc ncon onsc scio ious us pe pers rsis istin ting g re repr pres esen enta tati tion on of th thee st stru ruct ctur uree an and d sh shap apee of one’s one ’s body bo dy.. (d)   A ca (d) cano noni nica call rep epre rese sent ntat atio ion n of wh what at bo bodi dies es in ge gene nera rall lo look ok or fe feel el li like ke.. (e)   A knowledge of one’s own specific appearance. (e)

 

Bodies and Persons

 

29

Explic licit it con concep ceptua tualiza lizatio tions ns of the bod body y, acq acquir uired ed soc social ially ly or aca academ dem-(f)   Exp (f) ically (e.g., that one has a liver). (g) Emotional (g)  Emotional attributions toward one’s body, some of what are tacit and socially determined. (h) Cultural (h)  Cultural symbolizations of the body. (i)   The neuronal vehicles for some of the contents referred to above. (i) These very different aspects of bodily representation can, however, be adequately distinguished from each other. “Body schema” is the way in which the body actively integrates its posture and position in the environment me nt.. We do no nott or ordi dina nari rily ly se sens nsee ou ourr bo bodi dies es ma makin king g po post stur ural al ad adju just stme ment ntss as we perceive objects and events and move about in the world. Body schemas allow us to walk adroitly without bumping into or tripping over things, to follow and locate objects, to perceive shape, distance, and duration, and to catch a ball with accuracy. These mundane events all take place independent of our conscious thoughts of the body body.. Ourr fe Ou felt lt se sens nsee of mo move veme ment nt is re regu gula late ted d by ou ourr pr prop opri rioc ocep epti tive ve sy syst stem ems. s. Proprioception is neglected as an important embodied system, because it is not traditionally seen as an input i nput system for presenting the world to the mind. Sir Charles Sherrington (1906) called this system the “sixth sense.” The information proprioception provides comes from the nerve endings in muscles and joints, and partly also from those in the skin. The balance organ in the ear contributes information i nformation about one’s posture and position in sp spac ace. e. Ne Nerv rvee en endi ding ngss in th thee mu musc scle less gi give ve in info form rmat atio ion n ab abou outt th thee am amou ount nt and fluctuation of muscle tone and the length and tension of the muscles, and thus provides information about movement and the amount of force used. Nerve endings endings in the joints give information information about the movem movement ent and position of the joints, and thus about movement and posture. Stretch receptors in the skin, especially in the face, give information about facial expressions and movement in the speech eating.gives The information balance organ, together with information from neckand muscles, about global posture and position with respect to the horizontal plane. Sherrington emphasized how proprioception functions automatically and unconsciously and may operate when the brain is disconnected from the nervous system. All our movements and the maintenance of a posture req equi uirre a su subt btle le co coor ordi dina nati tion on of co coun untl tles esss mu musc scle less an and d jo join ints ts th that at ma make ke up

our body schema. Without immediate feedback from the sensory nerves about what the muscles and joints are doing, all of our movements and even the maintenance of our posture would go totally awry. The body schema, the continually updated, nonconceptual, nonconscious information about the body provides the necessary feedback for the execution of   both our gross motor programs and and their fine tuning (Gallagher, (Gallagher, 1995). Take, for example, a simple bodily action like li ke standing up straight. We We have ha ve kn know own n ho how w to do th that at si sinc ncee in infa fanc ncy y, an and d ne need ed no nott bo both ther er co cons nsci ciou ousl sly y

 

30

 

Embodiment and Cognitive Science

with the appropriate motor program to perform the action. Also, the fine tuning of this posture is provided for by the body schema. If our arms are slightly in front of the body, we have to lean back somewhat to compensate for the extra weight in front. If we carry something in front of  us, we have to compensate more. The compensating just happens; we don’t have to think about it. We We don’t even notice these small corrections, not in others and not in our own case. It is only when we see people with very large bellies or pregnant woman that we notice that they are leaning backward. All that information from the nerve endings in muscles and joint joints, s, together with the information information from the balance organ, organ, is needed. The body schema has to feed it to the motor program in time; otherwise we would fall over. But we don’t have to be bothered with it. It all happens automatically, so that we have our hands literally free for othe ot herr th thin ings gs.. Of co cou urs rsee th this is on only ly go goees for mo moto torr ta task skss th that at ar aree no nott to too o co commplex pl ex,, or th that at we ha have ve ma mast ster ered ed so some me ti time me ag ago. o. Ac Acqu quir irin ing g ne new w mo moto torr sk skil ills ls,, such as learning to drive or play the violin, requires conscious effort and attention. Our body schemas may be supramodal and used in comprehending  body position information about both self and others. One study in support po rt of th this is id idea ea sh show owed ed pa part rtic icip ipan ants ts th thee su succ cces essi sive ve po posi siti tion onss of a mo mode dell in either the same or differ different ent body positions (Reed & Farah, 1995). The partici ti cipa pant nts’ s’ ta task sk wa wass to mo move ve ei eith ther er th thei eirr ar arms ms or th thei eirr le legs gs as th they ey vi view ewed ed th thee first picture. Participants better detected changes in the model’s position when these changes were identical to the body movements made by the participants. Another study showed that the particular part of the participant’s body being moved determined the participant’s ability to detect changes in the model’s body positions. People’s body schemas appear to have internal organization, with different representations from different  body parts. Using one part of the body schemas to monitor one’s own movement automatically focuses attention on the corresponding parts of  other bodies that we are watching. One complex study employed a modified version of the “alien hand” para pa radi digm gm to ex exam amin inee th thee co cont ntri ribu buti tion on of bo body dy sc sche hema mass to se self lf-r -rec ecog ogni niti tion on (van den Bos & Jeannerod,  2002). A participant and experimenter sat at opposite ends of a table. Each person placed her right hand, which was gloved, on the table so that the hands faced each other. The participant’s hand, however, was hidden under a screen with a mirror. A video camera filmed the mirror image that was displayed on the screen, which created thee im th impr pres essi sion on fo forr th thee pa part rtic icip ipan antt th that at sh shee wa wass lo look okin ing g di dirrec ectl tly y at th thee ta tabl blee

with the two hands. Each trial began with the participant’s and experimenter’s hands in fists fis ts.. A si sign gnal al pr prom ompt pted ed th thee pa part rtic icip ipan antt to mo move ve ei eith ther er th thee th thum umb b or in inde dex x finger finger. . The experimenter same(different-movement fi nger as the participant finger (samemovement condition) or moved anotherthe finger condition),

 

Bodies and Persons

 

31

After  1  second, the screen was turned off and an arrow appeared pointing to the location of the screen where the hands had been displayed. The participant’s task was to judge whether the hand she had seen at the location of the arrow was her own hand or the experimenter’s. There was one other important factor in this study. The image of the hand on the screen was also rotated in varying degrees or not rotated. When there was no rotation, the spatial orientation of the participant’s own hand was consistent with the position of her body. When the image was rotated by   90  degrees, the orientation of the participant’s, as well as the experimenter’s, hand was incongruent with the participant’s body orie or ient ntat atio ion. n. In th thes esee tw two o ca case ses, s, bo both th ha hand ndss we werre se seen en as “a “ali lien en.” .” Wh When en th thee image was rota rotated ted 180  degrees, the experimenter’s hand was congruent with the participant’s body orientation, and the participant’s hand was congruent with the experimenter’s body orientation. This condition made itseemasiftheexperimenter’shandbelongedtotheparticipant’sbody,and the participant’s own hand was seen as if it belonged to the experimenter. experimenter. Van den Bos and Jeannerod first examined the effect of action cues on self-recognition. The participants almost always recognized their own hands in the differe different-movement nt-movement conditions, but made significantly more errors in recognizing their own hands when the experimenter performed the same finger movement. Thus, self-recognition becomes more difficult when fewer action cues were available. Rotation of the image also affected participants’ self-recognitions, but only on ly in th thee sa same me-m -mov ovem emen entt co cond ndit itio ion. n. Th Thus us,, pa part rtic icip ipan ants ts ma made de th thee fe fewe west st (15%) errors when their hand orientations were congruent with their own  body orientations, a higher degree of errors (24%) when the orientations of th thei eirr ha hand ndss we werre in inco cong ngru ruen entt wi with th bo body dy or orie ient ntat atio ion n (a (att 90 degr degrees ees rota rota-tion), and the most errors (35%) when the experimenter’s hand was congruent with the participants’ body orientations ( 180 degree rotation). The higher error rate in the   180  degree rotation condition shows that body schema clearly contributes to self-recognition, especially when other action cues are absent. Studies like this reveal, again, that body schemas are integral to both how we move and recognizing who we are as unique persons. Not surprisingly, the environment affects people’s intuitions about their bodies. People perceived their arm lengths to be greater when the outstretched arm approached a barrier (a wall) than when the arm was outstretched in an op open en sp spac acee (a ha hallw llway ay)) (S (Sho hont ntz, z, 1969). Pe Peop ople le al also so es estim timat ated ed ar arm m le leng ngth th to be greater when they were instructed to point to objects (Shontz, 1969). When Wh en as aske ked d to es esti tima mate te he head ad wi widt dth, h, pe peop ople le ov over eres esti tima mate ted d le less ss wh when en th thei eirr faces were touched by someone else than when they were not touched. These findings demonstrate how the body’s boundaries expand and contract in different contexts and tasks. Once again, there is no bodily per-

ception without a world in which the body moves. Any change in the

 

32

 

Embodiment and Cognitive Science

environment brings about some, even very slight, change in our experience of the body. Body Bo dy sc sche hema mass ar aree no nott is isom omor orph phic ic to to,, or ca can n be ex expl plai aine ned d by ph phys ysio iolo logy gy.. Our bodies react intelligently in response to the environment in complex ways that resist description by a single mechanism or reflex. We actively organize our embodied experience given practical concerns. My running across be explained in terms of the physiological activityaoffield my cannot body (Gallagher,  1995simply ). Instead, the pragmatic aim behind my movement (e.g., run to catch a ball in a game of baseball) explains my action, at the same time that the physical environment, my previous experience at catching balls, and even the rules of baseball shape the way I move my body body.. Thus, a person’s individual experience, and the personal and cultural reasons for them, give rise to different body schemas that cannot be explained solely in biological terms. Body image refers to conscious representations of the body, including how ho w th thee bo body dy se serv rves es as an ob obje ject ct of fe feel elin ings gs an and d em emot otio ions ns,, su such ch as wh whet ethe herr we experience ourselves as fat, thin, tired, and so on (Gallagher,  1995). People’s subjective evaluations of their own bodies, along with their associated feelings and attitudes, have been extensively studied in terms of the relationship between body image and satisfaction with individual  body parts, the relation of body image and eating disorders, and the relation of physical variables to body image (Fisher,  1990). A typical body imagequestionnaireasksparticipantstogiveratingsfor 19 items items,, includ including ing self-perceptions of whether their bodies are healthy/unhealthy, physically attractive/unattractive, attractive/ unattractive, sources of pleasure/ pleasure/displeasure, displeasure, something to be hidden/shown, hidden/ shown, calm/ calm/nervous, nervous, old/ old/young, young, frail/ frail/robust, robust, energetic/ energetic/not not energetic, and so on (Koleck, Bruchon-Schweitzer, Bruchon-Schweitzer, Cousson-Gelie, Gillard, & Quintard, 2002). These questionnaires often find that body satisfaction is associated with sex, health, and current and future emotional adjustment. There are important interactions between body schema and body image. For instance, the unconscious operation of the body schema influences significant aspects of our conscious experiences of the body. Some research demonstrates that changing how the body performs clearly affects how people perceive their bodies, but also colors people’s perception of space and external objects. Thus, people’s body size is consistently overestimated relative to their size estimates for other objects (Gardner, Martinez, & Sandoval,  1987). Various studies show that body schemas affect spatial perception and perception of objects. Changes in posture, mobility, physical ability, and other examples associated with the body schema imposed by abnormality, disease, or illness (e.g., obesity, rheumatoid toi d art arthri hritis tis,, mul multip tiple le scl scler erosi osis), s), or by tem tempor porary ary phy physic sical al cha change ngess (su (such ch as pregnancy), affect affect perceptual, cognitive, and emotive aspects of body image. For example, degeneration of body function and changes in mobility

 

Bodies and Persons

 

33

lead to decreases in the senses both of body integrity and of strong body  boundaries (Gardner (Gardner et al., 1987). Various studies also show that exercise, dance, and other embodied practices affect people’s emotional stances toward their body images (Asci, 2003). Disordered Bodies

Neurological damage cantheir have terrible effects on people’s perceptions of  their own bodies and of bodily experiences. When people’s perceptions of their bodies are disturbed, they suffer mental/emotional disarray.. For example, people with cerebral palsy often experience involuntary ray move mo veme men nts of th thee li limb mbs, s, and som omeeti time mess cl clai aim m th that at a bod ody y pa parrt is not th theeir own. Once again, the ability to move one’s body intentionally with some sense of being able to predict what happens next seems critical to identification of oneself with one’s body. One of the most devastating disorders that a person can experience, apart apa rt fr from om par paraly alysis sis,, is the los losss of pr propr oprioc iocept eption ion.. Con Consid sider er the fam famous ous cas casee of Ian Waterman (Cole, 1995). Waterman fell sick at age  19, became very weak, and could not walk or maintain an upright position and his speech wass sl wa slur urrred ed.. He so soon on se seem emed ed to be pa para raly lyze zed, d, bu butt th ther eree wa wass no noth thin ing g wr wron ong g with his muscles or motor neurons. Even when lying in bed, Waterman could move his arms and legs in all directions. But he seemed to have no cont co ntrrol ov over er hi hiss mo move veme ment ntss an and d lo lost st al alll se sens nsee of to touc uch h an and d pr prop opri rioc ocep epti tion on from fro m the neck down. All the larg largee sens sensory ory nerves that send infor information mation from the periphery to the brain had been destroyed. Waterman was left with feelings of deep pain, of heat and cold, and of fatigue. But he did not have a single feeling of the position and posture of his body, or a feeling of touch on his skin. If he was not looking, Waterman Waterman could not say where his arms and legs were positioned. Waterman eventually taught himself to sit up by consciously planning small action he needed to do. For instance, he first tried to use the muscles of his abdomen as if he were about to perform a sit-up. But when this plan did nothework. Waterman that he needed to lift his up first. When did this and wasrealized first successful, Waterman Waterman was sohead pleased that th at he fo forrgo gott to th thin ink k ab abou outt hi hiss mo movm vmen ents ts an and d sl slum umpe ped d ba back ckwa warrds ds.. Ov Over er many months of laborious practice, Waterman Waterman learned to sit up, eat, dress himself, write, and then even walk. Although his destroyed nerves never got better, he employed visual feedback to consciously plan each bodily movement. However, in the dark, Waterman could not see himself or his surroundings, and was as helpless as he was during the onset of his illness. Wat ater erma man’ n’ss ab abil ilit ity y to us usee th thee vi visu sual al se sens nsee to su subs bsti titu tute te fo forr mu musc scle le se sens nsee indicates that in sighted people the two senses may often be combined to

 

34

 

Embodiment and Cognitive Science

allow judgments of movement. Deafferented patients such as Waterman (peo (p eopl plee wh whos osee se sens nsor ory y ne nerv rves es le lead adin ing g fr from om th thee pe peri riph pher ery y to towa ward rd th thee br brai ain n do not function any more) have a different awareness of their bodies than we have, because we normally do not need to rely so heavily on vision for our proprioceptive behavior. Although these patients’ body schemas are almost nonexistent, their body images remain remarkably intact because they th ey ha have ve an aw awar aren enes esss of ho how w th they ey lo look ok an and d ho how w mu much ch sp spac acee th they ey oc occu cupy py.. These patients experience pain immediately as located within particular places within theperception frame of their body Butproprioception. their body images arise from their visual alone andimage. not from One notable aspect of the disorder disordered ed body is that people still feel body parts even after the part has been removed. One study of  300  300  amputees in prisoner-of-war camps during World War II showed that  98 percent experienced a “phantom limb,” usually in the form of a pleasant tingling sensation (Henderson & Smyth,  1948). Some amputees, however, experience the missing limb as having pain. There are various speculations about phantom limb experiences. Resear se arch ch wi with th no norm rmal al ad adul ults ts ha hass sh show own n th that at si simp mply ly lo look okin ing g at a mo movi ving ng li limb mb can ca n cr crea eate te a se sens nsee of vo volu lunt ntar ary y mo move veme ment nt in th thee ob obse serv rver er (R (Ram amac acha hand ndra ran n & Blakeslee,  1998). People can even be fooled into thinking that someone else’s hand, for example, was their own. One study had people don a glove, insert their hand into a box, and then on a signal, draw a line on a piec pi ecee of pa pape perr (N (Nie iels lsen en,, 1963). Ho Howe weve verr, th thee pa part rtic icip ipan ants ts di did d no nott kn know ow th that at the hand they saw in the box was actually a mirror reflection of another person’s hand, also gloved and holding a pen in the exact place where they would expect their hand to be. When the signal was given, and the imposter hand drew a line that varied from the line the participant had  been instructed to draw, draw, participants typically adjusted adjusted their own arms to compensate for the observed arm’s initial trajectory trajectory.. This Th is “a “ali lien en ha hand nd”” pr proc oced edur uree ha hass be been en em empl ploy oyed ed to st stud udy y th thee ex expe peri rien ence ce of ph phan anto tom m lim limb b mo move veme ment nt (R (Ram amac acha hand ndra ran n & Bl Blak akes esle lee, e, 1998). Ind Individ ividuuals with a phantom arm placed both the phantom and their other real arm into a mirror box. They saw a reflection of the real arm in the place where the would be if it werethe real. When the moved the real armphantom voluntarily, he experienced phantom armperson as moving voluntarily as we well ll.. Th Thus us,, th thee vi visi sibl blee ha hand nd gu guid ided ed th thee ex expe peri rien ence ce of th thee ph phan anto tom. m. Ev Even en when wh en an ex expe peri rime ment nter’s er’s ar arm m ap appe pear ared ed in pl plac acee of th thee ph phan anto tom, m, mo move veme ment nt of that arm created the feeling that the phantom limb itself was moving. Stud St udie iess su such ch as th this is de demo mons nstr trat atee th that at wi will llfu full bo body dy mo move veme ment nt ca can n be ex expe pe-rien ri ence ced d si simp mply ly by wa watc tchi hing ng an any y bo body dy mo move ve wh wher eree on one’ e’ss bo body dy ou ough ghtt to be be.. People’s subjective experience of then disordered bodies show that  body schemas are not equivalent to physiology (Gallagher (Gallagher,,  1995). There is nothing in the physiology of an amputated leg that gives some patients

 

Bodies and Persons

 

35

the feel el of th thei eirr rea eall le legs gs be befo forre th thes esee we werre am ampu puta tate ted. d. In Inst stea ead, d, th thee mi miss ssin ing g the fe limb remain part of the individual’s body schema that continues to shape how that person moves and feels.

Understanding Other Bodies

When we see another person, we do not perceive his or her body as a mere physical thing, but rather as a living body like our own. There may  be a deep connection between the mental repre representation sentation of posture, the movement of one’s own body, and the perception of posture and movement of other bodies. Ideomotor action refers to body movements that tend to arise in observers as they watch other people perform specific action ti ons. s. Fo Forr ex exam ampl ple, e, av avid id sp spor orts ts fa fans ns of ofte ten n rep epor ortt te tens nsin ing g th thei eirr mu musc scle less an and d movi mo ving ng th thei eirr ar arms ms an and d le legs gs as th they ey obs bser erve ve,, say ay,, a fo foot otb bal alll gam amee on te tele levi vi-sion. Psychologists have debated the reasons for this kind of “ideomotor action” since the 19th century (Carpenter, (Carpenter, 1874; James, 1890). Contemporary research suggest that simulation mechanisms provide the common code between perception and action, such that perceiving an action induces the production of a similar act, or the urge to act, in the observer (Knuf, Aschersleben, & Prinz,  2001;  see Chapter  3 ). When we see someone perform an action, the same motor circuits that are activated when we perform such actions are concurrently activated (i.e., an “as-if-body” loop, Damasio, 1999, 2003). Because of this neural-sensation matching to a self-p sel f-perf erform ormed ed act action ion,, we und unders erstan tand d oth other’ er’ss mov moveme ements nts as goa goal-d l-dir irect ected ed actions. More generally, shared representations of perception and action may underlie social cognition and intersubjectivity (Gergely & Watson, 1999; Rochet, 2001; Tr  Trevarthen evarthen,, 1977). Understanding other people requires, among other things, the capacity to be em empa path thet etic ic.. Em Empa path thy y is no nott ju just st un unde ders rsta tand ndin ing g an anot othe herr pe pers rson on’s ’s pa parrticul tic ular ar ex expe peri rien ence cess (s (sad adne ness ss,, jo joy y, an and d so on on), ), bu butt is th thee ex expe peri rien ence ce of an anot othe herr as an em embo bodi died ed su subj bjec ectt of ex expe peri rien ence ce li like ke on ones esel elf. f. On Onee rec ecen entt th theo eory ry of em em-pathy suggests that attending to an object’s state (e.g., looking at another person) automatically activates the subject’s representations of the state, situ si tuat atio ion, n, an and d ob obje ject ct,, an and d th that at ac acti tiva vati tion on of th thes esee re repr pres esen enta tatio tions ns au auto toma mattically or generate associated automatic and somatic responses, unlessprimes inhibited (Preston the & de Waal, 2002 ). According to this model, various phenomena such as emotional contagion, cognitive empathy, guilt, and an d he help lpin ing g ar aree si simi mila larr in th that at th they ey rel ely y on a pe perrce cept ptio ionn-ac acti tion on me mech chan anis ism m (PAM). Activating shared representations between perception and action is clearly an important element in the experience of empathy. But empathy is deeply grounded in the experience of our lived bodies, and this experience enables us to directly recognize others, not as bodies endowed with

 

36

 

Embodiment and Cognitive Science

minds, but as persons like us (Gallese, Ferari, & Umilta, 2002). Other people’s ple ’s act action ions, s, sen sensat sation ions, s, and emb embodi odied ed exp experi erienc ences es bec become ome mea meanin ningfu gfull to us,, pr us prec ecis isel ely y be beca caus usee we sh shar aree th thes esee wi with th ot othe hers rs.. Bu Butt ho how w ca can n su such ch sh shar arin ing g of experience be possible? One possibility is that the mechanism for this sharing of experience is simulation (Gallese et al., 2002). By modeling some behavior, our actions prov pr ovid idee a si simu mula late ted d rep eprres esen enta tati tion on of th thee sa same me pr proc oces esss th that at ca can n be us used ed to prod pr oduc ucee it it,, on th thee on onee ha hand nd,, an and d to de deccod odee it wh wheen per erfo forrme med d by so some meon onee

else, on the other. These “as if” simulation mechanisms may underlie a wide range of processes as diverse as action perception and imitation (as simulatio simula tion n of the obs observ erved ed act action ion), ), emo emotio tion n per percep ceptio tion n (as sim simula ulation tionof of the percei per ceived ved emo emotion tion), ), and min mindr dread eading ing.. Sim Simula ulatio tion n the theory ory hol holds ds tha thatt we unders de rsta tand nd ot othe her’ r’ss th thou ough ghts ts by pr pret eten endi ding ng to be in th thei eirr “m “men enta tall sh shoe oes” s” an and d  by using our own mind/body as a model for the minds of others (Gallese & Goldman, 1998). There are dedicated brain structures, called “mirror neurons, ro ns,”” tha thatt und underp erpin in a dir direct ect,, aut automa omatic tic,, non nonpr predi edicti ctive, ve, and non noninf infer erent ential ial simulation mechanism, by means of which the observer would be able to recognize, understand, and imitate the behavior of others. Research suggest ge stss th that at a mi mirrror ma matc tchi hing ng sy sysste tem m co coul uld d be at th thee bas asis is of ou ourr ca cap pac acit ity y to perceive in a meaningful way, way, not only the actions, but also the sensations and an d em emot otio ions ns of ot othe hers rs (G (Gall alles ese, e, 2001). For exa exampl mple, e, sin single gle-ne -neur uron on re recor cordding experiments in humans have demonstrated that the same neurons become active when the subject either feels pain or observes others feeling pain (Hutchinson, Davis, Lozano, Troby, & Dostrovsky, 1999). Neuropsychological studies demonstrate that individuals with damage to the right somato som atosen sensor sory y cor cortic tices, es, whi which ch ar aree cri critic tical al to und unders erstan tandin ding g bod body y map maps, s, experience defects in emotion and feeling and in feeling empathy for others (Adolp (Ad olphs, hs, Dam Damasi asio, o, Tran ranel, el, Coo Cooper per,, & Dama Damasio sio,, 2000). Th Thus us,, ha havi ving ng so some me sense of our own bodily responses is an important part of understanding, and feeling appropriate emotion for, for, other people’s experiences. Bodies and Culture

People and their bodies move in physical environments imbued with culture tu re.. Th Thee bo body dy sy syst stem em (it (itss an anat atom omic ical al st stru ruct ctur ure, e, its or orific ifices es or en entr tran ance cess an and d exists, substances itcultural secretes, its position, offersexperiences insightful analysis for the understanding systems. “The etc.) physical of the  body,, always modified by the social categories through which it is known,  body sustain a particular view of society” (Douglas,  1970: 93). Anthropologists have demonstrated in a variety of cultural settings how many elementary embodied experiences are shaped by local cultural knowledge and practice (see Csordas,  1994; Lambek & Strathern,  1998). The body is appreciated at ed fo forr its sy symb mbol olic ic pr prop oper ertie ties. s. Pe Peop ople le in inst still ill cu cultu ltura rall me mean anin ing g in into to bo bodi dily ly proce pr ocesse ssess suc such h as br breat eathin hing, g, blu blushi shing, ng, men menstr struat uation ion,, bir birth, th, sex sex,, cry crying ing,, and

 

Bodies and Persons

 

37

laughing, and value the products of the body (e.g., blood, semen, sweat, tears, feces, urine, and saliva) differently in changing cultural contexts. More complex bodily experiences ranging from nerves (Low,  1994), rape (Winkler, 1994), AIDS (Martin, 1994), and pain and torture (Jackson, 1994; Scheper-Hughs & Lock,  1987), to name just a few, have been studied by anth an thro ropo polo logi gist stss to ex expl plor oree th thee lin linka kage gess be betw twee een n em embo bodi dime ment nt an and d cu cultu ltura rall meaning. At the same time, anthropomorphism is a primary means by which small-scale societies use body parts, and their actions, to refer to an enormous range of houses, artifacts, animals, and plants (Tilley, 1994). Much Mu ch of th thee in inte terrest in cul ultu turral st stud udie iess is not in em emb bod odim imen entt per se, but thee wa th ways ys th that at at atte tend ndin ing g to th thee bo body dy br brin ings gs di diff ffer eren entt cu cultu ltura rall pr proc oces esse sess in into to  1998

focus (Lambek & Strathern, ). Rather than being a biological given, embodiment is a category of sociocultural analysis, often revealing complex dimensions of the interactions between bodies and personhood. Talk Talk of embodiment is situated in reference to topics such as health and illness, kinship, modes of production and exchange, gender and age hierarchies, langua lan guage ge pra practi ctices ces,, re religi ligious ous and pol politic itical al dis discip ciplin lines, es, jur jural al rul rules, es, per pervas vasive ive metaphors, spirit possession, historical experiences, and myths. People’s  bodies are more than surfaces for social inscription (i.e., the “body as text” metaphor), but incorporate cultural meanings and memories. Culture does not just inform embodied experience; embodied experience is itself culturally constituted (Csordas, 1994; Maalej, 2004; Strathern, 1996). Many embodied experiences are rooted in social-cultural contexts (Quinn, 1991). Fo Forr in inst stan ance ce,, th thee no noti tion on of CO CONT NTAI AINM NMEN ENT T (s (see ee Ch Chap apte terr 4), is based on one’s own bodily experience of things going in and out of the  body,, and of our bodies going in and out of containers. But containment  body is not just a sensorimotor act, but an event full of anticipation, sometimes surprise, sometimes fear, sometimes joy, each of which is shaped by the presence of other objects and people that we interact with. Certain aspects of se sens nsor ory y pe perrce cept ptio ion n ar aree em emer erge gent nt an and d de depe pend nden entt on cu cult ltur ure, e, wh which ich in influfluences enc es the emb embodi odimen mentt of dis dispos positio itions ns thr throug ough h eve everyd ryday ay pra practi ctice ce (Bo (Bord rdieu ieu,, 1977; Shore, 1996; Shweder, 1991; Strauss & Quinn, 1997). Several ethnographic studies demonstrate that vision in many societies is not the central perceptual mode. Instead, the “lower” senses are central to th thee me meta taph phor oric ic or orga gani niza zati tion on of ex expe peri rien ence ce (s (see ee Ch Chap apte terr 4). Th This is do does es not impl im ply y th that at pe peop ople le in va vari riou ouss cu cult ltur ures es ha have ve di diff ffer eren entt ph phys ysio iolo logi gies es,, bu butt on only ly that they weigh sensory information differently in how they think about their experiences and the world around them. For instance, among the Song So ngha hay y pe peop ople le of Ma Mali li an and d Ni Nige gerr, sm smel ell, l, ta tast ste, e, an and d so soun und d co cont ntri ribu bute te to th thee organizat org anization ion of their rel religious igious and philos philosophic ophical al expe experienc riences es (Stoll (Stoller er,, 1989). Song So ngha hay y so sorrce cerrer erss an and d gr grio iots ts le lear arn n ab abou outt po powe werr an and d hi hist stor ory y by “e “eat atin ing” g” it it,, ingesting odors and tastes, savoring textures and sounds. The stomach is considered the site of human personality and agency. Social relations are considered in terms of eating.

 

38

 

Embodiment and Cognitive Science

The Anlo-Ewe speaking people of West Africa do not emphasize strict distinctions between the five senses in the way that people from Western cultures seem to do (Geurts,  2002). In Anlo-Ewe cultural contexts, a discret cr etee ca cate tego gory ry de dema mark rkin ing g th thee fiv fivee se sens nsor ory y sy syst stem emss is no nott ti tigh ghtl tly y bo boun unde ded d or seen se en as a pa part rtic icul ular arly ly me mean anin ingf gful ul wa way y of cl clas assi sify fyin ing g ex expe peri rien ence ce or th theo eori rizzing about knowledge. Instead, in the Anlo-Ewe mind, sensations caused  by stimuli from external objects are epistemologically related to sensations that arise from internal somatic modes (such as interoception, which governs balance, movement, and proprioception). The Anlo cultural tradition does not have a theory of the senses, but has a coherent and fairly complex theory of inner states which link sensations to emotions, dispositions, sition s, and vocat vocations, ions, referre referred d to as “sese “seselalame lalame”” (a feeling in the body or flesh). Sese Se selal lalam amee is a cu cult ltur ural ally ly el elab abor orat ated ed wa way y by wh which ich An Anlo lo-E -Ewe we re read ad th thei eirr own bodies while simultaneously understanding them in relation to ob jects, the environment, environment, and the bodies bodies of those around them. For instance, instance,

the Anlo-Ewe people greatly emphasize the proprioceptive quality of balance. They are openly encouraged to actively balance their own bodies as infants, they balance small bowls and pots on their heads, and they carry books and desks on their heads when walking to and from school. Adults perceive balance as a defining attribute of mature individuals and the human species more generally. generally. But this attribute is not merely a physical characteristic of individuals, but a direct association between bodily sens se nsat atio ions ns an and d wh who o yo you u ar aree or wh who o yo you u ma may y be beco come me.. Th Thus us,, yo your ur ch char arac acte terr and your moral fortitude is established in the way you move. Thus, people are designated as moral or immoral through reference to the cultural categories that implicate and create sensory phenomena. Amon Am ong g th thee On Onge gee, e, a hu hunt ntin ing g an and d ga gath ther erin ing g pe peop ople le of th thee Lit Littl tlee Anda An dama man n Is Isla land nd in th thee Ba Bay y of Be Beng ngal al,, sm smel elll is th thee pr prim imar ary y se sens nsor ory y me medi dium um through which the categories of time, space, and person are conceptualized (Howes,  2003). Odor, according to the Ongee, is the vital force that animates all living, organic beings. Newborns are said to possess little scen sc ent. t. Ch Chil ildr dren en ac acqu quir iree mo morre ol olfa fact ctor ory y st strren engt gth h as th they ey ge gett ol olde derr. Th Thee od odor or that th at a pe pers rson on sc scat atte ters rs ab abou outt du duri ring ng th thee da day y is sa said id to be ga gath ther ered ed up du duri ring ng sleep by an inner spirit and returned to the body, making continued life poss po ssib ible le.. De Deat ath h oc occu curs rs wh when en on onee lo lose sess on one’ e’ss od odor or,, an and d on once ce de dead ad,, a pe pers rson on  becomes an inorganic inorganic spirit seeking out the odors of the living in order to  be reborn. In these ways, the Ongee life li fe cycle is conceived in terms of an olfactory progr progression. ession. Daily life is a constant game of olfactory hide-and-seek. Animals are killed to release their odors, while people try to hide their own odors from both other persons and animals. The Ongee speak of “to hunt” as “to release smell causing a flow of death.” When walking, the Ongee try to step in the footsteps of the person in front to confuse personal odors, making it difficult for spirits to track them down. They also screen their

 

Bodies and Persons

 

39

odors with smoke. theythe areodors traveling single file, thebehind. leader The will carry burning woodWhen to cover of allinthose walking Onge On geee al also so pa pain intt th them emse selv lves es wi with th cl clay ay to in inhi hibi bitt sm smel elll em emis issi sion on.. An On Onge geee with wi th a pa pain inte ted d bo body dy wi will ll de decl clar aree, “T “The he cla lay y pai aint nt ha hass bee een n go goo od! I fe feel el th that at my smell is going slowly and in a zig-zag manner like the snake on the ground!” (Pandya, 11993: 137). Spac Sp acee is un unde ders rsto tood od by th thee On Onge geee in te term rmss of th thee dy dyna nami micc en envi virron onme menntal flow of smells, and not as static physical dimensions. Thus, a village’s space will expand or contract depending on the olfactory ambiance, as when strong-smelling substances such as pig’s meat are present or the scent of seasonal flows wafts through the air. If asked to draw a map, an Ongee will depict a line of movement of scents from one place to another, another, rath ra ther er th than an th thee lo loca cati tion onss of th thee pl plac aces es th them emse selv lves es.. Th Thee On Onge geee me meas asur uree ti time me as a cycle of odors, or a “calendar of scents” (Radcliffe-Brown, 1964). These few examples illustrate how different cultures attach different meanings to the uses of different bodily senses, including proprioception, and this affects the way each culture imagines and represents the world. Whatitmeanstobeaperson,andideasaboutthekindsofpersonsthatexist, aredirectlytiedtothesensethataculturalgrouprecognizes,attendsto,and incorporates bodies into their ways of being in the world. The senses are

ways of embodying social categories (Geurts,  2002). Sensory experience, which whi ch dif differ ferss fr from om phy physio siolog logica icall mec mechan hanism isms, s, can cannot not,, the there refor fore, e, be defi defined ned universally,, but is always deeply influenced by cultural variation. universally Many cognitive scientists react negatively to anthropological claims about the cultural nature of bodily sensations. After all, people across cultures have similar biology,reflect as well as highly similar bodily and social interactions that presumably universal properties of embodiment. I agree with this sentiment, although I am quick to note that cognitive scientists still ignore how body experience is shaped by cultural practices that resist simple biological explanation. One way to embrace the role of  cultural activity into a theory of embodied cognition is to recognize, and study,, different levels of embodiment in thought, language, and action. study Levels of Embodiment

Cognitive scientists generally wish to uncover the neural and cognitive mechanisms that presumably subsume perception, thought, language, emot em otio ion, n, an and d co cons nsci ciou ousn snes ess. s. Th Thee es esse sent ntia iall li link nk of bo bodi dies es an and d pe pers rson onss do does es not imply that whole bodies are the only level at which to analyze and understand language and cognition. There are, in fact, three levels of  embodi emb odimen ment: t: the neu neural ral lev level, el, the phe phenom nomeno enolog logica icall con consci scious ous exp experi erienc ence, e, and the cognitive unconscious (Lakoff & Johnson, 1999). Neural embodiment concerns the structures that characterize concepts and cog cognit nitive ive ope operat ration ionss at the neu neuro rophy physio siolog logica icall lev level. el. Our con concep cepts ts and experience are fundamentally embodied within the brain. Yet the neural

 

40

 

Embodiment and Cognitive Science

level alone cannot explain the bodily basis of language and cognition. Brains notform simply receive input from the environment and provide output do in the of instructions to the body. Neural assemblies operate in relation to the entire body as it functions within concrete situations. The cognitive unconscious consists of all the mental operations that structure and make possible conscious experience, including the understanding and use of language. The cognitive unconscious makes use of  and guides the perceptual and motor aspects of our bodies, especially those that enter into basic-level and spatial-relation concepts. It includes alll ou al ourr un unco cons nsci ciou ouss kn know owle ledg dgee an and d th thou ough ghtt pr proc oces esse ses. s. Th Thee bo body dy is cr cruc ucia iall at this level, because all of our cognitive mechanisms and structures are grounded in patterns of bodily experience and activity. The phe phenom nomeno enolog logica icall lev level el is con consci scious ous,, or acc access essibl iblee to con consci scious ousnes ness. s. It consists of everything we can be aware of, especially our own mental states, our bodies, our environment, and our physical and social interaction ti ons. s. Th This is is th thee le leve vell at wh whic ich h we fe feel el ex expe peri rien ence ce,, of th thee wa way y th thin ings gs ap appe pear ar to us, and of qualia, that is, the distinctive qualities of experience such as the pain of a toothache, the taste of chocolate, the sound of a violin, or the redness of a ripe Bing cherry. Thes Th esee th thrree le leve vels ls ar aree no nott in inde depe pend nden entt of on onee an anot othe herr. Th Thee de deta tail ilss of th thee character of the cognitive unconscious and of conscious experience arise from fr om the nat natur uree of neu neural ral str struct uctur ure. e. We wou would ld not hav havee the spa spatia tial-r l-rela elatio tions ns conc co ncep epts ts we ha have ve wi with thou outt to topo pogr grap aphi hicc ma maps ps or or orie ient ntat atio ionn-se sens nsit itiv ivee ce cells lls.. The neural level significantly determines, together with experience of the

external world, what concepts can be and what language can be. Peop Pe ople le ar aree no nott ju just st br brai ains ns,, or ne neur ural al ci cirrcu cuit its. s. Ne Neit ithe herr ar aree th they ey me merre bu bunndles of qualitative experiences and patterns bodily interactions. are they just structures and operations of the of cognitive unconscious.Nor All three are present, and explanations at all three levels are necessary for an adequate account of the human mind. Not surprisingly, in my view, the three levels of embodiment together are constitutive of what it means for someone to be a human person with a particular identity and different cognitive abilities. Conclusion

People’s experiences of themselves as “persons” are clearly intimately related lat ed to the their ir or ordin dinary ary bod bodily ily exp experi erienc ences. es. Our sen senses ses of age agency ncy,, own owners ership hip of ou ourr me ment ntal al ac acts ts,, un unit ity y, an and d co cont ntin inui uity ty ar aree ti tigh ghtl tly y li link nked ed to th thee reg egul ular arit ity y of rec ecur urri ring ng bo body dy ac acti tivi vitie ties. s. No None ne of th this is co comp mple lete tely ly cl clos oses es th thee mi mind nd-b -bod ody y gap,norshouldwhatIdescribeherebeseenasanattempttoreducepersonhood to the body body.. Of course, the boundaries of the self and of personhood are not stable, but are shifting, permeable, and partly structured by social and environmental contingencies. But it is generally possible to conceive

 

Bodies and Persons

 

41

of personhood as an emergent property property of interactions i nteractions of the brain, body body,, and wor world. ld. The These se dyn dynami amicc cou coupli plings ngs sug sugges gestt tha thatt und unders erstan tandin ding g the “se “self” lf” and our sense of who we are as individual persons with controllable minds and bodies require special attention to these couplings, not just to  brains, bodies, or world as separable entities. Appreciating Appreciating that minds are rel elat ated ed to wh whol olee pe pers rson ons, s, an and d th that at pe pers rson onss in inte tera ract ct wi with th ea each ch ot othe herr an and d th thee envi en viro ronm nmen ent, t, pr prov ovid ides es th thee ke key y to un unloc locki king ng th thee se secr cret etss of ho how w pe perc rcep epti tion on,, cognition, and language are thoroughly embodied. Let us continue now with this exploration of the embodied mind.

 

3 Perception Perceptio n and Action

Perception is the ability to derive meaning from sensory experience in order to guide adaptive behavior. Human perceptual experience is often thought to arise from the input of information from the world through the five senses to different regions of the brain. Most philosophers and psychologists argue that perception is an inferential process that occurs in a series of steps. We do not come into direct contact with the environment,  but only some aspects of the environment impinge upon us. But traditional acco ac coun unts ts of ho how w we se see, e, he hear ar,, sm smel ell, l, ta tast ste, e, an and d fe feel el do no nott ac ackn know owle ledg dgee th thee importance of the entire human body as it moves through the world and engages in intentional action. This neglect of the body in action has led l ed to  both simplified views of perceptual experience, and, ironically ironically,, overly complex pl ex me mech chan anis isms ms to ac acco coun untt fo forr ho how w pe peop ople le pe perrce ceiv ivee ob obje ject ctss an and d ev even ents ts in the real world. My aim in this chapter is to explore the importance of em bodied action in psychological accounts of human perceptual experience and action. An Overview of Embodied Perception

An embodied view of perception assumes, as explained by biologist

Humberto Maturana (1980:   5), that “Living systems are units of interactions; they exist in an environment. From a purely biological point of  view vi ew,, th they ey ca cann nnot ot be un unde ders rsto tood od in inde depe pend nden entl tly y of th thee pa part rt of th thee en envi virron on-ment me nt wi with th wh whic ich h th they ey in inte tera ract ct,, th thee ni nich che; e; no norr ca can n th thee ni nich chee be de defin fined ed in inde de-pendently of the living system that occupies it.” Moreover, “when an observ se rver er cl clai aims ms th that at an or orga gani nism sm ex exhi hibi bits ts pe perrce cept ptio ion, n, wh what at he or sh shee be beho hold ldss is an organism that brings forth a world of actions through sensory motor correlations congruent with perturbation of the environment in which he or sh shee se sees es it [t [the he or orga gani nism sm]] to co cons nser erve ve its ad adap apti tion on”” (M (Mat atur uran ana, a, 1983: 60). The idea of “bring[ing] forth a world of actions” emphasizes a person’s  bodily activity that abolishes a linear causal link between perception and 42

 

Perception and Action

 

43

action. As psychologist-philosopher John Dewey argued over   100  years ago in his famous critique of stimulus-response theories of behavior, behavior, We begin not with a sensory stimulus, but with a sensorimotor co-ordination, the optical-ocular optica l-ocular.. . . . In a certain sense, sense, it is the movement movement which is primary primary,, and the sensation which is secondary, this movement of body, head, and eye movements, determ det ermini ining ng th thee qua qualit lity y of wh what at is ex exper perie ience nced. d. In ot othe herr wo word rds, s, th thee re real al be begi ginni nning ng is with wi th th thee ac actt of se seei eing ng;; it is lo look okin ing, g, an and d no nott a se sens nsat atio ion n of li ligh ght. t. (D (Dew ewey ey,, 1896: 137–8)

All human activity involves embodied correlations. It is misleading to sugges sug gestt tha thatt per percep ceptio tion n and act action ion ar aree dis discr crete ete,, ind indepe epende ndent nt pr proce ocesse ssess tha thatt are causally related in a linear way. Consider one of Dewey’s examples, where he argued that the quality of what is experienced depends on how we are already coordinated in some activity: If one is reading a book, if one is hunting, if one is watching in a dark place on a lovely night, if one is performing a chemical experiment; in each case, the noise has a very differ different ent psychical psychical value; value; it is a different different experien experience. ce. . . . What provides provides the ‘stimulus’ is a whole act, a sensorimotor co-ordination, it is born from it as its matrix matrix . . . the ‘stimu ‘stimulus lus’’ arises out of this this co-ordina co-ordinatio tion; n; it is born from it as a matrix, it represents as it were an escape from it. Unless the sound activity had  been present to some extent in the prior coordinates, it would be impossible for it now to come come to prominence prominence in conscious consciousness. ness. . . . We do not have first first a sound and and then activity of attention, unless sound is taken as mere neuron shock or physical even ev ent, t, no nott as co cons nsci ciou ouss va valu lue. e. Th Thee co cons nsci ciou ouss se sens nsat atio ion n of so soun und d de depe pend ndss up upon on th thee motor response having already taken place. place . (Dewey, (Dewey, 1896: 140)

For Dewey, then, the meaning of any perceptual experience and the response arises together are part of the individual’s “what I am doing now.” now .” This motto nicely captures the essence of embodied perception. In mo more re co cont ntem empo pora rary ry wr writ itin ings gs,, Gi Gibs bson on (1966, 1979) ar argue gued d tha thatt per percep cep-tual systems have evolved to facilitate our interaction with a real, threedimensional world. Perception does not take place in the brain of the perceiver, but rather is an act of the whole animal, the act of perceptually guided gui ded exp explor loratio ation n of the env envir ironm onment ent.. The fun functio ction n of vis vision ion,, for exa exampl mple, e, is to ke keep ep th thee pe perrce ceiv iver er in to touc uch h wi with th th thee en envi virron onme ment nt an and d to gu guid idee ac acti tion on,, not to produce inner experiences and representations. At any given momenttheenvironmentaffordsahostofpossibilities:Icouldgrasptheobject, sit on the chair, walk through the door. These are examples of affordances: relations of possibility between actor and animator (Gibson,  1966,  1979).

Affordances enable animals to recognize what prey they may eat, what pred pr edat ator orss ma may y po poss ssib ibly ly ea eatt th them em,, wh what at tr tree eess ma may y be cl clim imbe bed d to es esca cape pe da dannger, and so on. Imagine the color, texture, taste, and smell of a pineapple. These properties of the object are its affordances, which become variously salient given our particular interactions with pineapples. Our perception of affordances is relative to the perceiving object, so that, for example, in look lo okin ing g at a wi wind ndow ow on onee pe perrce ceiv ives es no nott ju just st an ap aper ertu ture re,, bu butt an ap aper ertu ture re th that at

 

44

 

Embodiment and Cognitive Science

presents the possibility of one’s looking through it (Bermudez, 1995). Perception and embodied action are, therefore, inseparable in the perceptionaction cycle (Neisser (Neisser,, 1976), in which exploration of the visual world, for example, is directed by anticipatory schemes for perceptual action. Percep Per ceptio tion-a n-acti ction on link linkage agess ar aree cer certai tainly nly con constr strain ained ed by the env envir ironm onment ent and can only be accurately described in terms of interactions of brain,  body,, and world. Most generally,  body generally, meaningful perception arises as a result of structural couplings of the organism and its environment (Thompson, Palacios, & Varela, 2002; Varela, Thompson, & Rosch,  1991). Consider, as  just one example, the case of color perception perception (Thompson (Thompson et al., 2002). The traditional view suggests that the function of color vision is to recover from the retinal images reliable estimates of the invariant distal properties of specified surface reflectance (i.e., the percentage of light that a surface refle re flect ctss at ea each ch wa wave vele leng ngth th). ). Bu Butt an en enac acti tive ve ap appr proa oach ch su sugg gges ests ts th that at di diff ffer er-ent animals have different phenomenal color spaces and that color vision does not have the function of detecting any single type of environmental prop pr oper erty ty.. Co Colo lorr pr prop oper erti ties es ar aree en enac acte ted d by th thee pe perrce cept ptuo uo-m -mot otor or co coup upli ling ng of  animals with their environments. In fact, the “prespecified world” we find in, say, say, low-dimensional models of surface reflectances is actually the world as described in relation to the sensory-motor capacities of higher primates. It is reasonable to specify the world in advance when studying our own capacities or those of animals ma ls li like ke us us.. Bu Butt it is no nott le legi giti tima mate te to ma make ke th this is sa same me mo move ve wh when en st stud udyi ying ng animals different from us. For example, visual discrimination for birds is not a cyclopean image reconstructed, but a contextualized specification according to avian sensory-motor activity. activity. This activity reveals what constitutes a relevant world for us, or any animal, not a reconstruction of the world as it appears visually to us. The enactive view of color perception starts with two important facts (Thompson et al.,  2002). First, many different animals (e.g., insects, fish,  birds, and primates) living in diverse environment, all with extremely different neural apparatus, all possess color vision. Second, color vision nonetheless varies both in discriminality and sensitivity. These two observations suggest that there are important differences in phenomenal color spaces for different animals. Animals’ different perceptual experience arises not only from their distinct neurophysiological make-up, but also from their evolutionary histories of environmental interactions. For example, bee color vision is shifted toward the ultraviolet, compared to other insect species, and bee color space includes novel hues. These facts can be explained by appealing to animal-environment coevolution. Bee color vision is sensitive to ultraviolet colors because it is advantageous for

 bees to detect flowers that have ultraviolet reflectances. Moreover Moreover,, flowers have ultraviolet reflectances because it is advantageous from them to be seen by bees.

 

Perception and Action

 

45

One of the colored colored objec objects ts that animal animalss discr discriminat iminatee is other animals. animals. Coloration affords an animal’s visibility, both to cospecifics and to mem bers of other species in its environment. It is not surprising that coloration is invo in volv lved ed in ca camo moufl uflag agee an and d ma many ny ki kind ndss of vi visu sual al re reco cogn gniti ition on (e (e.g .g., ., sp spec ecie iess recognition, sexual recognition, and recognition of motivational states). Understanding the relation among color vision, animal coloration, visual recognition, and animal communication demands recognition of a broad range of physiological, ecological, and evolutionary considerations, ranging from the physiological functions of pigmentation, to coordinate interand intraspecies animal interactions, to the coevolution of various behavioral patterns. Most generally, there is a circular and reciprocal process of interaction in which the structure of the environment constrains the activity of the organism, and the activity of the organism shapes the environment (Odling-Smee, 1988). As Levins and Lewontin (1985) suggested, “the environment and the organism actively co-determine each other” (p. 89). The Brain Alone Is Insufficient to Explain Embodied Perception

Understanding the embodied nature of perception requires looking beyond the brain, and into body-world interactions, when seeking causal explanations for perceptual experience. A traditional story told in most psychobiology textbooks is that the somatosensory region, including both subcortical and cortical neurons, specializes in processing somatic information. Primary sensory cortex has much in common with both primary auditory and primary visual cortex. First, tactile sense from the contralateral side of the body is represented in sensory cortex. Second, there is an orderly representation representation of one side of the body on the surface of the brain. This Th is pr proj ojec ecti tion on of th thee bo body dy on onto to th thee se sens nsor ory y co cort rtex ex is ca call lled ed a “s “som omat atot otop opic ic map” and presents a picture called the “homunculus,” or little man. The homunculus is sketched in Figure 3.1. The empirical data used to construct the homunculus model were developed by Wilfred Penfield, a Canadian neurosurgeon, who was a pioneer in surgical removal of epileptic foci in the brain as a treatment for intractable epilepsy (Penfield & Roberts,  1958). He performed surgery on pati pa tien ents ts un unde derr lo loca call an anes esth thes esia ia so th that at th thee pa pati tien entt co coul uld d co comm mmun unic icat atee an and d move voluntarily voluntarily Penfie Penfield ld would electrically electrically stimul stimulate ate the area thoug thought ht responsible for the patient’s epileptic seizures. When the areas of S1 were stimulated, patients often reported various tactile sensations in particular areas on the contralateral side of the body. body. This method allowed, Penfield to cr crea eate te a ma map p of th thee pr proj ojec ecti tion on of va varrio ious us par arts ts of th thee bod ody y ont nto o th thee br brai ain. n. The homunculus map looks completely different from an ordinary human body. The areas for the face, lips, and hands are massive when compared those for the feet, legs, and trunk. These larger areas of the human

 

 

46

W     i     H    r    a  n   s    t    F    d    i    n    g   T    h  u   e  r    s   m   N  e   b   c  k  

Embodiment and Cognitive Science

 S       E       A      h       lb             r m          o       u      o     w      dl               e      r     

T  H  r   i     K   u   p n n  e k    e

Toes

      k       k     c       d     a     m     e      e     w    r    o       N      H     A

    n Hip     u     r Leg       T

Genitals

   m      b    r      l    d    a    n      E    e    a    r    o     H     F

   r  s   e   g     b    n     i   m     F   u    h    T   y e

  E

B  r   o w  

E    y e  e  F a  a c  ce      e

 s e   N o

Motor cortex

Li ps Jaw

 Tongue  To

ng  lo w i in S wa l lo

Somatosensory cortex

 F a c e

 p s  L i p  h  Tee t h Gums a w  Ja  J Tongue

Phar Ph ary ynx Intra ab Intra  abdo domi min nal

figure 3.1.  Human primary cortex. A view of the surface of the brain, showing

the location of motor and sensory cortices.

primary sensory cortex corresponds to their increased tactile sensitivity in comparison to smaller areas of the cortex. For example, people can make muchfinersensorydiscriminationswiththeirfingertipsandlipsthanwhen the stimuli are presented to their bellies or legs. The discovery of somatosensory maps in the brain has led to the tendenc de ncy y in ma man ny di disc scip ipli line ness to as assu sume me th that at ma matt tter erss of th thee bod ody y ar aree be best st,, an and d only, understood in terms of brain states and neural activity. Consider research showing that different areas of the monkey’s brain (parietal cortex)  become activated when specific areas on the monkey’s hand have been touched (Merzenich et al., 1983). The internal tactile map discovered here here is cl clea earl rly y an anal alog ogou ouss in th that at co cont ntig iguo uous us ar area eass in th thee ha hand nd ar aree rep eprres esen ente ted d as contig con tiguou uouss ar areas eas in the mon monkey key’s ’s bra brain. in. The These se pr proce ocesse ssess ar aree pr presu esumab mably ly re re-late la ted d be beca caus usee th thee ne nerv rvou ouss sy syst stem em ph phys ysic ical ally ly co conn nnec ects ts th thee ha hand nd to th thee br brain ain.. But, in fact, there is no single bundle of neurons that link every conceivable spot on the monkey’s hand to a specific spot on the sensory map for the hand in the monkey’s brain. This research has only shown that if  you touch area X on the monkey’s palm, area Y in the money’s brain is activated. In this way way,, sensory maps only describe “what happens when”

 

Perception and Action

 

47

and not that “the hand is connected to the brain here” (Clancey,  1997). Showing the relation between parts of the body and parts of the brain does not tell us anything about “how processes involving nerve stimulation tio n an and d ne neur ural al ac activ tivat atio ion n ar aree oc occu curr rrin ing g wi with thin in a si sing ngle le,, co conn nnec ecte ted d ph phys ysic ical al system sys tem,, the mon monkey key’s ’s bod body” y” (Cl (Clanc ancey ey,, 1997: 83). Pi Pict ctur ures es of bo body dy pa part rtss an and d corresponding brain areas only portray a correlation of processes in one phys ph ysic ical ally ly clo close sed d sy syst stem em (t (the he mo monk nkey ey’s ’s ne nerv rvou ouss sy syst stem em)) an and d sh shou ould ld no nott be interpreted as showing that specific brain states alone contain sensory experi pe rien ence ce.. Th Thee ar area eass th that at ar aree la labe bele led d P1, P2, an and d so on on,, en enco comp mpas assi sing ng ne neur ural al patterns that categorize the activities of these areas, are not hardware that creates and sustain embodied activity, such as monkey hand sensations and movement. In rec ecen entt ye year ars, s, mo more re de deta taile iled d st stud udie ies, s, us usin ing g si sing ngle le-c -cel elll re reco cord rdin ings gs,, ha have ve show sh own n th that at in inst stea ead d of on onee ho homu munc nculu ulus, s, th ther eree ar aree fo four ur di dist stin inct ct so soma mato toto topi picc maps in primary sensory cortex (Nicolelis & Fanselow,  2002). These four maps differ in the kinds of stimuli that produce different responses (area 1, rapidly adapting skin receptors; area  2, deep pressure; area  3a, muscle strret st etch ch re rece cept ptor ors; s; an and d ar area ea 3 b, cutaneous stimulation of both transient and sustained sensations. Most importantly, this selectivity of different parts of the cortex is not fixed. Cells within any one map can respond to all the di dif ffe ferren entt ki kind nds s of st stim imul ulat atio ion n so th that at th the e en enti tirre ma map p ca can n bechanges dyna dy nami mica call lly y reorganized, depending on injury and experience. “These strongly suggest that normal somatosensory cortex is subject to territorial competition, to a self-organizing force that can alter its topography” (Merzenich et al., 1983: 50). None of this discussion, again, downplays the necessity of neural activity in the grounding of perceptual experience. Recent neuroimaging stud st udie iess of a wo woma man n bo born rn wi with thou outt le legs gs or fo forrea earm rmss de demo mons nstr trat ates es th that at bo body dy parts can be represented in sensory and motor cortices even when these parts have not been physically present (Brugger, Regard, & Shiffrar,  2000). Huma Hu man n br brai ains ns ha have ve cl clea earl rly y ev evol olve ved d to rep eprres esen entt bo body dy pa part rts. s. Yet we sh shou ould, ld, nonetheless, not assume a direct correspondence between a sensory/ neural map and perceptual experience. Perception is not simply a matter of mapping stimulations with brain states (Clancey,  1997). Moreover, perc pe rcep epti tion on is no nott ex expl plai aine ned d by so soma mato tose sens nsor ory y ac acti tivi viti ties es.. Th Thus us,, if an ir irri rita tant nt is injected into a nerve, this triggers a change in the activity of the nervous syst sy stem em.. Bu Butt mo most st sc scho hola lars rs wo woul uld d no nott ca call ll th this is ch chan ange ge in ac acti tivi vity ty a “p “per erce cepption ti on”” of th thee ir irri rita tant nt.. Pe Perrce cept ptio ion n ca can n on only ly be ex expl plai aine ned d wi with thin in th thee co cont ntex extt of   brain, body, body, and world interactions, and not by perturbations of neurons (Maturana, 1983). A different reason to question whether perception is merely a mapping of sensory stimulation onto neural activity comes from recent neuroscienneuroscientific re resea searc rch h on odo odorr per percep ceptio tion. n. The neu neuro rosci scient entist ist Walt alter er Fr Freem eeman an ask asked ed thee fo th foll llow owin ing g qu ques estio tions ns ab abou outt pe perrce cept ptua uall ex expe peri rien ence ce:: “W “Wit ithi hin n a fr frac actio tion n of 

 

48

 

Embodiment and Cognitive Science

seco cond nd af afte terr th thee ey eyes es,, no nose se,, ea ears rs,, to tong ngue ue,, or sk skin in is st stim imul ulat ated ed,, on onee kn know owss a se

the object is familiar and whether it is desirable or dangerous. How does such suc h re recog cognit nition ion,, whi which ch psy psycho cholog logist istss cal called led pr preat eatten tentiv tivee per percep ceptio tion, n, hap hap-pen so accurately or quickly, even when the stimulus are complex and the context in which they arise varies?” (Freeman,  1991: 78). Freeman arguess tha gue thatt the these se quic quick k per percep ceptio tions ns re reflec flectt sel self-o f-org rgani anized zed pr proce ocesse ssess inv involv olving ing the tendency of most collections coll ections of neurons throughout the cortex to shift quic qu ickl kly y fr from om on onee ac acti tivi vity ty ph phas asee to an anot othe herr in re resp spon onse se to ev even en sm smal alll in inpu puts ts.. This Th is se self lf-o -orrga gani nize zed, d, ch chao aoti ticc ac activ tivit ity y is no nott so sole lely ly,, an and d si simp mpli list stic ical ally ly,, a br brai ain n or neural process, but reflects the more global dynamics of interactions of   brain, body, body, and environment. environment. Freeman and associates applied this dynamical perspective to explain how trained rabbits recognize different smells (e.g., sawdust, banana) and  behave in differen differentt ways (e.g., chewing or licking). Using an EEG, they recorded large-scale ionic currents of the olfactory bulb and the olfactory cortex in responses to different smells, both during and after training. The EEG recordings revealed bursts of activity when the rabbit was inhaling. The bur burst st com compri prised sed syn synchr chrono onous us wav waves es of neu neural ral firi firings ngs at eac each h re recor cordin ding g site. However, However, the amplitude modulation (AM) varied in a particular way across the olfactory bulb, and not just within a small set of neurons, for each ea ch ty type pe of sm smeell ll.. Th Thee co con nto tour ur ma map p of am ampl plit itud udee wa wass al alwa ways ys th thee sam amee fo forr the same smell stimulus until the stimulus changed. When another smell was introduced, the contour maps of all the stimuli currently recognized changed to accommodate it, and the new pattern was stable until further smells were presented. For example, after the rabbit learned to recognize the banana smell, reexposure to sawdust led to the emergence of a new sawdust map. But each AM pattern is unique for each rabbit given its history, the shape of its body, the colors of its furs, and so on. AM patterns correspond to affordances by which the rabbit “in-forms” itself about possible bodily interactions with an odorant, such as whether it is something that can be eaten or something that should be feared (i.e., a predator). But the patterns are not representations of odorants, or signals thatspecifythepresenceoffoodordanger,becauseitisimpossibletomatch each AM with specific stimuli. The patterns are unique to the history of  the animal, because of its past experiences, which shaped the synaptic connections in the bulb’s neuropil. Freeman argued that the reentrant link  between the bulb and the cortex constitutes a coupled system in that the activity of the bulb is modulated by the activity of the cortex. Free Fr eema man n an and d co colle lleag ague ues’ s’ co comp mput uter er mo mode dell of th thee ol olfa fact ctor ory y sy syst stem em,, ba base sed d on “or “ordin dinary ary dif differ ferent ential ial equ equati ations ons,, des descri cribes besthe the dyn dynamic amicss of loc local al poo pools ls of  neur ne uron ons. s.”” Ex Expe peri rime ment ntss sh show owed ed th that at th thee mo mode dell de desc scri ribe bed d th thee EE EEG G da data ta,, in in-clud cl udin ing g bo both th th thee bu burs rsts ts or ne neur ural al ac acti tivi vity ty an and d th thee pr prop oper erti ties es of th thee co cont ntou ourr maps ma ps of th this is ac activ tivit ity y. Fr Free eema man n ar argu gued ed th that at th thee ra rapi pid d bu burs rsts ts of ne neur ural al ac acti tivit vity y alon al ong g th thee ra rabb bbit it’s ’s ol olfa fact ctor ory y sy syst stem em in re resp spon onse se to st stim imul ulii de demo mons nstr trat atee th that at

 

Perception and Action

 

49

the olfa fact ctor ory y sy syst stem em is a ch chao aoti ticc dy dyna nami mica call sy syst stem em.. Fo Forr ex exam ampl ple, e, th thee mo mode dell the ol predicted phase portraits of neural activity that most resembled strange attractors. Freeman speculated that perception in all sensory modalities involves repeated transitions between the strange attractors of the chaotic dynamical system of the brain, with each strange attractor representing a

particular meaningful stimulus. An act of perception is not the copying of an incoming stimulus. It is a step in a trajectory by which brains grow, recognize themselves, and reach into the environment to change to their own advantage” (1991: 85). Freeman’s olfactory model does not explicitly show sh ow ho how w od odor or rec ecog ogni niti tion on is ti tied ed to mo move veme ment nt.. Bu Butt hi hiss em empi piri rica call de demo monnstration shows that odor recognition, like all sensory processing, depends on a st strruc uctu tura rall co coup upli ling ng of pe perrce cept ptio ion n an and d ac acti tion on as pa part rt of th thee br brai ain, n, bo body dy,, and world dynamical interaction. AM patterns reflect an early stage in intention ten tional al beh behavi avior or,, or af affor fordan dances ces,, by whi which ch an ani animal mal inin-for forms ms its itself elf abo about ut whet wh ethe herr to ea eatt fo food od or ru run n fr from om a pr pred edat ator or gi give ven n a sp spec ecifi ificc od odor oran ant. t. Bu Butt th thee AM patterns are not representations of odorants, because of the impossi bility of matching these fluctuating, dynamic patterns with specific stimuli or with receptors that convey stimuli to the cortex. Instead, AM patterns are unique to an animal, arising out of past experiences, such as its movements (sniffing and licking), body shape, and color, which continuously shape and reshape how brains construct themselves. My ar argu gume ment nt ag agai ains nstt th thee re redu duct ctio ion n of pe perrce cept ptio ion n to ne neur ural al ac acti tivi vity ty do does es not at all imply that some areas of the brain may have evolved as part of  specialized perceptual systems. I only claim, following Freeman and others, that perception is not solely located in brain activity, activity, but must always  be situated in terms of more complex dynamic couplings involving the whole body in action. Moving the Body to Perceive

Perceptio Percep tion n can cannot not be und unders erstoo tood d wit withou houtt re refer ferenc encee to act action ion.. Peo People ple do not perceive the world statically, but by actively exploring the environment. For instance, if I move closer to the table in front of me, I see the textured line li ness in th thee wo wood od su surf rfac acee be bett tter er.. If I tu turn rn my he head ad,, I di dist stin inct ctly ly he hear ar th thee mu musi sicc playing softly on the stereo behind me. If I move over to the steaming cup of coffee on the counter, and lean over close by, I clearly smell the scent of coffee beans. Each bodily movement enables my sensory organs to do their work depending on my motivations and goals. Movement is essential to perception. As my eyes move, the structure of  the light around me (i.e., the optic array) changes and certain information previously unavailable (i.e., invariant information) now becomes present. The external structure is an ambient optic array that I manipulate when I move in it. When people merely touch an object, they understand little of  what is perceived unless they move their hands and explore its contours

 

50

 

Embodiment and Cognitive Science

textur turee (Gi (Gibso bson, n, 1962; St Ster eri, i, Sp Spel elke ke,, & Ra Rame meix ix,, 1993). By run unni ning ng th theeir and tex hands over “pictorially correct” raised surfaces, even the blind can understan st and d ma many ny sp spat atia iall re repr pres esen enta tati tion onss of de dept pth h (K (Ken enne nedy dy,, Ga Gabi bias as,, & Ni Nich chol olls ls,, 1991). Although our hands contain sensory transducers, the musculature with which we control movement allows us to explore objects in ways that make it easy to identify what is being felt. When we lift an object, this reveals something about its weight, rubbing our fingers across it tells us about its texture and overall shape, and squeezing it says something about its compressibility. To take a different example, without looking at

a rod, people can determine its length by wielding it (Turvey, Solomon, & Burton, 1989). The body movement alone generates sufficient perceptual information to specify the rod’s length. Indi In divi vidu dual alss po poss sses esss st stro rong ng as asso soci ciat atio ions ns be betw twee een n ob obje ject ctss an and d th thee ac actio tions ns comm co mmon only ly ca carr rrie ied d ou outt wi with th th them em (R (Ros osch ch et al al., ., 1976), wh whic ich h refl eflec ects ts th thee fa fact ct that th at an ob obje ject ct’s ’s fu func ncti tion on is in inti tima mate tely ly bo boun und d wi with th th thee ac acti tion onss we di dirrec ectt to it it.. Thee af Th affo ford rdan ance cess as asso soci ciat ated ed wi with th ob obje ject ctss ar aree so st stro rong ng th that at 3- and 4-year-old children may sometimes momentarily attempt to sit in toy chairs, get into toy to y ca cars rs,, or sl slid idee do down wn to toy y sl slid ides es,, de desp spite ite th thee fa fact ct th they ey ar aree ph phys ysic ical ally ly un unab able le to do so given their size (DeLoache, Uttal, & Rosengren, 2004). Objects are viewed as part of the same basic-level category to the extent they can be used for some interactional purposes (see Chapter  4 ). These associations are not restricted to high-level actions, such as writing with a pen, but are also apparent at the microscopic level, such as in the hand shape required to grasp a pen (Klatzky, Loomis, Lederman, Wake, & Fujita,  1993). People appl ap ply y va vari riou ouss ha hapt ptic ic te test stss to id iden entif tify y th thee di difffe ferren entt pr prop oper erti ties es of ob obje ject cts. s. On Onee study examined people’s ability to identify common objects on the basis of touch alone (Klatzky, Lederman, & Metzger,  1985). Blindfolded participants handled 100 common objects, each easily identifiable by name, such as toothbrush, paper clip, onion, fork, and screwdriver. Approximately 96% of the identifications were correct, and 94% of these occurred within 5 se seco cond ndss of ha hand ndlin ling g th thee ob obje ject ct.. Tou ouch ch de depe pend ndss qu quite ite ob obvi viou ousl sly y on aw awar areeness of the ways in which objects come into contact with one’s body and affect one’s body by giving rise to sensations. Indeed, a primary function of the cutaneous system is i s to employ stereotyped exploratory procedures (EPs), such as hand movements, to gain information about different kind of object properties (Lederman & Klatsky Klatsky,, 1990). An EP that applies pressurre to an ob su obje ject ct,, fo forr in inst stan ance ce,, mo most st ap appr prop opri riat atel ely y co conv nvey eyss ob obje ject ct ha harrdn dnes ess, s, whereas an EP employing lateral motion reveals aspects of object texture, and contour-following EP provides exact shape information. Explor Exp lorato atory ry pr proce ocedur dures es ar aree use useful ful in ide identi ntifyi fying ng obj object ectss eve even n whe when n the there re is no bo body dy co cont ntac actt be betw twee een n pe pers rson on an and d ob obje ject ctss (G (Gib ibso son, n, 1968). Fo Forr in inst stan ance ce,, we us usee st stic icks ks,, ra rake kes, s, sc scre rewd wdri rive vers rs,, ha hamm mmer ers, s, fis fishi hing ng rod ods, s, an and d te tenn nnis is ra rack ck-ets to act upon other objects. We We know that we are doing something to an obje ob ject ct th that at ha hass ca caus usal al co cons nseq eque uenc nces es ev even en th thou ough gh we ar aree no nott in di dire rect ct bo bodi dily ly

 

Perception and Action

 

51

contact with the object in the way, for instance, that my fingers touch the cofffe co feee mu mug g as I pu push sh it ac acrros osss th thee ta tabl ble. e. Em Empi piri rica call st stud udie iess de demo mons nstr trat atee th that at haptic exploration of objects using hand-held implements leads to reasonably accurate information about the objects (Bavac-Cikoja & Turvey,  1995; Lederman & Klatzky Klatz ky,, 2003). These studies do not imply simply that action is important for perception, but that there is no perception without action. Identifying People from Movement

People’s moving bodies are a key clue to our identification of them as people, both generally and specifically. An elegant demonstration of this is found conducted by ).ed In in one experiment, ligh li ghts ts we werrin e presearch lace la ced d at th the e ma majo jorr jo join ints tsJohansson of a per erso son n(1973 dreess dr ssed bla lac ck an and d pho ho-togr to grap aphe hed d in th thee da dark rk.. Vie iewi wing ng th thee li ligh ghts ts as st stat atio iona nary ry,, ob obse serv rver erss rep epor orte ted d

seeing seei ng on only ly ra rand ndom om ar arra rang ngem emen ents ts of do dots ts.. Ho Howe weve verr, if th thee pe pers rson on to wh whom om thee li th ligh ghts ts we werre at atta tach ched ed mo move ved d by wa walk lkin ing, g, ho hopp ppin ing, g, do doin ing g si sitt-up ups, s, or an any y other oth er fam familia iliarr act activi ivity ty,, obs observ ervers ers imm immedi ediate ately ly and unm unmist istake akenly nly saw a per per-son engaged in that activity. If the lights stopped moving, they returned to what appeared to be a random assemblage. Observers can detect the sex, se x, an and d ev even en th thee id iden enti tity ty,, of a pe pers rson on wa walk lkin ing g to wh whom om li ligh ghts ts ar aree at atta tach ched ed (Cutting, (Cuttin g, Prof Proffitt, fitt, & Kozlow Kozlowski, ski, 1978)  and distinguish between an actor walking normally and walking with a limp (Johansson, 1975). Even facial expressions, which involve elastic transformation, can be perceived from the movement of a few point-lights (Bassili, 1978). People are also better at identifying themselves from a point-light display pl ay of fig figur ures es wa walk lkin ing g th than an th they ey ar aree at rec ecog ogni nizi zing ng fr frie iend ndss an and d co colle lleag ague uess (Beardswo (Bear dsworth rth & Buckn Buckner er,, 1981). Pe Peop ople le ap appe pear ar to rec ecog ogni nize ze so some meth thin ing g th that at they were already familiar with, namely, how their own gait would look. This effect cannot be attributed to perceptual learning alone, as people see the gaits of friends and colleagues more than they do their own gait. A strong possibility is that the production of movement can generate a corresponding quasi-perceptual repr representation. esentation. In fact, perceptual judgments about human figures rely in part on the activation of one’s own body repre representation. sentation. Behavioral evidence demonstrates that the recognition of handedness of a visually presented hand depends on a covert recruitment of sensorimotor processes that are constrained by the neural structures controlling the side of the hand to be recognized (Parsons & Fox,  1998). Different studies have examined people’ pl e’ss ab abil ilit ity y to ju judg dgee th thee we weig ight ht of ob obje ject ctss be bein ing g li lift fted ed or ca carr rrie ied d by an anot othe herr person. People can extract weight information from videotapes (Valenti (Valenti & Costall,  1997), point light displays (Runson & Frykolm,  1983), and static photog pho tograp raphs hs (V (Vale alenti nti & Cos Costall tall,, 1997), an and d of ob obje ject ctss be bein ing g li lift fted ed or ca carr rrie ied. d. Onee po On poss ssib ible le ex expl plan anat ation ion fo forr th thes esee re resu sult ltss is th that at th thee pe perc rcep epti tion on of th thee mu musscular cul ar,, pos postur tural, al, and mov moveme ement nt cue cuess inv involv olved ed in lift lifting ing and car carryi rying ng re recr cruit uitss

 

52

 

Embodiment and Cognitive Science

stor ored ed kn know owle ledg dgee of ho how w on one’ e’ss bo body dy res espo pond ndss to lif lifti ting ng an and d ca carr rryi ying ng.. Th This is st recruited knowledge may then assist in perceptual judgments. Similarly, when wh en ap appa pare rent nt mo moti tion on is in indu duce ced d by a st stimu imulu luss pa patt tter ern n of tw two o se sequ quen entia tiall lly y pres pr esen ente ted d ob obje ject cts, s, th thee res esul ulti ting ng mo moti tion on fo foll llow owss th thee pr prin inci cipl plee of th thee sh shor orte test st path between two points. But when the two objects presented are human  body shapes, the resulting motion does not always follow the principle of  the shortest path (Heptulla-Chatterjee, Freyd, & Shiffrar, 1996; Shiffrar & Freyd, 1991). Instead, as if to automatically avoid impossible body movements, the motion takes longer paths and detours. Recent neuroimaging data da ta,, us usin ing g PE PET T, in indi dica cate tess mo moto torr sy syst stem em ac acti tivi vity ty du duri ring ng pe perrce cept ptio ion n of po posssibl si ble, e, bu butt no nott im impo poss ssib ible le,, hu huma man n mo move veme ment nt (S (Shi hiff ffra rarr & Pin Pinto to,, 2002). Mo More re-over, visual analysis of human movement differs from that of nonhuman movements, especially slower rates of display, and whenof the visual signal is connected withwith an observer’s internal representation possible human movement. Percei Per ceivin ving g oth other er peo people ple’s ’s act activi ivitie tiess oft often en act activa ivates tes the bra brain’ in’ss mot motor or sys sys-tem (Stevens et al.,  2000). Early studies showed shared mechanisms for action act ion per percep ceptio tion n and act action ion con contr trol ol in mon monkey keys. s. “Mi “Mirr rror or neu neuro rons” ns” in mon mon--

key ve key vent ntra rall pr prem emot otor or co cort rtex ex ar aree ac acti tive ve bo both th wh when en a mo monk nkey ey ob obse serv rves es a sp speecific action, such as someone grasping a food item, and when the monkey performs the same kind of action (DiPelligrino et al.,  1992; Gallese, 2000). Neurons in monkey premotor cortex discharge both when the animal performs a specific action and when it hears the corresponding action-related sound (Kohler, (Kohler, Keysers, Umilta, Fogassi, Gallese, & Rizzolatti, 2002). Other studies show that there are shared motor repr representations esentations for action, observation of another person’s actions, and imitation and mental simulation of action (Decety & Grezes,  1999; Rossi et al., 2002). Even the presumed goal of an observed action is recognized by observers and activates related motor cortices (Gallese et al.,  2002). A large meta-analysis revealed, for instance, that there are common activations sites in favor of  a functional equivalence between execution (e.g., sequential finger movement vs. rest), simulation (e.g., motor simulations of grasping objects vs. observing objects), and observation (e.g., observing grasping vs. observing in g ob obje ject ct). ). Th Thee mo most st ov over erla lap p in ac acti tiva vati tion on is ob obse serv rved ed in th thee su supp pple leme ment ntal al motor area, the dorsal premotor cortex, the supramarginal gyrus, and the superior parietal lobe) (Grezes & Decety, 2001). It is a mistake, however, to assume that perception of others in action is accomplished the activation shared motor representations in simply the brain. As noted inby Chapter  2, peopleofmay understand the actions of  othe ot hers rs by en enga gagi ging ng in a si simu mula lati tion on of th thee sa same me pr proc oces esse sess th that at ca can n pr prod oduc ucee the other person’s bodily actions. Beyond this possibility, there is clear evidence that our own physical actions, such as walking, running, coordinated finger action, and aspects of human speech, arise from the complex interactions of neural resources, bodily biomechanics, and external

 

Perception and Action

 

53

envi ronm nmen enta tall st stru ruct ctur ures es.. Pe Perc rcep epti tion on of ot othe herr pe peop ople le’s ’s bo bodi dily ly ac acti tion onss ma may y enviro also depend more on the dynamics of how certain movements are created from larger patterns of brain, body, and environmental interactions. For example, experiments in which participants perform a task using a handheld pendulum show a frequency of oscillation that matches the resonant frequency of the whole wrist-pendulum system (Kugler & Turvey,  1987). The same effect occurs in swinging a golf club or tennis racket or in rocking a car to get it out of the snow (Hatsopoulos & Warren,  1996). In each case, proprioceptive information (from a musculoskeletal system with an intrinsic spring-like dynamics) couples neural systems to bodily and/or environmental resources in a way that creates a larger autonomous dynami na mica call sy syst stem em.. Th Thes esee fin findi ding ngss do no nott su supp ppor ortt a tr trad adit itio iona nall vi view ew of bo bodi dily ly action that assumes a central representation of the movement, which then generates the behavior. Movement in Music Perception

Music perception is another domain in which perception is tightly associated with embodied movements, such as rhythmic gaits, breathing, and other locomotion phenomenon (Friberg & Sundberg,  1999;  Friberg, Sundberg, & Fryden,  2000;  Scruton,  1997). Shore and Repp (1995)  highlight the important fact that musical motion is first and foremost audible

human motion. They describe three levels of events awareness: the raw psychophysical perception of tone, the perception of abstract qualities of  tone apart from their source, and the apprehension of environmental ob jects that give rise to sound events. This last level is the “ecological level” of  perrce pe cept ptio ion n wh wher eree “t “the he li list sten ener er do does es no nott me merrel ely y he hear ar th thee so soun und d of a ga gall llop op-ing horse or bowing violin; rather rather,, the listener hears a horse galloping and a violinist bowing” (Shore and Repp,  1995:  59 ). In this ecological framework, the source of perceived musical movement, especially self-motion, is critical to a listener’s perceptual experiences, as is abundantly clear to listeners attending a live musical performance (Clarke, 2001). In this way, music perception involves an understanding of bodily motion – a kind of  empathetic embodied cognition. Behavioral studies demonstrate that musical structures have kinematic aspects that not only compel performers to modulate their tempo in specific ways, but also induces corresponding perceptual biases in musically trained listeners (Repp, 1998). Recent neurophysiological studies of music perception emphasized the major role of body motion music perception andhave production. Studies of brain-damaged patientsinwith lesions located in various regions of the brain show that the rhythmic component of an au audi dito tory ry im imag agee ca cann nnot ot be ac acti tiva vate ted d wi with thou outt rec ecru ruit itin ing g ne neur ural al sy syst stem emss known to be involved in motor activity, especially the planning of motor sequences (Carroll-Phelan & Hampson,  1996). These neuropsychological

 

54

 

Embodiment and Cognitive Science

data have allowed hypotheses about the induction of a sense of beat or pulse in terms of the so-called sensorimotor loop, which includes i ncludes the posterior parietal lobe, premotor cortex, cerebrocerebellum, and basal ganglia, Under this perspective, a perceived rhythm is literally an imagined move mo veme ment nt,, ev even en if th thee mu musc scul ulos oske kele leta tall sy syst stem em it itse self lf do does es no nott mo move ve (T (Tod odd, d, 1999). Thus, the act of listening to rhythmic music involves the same basic processes that generate bodily motion. Perceiving Causality from Movement

All physical movement is caused by ensembles of forces in action. The visual perception of object motion is strongly influenced by this causality. causality. Research on the visual perception of causality was initiated by the classic studies of Michotte (1963). Michotte examined whether and how people interpretthecausalityofobjectmotionbyaskingsubjectstodescribesimple films. Several of his studies focused on the interpretation of collisions. For exa exampl mple, e, Mic Michot hotte te pr propo oposed sed tha thatt peo people ple dir direct ectly ly per percei ceive ve “la “launc unchin hing” g” when one moving object contacts a second stationary object that is set in moti mo tion on af afte terr a br brie ieff de dela lay y. Pa Part rtic icip ipan ants ts in inva vari riab ably ly ex expe pect ct in th thes esee ca case sess th that at the motion of X caused the subsequent motion of Y. More recent research indica ind icates tes tha thatt obs observ ervers ers can mak makee fine dis discri crimin minati ations ons bet betwee ween n nor normal mal and physic phy sicall ally y imp imposs ossibl iblee col collis lision ionss (Ka (Kaise iserr & Pr Profit ofitt, t, 1987). Alt Althou hough gh Mic Michot hotte te argued that the perception of causality did not depend on experience, subsequent studies have suggested the opposite (Kaiser & Proffitt,  1984). For instance, practically all of the studies on perceiving causality assume that the visual system is the source of these perceptual judgments. But

people peop le’s ’s ex expe peri rien ence cess of th thei eirr ow own n bo bodi dies es in ac acti tion on,, mo movi ving ng ag agai ains nstt ob obje ject cts, s, aree su ar surrel ely y a cr cruc ucia iall fa fact ctor or as we well ll.. Mi Mich chot otte te as assu sume med d th that at di dirrec ectt pe perrce cept ptio ion n of ca caus usal alit ity y co coul uld d occ ccur ur in a si sing ngle le ev even ent, t, su such ch as in th thee col olli lisi sion on of bi bill llia iarrd  balls, without the mediation of specific prior experience. In fact, Immanuel Kant (1787/1929) strongly argued that infants and young children could not learn causal relations without having some prior notion of causality. But embodied activity, including both the effect of whole bodies moving against one another and physical objects, and smaller events such as the effect of moving one’s tongue and lips on one’s mother’s breast, may be fundamental to perceiving causality from movement (see Chapter 7). Of course, we learn to reason causally through vision as well, such as when we see event 2 following event 1, and no event  2  is ever witnessed without event  1  first being seen. Yet vision is not the only sensory input into theperception acquisition(White, of causation, as causality also fundamentally haptic  1999). Consider how I use my hands torests pushona mug across the surface of a table. Haptic perception teaches me about the proportions of the mug, such as its height, its hardness, its texture, and something of its size and shape. Understanding these specific properties

 

Perception and Action

 

55

of the mug is not critical, however, as my primary understanding of the even ev entt is my pu push shin ing g th thee mu mug. g. My un unde ders rsta tand ndin ing g of th this is fa fact ct ar aris ises es fr from om th thee inte in tera ract ctio ion n of th thee mo moto torr an and d se sens nsor ory y qu qual alit itie iess of my pu push shin ing g ac acti tion on.. I ma may y  be mistaken about the height, weight, and volume of the mug, as well as even whether it is a mug, but I clearly recognize that I am the one pushing the object across the table. The correlation between my pushing and the information given by the haptic system enables me to state without doubt that I am pushing the mug. In this way, action upon objects haptically perceived meets all the conditions for causal realism (White,  1999). My  judgment of a causal relation between my action and the mug moving is not just due to a spatio-temporal link between the action and the mug’s eventual behavior. Yet I know that I am the cause of my mug’s behavior, a fact that I understand directly from my perceptual/embodied experience. Our ability to make causal inferences about the interaction of people with each other and objects does not depend on our understanding of an abstract rule or concept for causality. Instead, we draw causal inferences from our direct embodied experience with other people and objects, and perceive causality, even when mistaken, by applying our own embodied experience of how we act upon objects. We feel our hair move because the wind blows, we lose our balance because somebody forcefully knocks against us, and we feel our knees buckle slightly because a heavy object is placed upon our shoulders. In each case, we experience the mechanical action of other things acting upon us because of haptic perception, in the same sa me wa way y th thaat we kn kno ow th that at ou ourr mo move veme ment ntss ar aree th thee ca caus usee of ot oth her ob obje jeccts moving when we act upon them. Children acquire causal knowledge through their repeated direct perception of causal relations as they interact with objects and other people (see Chapter 7). Over time, this knowledge of causal relations can be used to interpret other events in which causal relations cannot be directly perceived through by our embodied experience. For instance, imagine that I

see another person pushing a coffee mug across a table (White, 1999). My perc pe rcep epti tion on of th thee pe pers rson on’s ’s ac acti tion onss is di diff ffer eren entt fr from om wh what at he or sh shee pe perc rcei eive vess haptically, because I do not know his or her embodied experience in the way wa y it is be bein ing g pe perc rcei eive ved. d. De Desp spit itee th this is in inco comp mple lete te in info form rmat atio ion, n, I st still ill in infe ferr a causal relation between the person’s action and the movement of the mug  based on my knowledge of causal relations in my own body movements when wh en ap appl plie ied d to ot othe herr pe peop ople le an and d ob obje ject cts. s. I dr draw aw th thee ca caus usal al in infe fere renc ncee ab abou outt the event I witness because of my own understanding of embodied action and its effects in the world. Perceiving Movement in Static Patterns

Even when viewing static visual objects or patterns, people tacitly recognize ni ze th thee prese sen nce of mo move veme men nt, at le leas astt so some me of whi hich ch is ti tied ed to th thee hu huma man n

 

56

 

Embodiment and Cognitive Science

 body in action. Studies reveal that people infer dynamic information about movement when perceiving static shapes, such as when reading handwriting and viewing pictures of the human body (Babcock & Freyd, 1988;   1995

  1988

Freyd Pantzer, and Viviari ;   Freyd,(2000 Pantzer, & Cheng, Formiddle example, Kandel,&Orliaguet, ) showed participants).the letter of a three-letter word as it was handwritten. The participants’ task was to guess which of two letters would be the third letter. People were most accurate when the trajectories of the cursive handwriting were consistent with wi th th thee mo move veme ment ntss us used ed to wr writ itee th thee wo word rds. s. Th Thus us,, pe peop ople le pe perrce ceiv ived ed ha hand nd-writing displays based on the gestures that produced them, and not just the static features of which letters are most likely to follow one another. Other studies show that the velocity of a moving dot seems to be uniform if (and only if) it actually follows the law governing movement production (Viviani & Stucchi, 1992a b; , b; Viviani, Baud-Bovy & Redolfi, 1997). This suggests, once more, that production-related knowledge is implicitly involved in perceptual processing, as well as in perceiving causality. causality. Another Anoth er insta instance nce of peop people le infer inferring ring movement from static images is seen in the phenomenon of representational momentum (see Chapter  5). Our memory for the spatial location of an object is biased in the direction of the object’s motion, even when the object is presented statically (Freyd, 1983). Fo Forr exa xamp mple le,, wh wheen pa part rtic icip ipan ants ts vi vieew a pi pict ctur uree of a ma man n ju jump mpin ing g off  a wall and are asked to remember the man’s position, their memory for his position is systematically biased forward on the trajectory of his jump (Freyd & Finke,  1984). Thus, our memory for the location of a moving object, such as a person jumping, depends upon the spatio-temporal characte ac teri rist stic icss of th thee mo move veme ment nt th that at ca caus used ed th thee ob obje ject ct to oc occu cupy py th that at pa part rtic icula ularr location locati on (Fr (Freyd, eyd, 1987). Stu tudi dies es al also so sh show ow th that at see eein ing g an ob obje ject ct in mo moti tion on fa fa-cilit ci litat ates es pe peop ople le’s ’s re reco cogn gnit itio ion n of ne new w ob obje ject ctss wh when en th thes esee ar aree or orie ient nted ed wi with thin in the path of the first object’s motion path (Kourtzi & Shiffrar, 1997). Moreover, parts of the same cortical areas involved in motion perception (e.g., medial temporal/medial superior temporal cortex) are activated during perc pe rcep epti tion on of ph phot otos os re repr pres esen enti ting ng im impl plie ied d mo moti tion on,, bu butt no nott im imag ages es th that at di did d not imply motion moti on (Kourtzi & Kanwisher Kanwisher,, 2000). Our assumptions about the ways objects interact, and our bodily ex-

periences of moving against objects, may also influence our perception of  difffe di ferren entt ob obje ject ct pr prop oper erti ties es.. Fo Forr in inst stan ance ce,, th thee de dent nt in a ne new w ca carr is so some meth thin ing g curved that calls attention to a force that may have acted on the object  by virtue of its asymmetric shape. We do not see the dent in the car door as a static perceptual feature, but infer movement of the door’s surface when it was contacted by some person or thing. Leyton ( 1992) outlined a grammar of the forces likely to have acted on an object to create different shapes shap es.. Pe Perc rcep epti tion on of as asym ymme metr trie iess an and d po poss ssib ible le ca caus usal al fo forc rces es tie tiess in ni nice cely ly with wi th th thee cl claaim th that at per erce cep pti tio on in inv vol olve vess no nott on only ly wh what at mi mig ght be do done ne to an object by acting upon it in a certain way, but what must have happened

 

Perception and Action

 

57

to an object to create the particular shape it has. Some of the ways people infe in ferr ca caus usal al in influ fluen ence cess on st stat atic ic ob obje ject ctss mu must st be du duee to th thei eirr ow own n pr prev evio ious us,, and anticipated future, actions against objects and the effects that occur thereafter. Are Perception and Action Separate Activities?

Psychologists traditionally assume that perception and action represent two different parts of the information processing system. Part of the evidence for the separation of perception and action comes from studies looking at dissociations of visual cognition and visually guided action (Stoerig & Cowey, 1997). For example, lesioned monkeys and human patientscanperformvisuallyguidedactionswithoutvisualawareness(blindsight) (Humphrey,  1974;  Weiskrantz,  1980). Blindsight is most easily observed in people who have patches of dead tissue in the primary visual cortex, creating a blind spot or scotoma. Although people with scotomas claim to have no visual awareness in their blind spot, they can report with almost   100% accuracy the direction an object moves across it, and can distinguish between horizontal and vertical lines. These participants typically suggest that they were just guessing, and are quite surprised when informed that their guesses were correct (Weiskrantz (Weiskrantz et al., al ., 1974). Blindsight effects have been noted in laboratory studies of people with normal nor mal vis vision ion (He (Heywo ywood od & Ken Kentri tridge dge,, 2000). Eye Eye-mo -movem vement ent stu studie diess sho show w that people with normal brain functions do not perceive a gap in a target duri du ring ng sa sacc ccad ades es,, bu butt co coul uld d st stil illl po poin intt to th thee ne new w lo loca cati tion on of th thee sa same me ta tarrge gett (Bridg (Br idgema eman n et al. al.,, 1979). Dif Differ ferent entstu studie diess exa examin mining ing per percep ceptio tion n of ill illusi usions ons indicate that the induced motion illusion (Bridgeman, Kirch, & Sperling, 1981)  and Tichner’s circles illusion (Aglioti, Goodale, & DeSouza,  1995) affect verbal reports, but not pointing responses. A similar dissociation is foun fo und d wi with th st stat atio iona nary ry st stimu imuli li,, kn know own n as th thee Ro Roel elof ofss ef effe fect ct,, in wh which ich pa part rtic ic-ipants are shown a target with a frame in complete darkness (Bridgeman, Peery, & Anand, 1997). Target Target and frame could be shifted asymmetrically to the left or right, and people often misperceive the target as going in the direction opposite the surroundings after it has been shifted. When there is no delay between stimulus exposure and the cue to either make a ver bal response about the target’s location or point to where the target has  been, all ten participants evidenced the illusion in their verbal responses,  but five did not do so when they gave pointing responses. In the  4- and

8 second delay conditions, however, however, this dissociation was not found, sug

gesting that the motor system has a very limited li mited short-term memory. memory. One poss po ssib ibil ility ity is th that at th thee tw two o gr grou oups ps of pa part rtic icip ipan ants ts (i (i.e .e., ., po poin inti ting ng vs vs.. ve verb rbal al re re-port po rt)) di did d no nott ne nece cess ssar arily ily fo foll llow ow di diff ffer eren entt ps psyc ycho holo logi gica call la laws ws,, bu butt sw switc itche hed d from motor to cognitive modes at different delays after stimulus offset (Bridgeman, 2000).

 

58

 

Embodiment and Cognitive Science

Finally, other studies report a dissociation between perceptual and visuomotor processing for stimuli more distant than those studied in the laboratory (Proffitt, Creem, & Zosh,  2001). Participants judged the inclination or steepness of hills, both out of doors and in a simulated virtual environment. The angle judgment was obtained with three response measures – verbal estimates, adjustments of aboard representation of thehand. hill’s cross section, and haptic adjustments of a tilt with an unseen The first two measures yielded large overestimates of hill incline, whereas the latter judgments were close to veridical. These data suggest that there may ma y be tw two o pa path thwa ways ys in th thee vi visu sual al sy syst stem em,, ea each ch of wh whic ich h ac acce cess sses es di difffe ferren entt internal maps of visual space. Milner and Goodale (1995;  following Ungerleider and Miskin,  1982) claimed that these assorted neuropsychological and experimental findings point to a functional distinction between major parts of the visual system, which have subsequently come to be dubbed the “what” and “where” systems. The ventral pathway, projecting from the primary visual cortex to the inferotemporal lobe, is referred to as the “what” system  because of its concern with object recognition or pattern discrimination. The dorsal pathway, from the primary visual cortex to the posterior parietal cortex, in contrast, is referr referred ed to as the “where” system, because of its invo in volv lvem emen entt in pr proc oces essi sing ng sp spat atia iall pl plac acem emen entt of ob obje ject cts. s. Th Thes esee tw two o sy syst stem emss are also described as the “cognitive” and “sensorimotor” visual systems (Paillard, 1987). Many scholars, however, take issue with the supposed distinction between the “cognitive” and “sensorimotor” visual systems and present evidence in favor of an integrated view of perception and action. Under this view vi ew,, pe perrce cept ptio ion n an and d ac acti tion on ma may y be tw two o as aspe pect ctss of on onee an and d th thee sa same me ne neur ural al and psychological process process (M¨ (Moller, ¨  1999). For example, participants in one study had objects placed in front of them which they had to pick up and place on a piece of paper (Creem & Proffitt,  2001). They did this alone, or while engaged in a ants semantic or spatial imagery task ta sk.. Wh When en pa part rtic icip ipan ts we werre(pair-associated aske as ked d to pi pick ck upmemory) obje ob ject ctss wi with thou out t a se seco cond ndar ary y task ta sk,, th they ey mo most st of ofte ten n rea each ched ed ar arou ound nd to pi pick ck th thee ob obje ject ctss up by th thei eirr ha hand ndle less in a ma mann nner er ap appr prop opri riat atee fo forr us usee of th thee ob obje ject ct.. Ho Howe weve verr, wh when en th thee co cogn gnit itiv ivee syst sy stem em wa wass ta taxe xed d by a co conc ncur urre rent nt se sema mant ntic ic ta task sk,, pa part rtic icip ipan ants ts ra rare rely ly pi pick cked ed up the objects appropriately. Yet there was little decrease in performance when whe n the spa spatial tialima imager gery y tas task k was per perfor formed med con concur curre rentl ntly y. A sec second ond stu study dy showed that when participants were presented with a purely visuospatial task (i.e., tracking a moving dot), the concurrent spatial imagery task interfered with grasping performance, but not the semantic task. In I n general, without the influence of the cognitive system, the visuomotor system cannot reach and grasp an object effectively. effectively. At least partial information from thee se th sema mant ntic ic sy syst stem em is ne need eded ed to gr gras asp p an ob obje ject ct ap appr prop opri riat atel ely y in a ma mann nner er

defined by its functional identity (e.g., grasping a spatula). These findings

 

Perception and Action

 

59

illustrate a necessary interaction between visual cognition and visually guided embodied action. Other researchers have obtained significant effects of perceptual illusions on action, contrary to some earlier work (Franz, 2001). Thus, there is no di difffe ferren ence ce in th thee si size ze of th thee pe perrce cept ptua uall an and d gr gras asp p il illu lusi sion onss if th thee pe perrce cepptual tu al an and d gr gras aspi ping ng ta task skss ar aree ap appr prop opri riat atel ely y ma matc tche hed d (F (Fra ranz nz et al al., ., 2001). One possible resolution of perceptual illusions affect visu vi sual ally ly gu guid ided ed ac acti tion onthe is todebate be mo morrover e prewhether cisse ab ci abou out t th thee lo loccus of thee il th illu lusi sion onss within the brain (Milner & Dyde, 2003). Unless the illusion really operates deep within the ventral stream (i.e., the “what” system), it is likely to influence activity in both the dorsal and ventral streams, thus not providing the best test of a perception and visuomotor dissociation. Some other compelling evidence against a “what” and “where” distinction comes from studies with brain-damaged patients. For instance, patients with damage to the human homologue of the dorsal stream have diffi di fficu cult lty y re reac achi hing ng in th thee co corr rrec ectt di dire rect ctio ion n fo forr ob obje ject ctss pl plac aced ed in di diff ffer eren entt po po-siti si tion onss in th thee vi visu sual al fie field ld co cont ntra rala late tera rall to th thei eirr le lesi sion ons. s. Pa Patie tient ntss wi with th da dama mage ge to this region of the cortex often show an inability to rotate their hands or open their fingers properly to grasp objects placed in front of them. But thes th esee pa patie tient ntss ca can n de desc scri ribe be th thee or orie ient ntat atio ion, n, si size ze,, sh shap ape, e, an and d re rela lativ tivee sp spat atia iall location of the very objects they are unable to grasp correctly (Goodale, et al. l.,, 1998). Pa Pati tien ents ts wi with th da dama mage ge to th thee ve vent ntra rall st strrea eam m sh show ow th thee op oppo posi site te pattern of deficits (Goodale et al., 1998). Good Go odal alee an and d Mu Murp rphy hy (2000) su sugg gges estt th that at on onee sh shou ould ld no nott th thin ink k ab abou outt th thee dorsal stream as a system for spatial vision per se, but rather as a system for the visual control of skilled action. Both streams process information about the orientation, size, and shape of objects and about their spatial relations. Both streams are also subject to modulation by attention. But each ea ch st strrea eam m de deal alss wi with th th thee in inco comi ming ng vi visu sual al in info form rmat atio ion n in a di difffe ferren entt wa way y. The ventral stream transforms visual information into perceptual representations that embody the enduring characteristics of objects and their spatial relations withwhich each other. visual transformations carried out in the dorsal stream, util ize The utilize moment-to-moment information about the disposition of objects within egocentric frames of references, mediate the control of goal-directed acts. Althou Alt hough gh vis visual ually ly con contr troll olled ed act action ion inv involv olves es neu neural ral pat pathwa hways ys dif differ ferent ent from those underlying explicit perceptual judgments, leading to dissociations between perceptual and motor performance, perception and action may be similar to the extent that the tasks used to assess them depend on the same visual information (Smeets & Brenner, Brenner, 1995; Vishton et al., 1999). For ins instan tance, ce, inf inform ormati ation on use used d in per percep ceptua tuall jud judgme gments nts of sel self-m f-moti otion on (i. (i.e., e., retinal flow) is also used to control steering with a joystick (Li & Warren, 2002). Thus, perception and action are not different entities, but are two aspects of behavioral control (Kotchoubey (Kotchoubey,, 2001). The dorsal system may

 

60

 

Embodiment and Cognitive Science

 be responsible for the fast modulation by optical information of action such as reaching and grasping (Green, 2001). The ventral system does not support perception, as distinct from action, but controls extended actions that unfold over longer time scales, thus drawing on optical information over larger spatial scales than simple, fast limb movements. Perceptual scientists continue to debate the merits of the two-visualsyst sy stem emss hy hypo poth thes esis is.. Bu Butt it se seem emss cl clea earr th that at th thee or orig igin inal al ve vers rsio ion n su sugg gges esti ting ng a sharp separation of visual perception and visually guided action has  been much weakened in recent years. One new proposal greatly diminishes the distinction between visual representation and the control of actions (Ellis & Tucker,  2000). According to this idea, the representation of  a visua visuall objec objectt includ includes es not only description description of its visual pro propert perties, ies, but also encodings of actions relevant to that object. For instance, a mental representation repr esentation of a wine glass is, at least in part, constituted by its making available reaching and grasping, and all the other things one has learned to do with a wine glass. Resear Res earch ch sup suppor portin ting g thi thiss the theory ory com comes es fr from om stu studie diess on sti stimul mulususresponse compatibility, compatibility, where stimulus properties of objects quickly give rise to response codes (Simon, 1969, 1996). The Simon effect, for instance, invo in volv lves es res espo pons nses es th that at ar aree ma made de mo morre qu quic ickl kly y wh when en th thee lo loca cati tion onss of st stim im-ulus ul us an and d res espo pons nsee co corr rres espo pond nd th than an wh when en th they ey do no not, t, de desp spit itee th thee fa fact ct th that at the location of the stimulus is not relevant to the task (Simon & Ruddell, 1967). For example, suppose that a left response is to be made to the letter H and a right response to th thee letter S. An H displayed to th thee left of fix fixaation yields faster responses than the same stimulus displayed to the right of  fixation. Many studies have explored S-R compatibility in regard to the impact of perception on action. Consider the following study (Brass et al., 2000). Participants were asked either to learn to respond differentially to nons no nspa patia tiall cu cues es (e (e.g .g., ., a 1 or 2) or to res espo pon nd by im imit itaati tin ng on onee of tw two o fin finge gerrmovement stimulus cues. If the participants were presented with a 1, they werre as we aske ked d to res espo pond nd wi with th th thei eirr in inde dex x fin finge gerr, an and d a 2 cu cued ed a re resp spon onse se wi with th the middle finger. finger. In another condition, when participants saw a movie of  an finger, were askedSeeing to move their of index finger, finger and similarly for index a movie of a they middle finger. movies actions to, imitate as the stim st imul ulus us sp sped ed up rea eact ctio ion n ti time mes, s, as ex expe pect cted ed if pe perrce cept ptio ion n an and d pr prod oduc ucti tion on of ac acti tion onss ar aree li link nked ed.. Mo Morreo eove verr, be bein ing g as aske ked d to res espo pond nd to th thee mo movi viee of th thee index finger by moving one’s own middle finger slowed down reaction times, an interference effect caused by stimulus-response incompatibility. incompatibility. A different experiment showed that the more a stimulus was similar to a requ re quir ired ed ac actio tion, n, th thee fa fast ster er pe peop ople le res espo pond nded ed.. Th Thes esee da data ta ill illus ustr trat atee th thee ti tigh ghtt influence that perception has in forming motor movements. Other studies show that even task-irrelevant spatial location information elicits a congruent spatial response code (Eimer, 1995). But location is not the only feature of a visual object that elicits action-related response

 

Perception and Action

 

61

prop oper erti ties es,, ir irrrel elev evan antt of a go goal al.. Ac Acti tion on-r -rel elat ated ed fe feat atur ures es,, su such ch as si size ze,, sh shap ape, e, pr and an d or orie ient ntat atio ion, n, ha have ve si simil milar ar ef effe fect cts. s. Tuc ucke kerr an and d El Elli liss (1998) pre presente sented d pictures of objects to participants who judged whether they were shown in a normal or inverted vertical orientation. The participants responded accordingly by pressing a left or right key. The objects were also depicted in onee of tw on two o ho hori rizo zont ntal al or orie ient ntat atio ions ns,, di diff ffer erin ing g in te term rmss of wh whic ich h ha hand nd wo woul uld d  be optimal to use in reaching and grasping the object in that orientation. For instance, a teapot could be presented with its handle to the viewer’s left or right. Despite the irrelevance of horizontal orientation to the task, this variable influenced participants’ key press responses. If the hand of  response was the same as the optimal hand for reaching and grasping, implicit in the horizontal orientation, participants were faster and more accurate, compared to the incongruent case. Further studies on the relevance of visual perception on motor programs investigated interference effects effects of distractors on a grasping movement (Castiello,  1996). In this study, distractors were task-relevant for a secondary, nonspatial task, but they interfered with the kinematics of the main grasping task. A different study examined whether a nonrelevant prime picture influenced the latency of a subsequent grasping movement (Craighero, Fadiga, Rizzolatti, & Umilta,  1999). There was a reduction in the grasping latency when the prime picture depicted the to-be-grasped object, but not when the prime depicted a different object. Thus, visual perception of an object affects the programming of a movement that immediately following the perception. These results have been interpreted as supporting the “premotor theory of attention” in which spatial attention is constrained by motor processes (e.g., saccadic eye movements, arm movements) (Rizzolatti, Riggio, & Sheliga, 1994). Many Ma ny st stud udie ies, s, th then en,, de demo mons nstr trat atee th that at se seei eing ng an ob obje ject ct af affo ford rdss ac actio tions ns as as-sociat soc iated ed wit with h it. Sma Smalle llerr obj object ectss wit within hin arm arm’s ’s re reach ach af affor ford d gra graspi sping, ng, or mor moree spec sp ecifi ifica call lly y, a pa part rtic icul ular ar ki kind nd of gr gras asp. p. Fe Feat atur ures es of an ob obje ject ct su such ch as it itss lo loca ca-tion, shape, and orientation will lead to activation of specific components of reaching and grasping actions. Particular directions of reach, particular hand shapes, and particular hands will be activated by the sight of an ob ject within reach. These potentiated components of a grasping response are referred to as “micro-affordances” (Ellis & Tucker, 2000). Ellis and Tucker characterize these behavioral possibilities as dispositional properties of  an observer’s nervous system (also see Shepard, Shepard, 1984). Under the “microaffo af ford rdan ance ce”” vi view ew,, gr gras aspi ping ng in ge gene nera rall is no nott fa faci cilit litat ated ed by an ob obje ject ct (s (suc uch h as in Gibson’s affordances), but a specific grasp appropriate to that object, in cont co ntex ext, t, su such ch as a pa part rtic icul ular ar sh shap apee of th thee ha hand nd,, an and d a pa part rtic icul ular ar or orie ient ntat atio ion n of the wrist, and so on, is facilitated. Another new theory has been proposed for understanding the links  between perception and action or motor planning (Hommel et al., 2001). The “theory of event coding” (TEC) holds that cognitive representations

 

62

 

Embodiment and Cognitive Science

eventss (i.e., of any to-be to-be-per -perceive ceived-ind-in-the-w the-world orld incident in the distal of event environment) subserve not only representational functions (e.g., for perception, imagery, imagery, memory memory,, and reasoning) but action-related functions as

well (e.g., action planning and initiation). TEC claims that perceiving and action planning are functionally equivalent, because they are alternative ways wa ys of in inte tern rnal ally ly re repr pres esen enti ting ng ex exte tern rnal al ev even ents ts (o (orr mo morre pr prec ecis isel ely y, in inte tera racction tio n be betw twee een n th thee ev even ents ts an and d th thee pe perc rcei eive ver/ r/ac acto tor) r).. Pe Perrce ceiv ivin ing g th thee wo worl rld d is a proce pr ocess ss of act actual ually ly acq acquir uiring ing inf inform ormati ation on abo about ut the per percei ceiver ver-en -envir vironm onment ent rela re latio tions nshi hip, p, in incl clud udin ing g mo move veme ment ntss of ey eyes es,, ha hand nds, s, fe feet et,, an and d bo body dy.. Th Thee pr proocess of perceiving both presupposes and affords active behavior and performing an action both relies on and provides perceptual information. In this sense, perceptions, or stimulus codes, and actions, or response codes,  both represent represent the results results of, and the stimulus for, for, particular sensorimotor coordinations. TEC TE C ar argu gues es th that at cr cros osss-ta talk lk be betw twee een n pe perrce cept ptio ion n an and d ac acti tion on oc occu curs rs at tw two o levels – compensation and adaptation. Compensation refers to the fact that in order to interpret any change in the spatial distribution of signals at three receptive surfaces, animals must have a way to compensate for their own body movements. Thus, the system has to take into account the animal’s body movements before it can use the sensory signal to recover thee st th stru ruct ctur uree of th thee en envi viro ronm nmen enta tall la layo yout ut (B (Bri ridg dgem eman an,, 1983; Epstein, 1973; Shebilske, 1977). Adaptation refers to the flexibility of sensorimotor couplin pl ings gs an and d to th thee fa fact ct th that at pe perrce cept ptio ion n ca can n wi with thin in ce cert rtai ain n li limi mits ts be ed educ ucat ated ed  by action planning. For instance, studies of distorted vision have demonstrated that perception may teach action and action may teach perception at th thee sa same me ti time me (R (Red eddi ding ng & Wal alla lace ce,, 1997; Welch, 1978), ag again ain su sugg gges estin ting g that commensurate or identical representations underlie both perception and action (Van der Heijden, Mussler, & Bridgeman, 1999; Wolff, 1999). TEC’ TE C’ss co core re co conc ncep eptt is th thee “e “eve vent nt co code de”” th that at re repr pres esen ents ts th thee di dist stal al fe feat atur ures es of an event. These distal features are not specific to a particular stimulus or response, but register sensory input for various sensory systems and modulate the activation of different motor systems. Thus, distal feature codes refer to not solely to single dimensions of color, size or shape, but to compl complex ex embod embodied ied poss possibiliti ibilities, es, such as “sit“sit-on-ab on-ablenes leness.” s.” Even time and change might be repre represented sented by feature codes, so that events such as a “leftward motion” can be coded. Feature codes are not simply given but evolve evo lve and cha change nge tho though ugh the per percei ceiver ver’s/ ’s/act actor’ or’ss exp experi erienc ence. e. For ins instan tance, ce, a particular action may not always be coded as “left” or “right,” but will  be understood as, say say,, “left-of-body “left-of-body,” ,” “left-of-right index finger finger,” ,” and “lefthanded.” Cons Co nsid ider er,, fo forr ex exam ampl ple, e, a pe pers rson on rea each chin ing g fo forr a bo bott ttle le st stan andi ding ng in fr fron ontt of  him hi m or he herr. On Onee of th thee ma many ny po poss ssib ible le way ayss to an anal alyz yzee th this is sit itu uat atio ion n wo woul uld d  be to conceive of the bottle as stimulus and of the reaching movement as a suitable response. A successful response clearly requires that several

 

Perception and Action

 

63

features of stimulus and action planning match. For instance, the internal distance of the hand should be identical to the perceived hand-bottle distance; the internal grasp should be identical to the bottle’s perceived location. This matching task is easy because stimuli and to-be-performed responses share a large number of features. After all, action codes are already activated in the course of perceiving the stimulus. Various evidence supports the predictions of the TEC. First, neurologi-

cal fin cal findi ding ngss on mi mirr rror or-n -neu eurron onss su sugg gges estt a po popu pula lati tion on of ne neur uron onss th that at fu fulfi lfill ll  both perceptual and action-planning functions (Rizzolati et al., 1990; see Chapters 5 and 6). These neurons may provide the neuroanatomical substrate of the common codes assumed for stimulus perception and action response. Behavioral evidence consistent with TEC comes from various sources, incl in clud udin ing g a du dual al ta task sk ex expe peri rime ment nt in wh whic ich h pa part rtic icip ipan ants ts br brie iefly fly se seee a ma mark rked ed arrow while performing an already prepared, prepared, unspecific left or right keypress pre ss (Homm (Hommel, el, 1998). Aft After er the key keypr press ess,, par partic ticipa ipants nts jud judged ged the dir direct ection ion of the arr arrow ow,, which randomly randomly point pointed ed to the left or right side. Planning Planning and performing a left- or right-hand keypress requires requires integrating a LEFT or RIGHT code, respectfully, into the corresponding action plan. If so, this interpreted code should be most available for processing and coding a LEFT or RIGHT arrow, so that people should be effectively blind to the arrow that points to the same side as a response. Consequently, left-pointing key presses in Thu the s, way thate rightrig ht-poi pointi nting ngarrows arrows arr owsappear appear app earduring during dur ingleft-hand rightrig ht-han hand d key press pr esses. es. Thus, featur fea ture overlap between codes plan and stimulus should improve stroke processing. Indeed, this was what was found. In general, if the arrow pointed to the side of the response, it was about 10% less accurate than if the arrow direction and concurrent response response did not match (Museeler & Hommmel, 1997). A di diff ffer eren entt se sett of st stud udie iess ex exam amin ined ed ac acti tion on-p -per erce cept ption ion tr tran ansf sfer er by ha havi ving ng participants first make arm movements given verbal commands without visual feedback (Hecht, Vogt, & Prinz, 2001). The participants then made visu vi sual al ju judg dgme ment ntss of si simi mila larr pa patt tter erns ns.. A se sepa para rate te gr grou oup p of pa part rtic icip ipan ants ts pe perrformed the visual task first and then the motor one. The studies showed that there was transfer of both perception to action and action to perception. Th tion. Thes esee fin findi ding ngss ar aree co cons nsis iste tent nt wi with th th thee cl clai aim m th that at pe perrce cept ptio ion n an and d ac acti tion on share a common repr representational esentational code. Other experimental paradigms offer data in support of TEC (Prinz, 1997). In Indu duct ctio ion n pa para radig digms ms,, in incl clud udin ing g sp spat atia iall co comp mpat atib ibil ility ity ta task skss an and d se sennsorimo sor imotor tor syn synchr chroni onizat zation ion,, sho show w tha thatt cer certai tain n vis visual ual sti stimul mulii ind induce uce par partic ticuular actions by virtue of similarity. similarity. Interference paradigms showed mutual interference between the perception of ongoing events and the perception and control of ongoing action. Hecht et al. claim that these kinds of  kinesthetic-visual transfer are most likely due to visuomotor-kinesthetic

 

64

 

Embodiment and Cognitive Science

matc tchi hing ng,, wh whic ich h su sugg gges ests ts th that at pe perrce cept ption ion an and d ac actio tion n sh shar aree co comm mmon on me mech ch-ma anisms, exactly what is predicted by TEC. In general, there is emerging evidence that perception and action are deeply dee ply int interr errela elated ted,, and pos possib sibly ly sha share re com common mon neu neuro rolog logica icall mec mechan hanism isms. s. This work is clearly consistent with an embodied view of perception, as opposed to traditional accounts that sharply distinguish perception from embodied action. Perception as Anticipated Embodied Interaction

One reason that many scholars argue against a strict divide between per-

ception and action is that perceiving an object without touching it partly invo in volv lves es im imag agin inin ing g ho how w it ma may y be ph phys ysic ical ally ly ma mani nipu pula late ted d (N (New ewto ton, n, 1996; O’Regan, 1992). This perception-action coupling suggests that perceiving an object requires people to conjecture something that if pulled would  bend, if thrown would knock something else aside, and if turned would reve veaal an anot othe herr si side de.. I see an ob obje ject ct an and d im imag agin inee ho how w I mi migh ghtt us usee it wit ith hou outt doing so. For example, I understand the chair in the corner of the room as something I could potentially sit on or stand on or lift to ward off a snarling lion if I walked over to it. This idea can be extended to all objects and physical events in the world. Thus, I perceive the leaves covering my yard as something I could go rake up if I had the right tool to do so. In this way,, perceiving something is not simply a visual experience, but involves way nonvisual, experiences such asofsmells, sounds, andspecific movement of  one’s entiresensory body, such as the feelings readiness to take action upon up on th thee ob obje ject ct.. Un Unde derr th this is vi view ew,, pe perc rcep epti tion on is tig tight htly ly lin linke ked d to su subj bjun unct ctiv ivee thought processes (Ellis, 1995; Newton, 1996). Consider how a person perceives an object by employing various elementary subjunctives (Newton,  1996). An object that reflects light uniformly will be seen as potentially hard and slick to the touch, whereas the warm and fuzzy blanket reflects light unevenly. To see that something is flat is to see it as giving rise to certain possibilities of sensorimotor contingency.. To tingency To feel a surface as flat is precisely to perceive it as impeding or shaping one’s possibilities of movement. Each case of perception involves some so meon onee im imag agin inin ing g wh what at it wo woul uld d fe feel el li like ke to to touc uch h an ob obje ject ct,, gr gras asp p it wi with th thee ha th hand ndss, tu turrn it ov over er,, bit itee it it,, sm smeell it it,, an and d so on on.. De Deve velo lopm pmen enta tall ev evid ideenc ncee showss th show that at pr pres esch choo oole lers rs ca can n cl clas assi sify fy an anima imate te ob obje ject ctss de desp spite ite po post stur ural al va vari ri-atio at ion n am amon ong g th them em,, pr prim imar aril ily y be beca caus usee th thee ch chil ildr dren en co coul uld d im imag agin inee wa ways ys th that at each object many be physically manipulated without altering its identity (Becker & Ward, 1991). Object perception is not an event that happens to us; rather it is something that we do by looking at the object. Our looking at something is a goal-directed task that demands the coordination of head position and eye focus to bring the object into the visual field. To do this, the world is

 

Perception and Action

 

65

conceptualized in part as patterns of possible bodily interactions, or affordan da nce cess (e (e.g .g., ., ho how w we can mo move ve ou ourr han and ds an and d fin finge gers rs,, our le legs gs an and d bod odie iess, our eyes and ears, to deal with the world that presents itself). Under this perspective, eyes themselves do not see. But to see is to explore the environment by means of the exercise of one’s visual apparatus (e.g., one’s eyes). The activity of seeing thus depends on one’s awareness (at least sometimes) of one’s eye movements, also on head and body movements, and characteristic patterns of bodily sensations. Aree pe Ar peop ople le aw awar aree of bo bodi dily ly po poss ssib ibil ilit itie iess wh when en th they ey se seee ob obje ject cts? s? A gr grow ow-ing in g bo body dy of res esea earc rch h ha hass de demo mons nstr trat ated ed th that at pe peop ople le re read adily ily pe perc rcei eive ve ob obje ject ctss in terms of the possible bodily actions they afford. For instance, when observ se rveers ar aree as aske ked d to vi view ew st stai airs rs of dif iffe ferren entt hei eigh ghts ts an and d ju judg dgee th thee on onee th theey could ascend in a normal fashion, they were consistent and accurate with resp re spec ectt to th thei eirr ac actu tual al st stai airr-cl climb imbin ing g ab abil iliti ities es (i (i.e .e., ., ju judg dged ed cl clim imbi bing ng he heig ight htss were a constant proportion of leg length) (Warren, (Warren, 1984). Similar findings

have been reported for people’s judgments of sitting height (Marks et al., 1990), abilities to stand on or traverse different surfaces (Burton,  1992; Fitzpatrick et al.,  1994), judgments of the capabilities of different people (Stoffregen (Stoffre gen et al., 1997), grasping of real objects (van Leeuwen, Smitsman, & van Leeuwen, 1994), catching fly balls (Oudejams, Michaels, Bakker, & Dohne, 1996), the use of tools (Wagman & Carello,  2001), climbing walls (Boschker, Bakker, & Michaels, 2002), and the design of virtual reality environments (Smets et al., 1995). The results of these studies are consistent with the idea that anticipated bodily interactions are a significant part of  perceptual experience. Sensorimotor Contingency Theory

Sensorimotor contingency theory is a new development that strongly em braces the importance of real and anticipated bodily movement in perceptual experience (Noe & O’Reagan,  2002;  O’Regan & Noe,  2001;  also see Churchland, Ramachandran, & Sejnowski, 1994 for a compatible perspective). The basic premise of this theory is similar to that of “interactive visi vi sion on”” id idea ea in as asse sert rtin ing g th that at “v “vis isua uall ex expe peri rien ence ce is no nott so some meth thin ing g th that at ha happpens pe ns in in indi divi vidu dual als. s. It is so some meth thin ing g th they ey do do”” (N (Noe oe & O’ O’Re Reag agan an,, 2002: 567). Sensorimotor contingencies are a set of rules of interdependen interdependence ce between stim st imul ulat atio ion n an and d mo move veme ment nt.. Pe Perc rcei eive vers rs le lear arn n to ma mast ster er th thee wa ways ys th that at vi visu sual al info in form rmat atio ion, n, fo forr ex exam ampl ple, e, ch chan ange gess as a fu func ncti tion on of mo move veme ment nt wi with th res espe pect ct to the environment. Visual Visual experience is, therefor therefore, e, a temporarily external pattern of skillful activity. Consider the experience of driving a Porsche (Noe & O’Reagan, 2002). There is really no single feeling associated with driving a Porsche in the sense sen se tha thatt the there reis is no spe specia ciall bod bodily ily sen sensat sation ion tha thatt ari arises seswhe whenev never er som someon eonee is driving the car car.. Instead, the experience of Porsche driving is constituted

 

66

 

Embodiment and Cognitive Science

 by what a person does when he or she drives a Porsche, such as turning the wheel, shifting gears at differe different nt speeds, feeling the vehicle accelerate when the gas pedal is pressed, and even feeling the sensation of the wind  blowing through through one’s hair if driving a convertible. A person’s experience experience of driving a Porsche has no single defining sensation, but is grounded in knowledge of the sensorimotor contingencies governing the behavior of  the car. Seei Se eing ng,, in th thee se sens nsee of ha havi ving ng a vi visu sual al ex expe peri rien ence ce,, is lik likee Po Pors rsch chee dr driv ivin ing. g. Thus, the experience of seeing is like that of driving in being constituted  by all the things one does when one sees. Seeing a chair chair,, again, involves i nvolves that set of things that one can do with a chair, chair, such as sitting on it, holding it up to fend off a snarling tiger, or even moving one’s eyes around it to  better appreciate the chair’s red velvet covering. Each of these things can  be done when you visually interact with the chair chair.. Of course, there are sensations that arise from seeing a chair, such as feeling relieved; these alone do not define the fundamental elements of visual experience. They aree ju ar just st ac acci cide dent ntal al ad addd-on onss to th thee ac activ tivit ity y of se seei eing ng an ob obje ject ct in so some me sp spec ecifi ificc situation. One implication of sensorimotor contingency theory is that people

should not generally be aware of all aspects of the environment before them. People will experience only those aspects of the world to which they are attending. In fact, as suggested above, there is good experimental evidence that people often fail to notice changes in the environment that are quite large and in full view view.. These “change-blindness” effects occur in circumstances as simple as making saccades to more complex events such  1997 as changing the person to, 1997 whom one is (Levin & O’Re O’ Rega gan, n, Re Resn snic ick, k, & Cla Clark rk, ; Re Resn snic ick, k, talking O’Re O’ Rega gan, n, & Cla Clark rk,Simons, , 1997; Simons; & Chabris, 1999). For example, in one remarkable study, study, a person walking across a university campus was stopped by a person (the experimenter) holding a map who asked for directions to a specific location (Simons & Levin Levin,,  1997). During this conversation, two workmen carrying a door lengthwise walked between the subject and the person asking for directions. While the subject’s vision was blocked by the door, one of the workmen carrying the door quickly switched places with the person originally aski as king ng fo forr di dirrec ecti tion ons. s. Th Thus us,, in ju just st a fe few w se seco cond nds, s, th thee su subj bjec ectt wa wass no now w ta talk lk-ing in g to a di diff ffer eren entt pe pers rson on we wear arin ing g di diff ffer eren entt cl clot othe hes. s. On Only ly 50% of th thee su subj bjec ects ts noticed any change had occurred when they were asked moments later. A related phenomenon, called “inattentional blindness,” occurs when partic par ticipa ipants nts ar aree eng engage aged d in att attent ention ion-in -inten tensiv sivee tas tasks ks and fai faill to not notice ice whe when n extraneous stimuli are presented (Mack & Rock,  1998). For example, people were asked to judge which of the horizontal and vertical lines of a  briefly displayed cross was larger. larger. In one condition, an extra, unexpected element was presented. Participants were asked whether they saw anythin th ing g el else se be besi side dess th thee cr cros osss an and d we werre th then en gi give ven n a rec ecog ogni niti tion on te test st to as asse sess ss

 

Perception and Action

 

67

their perception of the extra element. Many participants failed to notice the extra element when they were just told to make judgments about the crossed lines. But when people were told to focus on anything else in the disp di splay lay,, th they ey ra rarrel ely y mi miss ssed ed se seei eing ng th thee ex extr traa el elem emen ent. t. Th Thus us,, pe peop ople le’s ’s ex expe pecctations about what they were seeing greatly influenced what they actually perceived. None No ne of th thes esee fin findi ding ngss ar aree co cons nsis iste tent nt wi with th an any y th theo eory ry th that at su supp ppos oses es th that at people construct full-blown  3 -dimensional representations of real-world environments. But the results are quite consistent with the idea that perception is a skill-based activity that fundamentally depends on eye, head, and body movements. To bring something into visual consciousness, one must do something (e.g., squint, lean forward, tilt toward the light, walk to a window) wi ndow) and not merely passively see. We We experience only the things we specifically attend to, depending on our current needs and goals. The rest of the world is simply not present because of the lack of focused attention. Unattended portions of the world seem to be there only because we can direct our bodily attention to them in various ways when needed. Once On ce we ha have ve do done ne so so,, th thee ri rich chly ly de deta tail iled ed in info form rmat atio ion n be beco come mess pa part rt of ou ourr conscious perception, as if the world had been there all along. Perception scient sci entist istss deb debate ate whe whethe therr cha change nge bli blindn ndness ess and ina inatte ttenti ntiona onall bli blindn ndness ess ar aree  better explained as inattentional amnesia (Wolff, (Wolff, 1999), or active suppression (Tipper,  1985). Nonetheless, people rarely see everything out there in the world unless they move their bodies in different ways to attend to

environmental information. One provocative consequence of sensorimotor contingency theory is that there may be no need for “binding” in the visual system. Many perceptuall sci ceptua scient entist istss bel believ ievee tha thatt the var variou iouss vis visual ual sub subsys system temss for vis visual ual sti stimmuli, must somehow be unified to explain coherent perceptual experience. For example, there may be particular cells (i.e., “grandmother” cells), or high hi ghly ly lo loca cali lize zed d co cort rtic ical al re regi gion onss (e (e.g .g., ., co conv nver erge genc ncee zo zone nes) s) th that at co comb mbin inee in in-format for mation ion per pertai tainin ning g to spe specifi cificc per percep cepts. ts. Mor Moreov eover er,, sep separa arate te cor cortic tical al ar areas eas that th at ar aree co conc ncur urre rent ntly ly ac acti tiva vate ted d du duri ring ng pe perc rcep eptu tual al an anal alys ysis is ma may y os osci cilla llate te in synchrony, with the synchrony providing the coherence, or unity, to perceptual experience (Brecht, Singer, & Engel,  1998). Yet the fact that object features seem to be part of a single object does not demand that all these features must be “represented” in a unified way way,, whether it be in a single  brain region, or in terms of various brain processes. Noe and O’Reagan argue that “what explains the conceptual unity of experience is the fact that experience is a thing one does, and one is doing with respect to concept ce ptua uall lly y un unifi ified ed ex exte tern rnal al ob obje ject ct”” (N (Noe oe & O’ O’Re Reag agan an,, 2002: 585). Un Unde derr th this is view, the classic problem of binding in perception may be dismissed as a pseudoproblem. O’Reagan and Noe (2001) further argue that the senses cannot be individuated by their distinct qualitative characteristics (i.e., only eyes see,

 

68

 

Embodiment and Cognitive Science

ears hear, and so on), but by different patterns of sensorimotor contingency by which they are governed. After all, it is possible to “see” the world through sensory systems other than vision. Consider one tactile vision substitution system (TVSS) (Bach-y-Rita, 1996; Kaczmarek & Bachy-Rita,  1995). In TVSS, optical images picked up by a camera (worn on the head) are transduced in such a way as to activate an array of stimulators (vibrators or electrodes) in contact with the skin (e.g., the abdomen, the back thigh, or most recently, the tongue). Optical images in this way produce localized patterns of tactile sensation. After an initial period of  training, both congenitally blind subjects and blindfolded normal participant pa ntss we werre ab able le to pe perrce ceiv ivee so some me si simp mple le di disp spla lays ys.. Th Thes esee pa part rtic icip ipan ants ts ev even en reported that after some training, they ceased to experience tactile sensation ti onss wh when en th they ey us used ed th thee TV TVSS SS de devi vice ce,, an and d ex expe peri rien ence ced d ob obje ject ctss as ar arra raye yed d  before them in three-dimension three-dimensional al space, just as captured by the camera. When Wh en th thee ca came mera ra pr pres esen ente ted d a ra rapi pidl dly y ap appr proa oach chin ing g ob obje ject ct,, fo forr in inst stan ance ce,, th thee rapi ra pid d ex expa pans nsio ion n of th thee TV ob obje ject ct co corr rres espo pond nded ed to ex expa pand ndin ing g ac acti tivi vity ty on th thee tact ta ctil ilee gr grid id,, ca caus usin ing g th thee pe pers rson on to im imme medi diat atel ely y du duck ck.. Pa Part rtic icip ipan ants ts “l “lea earn rn to makeperceptualjudgmentsusingvisualmeansofanalysissuchasperspective parallax, looming, and zooming and depth judgments” (Bach-y-Rita, 1996: 91). This kind of tactile perception enables participants to judge the shape, size, and number of objects and to perceive spatial relationships between them. With sufficient practice, participants are able to engage in tasks requiring skillful sensorimotor coordination (e.g., batting a ball or working on an as asse semb mbly ly li line ne). ). Pa Part rtic icip ipan ants ts mu must st,, ho howe weve verr, be ab able le to mo move ve th thee ca cammeraa on th er thei eirr ow own. n. En Enab abli ling ng pe peop ople le to ex expl plor oree fig figur ures es by sw swee eepi ping ng th thee ca cammeraa ov er over er a di disp spla lay y res esul ults ts in vi vibr brat atio ion n ch chan ange gess th that at ar aree ne nece cess ssar ary y fo forr me mean an--

ingful tactile images to be perceived. Of course, tactile vision is not nearly as ef effe fect ctiv ivee as rea eall vi visi sion on.. Bu Butt th thee wo work rk on TV TVSS SS pr prov ovid ides es an anot othe herr ex exam ampl plee of the power of sensorimotor contingencies in perceptual processes. A final source of evidence supporting the idea of sensorimotor contingencies in perception comes from a case study of a blind patient whose sight was attempted to be restored by surgical removal removal of cataracts (Gregory & Wallace, 1963). In fact, this surgery does not restore restore sight. One pati pa tien ent’ t’ss ex expe peri rien ence ce wa wass de desc scri ribe bed d in th thee fo foll llow owin ing g wa way: y: “H “Hee he hear ard d a vo voice ice coming from in front of him and to one side: he turned to the source of the sou so und an and d sa saw w a “bl blur ur.” .” He rea eali lize zed d th that at th this is mi migh ghtt be a fa face ce,, Up Upon on ca carrefu full questioning, he seemed to think that he would not have known that this was a face if he had not previously heard the voice and known that voices come from faces” (Gregory & Wallace,  1963:   122). The patient acquired some so me fo form rm of vi visu sual al se sens nsat atio ion, n, or im impr pres essi sion on,, bu butt ha had d no nott ye yett ac acqu quir ired ed th thee abilit abi lity y to see see,, pr presu esumab mably ly bec becaus ausee new newly ly acq acquir uired ed vis visual ual imp impre ressi ssions ons wer weree not yet int integr egrate ated d wit with h pat patter terns ns of sen sensor sorimo imotor tor con contin tingen gency cy gov govern erning ing the

 

Perception and Action

 

69

occurrence of these sensations (Noe, 2004). Although surgery restores the mechanisms enabling visual sensations, it cannot restore the linkages between impressions and bodily movement required for correct vision. It should not be surprising, then, that blindness can result by simply severing in g vi visu sual al ar area eass co conn nnec ecte ted d to th thee mo moto torr ar area eass (N (Nak akam amur uraa & Mis Mishk hkin in,, 1980). Even Ev en wh when en th thee se sens nsor ory y ap appa para ratu tuss an and d as asso soci ciat ated ed co cort rtic ical al re regi gion onss fo forr vi visi sion on are intact, there is no vision without activation in the motor cortex. Case studies such as this support the general claim that perception of objects and space is based on a person’s anticipation of the sensory consequences of actions that could be performed in a given situation. Robotics

Traditional artificial intelligence (AI) models assume that intelligence is largel lar gely y und unders erstoo tood d in ter terms ms of man manipu ipulat lating ing car carefu efully lly con constr struct ucted ed int intern ernal al modelsonofbuilding externalmodels reality.of This ledfrom to a quest fordata intelligent  based the has world sensory and themachines development of algorithms to “reason” about the world using these models. Most of the AI work tries to capture important aspects of high-level cognition, such as reasoning and language understanding. But people in a related  branch of AI called “artificial life” (AL) have lowered their aim to develop machines that could exhibit insect-like intelligence. These scholars argue that th at th thee ma majo jorr pa part rt of na natu tura rall in inte tell llig igen ence ce is cl clos osel ely y bo boun und d up wi with th th thee ge genneration of adaptive behavior in the harsh, unforgiving environment most animals inhabit. Many of the new robots being constructed within AL are based on two fundamental principles: situatedness and embodiment (Brooks, 2002). “A situ si tuat ateed cr crea eatu turre or rob obot ot is one th that at is em emb bed edd ded in th thee wo worrld ld,, an and d wh whic ich h does not deal with abstract descriptions, but through its sensors with the here and now of the world, which directly influence the behavior of the crea cr eatu turre. An em embo bodi died ed cr crea eatu turre or ro robo bott is on onee th that at ha hass a ph phys ysica icall bo body dy an and d expe ex peri rien ence cess th thee wo worl rld, d, at le leas astt in pa part rt,, di dirrec ectl tly y th thrrou ough gh th thee in influ fluen ence ce of th thee

world on that body body.. A more specialized type of embodiment occurs when the full extent of the creature is contained with that body” (Brooks,  2002: 52–3). Take, for example, a robot that moves around in the world by avoiding obstacles, finding doorways, and identifying objects, such as soda cans, that it can pick up. Traditional robots developed for this purpose see the world in order to get a full  3-dimensional model, and only then construct an in inte tern rnal al pl plan an fo forr mo movi ving ng th thrrou ough gh th thee en envi viro ronm nmen ent. t. On Once ce th this is fu fullll-bl blow own n plan has been created, instructions are sent to the body as to what movements the robot should do. Thus, a high-level planner calls lower-level motor modules, when required, to act in particular ways.

 

70

 

Embodiment and Cognitive Science

thee ot othe herr han and d, mo morre rec eceent AL rob obo ots st staart by mo movi vin ng fir firsst, an and d us usee On th their own activity as a guide to understanding the environment. Brooks (1991) coined the slogan “the world is its own best model” to capture the essence of this idea. The overall behavior of the system is a result of variouss au ou auto tono nomo mous us ac activ tiviti ities es ov over erri ridi ding ng ea each ch ot othe herr, an and d no nott of th thee sy syst stem em as a whole making a global decision based on centrally held internal representations of the world. Each layer of the system is sensitive to specific parts of the environment. Although observers may interpret the robot’s overall behavior as meaningful, there is no “meaning” programmed within the system, apart from the specific behaviors of its various parts in responsee to local environmenta spons environmentall condi conditions tions (e.g., activi activity ty patte patterns rns that are  bootstrapped by tight stimulus-response stimulus-response feedback loops, loops, evolved to meet particular environmental constraints). For instance, an early robot, “Genghis,” was a big six-legged insect whose behavior was not organized in a single program, but in terms of  fifty-one tiny parallel programs (Brooks,  2002). Three of these programs allowed Genghis topond scramble around rough mostly its ba balan lance ce, , an and d res espo nd to ob obst stac acle less inover itss wa it way y. Th Theeterrains, six si x py pyro roel elec ectr tric ickeeping sens se nsor orss arrayed at the front of the robot allowed it to sense the presence of heatemitti emi tting ng mam mammal mals. s. Whe Whenev never er Gen Genghi ghiss enc encoun ounter tered ed a hea heat-e t-emitt mitting ing sme smell, ll, it wo woul uld d ch chas asee th thee so sou urce of th thee sm smeell ove verr wh what ateeve verr te terr rrai ain n wa wass in fr fro ont of  it. Of course, Genghis never knew the actual content of the heat-emitting smell, it just followed it. Genghis never planned ahead its trajectory or its every move. Its behavior is an emergent activity plan with a structure that lies at a level above that of its individual reflexes. A la late terr rob obot ot,, kn know own n as “H “Her erbe bert rt,” ,” mo move ved d ar arou ound nd a la labo bora rato tory ry en envi viro ronnment and collected soda cans, not by detailed advance planning, but by very ve ry su succ cces essf sful ully ly us usin ing g a co colle llect ctio ion n of co coar arse se se sens nsor orss an and d si simp mple le,, rel elat ativ ivel ely y independent, behavioral routines (Connell,  1989). Basic obstacle avoidance was controlled by a ring of ultrasonic sound sensors that brought the robot to a halt if an object was in front of it. General locomotion (randomly dom ly dir direct ected) ed) was int interr errupt upted ed if Her Herber bert’s t’s sim simple ple vis visual ual sys system tem det detect ected ed a roughly table-like outline. At this point, a new routine kicked in and the tablesurfacewassweptusingalaser.Iftheoutlineofacanwasdetected,the whole robot rotated until the can was centered in its field of vision. This physical action simplified the pick-up procedure by creating a standard

action acti on-f -fra rame me in wh whic ich h th thee rob obot ot ar arm, m, eq equi uipp pped ed wi with th si simp mple le to touc uch h se sens nsor ors, s, gently skimmed the table surface ahead. Once a can was encountered, it could be grasped and collected and the robot then moved on. Notice, again, that Herbert succeeds without using any conventional planning techniques and without creating and updating any detailed inner model of the environment. Herbert’s world is composed undifferentiated obstacles and rough table-like and can-like outlines. of Within this world the robot also exploits its own bodily actions (rotating the torso to

 

Perception and Action

 

71

center the can in its field of view) to greatly simplify the computational problems involved in eventually reaching for the can. Herbert is thus a simple example of both a system that succeeds using minimal representational resources and one in which gross motor activity helps streamline the perceptual routines. A different artificial life system, called “Creature,” decomposed a situated system into a numbe numberr of simple tasktask-achie achieving ving behaviors, behaviors, each of  which links specific sensory and motor capacities so that it may interact independently and reactively with properties of the t he surrounding environment (Brooks,  1991). The robot’s task-achieving behavior is not handled  by explicit goal-directed planning, but by layered layered control that is achieved  by building the lowest level task-achieving behavior behavior,, debugging its operation ti on,, th then en bu buil ildi ding ng an anot othe herr on th this is fo foun unda dati tion on,, an and d so on on.. Fo Forr ex exam ampl ple, e, th thee robo ro bot’ t’ss rea eall-wo worl rld d ex expl plor orat ation ion ca can n be bu built ilt by st star artin ting g wi with th Le Leve vell 0:“donot come co me in into to co cont ntac actt wi with th ot othe herr ob obje ject cts. s.”” Ad Addi ding ng Le Leve vell 1, “wand “wander er aimles aimlessly sly,” ,” will produce moving around without hitting anything. With the addition of Le Leve vell 2, “v “vis isit it in inte tere rest stin ing g pl plac aces es”” (e (e.g .g., ., co corr rrid idor orss of fr free ee sp spac acee de dete tect cted ed by sensors), the directed robot’s behavior comes toPreadaptive look like exploration, without any goal or plan at that function. sensorimotor pairings  between vision and reaching, or between manual contact and retrieval to the mouth, interact independently with the robot’s experience of the environment to generate an illusion of hierarchically controlled sequencing and goal-directedness goal-directedness.. Another example of insect robotics is a society of robots designed to collec col lectt or oree sam sample pless (St (Steel eele, e, 1994). Th Thes esee rob obot otss co coop oper erat atee by dr drop oppi ping ng el elec ec-tronic bread samples along their travels, according to what they are sensing and doing doing.. By dropping dropping marke markers rs that meani meaningfull ngfully y chang changee the later  behavior of the system, the robots use objects in the world to repr represent esent their intera int eracti ctive ve exp experi erienc encee (se (seee Cha Chapte pterr 5). Th Thee ce cent ntra rall id idea ea is th that at de desc scri ript ptio ions ns of the patterns that the robots followed were not built in. Although the robots were designed to create a path, in order to pick up ore samples, the motion of the path was not represented as a plan inside the robot. The robots exhibited self-organized behavior simply by acting as individuals, reacting locally and following a hierarchy of rules for picking up ore samples. Steel’s self-organization design was based on the idea that emergent stru st ruct ctur ures es wi will ll gr grow ow fr from om th thee in inte tera ract ctio ion n of ma many ny el elem emen ents ts an and d th then en de deca cay y again until an equilibrium state is reached. As he summarized, “We will design a system of interacting robots where equilibrium behavior consists in ex expl plor orin ing g th thee te terr rrai ain n ar arou ound nd th thee ve vehi hicl cles es.. Th Thee pr pres esen ense se of th thee roc ock k sa sammples pl es co cons nsti titu tute tess a di dist stur urba banc nce. e. Th Thee de desi sirred di diss ssip ipat ativ ivee st stru ruct ctur uree co cons nsis ists ts of 

spatia spat iall st stru ruct ctur uree (i (i.e .e., ., a pa path th)) fo form rmed ed by th thee ro robo bots ts be betw twee een n th thee sa samp mple less an and d the vehicle. This structure should spontaneously emerge when rock samples are present, should enforce maximimize performance and). shou sh ould ld dis disap appe pear aritwh when en all sa samp mple lessitself have ha ve to been be en colle co llect cted ed”” (S (Ste teel els, s, 1990: 182

 

72

 

Embodiment and Cognitive Science

Both Brooks and Steele’s robots are examples of structured coupling  between the robot and aspects of the environment. For example, wall following, as in “Herbert,” is continuously enhanced by ongoing obstacle avoidance and forward movement, and ongoing wall hugging enables doorway entry. Most generally, coordinated movement is accomplished efficiently and directly by layered automata where states relate what the program has just done and what it is sensing now. The recent work on robotics provides a kind of existence proof demonstrating the intricate links between perception and action in producing meaningful behavior. Thes Th esee ro robo bots ts sh show ow,, si simi milar larly ly to th thee ex expe peri rime ment ntal al wo work rk wi with th hu huma man n pa part rtic ic-ipants, that perception does not precede action, because action is integral to perceptual performance. At the same time, AL robots illustrate that it is not always possible to define meaningful action apart from how an organism interacts with the environment. For instance, Ghengis engages in a type of behavior that appears to be “chasing prey,” but this phrase is only a convenient description of an emergent behavior, and not some internalized “mental” rule that the robot follows. “Chasing prey” emerges from actual embodied performance. High-level actions cannot, therefore,  be reduced to simple of movement, but may emerge fromand thethe interaction between thesequences robot’s various simple activities, or reflexes, particular environment to which it is adapted (e.g., Brook’s subsumption architecture). A very recent demonstration of an embodied agent more explicitly em braces ideas from dynamical systems theory (Beer (Beer,,  2003). This agent is capa ca pabl blee of vi visu sual ally ly di disc scri rimin minat atin ing g be betw twee een n ob obje ject ctss by ca catc tchi hing ng ci cirrcu cula larr ob ob- jects and avoiding diamond-shaped ones. The agent’s behavior was controlled through a dynamical “nervous system” that “evolved” for these purposes. A series of experiments showed that the agent foveated on and then scanned any object before either catching it or avoiding it, and made this discrimination primarily on the basis of the object’s width. Quite importan por tantly tly,, the age agent nt doe doess not re really ally kno know w the dif differ ferenc ences es bet betwee ween n the dif differ fer-ent types of objects, because its ability to engage in categorical perception is not located within a single subsystem. Instead, the agent’s adaptive behavi ha vior or is a pr prop oper erty ty of th thee en enti tirre co coup uple led d sy syst stem em (i (i.e .e., ., in inte tera ract ctio ions ns of br brai ain, n,  body,, and environment)  body environment) that it was specifically selected selected for. for. Even if it may  be possible to decompose the coupled brain, body body,, and environment systems to better understand how each participates in the overall behavior, it makes little sense to attribute simple cause and effect to these parts. As with any dynamical system, effects unfold in time to become causes, so that th at di disc scri rimi mina nati tion on be beha havi vior or re real ally ly ta take kess pl plac acee on an ex exte tend nded ed tr tran ansi sien entt of  the entire coupled system (Beer, 2003: 236). Thee le Th less sson on he herre fo forr co cogn gnit itiv ivee sc scie ienc ncee is th that at de desc scri ript ptio ion n of wh what at a pe pers rson on (or animal or mechanical object) does should not be reified as a naive intentional description, or assumed to be causally grounded in internal

 

Perception and Action

 

73

mental re mental repr prese esenta ntatio tions. ns. Ena Enacti ctive ve sys system tems, s, suc such h as the AL ro robot bots, s, sho show w how know kn owle ledg dgee is em embe bedd dded ed in a di dist stri ribu bute ted d fa fash shio ion n (b (bod ody y, se sens nsor ors, s, ac actu tuat ator ors, s, nervous/control systems, etc.), or even partly in the environment. One may argue, nonetheless, that these robots are merely reactive, living in the “her “h eree an and d no now w of th thee wo worl rld, d,”” an and d ar aree tr trul uly y no nott au auto tono nomo mous us.. Au Auto tono nomo mous us agents are situated and capable of learning from their experiences (and from evolutionary history) in ways that can be applied to solving new real-world problems. Some cognitive scientists voice concern over whether it will be possible to sc scal alee up th thes esee re reac acti tive ve sy syst stem emss to mo mode dell mo morre co comp mple lex x co cogn gnit itiv ivee be beha havviorss (Cl ior (Clark ark,, 1996; Ziemke, 1999). So Some me se self lf-o -org rgan aniz ized ed sy syst stem emss ha have ve tr trie ied d to ground an agent’s functions by connecting sensors and actuators to some central mechanism (e.g., a connectionist network or a classifier system). These systems allow the agent to adapt the central mechanisms on the  basis of the robot’s interaction with the environment. In this way way,, a selforganized autonomous agent may be grounded in experience (Beer, (Beer, 1997, 2003). Of course, the robot’s designers have a fundamental role in choosing in g a pa part rtic icul ular ar ar arch chit itec ectu turre (e (e.g .g., ., nu numb mber er of un unit its, s, la laye yers rs of a co conn nnec ecti tion onis istt network), which makes these “autonomous” robots not so autonomous. A different criticism of some AL robots is posed in terms of the “robot grounding” or “body grounding problem” (Sharkey & Ziemke, 1998). Despite embracing both embodiment and situatedness in designing enactive robots, most systems fail to capture the way bodily mechanisms are truly embedded in their environments. Biological self-organized agents are not designed and then inserted into some environment, because living organisms is ms em embo body dy a lon long g hi hist stor ory y of mu mutu tual al sp spec ecifi ifica cati tion on an and d st stru ruct ctur ural al co coup uplin ling g throughout evolution and an organism’s individual lifetime. Real bodies are rooted in the environment, and are not simply interfaces between internal controller and environment. Human bodies, to take one example, are tailored to perceive and act in a meaningful way because of the environments they inhabit. Thus, the embodied nature of current AL robots partly neglects the historical reality and environmental embodying of living in g bo bodi dies es in wh whic ich h bo body dy,, ne nerv rvou ouss sy syst stem em,, an and d en envi viro ronm nmen entt co coev evol olve ve an and d are mutually determined (Ziemke,  1999). Perception and action are more tightly integrated in biological organisms than is necessarily the case with most AL robots (see Nolfi & Floreano,  2000 for further discussion of this issue). A Dynamic Model of Intentional Action

 Just as perception is traditionally thought to precede action, so too is action thou th ough ghtt to or orig igin inat atee in pr prio iorr me ment ntal al in inte tent ntio ions ns.. Fo Forr in inst stan ance ce,, if yo your ur fr frie iend nd looked at you and purposefully yawned, you presume that she had some idea in mind that caused her to do as she did. But how can a thought or

 

74

 

Embodiment and Cognitive Science

mental state provide the causal foundation for human action? Although we feel that a conscious choice underlies purposeful yawning, there must  be some physical explanation for the yawn. As Aristotle famously argued, nothing can cause or move itself. Every physical event must result from some previous physical event (e.g., as efficient causes such as the movement that occurs when billiard balls strike one another), not from a mere conscious or mental act. But our ordinary experience, again, suggests that an intention to move our bodies results in our bodies actually moving in thee wa th ways ys th they ey do do.. Co Cont ntem empo pora rary ry ph phil ilos osop ophe hers rs,, wo work rkin ing g on th thee to topi picc of ac ac-tion theory, theory, continue to struggle over how the mind both causes behavior and continuously monitors and directs its effects. Recent experimental results show that people mistakenly believe that their conscious will directs thee ti th timi ming ng of th thei eirr mo moto torr ac acti tion onss wh when en,, in fa fact ct,, th thee br brai ain’ n’ss ac acti tion on-r -rea eadi dine ness ss potential seem to cause one’s conscious intention (Libet,  1985). Findin Findings gs such su ch as th thes esee ar aree di diffi fficu cult lt to rec econ onci cile le wi with th tr trad adit itio iona nall no noti tion onss of in inte tent ntio iona nall action.  Juarrero  Juarrer o (1999)  criticizes philosophers for failing to provide coherent answ an swer erss to th thee qu ques estio tion n of wh what at ca caus uses es in inte tent ntio iona nall be beha havi vior or.. Sh Shee ad adva vanc nces es the idea that intentional behavior, and its causes, is best characterized as a fluid, dynamic process taking shap shapee thr through ough the intera interactions ctions between  brains, bodies, and their environments. environments. Juarrero Juarrero adopts the perspective of  complex dynamic systems theory as a “theory-constitutive metaphor” for reconceptualizing mental causation, particularly in terms of how philosophers think of the causes for intentional action. Juarrero’s philosophical analys ana lysis is has obv obviou ious, s, pr profo ofound und imp implica licatio tions ns for cog cognit nitive ive sci scienc encee exp explan lanaation of causation and human action. One imp import ortant ant imp implic licati ation on of dyn dynami amical cal sys system temss the theory ory is tha thatt the int intenentio ti ons one fee eels ls to pu purrpo possef eful ully ly ya yaw wn, or ra rais isee on onee’s han and d to wa wave ve he hell llo o to a friend, frien d, res result ult fro from m a pers person’s on’s self-o self-orga rganizing nizing tende tendency ncy.. This selfself-org organizin anizing g strruc st uctu turre em embo bodi dies es a te tend nden ency cy fo forr so some meon onee to wa want nt to pu purp rpos osef eful ully ly ya yawn wn even before the desire to perform the action reaches awareness. A concrete illustration of this point is seen in the developmental work of Thelen and Smith (1994). Thelen and Smith argued that motor development in infants is not a maturational processes determined by some hard-wired genetic code. Instead, motor development is a process of dynamical selforganization that arises from the infant’s continuous interaction with its changing environment. For example, two infants started out with different inherent dynamics for reaching. One infant, Gabriel, flailed wildly and repeatedly as she reached for an object, yet another infant, Hannah, was far less physically active and carefully assessed the situation before reaching. Both infants learned to successfully reach objects within a few weeks of one another. Yet this fact does not imply that there must be a preprogrammed pattern for intentional action that simply unfolds over time, as suggested by

 

Perception and Action

 

75

Piaget Piag et (1975). In Inst stea ead, d, ea each ch ch chil ild d ge gene nera rate ted d di diff ffer eren entt so solu lutio tions ns fo forr su succ cces esssfull rea fu each chin ing. g. Th Thus us,, as a se self lf-o -org rgan aniz izin ing g sy syst stem em,, th thee in inte tera ract ctio ion n of an in infa fant nt’s ’s initial spontaneous dynamics and the environment facilitates each child’s movement from equilibrium to a transformation that establishes secondorder context-dependent constraints. These second-order constraints reshape, in different ways for each child, the springlike attractors for limb move mo veme ment nts, s, all allow owin ing g ea each ch ch child ild to su succ cces essf sful ully ly le lear arn n to re reac ach h fo forr an ob obje ject ct.. A dy dyna nami mica call ac acco coun untt of in inte tent ntio iona nall ac acti tion ons, s, su such ch as yo your ur fr frie iend nd’s ’s ya yawn wn,, starts idea thatisself-organized dynamical structures areadaptive globally stable with even the when there disorder at lower levels. Yet Ye t a complex system, such as your friend, can be driven from equilibrium toward insta bility because of the interaction of external circumstances and the system’s own internal dynamic processes. For example, feeling bored at a lecture can ca n pr prec ecip ipit itat atee in inst stab abil ilit ity y, no nott on only ly at th thee ne neur urol olog ogic ical al le leve vel, l, bu butt at th thee co coggnitive and emotional levels as well. By forming an intention, say to yawn whil wh ilee lo look okin ing g ov over er at an anot othe herr pe pers rson on,, a co cogn gnit itiv ivee ph phas asee ch chan ange ge ta take kess pl plac acee thatt dis tha dissip sipate atess the dis disequ equilib ilibriu rium. m. Bey Beyond ond re resto storin ring g dyn dynamic amic equ equilib ilibriu rium, m, the new intention’s restructured restructured contextual constraints reor reorganize ganize the semantic space into a more differentiated and complex set of options. In this way, by formulating a prior intention, people avoid the need to consider and evaluate every logical and physical possibility for action. Thus, once your friend forms the intention to let you know of her belief about the lectur lec ture, e, tha thatt cog cognit nitive ive re reor organ ganiza izatio tion n cir circum cumscr scribe ibess you yourr fri friend end’s ’s yaw yawnin ning g at you, rather than writing you a note, shaking her head, whispering to you, and so forth. Conceptualizing action from a dynamical system perspective explains why people need not explicitly decide something each time they act. The pers pe rson on’s ’s cu curr rren entt fr fram amee of mi mind nd au auto toma mati tica call lly y se sele lect ctss a su subs bset et fr from om th thee un un-limited other alternatives within her self-organized constraint-space. For instance, when your friend decides to inform you of her belief about the lecture, she does not need to explicitly formulate a decision or proximate intention about what to do. Her “choice” of yawning rather than doing some so meth thin ing g el else se (e (e.g .g., ., wr writ itin ing g a no note te,, ta talk lkin ing g al alou oud d to yo you) u) ca can n be “d “dec ecid ided ed””  by the interaction between her own dynamics and the environment environment as the process “moves downstream” (to use the dynamical language of moving through “landscapes”). For instance, your friend knows that her being in a lecture prevents her from saying something aloud, or perhaps even whispering. None of this, however, requires that she form an explicit intent te ntion ion re requ quir irin ing g ex expl plic icit it de deli libe bera ratio tion. n. Sh Shee ca can n ju just st de deci cide de to co comm mmun unic icat atee her belief about the lecture and the environmental constraints take care of  the fine-grained details of how this intention is manifested in real-world  behavior..  behavior A different phenomenological demonstration of how intentional action can be explained in dynamical terms is seen in David Sudnow’s ( 1978)

 

76

 

Embodiment and Cognitive Science

rema mark rkab able le ac acco coun untt of hi hiss ow own n ex expe peri rien ence cess le lean anin ing g to pl play ay ja jazz zz im impr prov ovis isaare tion on the piano. Sudnow began to master jazz improvisation by copying the sounds of jazz performers. But this proved to be extremely difficult  because Sudnow was unable to precisely identify the notes and tempo-

ral values in these master performances. Even when Sudnow came close to making an exact description description of a maste masterr ’s sounds, something something still did not sound right. The difficulty was that the sounds did not provide the right information to guide his actions in moving his hands appropriately across the keyboard. A breakthrough occurred when an accomplished pian anis istt ur urge ged dthat Sudn Su dnow ow to prod pr oduc ucee a sm smal l nu numb mber er of si simp mple scal sc ales es an and chor ch ord d sequences were characteristic ofallthe jazz sound helewished todexhibit while playing. Sudnow Sudno w began to do this, patterning himself himself after the work of Jimmy Rowles. For instance, “There were these three diminished scales to begin with, each identified by refer reference ence to a theoretic system that related its use to four of the twelve dominant chords, so on my thinking there was a ‘cognitive map,’ each scale named by a starting place, each related to its class of chords” (Sudnow,  1978:   21). By repeatedly performing these scales, Sudnow came upon a remarkable result. “I recall playing one day, day, and finally as I set out into a next course of notes, after a lift-off had occurred, that I was expressively aiming for the sounds of these particular notes, that the sounds seemed to creep up into my fingers, that the depression of the keys realized a sound being prepared for on the way down” (Sudnow,  1978:  37 ). In other words, by engaging in a structured activity, reliable correlations emerged between the motor activities involved and the perceptual input deriving from interactions with a specific environment. Sudnow correctly noted that the “knowledge” that emerged from his interactions with the piano was not explicit, context-free, musical knowledge. “As I found the next sounds coming up, as I set out into the course of notes, it was not as if I had learned about the keyboard so that looking dow do wn I cou ould ld te tell ll wh what at a rega garrded no note te wo woul uld d so soun und d li like ke.. I do no nott hav avee th thaat skill, nor do many musicians. I could tell because it was the next sound,  because my hand was so engaged with the keyboard that it was given a(Sudnow, setting settin g of sounding places its own configurati config uration on and of potentialit poten tialities” ies”  1978 :  45 ). Thus, theinsituat situated, ed, embodied activity performin perf orming g scales and chord sequences enabled Sudnow to become familiar with the attendant sounds and physical sensations and the correlations between them. Furthermore, Sudnow insisted that the same key, struck from different approaches with different intent, would give off a slightly differen differentt soun so und, d, be beca caus usee in pe perf rfor ormi ming ng it it,, he wa wass “g “goi oing ng fo for” r” th that at pa part rtic icul ular ar so soun und. d. A model for any note may have an objective classification, but that does nott pr no prov ovid idee th thee dy dyna namic mical al ba basi siss of th thee se sens nsor orim imot otor or me mech chan anis isms ms in invo volv lved ed in making that sound happen.

 

Perception and Action

 

77

this is wa way y, in inte tent ntio ions ns ne need ed no nott be vi view ewed ed as in inde depe pend nden entt me ment ntal al ev even ents ts In th that cause behavior, but are best characterized as dynamic processes em bedded in a physical, historical, and social world, including those of a person learning a skilled activity like playing jazz improvisation. Sudnow’s phenomenological analysis of his learning suggests that skilled performance is not generated from a prior mental decision to act in a particular way that is independent of his ongoing piano-playing behavior. Instead, skilled human action may arise from how the individual’s frame of mind

automatically selects a subset of behaviors (i.e., hitting the right notes in thee ri th righ ghtt wa way y to pr prod oduc ucee th thee ri righ ghtt so soun unds ds)) fr from om th thee un unli limi mite ted d al alte tern rnat ativ ives es within the self-organized environment interactions. constraint space that is defined by the personConclusion

Perception Perceptio n is not a kind of infor information mation processing processing that sampl samples, es, selects, or points out features of an independent object world. People do not first perceive to create a full-scale internal model of the world that is then used to generate appropriate action. Instead, perception involves bodily movements of various sorts and the anticipation of action when adapting to environmental situations. In this way, perception establishes a reciprocal relat re lation ion in the phy physic sical al coo coord rdina inatio tion n of the or organ ganism ism wit with h the env envir ironm onment ent.. This kind of body-world structural coupling is fundamentally grounded in self-movement. An embodied approach to perception and action sees thes th esee as dy dyna namic mical ally ly in inte tert rtwi wine ned, d, in th that at th thee ph phys ysic ical al pr prop oper erti ties es of th thee re real al world are not entities to be statically perceived, but are opportunities for action. Exactly how do perception and action interrelate? Many scholars talk of perception-action linkages in terms of “shared representations.” But there is one functional scheme that provides a broader dynamical account for perception-action couplings. An overview of this is seen in Figure  3.2 (Viviani,  2002). This model includes two sets of hierarchically organized oscillators (or nonlinear resonators) that generate percepts and organize action. Each oscillator is tuned to respond maximally to just one category of st stim imul uli, i, ye yett is fle flexi xibl blee en enou ough gh to res espo pond nd to st stim imul ulii of di difffe ferren entt ty type pes, s, or st stim imul uliai th that at ar areecollection inco in comp mple lete teresonant or co corr rrup upt. t. Th Thee pe per cept ptua ual l se sett is than assu as sume med to in in-clude richer of modes, orrce limit cycles, thedmotor set. Thus, the behavior of the perceptual system converges toward toward certain  basins of attraction more readily than is the case with the motor set. Each sett ha se hass a fu full ll co comp mple leme ment nt of wi with thin in-s -set et co coup upli ling ngs, s, so some me of wh whic ich h ar aree ge gene nettically determined, with others acquired from experience. These within-set couplings provide for the activation and organization of individual components in their respective domains. The perceptual mode that prevails at any moment depends on the sensory inflow, although the “winner” is

 

78

 

Embodiment and Cognitive Science

figure 3.2.  A func functio tional nal sch scheme eme for describ describing ing mot motor–p or–perc ercept eptual ual int interac eractio tions; ns; full

description in the text.

strongly influenced by the couplings within the motor set. Through these couplings, resonant modes are induced in the motor oscillator even when there is no direct activation from any conscious will to move. There may even ev en be no rea easo son n to as assu sume me an any y di difffe ferren ence ce be betw twee een n pe perrce cept ptua uall an and d mo moto torr oscillators, because these may originate in the same physiological mechanism. Regardless of whether activation spreads from perception to action, or the other way, way, the end product is an integrated superordinate limit cycle, cl e, or ba basi sin n of at attr trac acti tion on.. In fa fact ct,, so some me fa fami mili liar ar pe perrce cept ptua uall ex expe peri rien ence ce ma may y  be precluded in individuals in whom the motor set of oscillators are not functional. At the same time, suppressing perceptual resonance will alter motor functions, to some extent, both because some reaffe reafference rence would be missing, and because some global resonant modes would be unavailable. This Th is sp spec ecul ulat ativ ivee pr prop opos osal al ca can n ac acco coun untt fo forr se seve vera rall as aspe pect ctss of pe perc rcep epti tion on-acti ac tion on in inte tera ract ctio ion, n, in incl clud udin ing g mir mirro rorr ne neur uron ons, s, th thee ro role le of an antic ticip ipat ator ory y ac actio tion n in perception, why individuals with impaired motor cortex may suffer from vision problems despite having intact visual systems, and different empirical effects demonstrating the two-way influence of perception and action. Conceiving of perception and action as part of a dynamical system emphasizes the property that small changes in their parameters may result in abrupt (catastrophic) changes in the qualitative type of its attractors. This featureand of dynamical systems provides a way ofsuggests incorporating  both continuity discontinuity within the system and ways of describing behavior in terms of striving for equilibrium equili brium as it unfolds in real-time. Most generally, generally, this functional scheme explains many of the dynami na mica call pe perrce cept ptio ionn-ac actio tion n co coup upli ling ngss th that at ar aree at th thee ve very ry he hear artt of em embo bodi died ed perception.

 

4 Concepts

A wonderful illustration of the importance of embodiment in ordinary concepts is evident in how Apache Indians in North America name the parts of automobiles (Basso, 1990). In the Couer d’Alene Indian language,

the ti the tirres of a car or tr tru uck be beco come me “w “wrrin inkl kled ed fe feeet, t,”” a ref efeerenc ncee to th thee pa patt tteern on their treads. The new knowledge of automobiles is likened to the old knowledge of the body. Basso has described an entire system of naming the parts of motorized vehicles in the language of the Western Apache of  east-central Arizona. The Western Apache have extended the names for the body parts of humans and animals to refer to the parts of automobiles and pickup trucks. In this structural metaphor, metaphor, the hood became the nose (“bichih”), the headlights became the eyes (“bidaa”), and the windshield  became the forehead (“bita”). The term for the face (“binii”) was extended to the whole area extending from the top of the windshield to the front  bumper,, so this term included the nose/hood and forehead/winds  bumper forehead/windshield hield as subparts. The front wheels became the hands and arms (“bigan”), and the rear wheels became the feet (“bikee”). All the items under the hood were classified as parts of the innards (“bibye”). Under the hood, the  battery became the liver (“bizig”), the electrical wiring the veins (“bit qqs”) qqs ”),, th thee ga gass ta tank nk th thee st stom omac ach h (“ (“bi bibi bid” d”), ), th thee di dist stri ribu buto torr th thee he hear artt (“ (“bi bijii jii”) ”),, the radiator the lung (“bijii izole”), and the radiator hoses the intes intestines tines (“bich’i”). There is an underlying conceptual metaphor, metaphor, MOTOR VEHICLES ARE HUMA HU MAN N BO BODIE DIES, S, wh whic ich h ex expr pres esse sess co corr rres espo pond nden ence cess be betw twee een n th thee pa part rtss of  hu huma man n bMAN eing ei ngs s an and dDIES the th e pa par rtstaph of car ca and an truc tr ucks In th the OTO O RkVE VEH HIC ICL Ltor ESr ARE AR E HU HUMA N BODI BO ES meta me phor or,r,sth the edth thin ing g ks. of.wh whic ich he M weOT spea sp eak (the (t he moto mo vehi ve hicl cle) e),, wi with th it itss co cons nsti titu tuen entt pa part rtss an and d re relat latio ions ns (i (its ts co cogn gniti itive ve to topo polo logy gy), ), is thee ta th tarrge gett do doma main in,, wh wher erea eass th thee th thin ing g wi with th wh whic ich h we ref efer er (h (hum uman an bo bodi dies es), ), with its own constituent parts and relations, is the source domain. The naming of vehicle parts with the names of human body parts preserved thee hi th hier erar arch chic ical al co cogn gnit itiv ivee st stru ruct ctur uree of re rela latio tions nshi hips ps am amon ong g th thee pa part rtss so th that at 79

 

80

 

Embodiment and Cognitive Science

 both the car’s body and the human body have “innards” that include, for instance, “liver.” Some scholars may argue that how people in one culture name parts of automobiles may not be the best place to find evidence of embodied concepts. But the Coeur d’Alene names for automobiles reveal a deepseated cognitive imperative to make sense of the world in terms of our  bodies (and in terms of embodied metaphor). By recognizing how parts of an automobile have a metaphorical relationship to human body parts (and functions), the Western Apache have created something new using their imaginative processes to the fullest. The Western Apache have in a sense reordered their experiences by looking at something differently than before. Yet this reordering of experience is not merely clich´ed, ed, or a one-shot mapping. Instead, there is a complex system of metaphorical correspo corre sponde ndence ncess tha thatt is gr groun ounded ded in peo people ple’s ’s or ordin dinary ary exp experi erienc ences es of the their ir own bodies. This chapter explores the importance of embodied experience in concepts and conceptual structure. I argue that significant aspects of both concrete and abstract concepts arise from, and continued to be structured in terms of, pervasive patterns of embodied activity. There is a growing literature from cognitive psychology and cognitive linguistics to support this contention. Much of this work is focused on inferring the existence

of embodiment from the analysis of linguistic statements and how people interp int erpre rett the them m (al (also so see Cha Chapte pters rs 5 and 6). Alt Althou hough gh tra tradit dition ional al exp experi erimen men-tall ps ta psyc ycho holog logis ists ts ar aree of ofte ten n sk skep epti tica call ab abou outt ma maki king ng cl claim aimss ab abou outt co conc ncep eptu tual al structures from from analyses of how people speak, the linguistic evidence demand ma ndss ex expl plan anat atio ion n wi with thin in a mo morre ge gene nera rall ps psyc ycho holo logi gica call ac acco coun untt of hu huma man n conceptual systems. As will be seen, recognition of the body is central to the study and description of concepts. Traditional Views of Concepts

The traditional view in psychology and philosophy argues that concepts are stored mental representations that enable people to identify objects and events in the real world. Classical theories assume that rules describe the objects in a category independent of situations (Bruner, Goodnow, & Austin,  1956). For example, a rule might attempt to capture the physical properties of chairs that are necessary and sufficient for membership apart apa rt fr from om the con contex texts ts in whi which ch cha chairs irs app appear ear.. Peo People ple pr presu esumab mably ly ide identi ntify fy certain features or attributes of objects, such as “that object has four legs and bar barks, ks,”” tha thatt mat match ch pr pre-e e-exis xistin ting g sum summar mary y re repr prese esenta ntatio tions ns in lon long-t g-term erm memo me mory ry,, su such ch as “A do dog g ha hass fo four ur le legs gs an and d ba bark rks. s.”” In th this is wa way y, co conc ncep epts ts an and d categories are defined by their relations to objects in the external world. Most Mo st th theo eori ries es of co conc ncep epts ts al also so as assu sume me th that at th ther eree is a si sing ngle le am amod odal al sy symb mbol ol to rep eprres esen entt a pr prop oper erty ty ac acro ross ss di diff ffer eren entt ca cate tego gori ries es.. Fo Forr ex exam ampl ple, e, th ther eree mu must st

 

Concepts

 

81

 be a conceptual symbol for the property “red” that is the same attribute in concepts as different as apples, wine, and fire trucks. Amodal symbols are language-independent, context-independent, and disembodied. Concepts are furthermore hierarchically organized in terms of their ob jective properties. “Chairs,” for instance, fall into the middle mi ddle level of the hierarchy with furniture as a superordinate level and specific chairs (e.g., rocking chairs) associated with the subordinate level. These “facts” presumably reflect the nature of things as they objectively exist in i n the world. Psychological theories try to explain the process of identifying any instance as belonging to a particular category and the way conceptual informati ma tion on is st stru ruct ctur ured ed in me memo mory ry to fa faci cilit litat atee un unde ders rsta tand ndin ing g of th thes esee ob obje ject ctss and events. A common method for uncovering the contents of any concept ce ptua uall re repr pres esen enta tati tion on is to pr pres esen entt pa part rtic icip ipan ants ts wi with th a wo word rd fo forr so some me co conncept and have them verbally list features typically true of it. In property verification tasks, participants read the word for a concept and verbally say whether a second word specifies a property true of the concept (e.g., “bird “bi rd”” – “wi “wings ngs”). ”). Psy Psycho cholog logist istss gen genera erally lly ass assume ume tha thatt in com comple pletin ting g the these se tasks, people access structured feature lists, propositions, frames, and semantic nets that contain only amodal features. The example of a conceptual category judged best is called a “prototype.” Categories are assumed to be mentally represented in terms of prototy to type pes, s, wi with th th thee de degr gree ee of ca cate tego gory ry me memb mber ersh ship ip be bein ing g de dete term rmin ined ed by th thee degree of similarity to the prototype. For example, sparrows are closer to thee pr th prot otot otyp ypic ical al bi birrd fo forr Am Amer eric ican anss th than an ar aree pe peng ngui uins ns or os ostr tric iche hes, s, an and d th this is make ma kess it ea easi sier er fo forr pe peop ople le to ve veri rify fy st stat atem emen ents ts su such ch as “A sp spar arrrow is a bi birrd”

than to verify “A penguin is a bird” (Rosch, 1975; Rosch & Mervis, 1975). These empirical results are not due simply to some exemplars being more common than others, because even rare instances of a category may be clos cl oser er to th thee pr prot otot otyp ypee th than an mo morre fr freq eque uent nt ex exam ampl ples es.. Th Thus us,, pe peop ople le ra rate te ra rare re items of furniture such as “love seat,” “davenport,” and “cedar chest” as  being better exemplars of the category “furniture” than they do frequently encountered objects such as “refrigerator” (Rosch, 1975). Problems with the Traditional View

There are several major problems with the traditional view of concepts. First, evidence shows that people represent represent certain properties quite differently en tly in di difffe ferren entt co cont ntex exts ts.. Fo Forr in inst stan ance ce,, th thee co conc ncep eptt of “fi “firre” is pr pres esum umab ably ly anabstractrepresentationthatarisesfromallconcreteinstancesofhowfires are understood in specific contexts. Yet Yet typicality judgments often vary as a function of context (Roth & Shoben,  1983). “Tea” “Tea” is judged to be a more typical beverage than “milk” in the context of secretaries taking a break,  but the opposite is true in the context of truck drivers taking a break. “Birds” that are typical from an American point of view, such as robins

 

82

 

Embodiment and Cognitive Science

and eagles, are atypical from the point of view of an average Chinese citizen (Barsalou & Medin, 1986). These findings suggest that prototypes are clo closel sely ys tie tied d to individ ind ividual ual contex texts ts and are ar e not necess arily ly abstra abs ctke repr re prese esenn-r tati ta tion ons that th at emer em erge ge from fr om con spec sp ecifi ific c in inst stan ance ces s ofnec any an yessari conc co ncep ept. t. Ttract o ta take anot an othe her example, the property “red” is represented differently in apples, lettuce, pota po tato toes es,, an and d wi wine ne (H (Hal alff ff,, Or Orto tony ny,, & An Ande ders rson on,, 1976). Pe Peop ople le li like kely ly rep eprresent se nt th thee sa same me pr prop oper erty ty lo loca cally lly in di diff ffer eren entt co cont ntex exts ts,, ra rath ther er th than an gl glob obal ally ly as a sing si ngle le sy symb mbol ol (i (i.e .e., ., th thee lo loca call fo form rm as assu sump mptio tion) n) (S (Sol olom omon on & Ba Bars rsal alou ou,, 2001). Concepts are not direct reflections of things in nature, contrary to the traditional view. view. Concepts do not directly preserve aspects of the external objects they refer to, contrary to the assumption that concepts are amodal symbols. For example, certain cognitive categories in the middle of taxonomic hierarchies (i.e., basic-level categories) can be explained in terms of certain nonobjective properties. Empirical research has shown that the  basic level is special for the following reasons reasons (Lakoff, 1987): (1)  It is the highest level at which category members have similarly perceived overall shapes. For example, you can recognize a chair by its overall shape. But there is no overall shape that you can assign to a ge gene nera rali lize zed d pi piec ecee of fu furn rnit itur uree su such ch th that at yo you u co coul uld d rec ecog ogni nize ze th thee category from the shape. (2)  It is the highest level at which a single mental image can represent the entire category. You can form a mental image of a chair. You can gett me ge ment ntal al im imag ages es of op oppo posi sing ng ca cate tego gori ries es at th this is le leve vel, l, su such ch as ta tabl bles es and beds. But you cannot get a mental image of a general piece of  furn fu rnitu iturre th that at is no nott so some me pa part rtic icula ularr pi piec ecee of fu furn rnitu iturre, su such ch as a ta tabl blee or bed. (3)  It is the highest level at which a perso person n uses similar motor actions actions in in inte tera ract ctin ing g wi with th ca cate tego gory ry me memb mber ers. s. Pe Peop ople le ha have ve mo moto torr pr prog ogra rams ms forr in fo inte tera ract ctin ing g wi with th ob obje ject ctss at th thee ba basi sicc le leve vell – in inte tera ract ctin ing g wi with th ch chai airs rs,,

tables, and beds. There are no motor programs for interacting with generalized pieces of furniture. (4)   It is th thee le leve vell at wh whic ich h mo most st of ou ourr kn know owle led dge is or org gan aniz ized ed.. Thi hink nk of  alll th al that at yo you u kn know ow ab abou outt ca cars rs ve vers rsus us al alll th that at yo you u kn know ow ab abou outt ve vehi hicl cles es.. You know a handful of things about vehicles, but a huge number of things about cars. It is at the basic level that most of our useful information and knowledge is organized. Thesee ob Thes obse serv rvat atio ions ns ex expl plai ain n wh why y th thee ba basi sicc le leve vell of ca cate tego gori ries es ha hass pr prio iori rity ty over the superordinate and subordinate levels. Most generally, the basic level is the level at which people interact optimally with their environment, giv men iven en th thee ki kin nds of bod odie iess an and d bra rain inss th that at th theey hav avee and th thee env nvir iron on-ments they inhabit. For example, to decide that a particular couch belongs to the category “things that can fit through the front doorway,” a good strategy is to manipulate an analog representation of the couch’s shape in relation to an analog representation of the doorway (Barsalou,  2001).

 

Concepts

 

83

Representations that preserve perceptual properties, as suggested above, aree oft ar often en mor moree ef effici ficient ent tha than n pur purely ely sym symbol bolic ic re repr prese esenta ntatio tions, ns, bec becaus ausee the they y do not require external constructs to ensure proper inferences (Barsalou, 1999a). These facts can only be explained in terms of human embodiment and raise serious doubts about whether prototypes are abstract, pre-

existing conceptual repr representations. esentations. A related problem for the traditional view of concepts is that prototype theo th eory ry as assu sume mess th that at ca cate tego gory ry me memb mber ersh ship ip is de dete term rmin ined ed by wh whet ethe herr so some me candidate is sufficiently similar to the prototype, or a set of already represented examples, where similarity is based on matches and mismatches of independent, equally abstract features. Yet similarity does not explain many kinds of prototype effects. Thus, goal-derived categories, such as “foods to eat while on a diet” and “things to take on a camping trip,” reveal the same typicality effects as do other categories (Barsalou,  1983, 1985, 1989, 1991). The basis for these effects is not similarity to some prototype, but rather similarity to an ideal. For instance, typicality ratings for the category of “foods to eat while on a diet” are determined by realworld knowledge such as how clearly each example conforms to the ideal of zero calories. This real-world knowledge, including temporary analog representations, repr esentations, is used to reason about or explain properties of concepts, not simply to match them to some pre-existing, abstract prototype. Even though categories such as “things to take on a camping trip” have prototypic structure, such a structure does not exist in advance because the category is “ad hoc” and not conventional (Barsalou, 1990). In fa fact ct,, th ther eree ar aree ma many ny wa ways ys in wh whic ich h pr prot otot otyp ypes es ma may y be fo form rmed ed (R (Ros osch ch,, 1999). Some may be based on statistical frequencies, such as the mean or number (for family resemblance structures) of various attributes. Others appear to be ideals made salient by factors such as physiology (good color, form),formal social structure (ideal foods to eat good on a diet), structures(president, (multiplesteacher), of ten ingoals the decimal system), and individual experience (the first learned, the most recently encountered, or one made salient because it is especially meaningful, emotiona tio nal, l, or in inte tere rest stin ing) g).. Pr Prot otot otyp ypes es ar aree no nott su summ mmar ary y ab abst stra ract ctio ions ns ba base sed d on a

few de few defin finin ing g at attr trib ibut utes es,, bu butt ar aree ri rich ch,, im imag agis isti tic, c, se sens nsor ory y, fu full ll bo bodi died ed me ment ntal al events. One re reaso ason n cog cognit nitive ive sci scient entist istss mis mistak takenl enly y bel believ ievee tha thatt con concep cepts ts mus mustt be pre-e pr e-exis xistin ting g men mental tal str struct uctur ures es is tha thatt the they y com commit mit the “ef “effec fects ts = structures” fallacy (Gibbs,  1994;  Gibbs & Matlock,  2000;  Lakoff,  1987). This fallacy reflects the belief that the goodness-of-example ratings, for example, obtained in psychological experiments are a direct reflection of degree of  category membership. But the “effects   =  structures” interpretation cannot account for many of the types of data reviewed above, especially the problems of complex categorization (Gibbs,  1994). In fact, many kinds of  prototype effects can be explained by other principles (e.g., metaphoric and metonymic reasoning) that do not assume that the effects obtained in

 

84

 

Embodiment and Cognitive Science

experiments reflect the structure of pre-existing knowledge. For example, many instances of complex categorization have prototype effects that are grounded in reference-point reasoning that reflects cases where a part of  some so me ca cate tego gory ry st stan ands ds fo forr th thee en enti tire re,, mo more re co comp mple lex x ca cate tego gory ry (L (Lak akof off, f, 1987). Thus,, th Thus thee pr prot otot otyp ypica icall ca case se of a “h “hou ouse sewi wife fe mo moth ther er”” is us used ed to rea easo son n ab abou outt the more complex category of “mother “mother,” ,” because of the t he salience of nurturanc ra ncee to th thee co conc ncep eptt of mo moth ther er.. Ot Othe herr in inst stan ance cess of mo moth ther ers, s, su such ch as “u “unw nwed ed moth mo ther ers, s,”” “s “sur urrrog ogat atee mo moth ther ers, s,”” “s “sta tage ge mo moth ther ers, s,”” an and d so on on,, ar aree al alll de deri rive ved d from the central, prototypical case. But this process of part-for-whole reasoning creates the less prototypical instances in the moment of thinking, and an d th they ey ar aree no nott si simp mply ly rea ead d of offf an en endu duri ring ng co conc ncep eptu tual al ma map p in lo long ng-t -ter erm m memory. Much research points to the flexibility of concepts, which is difficult to reco re conc ncile ile wi with th tr trad adit itio iona nall vi view ewss of co conc ncep epts ts as ab abst stra ract ct,, di dise semb mbod odie ied d sy symm bols. One set set of studies asked people to provide definitions for categories, such as bachelor, bird, and chair (Barsalou, 1995). An analysis of the overlap la p in th thee fe feat atur ures es pa part rtic icip ipan ants ts pr prov ovid ided ed fo forr a gi give ven n ca cate tego gory ry rev evea eale led d th that at on average only 47% of the features in one person’s definitions for a category existed in another person’s definition. A great deal of flexibility also exists within individuals when they are asked to provide definitions for conc co ncep epts ts.. Wh When en pa part rtic icip ipan ants ts in th thee ab abov ovee st stud udy y re retu turn rned ed 2 we week ekss la late terr an and d defined the same categories again, only  66 % of the features noted in the first fir st se sess ssio ion n we were re pr prod oduc uced ed ag again ain in th thee se seco cond nd se sess ssio ion. n. Th Thes esee res esult ultss in indi di-cate that that substantial flexibility exists in how a person conceptualizes the same category on differen differentt occasions (Barsalou, 1995). The significant flexibility shown by many experiments on defining categories arises not from differences in knowledge, but from differences in thee re th retr trie ieva vall of th this is kn know owle ledg dgee fr from om lo long ng-t -ter erm m me memo mory ry.. On di diff ffer eren entt oc occa ca-sion si ons, s, di diff ffer eren entt in indi divi vidu dual alss re retr trie ieve ve di diff ffer eren entt su subs bset etss of fe feat atur ures es fr from om th thei eirr extensive ofs aofcategory. the same way, an individual may ret etri riev evee di difknowledge ffe ferren entt as aspe pect cts hiss or he hi herrIn know kn owle ledg dgee of a ca cate tego gory ry on di difffe fer ren entt occasions. For instance, the statement “The Christmas bird fed 12 people” make ma kess en ency cycl clop oped edic ic in info form rmat atio ion n ab abou outt tu turk rkey eyss an and d ge gees esee mo most st ac acce cess ssib ible le,, wher wh erea eass se seab abir irds ds wo woul uld d be mo most st ac acce cess ssib ible le gi give ven n a st stat atem emen entt su such ch as “T “The he  bird followed the boat boat out to sea.” Our ordinary experience with objects in the real world does not demand that we classify them. For example, my interactions with beds in-

volve sleeping in them, moving them, and making them up, as well as, under special circumstances such as being in a psychology experiment, classi cla ssifyi fying ng the them m as bel belong onging ing to the cat catego egory ry of “fu “furni rnitur ture” e” (Mu (Murph rphy y, 2002; Ross, 1999). People may certainly learn a good deal about beds from interacting with them in different ways that surely influence their concepts for “beds.” One set of studies had people categorize patients based on their symptoms and then prescribe appropriate medicines based on these

 

Concepts

 

85

symptoms (Ross, 1990). The symptoms that were most critical to the prescriptions were as important as the symptoms themselves when people categorized the patients into different groups. Thus, what people learn from their interactions with items is incorporated into their concepts, and consequently affects categorization. Not surprisingly, other work shows that people who actually work with category members (e.g., tree experts) develop representations representations that reflect their interactions with objects (Medin et al., 1997). This is why, for example, biologists and laypersons have different concepts of tree types. These studies on the effect of category use on categorization tasks illustrate tra te how peo people ple’s ’s ric rich h int intera eracti ctions onswith withthe the wor world ld sha shape pe kno knowle wledge dge acq acquiuisition and representation. Psychologists conducting these studies discuss people’s “interactions with the world” as a kind of “background knowledge.” But some of these studies may reflect something of people’s em bodied understanding understanding of objects and events, and not simply simply their abstract abstract knowledge or beliefs. Consider one set of experiments in which people were shown pictures of people performing actions on objects (Pazzani,  1997). Each picture depicted an adult or child doing an action on an uninflated balloon that was either large or small, and either yellow or purple. The action required was either to drop the balloon in a cup of water or to stretch it. Participants had to learn one of two types of categories, disjunctive and conjunctive. The disjunctive category was based on the rule “the person must be an adult OR the action must be stretching the balloon.” The conjunctive category was defined by the rule “the color must be yellow AND the balloon must be small.” Earlier research indicated that conjunctive categories are easier to learn than disjunctive ones (Bruner, Goodnow, & Austin,  1956). But participants here were also instructed either to learn “Category Alpha” or to identify those balloons that would inflate. “Category Alpha” instructions presumably would not activate people’s background knowledge about balloons, but the second task should prompt participants to think about inflating balloons. In fact, when people received the “Category Alpha” instructions, they found the conjunctive category easier to learn than the disjunctive one. But when given the inflate instructions, participants learned disjunctive categories much faster, even more easily than did the Alpha people learning l earning conjunctive categories. This pattern of data demonstrates the strong effect that background knowledge (i.e., about balloon inflation) can have on category learning. Yet people probably do not have stored, declarative knowledge about the exact properties associated with balloon inflation. Alternatively, Alternatively, being instructed to identify which balloons most easily inflate prompts people to run embodied simulations simulations about inflati inflation on given each balloon. My claim

is that discussions of background knowledge effects on category learning may, in some cases, show the importance of embodied simulations,

 

86

 

Embodiment and Cognitive Science

rath ther er th than an th thee ac acti tiva vati tion on of pr pres esto tore red, d, ab abst stra ract ct,, de decl clar arat ativ ivee kn know owle ledg dge, e, in ra making categorization judgments. These observations suggest that concepts are temporary constructions in wo work rkin ing g me memo mory ry,, ba base sed d on em embo bodi died ed si simu mula lati tion ons, s, an and d no nott st stab able le st stru rucctures stored in long-term memory. memory. One possibility is that concepts may be defined as statistical patterns in sensory-motor systems that take different forms in different context. Human conceptual systems evolved to support embodied action in the environment. Let us consider this idea in more detail. Perceptual Symbols

A significant new goal in cognitive science is to establish how perceptual processes guide the construction of concrete and abstract concepts. The  best example of this work is the development development of the theory theory of “perceptual “perceptual symbol systems” (Barsalou, 1999a, 2003). Perceptual symbols are derived from the representations generated from perceptual input systems, but are acquired by performing operations on perceptual representations and are similar to these operations. Thus, perceptual symbols are schematic, yet maintain some of the structure of the perceptual representations representations from which wh ich th they ey we were re de deri rive ved. d. Un Unlik likee am amod odal al co conc ncep epts ts,, pe perrce cept ptua uall sy symb mbol olss ar aree nonarbitrary,, given their similarity to the objects they represent. nonarbitrary Perceptual symbols are also multimodal and include information from the five senses, along with proprioceptive and kinesthetic information. Perceptual symbols are not necessarily conscious images, but are unconscio sc ious us st stat ates es of pe perrce cept ptua uall sy syst stem emss sp spec ecifie ified d ne neur urall ally y (B (Bar arsa salo lou, u, 2003). For inst in stan ance ce,, th thee re repr pres esen enta tati tion on of a ch chai airr mig might ht be sp spec ecifie ified d as a co confi nfigu gura rati tion on of neurons active in the visual system, rather than as a conscious mental image. These perceptual representations are not necessarily holistic, but can reflect selective aspects of a perceptual state extracted via selective attention and stored in long-term memory (see sensorimotor contingency theory in Cha theory Chapte pterr 3). Th Thus us,, se sele lect ctiv ivee at atte tent ntio ion n mi migh ghtt fo focu cuss on th thee fo form rm of an obje ob ject ct,, st stor orin ing g on only ly it itss sh shap apee in me memo mory ry,, an and d no nott its co colo lorr, te text xtur ure, e, po posi sitio tion, n, size, and so forth. This schematic extraction process not only operates on sens se nsor ory y st stat ates es,, bu butt al also so wo work rkss on in inte tern rnal al me ment ntal al ev even ents ts,, ex extr trac actin ting g as aspe pect ctss of representation states, motivational states, and emotions. When there is no sensory input, activation of the conjunctive neurons partially re-enact, or reinstantiate, the earlier visual stimuli. These re-enactments, or simulations, are specific skills that serve as the foundational mechanism for providing context-specific representations of a category. Perhaps the most interesting aspect of perceptual symbol theory is the idea that conceptual processing involves sensorimotor simulations. Under this view, concepts are not understood and stored as abstract, disembodied symbols, because crucial elements of relevant perceptual perceptual and sensorimotor information are used in conceptual processing.

 

Concepts

 

87

In fact, empirical evidence supports the claims of perceptual systems theory. First, evidence from cognitive neuroscience suggests that concepts are grounded in sensory-motor regions of the brain (Damasio, 1989; Damasio & Damasio, 1994; Gainotti et al., 1995; Martin, Ungerleider, & Haxby,  2000;  Martin et al.,  1996;  Pulvermueller,  1999;  Rosler, Heil, & Hennighausen,  1995;  Tranel, Damasio, & Damasio,  1997;  Warrington & Shallice, 1984). For example, functional imagery studies demonstrate that processing man-made objects activates the left ventral premotor cortex (Ger (G erla lach ch,, La Law w, & Pa Pauls ulson on,, 2002; Gr Graf afto ton, n, Fa Fadi diga ga,, Ar Arbi bib, b, & Ri Rizz zzol olat atti ti,, 1997; Martin et al.,  1996). Comprehension of man-made objects may therefore depend on motor-based knowledge of object utilization (action knowledge). In one case study, for example, patients with semantic dementia were better able to name objects when they also demonstrated, or mimed, the object’s use than when simply asked to name the object alone (Coc (C occi ciaa et al al., ., 2004). Ot Othe herr ne neur uroi oima magi gini ning ng st stud udie iess sh show ow th that at ac acti tion on kn know owlledge is important for the processing of manipulatable objects in general, regardless of whether these are man-made (articles of clothing) or not (fruits/vegetables) (Gerlach et al.,  2002). This latter finding is contrary to theories that maintain that concepts are represented in taxonomic categories gor ies,, but is con congr gruen uentt wit with h per percep ceptua tuall sym symbol bol the theory ory,, in whi which ch cat catego egorie riess differ in the weight they put on different forms of knowledge in varying tasks. One challenge for psychological theories of concepts concerns productivity tiv ity,, or the abi abilit lity y to com combin binee re repr prese esenta ntatio tions ns to for form m mor moree com comple plex x str strucucture tu res. s. Tra radi ditio tiona nall th theo eori ries es ha have ve di diffi fficu cult lty y ex expl plai aini ning ng ho how w co conc ncep epts ts ma may y be comb co mbin ined ed in na natu turral wa ways ys.. Fo Forr in inssta tanc ncee, A an and d B ma may y be ea eassil ily y bl blen ende ded d to to-geth ge ther er,, bu butt X an and d Y ca cann nnot ot.. Pe Peop ople le’s ’s un unde ders rsta tand ndin ing g of ho how w co conc ncep epts ts me mesh sh together in the real world also constrains their conceptual understanding of events. Consider the expressions “The lamp is above the table” and “The table is below the lamp” (Solomon,  1997). These differe different nt sentences expr ex pres esss di difffe ferren entt sc sche hema mati ticc im imag ages es.. Fo Forr “T “The he la lamp mp is ab abov ovee th thee ta tabl ble, e,”” ou ourr embodied understanding of “above” is a schematic image having a top and a bottom region where our attention is focused on the top. For “The table is below the lamp,” our understanding of “below” also contains a top and bottom regions, but, this time, our attention is focused on the bottom. to m. Th Thes esee di diff ffer eren entt co cons nstr trua uals ls of th thee rel elat atio ions nshi hip p be betw twee een n th thee “l “lam amp” p” an and d “table” are created by meshing types from memory with tokens from the percei per ceived ved env envir ironm onment ent.. Und Under er thi thiss vie view w, we cr creat eatee dif differ ferent ent pr propo oposit sition ional al construals const ruals by deter determining mining which pattern of action from memory can be used to mesh with extant properties of the environment (Barsalou, 1999 b  b;; Glenberg, 1997). Perceptual systems theory assumes that concepts can be combined according to the constraints by which objects can be physically manipulated in the world (Barsalou, in press). One study in support of this idea exam ex amin ined ed th thee fe feat atur ures es pa part rtic icip ipan ants ts rep epor orte ted d fo forr no noun un ph phra rase sess in wh whic ich h th thee

 

88

 

Embodiment and Cognitive Science

same modifier (e.g., “half”) revealed the insides of an object (e.g., “half  watermelon”), or kept the insides of an object hidden (e.g., “half smile”) (Wu & Barsalou, 2001). Participants produced more internal features (e.g., red, seeds) when describing noun phrases revealing the insides of an ob ject than for noun phrases that kept the insides occluded. These findings support the idea that participants mentally simulate how concepts (“half” with wi th “w “wat ater erme melo lon” n” or “s “smi mile le”) ”) mi migh ghtt be co comb mbin ined ed in th thee rea eall wo worl rld d wh when en they are asked to describe the features of different objects. A different series of studies on conceptual combinations examined the role of perceptual similarity in categorization judgments (Solomon & Barsalou,  2001). These studies manipulated perceptual similarity across different tests. One participant, for example, first verified the conceptproperty pair “PONY-mane,” and later either “HORSE-mane” or “LIONmane.” If people perceptually simulate the concept to verify the property, property, then th en th they ey sh shou ould ld be fa fast ster er to ve veri rify fy th thee “H “HOR ORSE SE-m -man ane” e” se sequ quen ence ce th than an th thee “LION-mane “LION -mane”” seque sequence, nce, because horse manes are mor moree simila similarr to pony manes than are lion manes. This is exactly what happened. When participants process the “HORSE-mane” pair, they are reminded of the earlier pair involving manes, which either facilitates or inhibits priming depending in g on th thee ty type pe of ma mane ne su sugg gges este ted. d. Th Thes esee fin findi ding ngss de demo mons nstr trat atee ho how w pe peop ople le performing verification tasks perceptually simulate a concept, and do not simply activate its abstract features. Another set of studies showed that verifying a property in the auditory modality (e.g., “BLENDER-loud”) is slower after verifying a property in a differentt modality (e.g., “CRANBERRY-tart”) than after verifying a propdifferen erty ert y in the sam samee mod modali ality ty (e. (e.g., g., “LE “LEA AVES VES-ru -rustl stling ing”) ”) (Pe (Peche cherr, Zee Zeelen lenber berg, g, & Barsalou, 2003). Th Thus us,, sw switc itchi hing ng mo moda dalit litie iess in co conc ncep eptu tual al pr proc oces essi sing ng ta take kess effort, which is clearly contrary to the idea that conceptual knowledge is represented in an amodal manner. Percep Per ceptua tuall sym symbol bolss the theory ory ass assume umess tha thatt eve even n whe when n par partic ticipa ipants nts re recei ceive ve nonp no npic icto tori rial al ma mate teri rial alss an and d ar aree no nott as aske ked d to us usee im imag ager ery y, th they ey ne neve vert rthe hele less ss perform perceptual simulations spontaneously spontaneously.. Concepts arise as on-thefly si simu mula latio tions ns of ev even ents ts (B (Bar arsa salo lou, u, 2002, 2003). In st stan anda dard rd ca cate tego gori riza zati tion on tasks, for example, people first simulate referents of the concept perceptually and then scan their simulations to produce the required information (e.g (e .g., ., lis listt th thee fe feat atur ures es of “c “cha hair irs” s”). ). Pa Part rtic icip ipan ants ts’’ re resp spon onse sess in ty typi pica call ca cate tego go-rization tasks do not, therefore, reflect the pure amodal contents of some pre-existing, taxonomic mental representation. This conclusion is consistent with related work showing that when people are shown an object to be categorized, they are often reminded of a superficially similar ob ject. Once reminded, they try to come up with an abstract description of  the category that encompasses both objects (Ross, Perkins, & Tenpenny, 1990). Another implication of perceptual symbol theory is that if a conceptualization attempts to simulate a perceptual experience, then it should

 

Concepts

 

89

typically simulate a situation, situati on, because situations are intrinsic parts of percept ce ptua uall ex expe peri rien ence ce.. Fo Forr in inst stan ance ce,, im imag agin inin ing g a ch chai airr in a li livi ving ng roo oom m ev evok okes es

a very different chair than imagining a chair in a jet. One study nicely illust lu stra rate tess ho how w pe peop ople le ima imagin ginee th them emse selv lves es in co conc ncrret etee si situ tuat atio ions ns to pr prod oduc ucee exempl exe mplars ars of con concep cepts ts (V (Vall alleeee-T Tour ourang angeau eau,, Ant Anthon hony y, & Aus Austin tin,, 1998). Pa Parrticipants generated exemplars from common taxonomic categories, such as furniture and fruits, and from ad hoc categories such as “things dogs chase” and “reasons for going on a holiday.” Afterward, participants described the strategies they used in generating these examples. Several kinds of strategies were reported. “Experiential mediation” involved vol vedre retri trievi eving ng an aut autobi obiogr ograph aphica icall mem memory ory of a sit situat uation ion tha thatt con contai tained ned individuals from the target category, and then reporting the category to which this individual belonged. When generating types of fruit, for example, participants first retrieved a memory of a grocery store, scanned across it, and reported the types of fruit present in the produce section. “Semantic “Sema ntic mediation,” mediation,” on the other hand, involv involved ed first ret retrievin rieving g a detached taxonomy that contained the target category and then reporting its subcategories. Thus, when generating examples of fruit, people first retrieved the fruit taxonomy and then reported subtypes, such as tropical fruit, dried fruit, and citrus fruit. Analysis of participants’ self-reported strategies showed that people us used ed “e “exp xper erie ient ntia ialcommon l me medi diat atio ion” n” ab abou outt th thr ree tim times es ascategories. ofte of ten n as “s “sem eman ticc surme-me diation” for both taxonomic and ad hoc It isanti not pris pr isin ing g th that at si situ tuat atio ions ns ar aree im impo port rtan antt fo forr ad ho hocc ca cate tego gori ries es,, gi give ven n th that at th thes esee catego cat egorie riess ari arise se out of goa goal-d l-dir irect ected ed act activi ivity ty in spe specific cificcon contex texts. ts. Muc Much h mor moree surprising is that concrete situations were reported just as often for common mo n ta taxo xono nomi micc ca cate tego gori ries es,, su sugg gges estin ting g th that at th they ey,, to too, o, ar aree or orga gani nize zed d ar arou ound nd situations. In ge gene nera ral, l, re rese sear arch ch rel elat ated ed to pe perc rcep eptu tual al sy symb mbol ol th theo eory ry su sugg gges ests ts a ne new w view of concepts that explains the perceptual grounding for the creation and retrieval of conceptual knowledge (see Barsalou,  2001). A conceptualization of a category typically includes background information based partly on people’s embodied simulations or acting in life-like situations. This re-enactment or simulation is not necessarily complete. But each conceptualization represents a category in a way that is relevant to the background situation, such that differe different nt conceptualizations represent the category differently. In this way, perceptual symbols theory suggests how concepts arise in the moment from a tight coupling of cognitive and motoric processes. Finally, perceptual symbols are involved with the representation of  abst ab stra ract ct co conc ncep epts ts.. Co Conc ncep epts ts su such ch as tr trut uth, h, be beau auty ty,, an and d vi virt rtue ue ar aree no nott un unde derrstoo st ood d as si sing ngle le im imag ages es,, an and d ea each ch ma may y ha have ve mu mult ltip iple le rep eprres esen enta tati tion onss in di diffferren fe entt co cont ntex exts ts.. Th Thee co conc ncep eptt of “b “bea eaut uty” y” ma may y in invo volv lvee ou ourr ab abili ility ty to pe perc rcep ep-tual tu ally ly id iden entif tify y in inst stan ance cess by th thei eirr ap appe pear aran ance ce an and d th thee in inte tern rnal al st stat ates es,, su such ch as emotional feelings, that they produce in us. Abstract conceptual relations

 

90

 

Embodiment and Cognitive Science

may be characterized as the manipulation of perceptual symbols. For instance, consider the idea of a “counterfactual” (Prinz & Barsalou, 2002). A counte cou nterfa rfactu ctual al tho though ughtt is a way of for formin ming g per percep ceptua tuall sim simula ulatio tions. ns. You can have a counterfactual thought if you form a simulation representing some nonactual state of affairs, P, and you are disposed, upon doing so, to add a representation of another state of affairs, Q, to your repre representation. sentation. This

is a way of representing the fact that Q could not have occurred if P had  been the case. Counterfactual thought need not, then, depend on specific images, but is one way of engaging in a kind of perceptual simulation. Explaining abstract concepts and how they arise in the mind remains onee of th on thee gr grea eate test st ch chal alle leng nges es fo forr co cogn gnit itiv ivee sc scie ienc nce. e. Pe Perc rcep eptu tual al sy symb mbol olss ar aree on the right track, for the most part, in describing how abstract concepts may ma y ha have ve mu multi ltipl plee rea eali liza zati tion ons, s, as is th thee ca case se fo forr al alll co conc ncep eptu tual al pr proc oces essi sing ng.. Moreover, this view also correctly claims that abstract ideas have a perceptual, perhaps embodied, basis. As we will now see, the growing belief  that concepts are tied to perceptual symbols is most dramatically voiced in cognitive linguistics. Image Schemas and the Metaphorical Nature of Abstract Concepts

Cognitive Cognit ive lin lingui guisti stics cs emb embrac races es the imp import ortant ant ide ideaa tha thatt ling linguis uistic tic str struct uctur ures es are related to, and motivated by, human conceptual knowledge, bodily experience, andcognitive the communicative functionslook of discourse. Unlike generative linguists, linguists explicitly for language-mind and language-mind-body linkages in their descriptions of linguistic structure andlinguisticbehavior.Akeypartofthisworkclaimsthatmanyofourconcepts cep ts ar aree gr groun ounded ded in, and str struct uctur ured ed by by,, var variou iouss pat patter terns ns of our per percep ceptua tuall interactions, bodily actions, and manipulations of objects (Johnson,  1987; Lakoff, 1987; Lakoff & Johnson, 1999; Talmy, 1988, 2000). Specific patterns of force dynamics underlie our embodied understandings of abstract concepts (T (Talmy almy,, 1988, 2000). Fo Forc rces es ar aree vi view ewed ed as ph phys ysic ical al,, em embo bodi died ed en enti titie tiess (an (a n ag agon onis ist) t) ac acti ting ng in co comp mpet etiti ition on ag agai ains nstt ot othe herr fo forrce cess (a (an n an anta tago goni nist st), ), wi with th each ea ch en enti tity ty ha havi ving ng va vary ryin ing g st stre reng ngth thss an and d te tend nden enci cies es.. We un unde ders rsta tand nd th thes esee entities primarily from our own bodily experiences such as pushing and  being pushed, moving objects, and feeling the forces acting within our  bodies as we move about the environment. These patterns are experiential gestal ges talts, ts, cal called led “im “image age sch schema emas,” s,” tha thatt eme emerg rgee thr throug oughou houtt sen sensor sorimo imotor tor activity as we manipulate objects, orient ourselves spatially and temporally, temporally, and direct our perceptual focus for various purposes. Image schemas can generally be defined as dynamic analog representations of spatial relations and movements in space. Even though image schemas are derived from perceptual and motor processes, they are not themselves sensorimotor processes. Instead, image schemas are “primary mean me anss by wh whic ich h we co cons nstr truc uctt or co cons nsti titu tute te or orde derr an and d ar aree no nott me merre pa pass ssiv ivee

 

Concepts

 

91

receptacles into which experience is poured” (Johnson,  1987:  30 ). In this way, image schemas are different from the notion of schemata traditionally al ly us used ed in co cogn gnit itiv ivee sc scie ienc nce, e, wh whic ich h is of ab abst stra ract ct co conc ncep eptu tual al an and d pr prop opos osiitional event structures (see Rumelhart,  1980). By contrast, image schemas are imaginative, nonpropositional structures that organize experience at the level of bodily perception and movement. Image schemas exist across all perceptual modalities, something that must hold for there to be any sensorimotor coordination in our experience. As such, image schemas are at once visual, auditory, kinesthetic, and tactile. At the same time, image sche sc hema mass ar aree mo more re ab abst stra ract ct th than an or ordi dina nary ry vis visua uall me ment ntal al im imag ages es an and d co cons nsis istt

of dynamic spatial patterns that underlie the spatial relations and movement found in actual concrete images. Studies in cognitive linguistics suggest that at least two dozen different image schemas and several image schema transformations appear regularly in people’s everyday thinking, reasoning, and imagination (Johnson, 1987;  Lakoff,  1987). Among these are the schematic structures of CONT C ONTAINER, AINER, BALANCE, SOURCESOURCE-P PATH-GOAL, PATH, CYCLE, C YCLE, ATTRACTION, TRACTI ON, CENTE CENTER-PER R-PERIPHER IPHERY Y, and LINK. These image schem schemas as cover a wide range of experiential structures that are pervasive in experience, have internal structure, underlie literal meanings, and can be metaphorically elaborated to provide for our understanding of more abstract conceptual domains. Consider the SOURCE-PATH-GOAL schema. This schema first develops as we le ops lear arn n to fo focu cuss ou ourr eye yess and tr trac ack k fo forrms as th they ey mo mov ve th thrrou ough ghou outt ourr vi ou visu sual al fie field ld.. Fr From om su such ch ex expe peri rien ence ces, s, a rec ecur urri ring ng pa patt tter ern n be beco come mess ma mannifest in tracking a trajectory from point A to another point B. Later on, as we move our bodies in the real world, ranging from experiences of reaching for objects to moving our entire bodies from one location to another, moree varie mor varied d SOUR SOURCE-P CE-PA ATH-GO TH-GOAL AL expe experienc riences es becom becomee salie salient. nt. Althou Although gh SOURCE-PATH-GOAL experiences may vary considerably (e.g., many objects, shapes, types of paths traveled), the emergent image-schematic structure of SOURCE-PATH-GOAL supports literal meanings, such as seen in “He walked across the room to the door,” and can be metaphorically projected onto more abstract domains of understanding and reasoning (Johnson, 1987). This metaphorical mapping preserves the structural characteristics or the cognitive topology of the source domain (Lakoff, 1990). Thus, the SOURCE-PATH-GOAL schema gives rise to conceptual metaphors such as PURPOSES ARE DESTINATIONS, DESTINATIONS, which preserve the main ma in st stru ruct ctur ural al ch char arac acte teri rist stic icss of th thee so sour urce ce do doma main in (i (i.e .e., ., SO SOUR URCE CE-P -PA ATH TH-GOAL). English is replete with conventional expressions that illustrate this underlying metaphorical conceptualization. For instance, we start off to get our Ph.D.s, but along the way we get sidetracked or led astray, and are diverted from our original goal. We try to get back on the right path and

 

92

 

Embodiment and Cognitive Science

to keep the end in view as we move along. Eventually we may come a long way and reach our goal (Johnson, 1993). It is simply not an arbitrary fact of English that we talk about our lives and careers in terms of sources, paths, and goals; rather rather,, we metaphorically conceptualize our experiences through very basic, bodily experiences in the world that are abstracted to form higher-level metaphoric thought. This way of talking about experiencee sho enc shows ws how the PUR PURPOS POSES ES ARE DES DESTIN TINA ATIO TIONS NS met metaph aphor or,, re resul sultin ting g from a very basic image-schematic structure, is constitutive of our understanding of intentional action. Some creative instantiations of the SOURCE-PA SOURCE-PATH-GOAL schema are seen se en in po poet etry ry.. Co Cons nsid ider er an ex exce cerp rptt fr from om a po poem em,, by th thee Ch Chil ilea ean n po poet et Pa Pabl blo o Neruda, titled “Ode and Burgeonings” (Neruda, 1972): My wild girl, we have had to regain time

and march backward, in the distance of our lives, kiss after kiss, gathering from one place what we gave without joy, joy, discovering in another the secret road that gradually brought your feet closer to mine.

This is one of Neruda’s great love poems. Speaking of love seems to stretch the limits of language, l anguage, one reason that we appreciate the works of  poets who find expression for such experiences. The above lines illustrate unique, poetic instantiations of how we metaphorically conceptualize our love experiences partly in terms of journeys motivated by the SOURCEPATH-GOAL image schema. The poet talks about going “backward, in the distance” (i.e., the path) of his love relationship with the “wild girl,” stop st oppi ping ng at th thos osee pl plac aces es th that at “w “wee ga gave ve wi with thou outt jo joy” y” to fin find d “t “the he se secr cret et ro road ad”” that brought true unity and happiness. Although these phrases are novel, they the y ar aree re relat lated ed in met metaph aphori orical cal way wayss to mun mundan danee exp expre ressi ssions ons peo people ple oft often en use to talk about love and love relationships. For instance, consider the following list of conventional expressions: “Look how far we’ve come.” “It’s been a long, bumpy road.” “We’re “W e’re at a crossroads.” “We may have to go our separate ways.” “We “Our marriage is on the rocks.” “We’re “W e’re spinning our wheels.”

These (and other) conventional expressions cluster together under one of th thee ba basi sicc me meta taph phor oric ical al sy syst stem emss of un unde ders rsta tand ndin ing: g: LO LOVE VE IS A JO JOUR URNE NEY Y (Lakoff & Johnson,   1980). This conceptual metaphor involves a tight mapp ma ppin ing g ac acco corrdi ding ng to wh whic ich h en enti titi ties es in th thee do doma main in of lo love ve (e (e.g .g., ., th thee lo love vers rs,, their common goals, the love relationship, etc.) correspond correspond systematically

 

Concepts

 

93

to entities in the domain of a journey (e.g., the traveler, the vehicle, destinations, etc.). Various Various correspondences arise when we think of love as a  journey.. Among these are the ideas that the person in love is a traveler  journey traveler,, the goal of ultimate love is a destination, the means for achieving love are routes, the difficulties one experiences in love are obstacles to travel, and the progress in a love relationship is the distance traveled. It is not an arbitrary matter to speak of love relationships as being at crossroads, on the rocks, or having been on a long, bumpy road. Instead, we understand that each of these expressions is appropriate to use in talking about love relationships precisely because of our common metaphorical transformation whereby love is conceptualized as being like a physical journey and our understanding of journeys is closely linked to the SOURCE-PATH-GOAL image schema. This discussion of the SOURCEPATH-GOAL schema shows that there are direct connections between recurring bodily experiences, metaphorically understood abstract concepts, and both conventional and creative language that refers to these abstract concepts. Another illustration of an image schema and how its internal structure

is projected onto new abstract domain via metaphor is seen by consider ing the BALANCE schema (Johnson,  1987). The idea of balance is something that is learned “with our bodies and not by grasping a set of rules” (Johnson, 1987: 74). Balancing is such a pervasive part of our bodily experience that we are seldom aware of its presence in everyday life. We come to know the meaning of balance through the closely related experiences of bodily equilibrium or loss of equilibrium. For example, a baby stands, wobbl wob blees, an and d dr drop opss to th thee flo floo or. It tr trie iess ag agai ain n an and d aga gain in,, as it le leaarn rnss ho how w to maintain a balanced erect posture. A young boy struggles to stay up on a twotw o-wh whee eele led d bi bicy cycl clee as he le lear arns ns to ke keep ep hi hiss ba bala lanc ncee wh whil ilee ri ridi ding ng do down wn th thee street. Each of us has experienced occasions when we have too much acid in ou ourr sto toma mach chs, s, whe hen n ou ourr han ands ds ge gett co cold ld,, ou ourr he head adss fe feeel to too o hot ot,, ou ourr bla laddders feel distended, our sinuses become swollen, and our mouths feel dry. dry. In th thes esee an and d nu nume merrou ouss ot othe herr wa ways ys we le lear arn n th thee me mean anin ings gs of la lack ck of ba bala lanc ncee or equilibrium. We We respond to imbalance and disequilibrium by warming our hands, giving moisture to our mouths, draining our bladders, and so forth until we feel balanced once again. Our BALANCE image schema emerges, then, through our experiences of bodily equilibriums and disequilibrium and of maintaining our bodily systems and functions in states of equilibrium. Ourr BA Ou BALA LANC NCE E im imag agee sc sche hema ma,, to co cont ntin inue ue wi with th th this is ex exam ampl ple, e, su supp ppor orts ts unde un ders rsta tand ndin ing g of li lite tera rall ex expr pres essi sion onss su such ch as “H “Hee ba balan lance ced d th thee we weig ight ht on hi hiss shoulder” and is metaphorically elaborated in a large number of abstract domains of experience (e.g., psychological states, legal relationships, formal systems) (Johnson,  1991). In the cases of bodily and visual balance, there seems to be one basic scheme consisting of a point or axis around which forces and weights must be distributed so that they counteract or

 

94

 

Embodiment and Cognitive Science

 balance off one another. another. Our experience experience of bodily balance balance and the percepperception of balance is connected to our understanding of balanced personalities, balanced views, balanced systems, balanced equilibrium, the balance of power, the balance of justice, and so on. In each of these examples, the ment me ntal al or th thee ab abst stra ract ct co conc ncep eptt of ba balan lance ce is un unde ders rsto tood od an and d ex expe peri rien ence ced d in terms of our physical understanding of balance. Image schemas have intern te rnal al lo logi gicc or st stru ruct ctur uree th that at de dete term rmin inee th thee rol oles es th thes esee sc sche hema mass ca can n pl play ay in strruc st uctu turi ring ng va vari riou ouss co conc ncep epts ts an and d in pa patt tter erns ns of rea easo soni ning ng.. It is no nott th thee ca case se that a large number of unrelated concepts (for the systematic, psychological, ca l, mo mora ral, l, le lega gal, l, an and d ma math them emat atic ical al do doma main ins) s) al alll ju just st ha happ ppen en to ma make ke us usee of  thee sa th same me wo worrd “b “bal alan ance ce”” an and d rel elat ated ed te term rmss (J (Joh ohns nson on,, 1991). Ra Rath ther er,, we us usee the same word for all these domains because they are structurally related  by the same sort of underlying image schemas, and are metaphorically elaborated Considerfrom nowthem. the pervasive bodily experience of MOMENTUM. We expe ex peri rien ence ce vis visua uall mo mome ment ntum um wh when en we se seee he heav avy y mo movi ving ng th thin ings gs co cont ntin inue ue to move even when encountering other objects. We experience kinesthetic momentum both when we are the object that the heavy moving thing encoun co unte ters rs an and d wh when en we ar aree th thee he heav avy y mo movin ving g th thin ing. g. We ex expe peri rien ence ce au audi dito tory ry momentum both as a correlate of visual and kinesthetic momentum and independently, as when thunder builds up to a crescendo. We even experien ri ence ce in inte tern rnal al mo mome ment ntum um as wh when en ce cert rtai ain n bo bodi dily ly fu func ncti tion onss bu buil ild d up su such ch

that they cannot be stopped. We We abstract out of all of these similar experi ences those aspects of form that they have in common or that are similar, which we refer to through language as momentum. The MOMENTUM image schema serves as the embodied basis for several abstract, metaphorical concepts. Consider the following utterances: “I was bowled over by that idea.” “Wee have too much momentum to withdraw from the election race.” “W ra ce.” “I got carried away by what I was doing.” “Once he gets rolling, you’ll never be able to stop him talking.”

These utterances reflect how the image schema for MOMENTUM allows discussion of very abstract domains of cognition, such as political support, control, arguments, and talking about physical objects moving with momentum. Imagine, for example, what it looks like for someone to  be bowled over by an idea. This does not make literal or physical sense give gi ven n th that at id idea eass ar aree on only ly ab abst stra ract ct en enti titi ties es.. Yet mo most st pe peop ople le rea eadi dily ly im imag agin inee a scene in which some physical force hits a person who is standing, thus caus ca usin ing g th that at pe pers rson on to fa fall ll ov over er as th thee ph phys ysic ical al fo forc rcee co cont ntin inue uess pa pass ssin ing g ov over er the person’s prone body. Many people appear to base their imagistic unders de rsta tand ndin ings gs of st stat atem emen ents ts su such ch as “I wa wass bo bowl wled ed ov over er by th that at id idea ea”” ba base sed d on their own embodied experiences of being running into, and sometimes run over, by other people or objects.

 

Concepts

 

95

Finally, consider the embodied roots of another salient image schema, STRAIGHT STRAI GHT (Cienk (Cienki, i, 1998). Th Thee te term rm “s “str trai aigh ght” t” is em empl ploy oyed ed in ma many ny ph phys ys-ical and abstract ways. For instance: “The straight edge of the table.” “Stand up straight.” “I can’t think straight.” “It rained for three days straight.” “Tell “T ell it to me straight.” “Let me get this straight.” “He’s not straight, but gay gay.” .” “I couldn’t keep a straight face.”

Why do we us Why usee “s “str trai aigh ght” t” in th thes esee ra rath ther er di difffe ferren entt wa ways ys?? Th Thee co conc ncep eptt of  straight has an important role in our sensory experience. Research shows that collinearity of points or elements in a visual pattern has an important role in visual perception (Foster, 1982). For example, classic Gestalt studies on empirical grouping, visual detection in moving fields, visual acuity in movement, visual texture discrimination, and visual discrimination of   briefly presented dot figures all show that perception of straightness is a fundamental property of how we see and make sense of visual events. Straight lines are more easily and quickly seen then curved lines. Horizontal and vertical straight lines are especially more easily perceived than are oblique straight lines (Attneave & Olson, 1967). These findings partly explain why one’s form of reference can be a form of gestalt (e.g., “The straight edge of the table”) or can be an orientation, such as vertical (e.g., “The picture on the wall is not straight”) (Zubin & Choi, 1984). Beyond the importance of straightness in visual perception, people ex-

perience positive correlation between things beingtoin order. Fora instance, people waiting for anstraightness event, suchand buying a ticket a movi mo vie, e, us usua uall lly y sta tand nd in su such ch a way as to fo forrm a str trai aigh ghtt li line ne.. Th Ther eree is al alsso a strong relationship between straightness and perceived solidness. Objects that are curved and soft, such as clothes, do not form solid containers in the way that objects with straight lines do, such as boxes. Image schemas do not simply exist as single entities, but are often linked together to form very natural relationships through different “image schema transformations.” Image schema transformations have been shown to play a special role in linking perception and reason. Among the most important image schema transformations are the following (Lakoff, 1987: 443): (a)   “Path-focus (a) to end-point Follow Follow, , in imagination, the path of  a moving object, and thenfocus”: focus on the point where it comes to rest, or where it will come to rest. (b)   “M (b) “Mul ulti tipl plex ex to ma mass ss”: ”: Im Imag agin inee a gr grou oup p of se seve vera rall ob obje ject cts. s. Mo Move ve aw away ay (in your mind) from the group until the cluster of individuals start

 

96

 

Embodiment and Cognitive Science

to become a single homogeneous mass. Now move back down to the point where the mass turns once again into i nto a cluster. (c) “Following (c)  “Following a trajectory”: As we perceive a continuously moving object,wecanmentallytracethepathithastraversedorthetrajectory it is about to traverse. (d) “Superimposition”:Imaginealargesphereandasmallcube.Increase (d)  “Superimposition”:Imaginealargesphereandasmallcube.Increase the size of the cube until the sphere can fit inside it. Now reduce the size of the cube and put it within the sphere. Each image schema transformation reflects important aspects of our visual, auditory, or kinesthetic bodily experience. To illustrate, consider how these transformations might apply to our earlier example of the image schema for balance or equilibrium. equil ibrium. A situation where several of these tran tr ansf sfor orma mati tion onss in inte tera ract ct wi with th th thee ba bala lanc ncee im imag agee sc sche hema ma is th that at of ha hand ndlin ling g a group of animals. In order to successfully control and navigate a large number of animals, cattle or sheep perhaps, one needs to maintain the cohesiveness of the group. If a portion of the herd begins to drift apart from the who whole, le, an ins instan tance ce of the mul multip tiplex lex to mas masss tra transf nsform ormati ation, on, equ equilib ilibriu rium m hass be ha been en lo lost st an and d ac acti tion on mu must st be ta take ken n to res esto torre it it.. Su Such ch a co corr rrec ecti tive ve ac acti tion on requires that the path of the drifters be ascertained, following a trajectory, trajectory, and that their destination be determined and “headed off,” path-focus to end-point focus. There are many examples such as this that illustrate the role of image schemas and different transformations in structuring our understanding of real-world phenomena. This discussion discussion of image schemas schemas and metaphor runs contrary contrary to the popular view that there is some abstract similarity existing between literal and metaphorical concepts, such as our understanding of difficulty in terms ter ms of hea heavy vy phy physic sical al wei weight ghtss (Mu (Murph rphy y, 1996). Th Ther eree is no nott an ob obje ject ctiv ivel ely y si simi mila larr se set t of at attr trib ibut utes es forr co fo conc ncep epts tsconnect such su ch as di diffi fficu cult lty ydispositions,” and an d ph phys ysic ical al we weig ight ht,, nor are there similar features that “sunny “bright words, wor ds,”” and “ra “radia diant nt smi smiles les.” .” Con Concep ceptua tuall met metaph aphor or the theory ory dem demons onstra trates tes,,

alternatively, that concepts from different domains are related to one another by virtue of how people are physically constituted, their cognitive abilities, and their interactions with the world. Thes Th esee de deve velo lopm pmen ents ts on th thee im impo port rtan ance ce of im imag agee sc sche hema mass in st stru ruct ctur urin ing g abstract concepts have led scholars, in several disciplines, to examine the embo em bodi died ed na natu turre of va vari riou ouss ab abst stra ract ct id idea eass an and d ev even ents ts.. Pr Pres esen ente ted d be belo low w ar aree severa sev erall ext extend ended ed exa exampl mples es of re resea searc rch h tha thatt pr provi ovides des add additio itional nal sup suppor portt for the prominence of image schemas and metaphors in structuring abstract concepts. Thinking

Metaphor plays an essential role in how people conceive, and talk about, thinki thi nking. ng. Cog Cognit nitive ive lin linguis guistic tic stu studie diess dem demons onstra trate te tha thatt the there re is an ext extens ensive ive

 

Concepts

 

97

subsystem of metaphors for mind, centered on the idea that THE MIND IS A BODY (Lakoff & Johnson, 1999; Sweet  Sweetser ser,, 1990). The mapping of the  body onto mind gives rise to the following submetaphors: THINKING IS PHYSICAL FUNCTIONING IDEAS ARE ENTITIES WITH AN INDEPENDENT EXISTENCE THINKING OF AN IDEA IS FUNCTIONING PHYSICALLY WITH RESPECT TO AN INDEPENDENTLY EXISTING ENTITY More sp Mor spec ecifi ifica call lly y, th ther eree ar aree fo four ur ex exte tens nsiv ivee sp spec ecia iall ca case sess of th this is me meta taph phor or such suc h tha thatt thi thinki nking ng is und unders erstoo tood d as fou fourr dif differ ferent ent kin kinds ds of phy physic sical al functio fun ctionin ning: g: mov moving ing,, per percei ceivin ving, g, man manipu ipulat lating ing obj object ects, s, and eat eating ing (La (Lakof koff  f  & Johnson,  1999). Consider first the idea that THINKING IS MOVING. This metaphor gives rise to a complex set of mappings: THINKING IS MOVING (e.g., “My mind was racing”) IDEAS ARE LOCATIONS LOCATIONS (e.g., “How did you reach that conclusion?”) REASON IS A FORCE (e.g., “He was forced to accept the plan”) RATIONAL THOUGHT IS MOTION THAT IS DIRECT, DELIBERATE, STEP ST EP-B -BYY-ST STEP EP,, AN AND D IN AC ACCO CORD RD WI WITH TH TH THE E FO FORC RCE E OF RE REAS ASON ON (e.g (e .g., ., “D “Don on’t ’t sk skip ip an any y st step epss in fig figur urin ing g ou outt ho how w to so solv lvee th that at pr prob oble lem” m”)) BEING UNABLE TO THINK IS BEING UNABLE TO MOVE (e.g., “I’m stuck”) A LINE OF THOUGHT IS A PA PATH (e.g., “You should pursue that line li ne of thought”) THINKING ABOUT X IS MOVING IN THE AREA AROUND X (e.g., “I’ve been pursuing this topic for some time”) COMMUNICATING COMMUNICA TING IS GUIDING (e.g., “He led me to that new idea”) UNDERSTANDING IS FOLLOWING (e.g., “I follow what you are saying”) RETH RE THIN INKI KING NG IS GO GOIN ING G OV OVER ER TH THE E PATH AG AGAI AIN N (e (e.g .g., ., “I ne need ed to go  back and consider that again”) again”) The metaphor THINKING IS PERCEIVING also has a complex set of  mappings, including the following:

THINKING IS PERCEIVING (e.g., I am trying to see what you are saying”) IDEAS ARE THINGS PERCEIVED (e.g., “The idea became clear”) KNOWING IS SEEING (e.g., “I finally see what you are saying”) ATTEMPTING TO GAIN KNOWLEDGE IS SEARCHING (e.g., “I am looking for the right plan”) AN AID TO KNOWING IS A LIGHT SOURCE (e.g., “He shed light on the new theory”) BEIN BE ING G IGN IGNOR ORAN ANT T IS BE BEIN ING G UN UNAB ABLE LE TO SE SEE E (e (e.g .g., ., “S “She he ha hass bl blin inde ders rs on”)

 

98

 

Embodiment and Cognitive Science

DECEPTION IS PURPOSEFULLY IMPEDING VISION (e.g., “He put up a smokescreen smokescreen”) ”) DIRECTING ATTENTION IS POINTING (e.g., “Let me point out the  benefits of this plan”) BEING RECEPTIVE IS HEARING (e.g., “He is deaf to what you are saying”) AGREEING IS SMELLING (e.g., “Something doesn’t smell right about this theory”) PERSONAL PREFERENCE IS TASTE (e.g., “That is a sweet idea”) Manipu Mani pula latin ting g ob obje ject ctss is an anot othe herr bo bodi dily ly ac acti tion on us used ed to un unde ders rsta tand nd th thin inkking. This THINKING IS OBJECT MANIPULATION metaphor has many mappings, including the following: THINKING THINKI NG IS OBJ OBJECT ECT MAN MANIPU IPULA LATIO TION N (e. (e.g., g., “Le “Let’s t’s tos tosss ar aroun ound d som somee ideas”) IDEAS ARE MANIPULABLE OBJECTS (e.g., “Let’s reshape that idea”) COMMUNICATING IS SENDING (e.g., “We exchanged ideas”) UNDERSTANDING UNDERST ANDING IS GRASPING (e.g., “She easily grasped the difficult concept”) INABILITY TO UNDERSTAND IS INABILITY TO GRASP (e.g., “The idea is hard to get hold of”) THE STRUCTURE OF AN IDEA IS THE STRUCTURE OF AN OBJECT (e.g., “That idea has many sides to it”) ANAL AN ALYZ YZIN ING G IDE IDEAS AS IS TAK AKIN ING G AP APAR ART T OB OBJE JECT CTS S (e (e.g .g., ., “H “Hee to tore re ap apar artt the argument”) The final embodied metaphor for thinking is ACQUIRING IDEAS IS EATING. EA TING. Its entailments include the following: ACCE AC CEPT PTIN ING G ID IDEA EAS S IS EA EATIN TING G (e (e.g .g., ., “H “Hee sw swal allo lowe wed d th that at id idea ea wh whol ole” e”)) INTERESTINIDEASISAPPETITEFORFOOD(e.g.,“Hehasanappetite for learning”) GOOD GO OD ID IDEA EAS S AR ARE E HE HEAL ALTH THFU FUL L FO FOOD OD (e (e.g .g., ., “T “Tha hatt is a sa savo vory ry idea”) DISTURBING IDEAS ARE DISGUSTING FOODS (e.g., “That idea is shit”) UNAPPEALING IDEAS ARE FLAVORLESS FOODS (e.g., “That is a  bland theory”) CONSIDERING IS CHEWING (e.g., “Let’s chew on that idea for a

 bit ) ACCEPTING IS SW SWALLOWING ALLOWING (e.g., “I can’t swallow that idea”) FULLY COMPREHENDING IS DISGESTING (e.g., “That’s too much for me to digest”) COMMUN COM MUNICA ICATIN TING G IS FEE FEEDIN DING G (e. (e.g., g., “He was wasfed fed sev severa erall new ide ideas” as”))

 

Concepts

 

99

These four metaphors are extremely common in discourse and demonstrate how abstract concepts can be made concrete through comparison to bodily action. But these metaphors are not special to English speakers; linguistic analyses have shown that these same metaphors are also found in Chinese (Yu,  2003). Consider just a few examples of these metaphors (original expression, literal translation, and colloquial meaning): THINKING IS MOVING “si-lu” (thinking route/path) “train of thought” “yun-t “yu n-tou ou zhu zhuanan- xia xiang” ng” (di (dizzy zzy-he -head, ad, los losing ing dir direct ection ion)) “co “confu nfused sed and disoriented” “fan-si” (reverse-think) “engage in introspection, self-examination” “zhui-su” (chase-trace) “trace back to, recall” THINKING IS PERCEIVING/SEEING “kan-fa” (see-method) “a way of looking at things” “kan-chuan” (see-penetrate) “see through something” “kan-di” (see-low) “look down on, belittle” “kan-qing” (see-light) “underestimate” THINKING IS OBJECT MANIPULATION MANIPULATION “sixiang jiaoliu” (exchange of thoughts/ideas) “exchange of ideas” “pao zai nao hou” (toss at brain back) “ignore idea” “wa-kong xinsi” (dig-empty thoughts/ideas) “rack one’s brain” “sixiang geda” (thought knot) “a hang-up on one’s mind” ACQUIRING IDEAS IS EATING “chen-fu guannian” (stale-rotten idea/concept) “an outworn idea” “sou zhuyi” (spoiled ideas/suggestions) “a lousy idea, a stupid suggestion” “ru-ji shi-ke” (like-hungry) “acquiring ideas with great eagerness” “sou-chang gua-du” (search-intestines) “search intently for an idea” It is not surprising that thinking is conceptualized in embodied ways across cultures because of the prominence that moving, perceiving, manipulating objects, and eating have in people’s everyday lives. As Yu (2003:  162 ) concludes, “The fact that distinct languages show metaphors in a systematic way supports the cognitive status of these metaphors as primarily conceptual, rooted in common human experiences” (also see Neumann, 2001). Linguistic Action

Abstract Abstra ct con concep cepts ts suc such h as tru truth, th, ide ideas, as,jus justic tice, e, and fri friend endshi ship p ar aree als also o talk talked ed about in concrete ways, as if they are items that can be physically manip-

ulated. For example, in his poem “Ultimately” (Hemingway (Hemingway,, 1960), Ernest

 

100

 

Embodiment and Cognitive Science

Hemingway writes of truth as if it is something physical that can actually  be spit out: He tried to spit out the truth; Dry-mouthed at first, He drooled and clobbered in the end; Truth dribbling his chin.

Our understanding of truth as something that can be spit out, in the above case with great difficulty for the person trying to speak honestly, is dependent on some recognition of a metaphorical idea (e.g., truth as a substance that can be ingested and regurgitated when needed). Again, there is an essential tacit connection between human embodiment, especially embodied action, and how we think about different physical and nonphysical concepts. Writers elaborate on these body-based metaphoricall co ca conc ncep epts ts in ne new w, cr crea eati tive ve wa ways ys th that at or ordi dina nary ry re read ader erss un unde ders rsta tand nd gi give ven n their own embodied experiences. The excerpt from Hemingway’s poem, shown above, indicates one instance of how people refer to linguistic action (e.g., speaking about the truth) in terms of associations to bodily actions (e.g., spitting out something thin g th that at ha hass be been en in inge gest sted ed). ). Ex Expe peri rien ence cess of th thee bo body dy in ac acti tion on,, es espe peci cial ally ly those related to oral and head activities and their related body parts, provide a critical source domain for structuring a variety of speech events. For example, one analysis of  175  175  body-part metaphors in a large corpus show sh owed ed th that at bo body dy pa part rtss an and d bo bodi dily ly fu func ncti tion onss ar aree es esse sent ntia iall so sour urce ce do doma main inss forr ch fo char arac acte teri rizi zing ng pe peop ople le’s ’s de desc scri ript ptio ion n of ta talk lk (G (Goo ooss ssen enss et al al., ., 1995). Th Ther eree are several ways in which people’s understanding of human embodiment is metaphorically projected to structure linguistic action. Thefirstwayinvolvesbodypartsthatplayaroleinspeaking,butthatare putt to a di pu diff ffer eren entt us usee (e (e.g .g., ., ea eati ting ng an and d br brea eath thin ing) g).. Fo Forr in inst stan ance ce,, ph phra rase sess su such ch as “feed” and “force/ram/thrust something down someone’s throat” depicts specific interactions between two people in which the speaker transmits something something to a listen listener er,, and the listen listener er obtains the infor information mation by eating it. A second group characterizes speaking as eating (or part of the process of eating), such as “chew the fat” or “chew the rag” (i.e., “to chat or complain”). Given that both fat and rags can be chewed a long time, with little nutritional value coming from these activities, these idiomatic phrases express the idea of talking about something a long time with little new information to be gained from the experience. Thee ph Th phra rase se “e “eat at on one’ e’ss wo worrds ds”” (i (i.e .e., ., to ad admi mitt th that at on onee ha hass sa said id so some meth thin ing g in error) illustrates a different metaphor through linguistic action. By referring to the directionality of eating (i.e., ingesting), compared with the direct dir ection ionali ality ty of spe speaki aking ng (i. (i.e., e., ext exteri eriori orizin zing), g), the these se idio idioms ms exp expre ress ss the ide ideaa that th at th thee sp spea eake ker’ r’ss wo worrds we were reso some meho how w de dest stro roye yed d by ma maki king ng th them em go ba back ck

 

Concepts

 

101

to the place from which they arose. This hypothetical action renders the speaker’s utterance mute because it no longer has the originally intended effect upon the audience. On the other hand, the term “regurgitate” (i.e., to report what one has already heard or learned) depicts the same direction of action as does speech and expresses the idea that the speaker has once ingested some idea, but not quite digested it completely (like people with food or liquid) liqui d) so The that experience it can indeed thrown up and outmany of themetaphorical body. ofbe breathing underlies phrases for linguistic action, in part because breathing is very much a part of speaking (e.g., “He breathed words of love into her ear”). The phrase “waste one’s breath” characterizes the air one breathes as a valuable resource, one that is essential to proper bodily functioning, which should not be expended needlessly. To “cough up” something is to remove a substance (bloo (b lood, d, ph phle legm gm)) th that at ca caus uses es bo bodi dily ly,, an and d of ofte ten n br brea eath thin ing, g, di disc scom omfo fort rt.. Wh When en a speaker “chokes back” something, he or she attempts to prevent somethin th ing g fr from om es esca capi ping ng th thee bo body dy,, th thus us ex expr pres essi sing ng th thee id idea ea of so some meon onee ex exer ertin ting g great control over what he or she has started to say. Thee me Th meta taph phor or “s “spi pitt ou out” t” re refle flect ctss th thee id idea ea th that at th thee sp spea eake kerr ha hass so some meth thin ing g of value in the body, which through effort he or she is able to gather up (like spit or phlegm) and say (or expectorate). Various expressions for linguistic action center on the movement of the visible speech organs. “Keeping one’s mouth shut,” “opening one’s lips,” and “closed lipped” describe positions of the mouth and lips to stand for either the presence or absence of speech. Saying something “tongue in cheek” or “to lie through one’s teeth” also express different types of  linguistic actions (e.g., in gesture, lie) in which contours of the face and mouth metaphorically structure our understanding of what a speaker is communicating. The bod bodily ily pos postur turee and exp experi erienc encee of lis listen teners ers cap captur tures es som someth ething ing abo about ut how linguistic actions are understood. When someone “turns a deaf ear” or when something “goes in one ear and out the other,” it is clear that the listene liste nerr is no not t de dedi dica cati ting ng th the e ri righ ghtt bo body dylinguistic part pa rt of su succ cces essf sful ul many comm co mmun unic icat atio ion. n. Beyond the embodied character of actions, aspects of  nonv no nver erba ball co comm mmun unic icat atio ion n res estt on bo bodi dily ly ac acti tion ons. s. “P “Pat atti ting ng so some meon onee on th thee  back,” “bringing/bending the knee to someone,” and “tipping someone the win wink” k” ar aree eac each h phy physic sical al act action ionss tha thatt re reflec flectt an ind individ ividual ual’s ’s app appre recia ciatio tion, n, respe re spect, ct, or fri friend endshi ship p wit with h ano anothe therr. The These se non nonver verbal bal act action ionss som someti etimes mes coocccur wi oc with th sp spee eecch, bu butt ca can n al alsso st stan and d al alo one ne.. “T “To o pat on oneese self lf on th thee ba bacck” is a difficult, even ridiculous action to perform, one reason that it expresses the unacceptability of praising oneself. Our sensory apparatus also plays an important role in various aspects of our metaphorical conceptualization of speaking. For instance, the metaphor “sniff” (i.e., to say something in a complaining manner) rests

 

102

 

Embodiment and Cognitive Science

on the embodied experience that the act of perceiving something with our noses is often accompanied by a special noise (i.e., sniffing). The sniffing noise represents that a person has perceived something of value, which gets transferred to the idea that a listener has understood something of  subs su bsta tanc nce. e. Sn Snif iffin fing g no nois ises es ar aree of ofte ten n ma made de wh when en so some meth thin ing g ob obje ject ctio iona nabl blee is smel sm elle led d an and d th this is ge gets ts ma mapp pped ed in into to th thee do doma main in of lin lingu guis isti ticc co comm mmun unic icat atio ion n to express the idea that the listener has just comprehended an unpleasant idea. Boththe people and animals noses into to things,” which characterizes positioning of the“poke bodytheir in preparation smelling something (i.e., dogs into holes in the ground, people into pots on the stove). The meta me taph phor oric ical al sa sayi ying ng th that at “h “hee po poke ked d hi hiss no nose se in into to ot othe herr pe peop ople le’s ’s bu busi sine ness ss”” suggests that a person has positioned himself to obtain, usually hidden, information. In a related way, “one puts out feelers” as preparation to perceiving something, which metaphorically is understood in the speech domain as putting oneself in a position to obtain information. Vio iole lent nt ph phys ysic ical al ac actio tion n pr prov ovid ides es a ri rich ch so sour urce ce do doma main in fo forr ch char arac acte teri rizi zing ng many man y kin kinds ds of ling linguis uistic tic act action ions. s. The spo sport rt of box boxing ing,, in par partic ticula ularr, pr provi ovides des the emb embodi odied ed act action ionss und underl erlyin ying g man many y lin lingui guisti sticc con concep cepts ts suc such h as “pu “pullin lling g one’s one ’s pun punche ches,” s,” “sp “sparr arring ing,” ,” or “be “beatin ating g som someon eonee to the pun punch. ch.”” In “pu “pullin lling g one’s punch,” speakers soften the impact of what they say for listeners. When speakers “spar” with listeners, the interaction is less serious, more playful, than is a full-fledged fight. And when speakers “beat someone to the punch,” they make a point or argument before their listeners do. A di diff ffer eren entt se sett of vi viol olen entt ac acti tion onss us used ed to co conc ncep eptu tual aliz izee lin lingu guis istic tic ac acti tion onss include “rap someone over the knuckles” and “box someone’s ear.” In  both cases, the focus in on the painful sensation the t he listener experiences as a result of what speakers say. Other violent metaphors include “butt out,” “kick someone around” (e.g., Richard Nixon’s famous statement in 1962 to the press “You “You won’t have Nixon to kick around anymore”), “tear apart,” and “choke off.” These idiomatic phrases reflect different aspects of situations where one person, the speaker, uses authority over another, with some expressions, such as “tear apart” and “choke off,” highlighting theMe extreme nature of“m theake speaker’s actions. Meta taph phor ors s su such ch as “mak e th thee fu furr fly fly,” ,” “b “bac ackk-bi bite te,” ,” “s “sna nap p at at,” ,” “b “bit itee so some me-one’s head off,” and “jump down someone’s throat” take as their source domain different violent actions seen in the animal world, especially denoting the arbitrary, unnecessarily hostile, nature of the action given the situat sit uation ion.. Des Descri cribin bing g spe speake akers’ rs’ lin lingui guisti sticc act action ionss usi using ng any of the these se phr phrase asess strongly implies that the person is overreacting to something another individual has said or done. A completely different type of violent action motivates “eat your heart out” and “cut off your nose to spite your face.” Both metaphors express the tremendous self-inflicted pain a person experiences by his or her own action.

 

Concepts

 

103

Several metaphors for linguistic action focus on restricted movement. “Tongue “T ongue tied,” “hold your tongue,” and “bite your tongue” all refer to the silent consequences of being unable to speak, mostly through self-control.

Somewh Some what at re rela late ted d ar aree ph phra rase sess wh wher eree so some me ob obje ject ct is cl clum umsi sily ly ha hand ndle led, d, su such ch as “fumble,” or where one’s actions are awkward, as in “heavy-handed” or “left-handed compliment.” When a speaker successfully exchanges information with another, often in cases where a speaker offers a reward to someone else in exchange for something, he or she “hands it to someone.” A particularly interesting metaphor is “shoot one’s mouth off.” This phrase views speaking in terms of clumsily handling a gun, which accident de ntal ally ly ca caus uses es it to fir fire. e. Wh When en a sp spea eake kerr “s “sho hoot otss of offf hi hiss mo mout uth, h,”” he wa wast stes es a bullet and draws unwanted attention, implying that the person does not rea eall lly y kn know ow wh what at he is do doin ing g or ta talk lkin ing g ab abou out. t. On th thee ot othe herr ha hand nd,, sp spea eake kers rs who pay excessive attention to what they are saying “split hairs,” thereby comm co mmun unic icat atin ing g id idea eass th that at ar aree tr triv ivia iall or of offf th thee po poin intt of th thee co conv nver ersa sati tion on or argument. The embodied experience of walking motivates various speech actions, with different parts of walking movements being tied to specific ways of  speaking. When someone “backtracks” while speaking, he or she reverses directions on the path he or she initially started out on to correct what has already been stated. A different error arises when someone “puts his foot in hi hiss mo mout uth, h,”” in indi dica cati ting ng vi viaa th thee me meta taph phor or of a se seri riou ouss mi mish shan andl dlin ing g of th thee  body when walking that a grave mistake has been made in saying what was just previously said. There are different degrees of intensity in making speech acts. For instan st ance ce,, a wa warn rnin ing g or reb ebuk ukee ma may y be mi mild ld or st strron ong. g. Th Thee de degr gree ee of em emph phas asis is is particularly salient in some metaphors. For instance, in the metaphor “raise one’s eyebrows” (meaning “express surprise or displeasure”), the speech act must be mild as the embodied action is quite slight. Duration is important in “chew the fat” (a high value in duration) and “go in one ear and out the other” (which implies a very brief duration). The met metaph aphori orical cal str struct ucturi uring ng of lin lingui guisti sticc act action ion via sig signifi nifican cantt pat patter terns ns of embodied experience is, of course, tied to image schemas. For instance, the image schema BALANCE (i.e.,various a symmetrical arrangement forces around a point or axis) motivates phrases referring to a of person’s attempt to restore equilibrium of the body (and mind). When people say “get something off my chest,” they describe a forceful action to remove an impediment that causes imbalance. Speakers who get something off  their chests remove oppressive forces by merely talking to an appropriate person, often the person most responsible for placing the burden or impediment on the speaker. speaker. “Getting something off one’s chest,” just like “blowing off steam” and “coughing something up,” restores a sense of   balance or well-being to an individual. individual.

 

104

 

Embodiment and Cognitive Science

The ima image ge sch schema ema CON CONT TAIN AINMEN MENT T und underl erlies ies man many y met metaph aphori orical cal con con-cepts related to our understanding of linguistic action. For instance, our mouths, like our bodies, are experienced as containers, such that when the container is open, linguistic action is possible, and when it is closed, there is only silence. To be “closed-lipped” reflects the silent, closed container, and when one “bites one’s lip,” the closing of the mouth and lips is done quickly with gre great at for force. ce. When peopl peoplee “lie through their teeth,”

the container is perceived as a hiding place where true information re sides, but the container is somewhat defective and we can see through the speaker’s shameless attempt to lie about something when the truth can partly be seen. Some metaphors talk of entering the mouth container, container, as when “one puts words in someone’s mouth” or “forces/rams/thrusts something down someone’s throat,” with the more forceful entry into the container reflecting greater intensity of the speaker’s linguistic action. Embodied CONTAINMENT CONTAINMENT also refers to cases where where objects, or information, are removed from the mouth or head of a speaker, speaker, as in “He took the words right out of my mouth” and “pick someone’s brains,” both of  which wh ich imp imply ly th that at th thee sp spea eake kerr is a pe pers rson on po poss sses essi sing ng so some me va valua luabl blee ob obje ject ct’s ’s)) worth stealing. The importance of the PATH image schema is seen in the metaphors  based on walking, such as “backtrack,” where the directionality directionality of movement along some path must be reversed. PATH also is relevant to cases of reversed motion, as in the eating metaphors of “eat one’s words” and “eat “e at cr crow ow,” ,” wh whic ich h ar aree sp spec ecifi ificc in inst stan ance cess of th thee ge gene nera rall id idea ea of “t “tak akin ing g ba back ck one’s words” (i.e., moving words back along the conduit path on which a speaker first send them). Thee im Th imag agee sc sch hem emaa of FO FORC RCE E is ce cent ntrral to ma man ny of th thee me meta tap pho horrs bas aseed on vi viol olen entt bo bodi dily ly ac acti tion onss no note ted d ab abov ove. e. In mo most st of th thes esee in inst stan ance ces, s, th thee fo forrce is noticeable because of its extreme nature (e.g., “bite someone’s head off” and “snap at someone”). These selected examples clearly illustrate how image schemas connect thee do th doma main inss of em embo bodi died ed ac acti tion on wi with th th thee do doma main in of li ling ngui uist stic ic ac acti tion on.. Mo Most st generally,, this examination of metaphor and linguistic action reveals how generally people use their intuitive phenomenological sense of their bodies to make sense of, and structure, more abstract conceptual domains. Grammar and Spatial Concepts

Anothe Anot herr pl plac acee to fin find d em embo bodi dime ment nt in co conc ncep epts ts is in th thee st stud udy y of nu nume meri rica call system sys temss acr across osslan langua guages ges.. Alt Althou hough gh spe speake akers rs do not notor ordin dinari arily ly und unders erstan tand d the embodied character of the numerals in their language, a closer look at many numeral systems reveals that the linguistic labels for numerals are not arbitrary, but are often, in some languages particularly, motivated by embodied experience.

 

 

105

Concepts good od il illu lust stra rati tion on of th thee em embo bodi died ed ch char arac acte terr of so some me nu nume mera rall sy syst stem emss A go is Mam Mamuo, uo, a cen centra trall Afr Africa ican n Nik Niko-S o-Saha aharan ran lan langua guage ge (He (Heine ine,, 1997).Thefirst five numerals, 1 through 5, are etymologically opaque, as there appears to  be no explanation for the words that refer to them. But the terms for the rest of the numerals are motivated by reference to body parts. Numerals in Mamuo are divided into quinary blocks, based on the five fingers of  thee hu th huma man n ha hand nd.. Th Thus us,, 5 co cons nstit titut utes es th thee ba basi sicc nu nume mera rall an and d co coun unti ting ng st star arts ts over with each block of five entities. A second numeral base is  20, which reflects the total number of fingers and toes, which results in a vigesimal system (i.e., a system having 20 as its primary numerical base). The body-part model for numerals in Mamuo is not unique to this lan-

guage but is found in most languages of the world. The human hand is the most widely used model for structuring numeral systems, with the numeral 5 constituting the smallest recurrent base number (i.e., the num ber from which counting begins again) in languages throughout the world (Majewicz, 1981). Even the decimal system in English, for instance, relates to emb mbod odim imeent in te term rmss of th thee nu numb mber erss of fin finge gers rs on ou ourr tw two o ha hand ndss (i (i.e .e., ., a syst sy stem em ba base sed d on 10). So Some me la lang ngua uage gess ap appe pear ar to ha have ve nu nume mera rall sy syst stem emss un un-related to the body-part model. But embodiment still plays a role in many cases. Thus, in Sotho, the verb for “jump” denotes the numeral   6. Why might the verb “jump” be related to 6? The motivation for this is that one must “jump over from one hand to the other” in counting upward from 5  to  6. This embodied action of jumping to the other hand stands for the actual numeral 6. Some languages never explicitly mention the word “hand” in refer reference ence to any numerals, but, once again, there is an implicit relationship between the hand and numerals. For example, speakers of Api, a language of the New Hebrides, do not say anything explicit about the hand in their numerals for   6   through   9, because hand in implicit in the morpheme for “new.” Thus, six (“otai”), is “new one,” seven (“oluao”) is “new two,” eight (“otolu”) is “new three,” and nine (“ovari”) is “new four,” whereas ten (“luc luna”) is “two hands.” A spectacular example of how body parts motivate numeral systems is seen in parts of New Guinea, where there are linguistic terms only for the first fir st fiv fivee nu nume mera rals ls,, bu butt sp spea eake kers rs ca can n ref efer er ge gest stur ural ally ly to nu nume mera rals ls as hi high gh as 20 by using the fingers on one hand, then going up to the wrist, elbow, elbow, the upperr ar uppe arm, m, an and d ar arou ound nd ba back ck to th thee ot othe herr ha hand nd (G (Grree eenb nber erg, g, 1978). Cou Counti nting ng is not the only arithmetical operation motivated by embodied experience. Some So me la lang ngua uage ge mu mult ltip iply ly by “t “two wo ha hand ndss an and d tw two o fin finge gers rs,” ,” “t “two wo ha hand ndss an and d three fingers,” and so on (Stampe, 1976). Many languages describe spatial orientation in terms of body parts. In Yucatec, a Mayan language of Mexico, the body-part term for back (“paach”) denotes “behind,” front (“tian”) refers to “in front of,” eye (“eich”) denotes “inside,” and marrow (“tu’ u’”) is used to talk of “in”

 

106

 

Embodiment and Cognitive Science

(Goldap,  1992;  Stolz,  1994). This mapping of body part to spatial orientation is one of the source domains used in the experience of reference poin po ints ts.. Ot Othe herr mo mode dels ls in incl clud udee en envi viro ronm nmen enta tall la land ndma mark rkss an and d dy dyna nami micc co conncepts. But the body-part model (i.e., the body in its most upright position) is clearly the most widely used in conceptualizing space. Thee hu Th huma man n bo body dy pe perv rvad ades es th thee mo most st si sign gnifi ifica cant nt mo mode dell fo forr ta talk lkin ing g ab abou outt spatial orientation. But animal bodies also serve as structural templates for talking about and expressing spatial relations (i.e., the zoomorphic model). Chalcatango Mixtec have different terms for “human back” and “ani “a nima mall ba back ck,” ,” wh whic ich h sh shap apes es th thee wa ways ys di diff ffer eren entt ob obje ject ctss ar aree co conc ncep eptu tual aliz ized ed (Brugman (Bru gman & McCau McCauley ley,, 1986).Forexample,atableisconceivedofasananimalwhosebackisthetabletopandwhosebellyisthetable’sunderside.The topofawallisdescribedbythenameofananimalback.Despitesuchexamples of animal backs being used to refer to inanimate objects, no language conceptualizes of spatial relations in terms only of the zoomorphic model. Cross-linguistic research research demonstrates the pervasiveness of the human

 body in referring to spatial concepts (Heine,  1997). Consider some examples from the following set of prepositions: The spatial concept “up” is described in terms of human body parts, especially by refer re ferenc encee to th thee hum human an han hand. d. In 87%ofallAfricanand 61% of all Oc Ocean eanic ic lan langua guage gess that used body parts for “up” terms, such as “above,” “up,” and “over,” the word “hand” is grammaticized for this specific purpose. Thespatialconceptof“down”ismostwidelydescribedinbothAfricanandOceanic language lang uagess usin using g env enviro ironme nmenta ntall land landmar marks ks (e.g (e.g., ., “ear “earth” th” and “gr “ground ound”). ”). How However ever,,  body parts still play an important importan t role in both language families in talking about “down “do wn”” co conc ncept epts. s. In Afr Afric ican, an, “b “but utto tock cks” s” an and d “an “anus us”” ar aree fr frequ equent ently ly me ment ntio ioned ned and aree se ar seen en in 85% of al alll la lang ngua uage gess th that at ha have ve a gr gram amma mati tici cize zed d mo morp rphe heme me fo forr “d “dow own. n.”” Ocea Oc eani nicc la lang ngua uage gess re refe ferr to ei eith ther er “f “foo oot” t” or “l “leg eg”” as th thee pr prim imar ary y (59%) body par artt in grammaticized morphemes for “down.” The concept of “front” is most frequently conceptualized in terms of the human face fa ce in 53% of the Afric icaan and 72% of th thee Oc Ocea eani nicc la lang ngua uage gess ex exam amin ined ed,, wi with th “e “eye ye”” and then “breast” also being widely seen in African languages. Environmental landmarks are rarely seen in linguistic expressions about “front.” The concept of “back” is, not surprisingly, discussed in terms of the human back, as seen in  78 % of the African and 95 % of the Oceanic languages. “Buttocks” “Buttocks” and “anus” are body parts widely seen in African expressions for the “back” concept. Once again, environmental landmarks are virtually absent as source domains for the “back” concept.

Spatia Spat iall de defin finit itio ions ns ar aree se seen en in ho how w Af Afri rica can n an and d Oc Ocea eani nicc la lang ngua uage gess ref efer er to th thee sp spat atia iall co conc ncep eptt of “i “in. n.”” Th Thee “b “bel elly ly/s /sto toma mach ch”” bo body dy pa part rt ac acco coun unts ts fo forr 92% of African expressions, with far fewer expressions referring to other  body parts such as “palm” (of hand) and “heart.” But Oceanic languages use a variety of body parts relatively equally in talk of “in,” including “tooth,” “body-skin,” “heart,” “liver “liver,” ,” and “bowels.”

 

Concepts

 

107

body’s extremities (e.g.,ofhand or“down,” arm) are “front,” rarely referre referred d toand in talkingThe about the spatial concepts “up,” “back,” “in” (Heine,  1989). Why might the extremities provide so little of the source domains in conceptualizing these spatial orientations? A good possibility is that our experience of the body’s extremities simply does not facilitate an understanding of these particular spatial concepts. But the concepts of  “left” and “right” refer to the extremities (i.e., “hand” to talk of left-right spatial orientation (Werner, 1904)). The location of the hand in relation to the rest of the body makes the hand more appropriate for the expression of “le left ft”” an and d “r “rig ight ht”” th thaan it do doees fo forr “up up”” an and d “d “dow own n.” On th thee ot othe herr han and, d, it is curious that the body parts “nose” and “knee-cap” are not used in speaking of “front.” One linguistic proposal argues that the following scale underlies the worrld wo ld’s ’s la lan ngu guag ages es in refe ferr rrin ing g to sp spaati tial al lo loccat atio ions ns:: do dow wn – up – in – fr fro ont –  back (Heine, Ulrike, & Hunnemeyer Hunnemeyer,,  1991). Under this view, if any one of the five concepts is conceived of in terms of a body-part word, then none of the concepts to its right may be derived from some nonembodied sour so urce ce,, su such ch as en envir viron onme ment ntal al la land ndma mark rks. s. Co Cons nseq eque uent ntly ly,, it is un unli like kely ly th that at a body-part term, such as “buttocks,” would be used to talk of “down,” and a landmark term for either “up,” “in,” “front,” or “back.” Similarly,

the body-part model would not be used in talking of “back” or “front” if  described in terms of environmental landmarks. A relevant distinction in mapping body parts is that the upper half of  the body plays a larger role in conceptualizing of spatial orientation than does the body’s lower half. This may be due to the upper body being perce cept ptua uall lly y mo morrconcepts. e dif diffe ferren enti tiat ated ed and an d mo more re salie sa lient nt fo for r sp spea eake rs wi wish shin ing g to talk ta lk about spatial For any given language, then, itkers seems more likely to call one’s toes “fingers of the foot” than to call one’s fingers “toes of  the hand.” Many East and Southeast Asian languages refer to the ankle bone as “foot-eye,” yet yet eyes are not described as “anklebones “anklebones of the head” (Matisoff, 1978; Schladt, 1997). The human body also provides a framework for talking about many kinds of abstract, schematic notions. Thus, reference reference to “top end” or “top” are done in terms of the head, “bottom end” in terms of the buttocks or foot, “opening” or “edge” in terms of the mouth, and “narrow sections” is understood as the neck or wrist. Different languages refer to objects in differ dif ferent ent way wayss acc accor ordin ding g to thi thiss gen genera erall bod body-p y-part art to abs abstra tractct-sch schema ematic tic set of relations. Thus, in Tzeltal, objects such as knives, pots, leaves, feathers, and plants have conceptual properties arising from different body parts (Levinson, 1994). Political Ideas

We have long been accustomed to the metaphor of the “body politic” ever since the work of philosopher Thomas Hobbes. The “body politic”

 

108

 

Embodiment and Cognitive Science

meta phor or is mo most stly ly th thou ough ghtt to op oper erat atee at a ge gene nera rall le leve vel, l, bu butt ne new w res esea earrch metaph reveals revea ls the dep depth th of thi thiss emb embodi odied ed exp experi erienc encee in str struct ucturi uring ng pol politi itical cal ide ideas. as. One analysis of the debates in the United States in 1990 over the Gulf War showed that several image schemas enabled people to reason about international politics (Beer, 2001). Balance is a central term in international relations. “Balance of power” expresses the shared wisdom of foreign policy. The terms “balance” and its cognates occur in the debate a total of  107  times. In the case of “balance,” we come to understand more clearly an entire complex of related application. Representative Peter Fazio (DOregon) uses “balance” to lay out the national pieces of the Gulf region and an d at atte temp mpts ts to st stru ruct ctur uree th thee fo forrce cess of th that at reg egio ion n on a ve very ry co comp mple lex x bo boar ard: d: “If we think about what is the long-term effect here, we have embraced Iraq to counter Iran. Now we are embracing Syria to counter Iraq. After we decapitated Iraq in this war, war, if that is what happens, what then is next in the region? How do we instill a new government in Iraq? How do we  balance the forces in the region? Will we have to occupy Iraq? Will we have to def efeend Ir Iraq aq ag agai ain nst Sy Syrria or Tur urk key or Ir Iran an in th thee nea earr fu futu turre in or orde derr to gain so-called or restore so-called balance in the region?” (CR, H-132). Blocka Blo ckage ge inc includ ludes es man many y sem semant antica ically lly re relate lated d ter terms ms suc such h as blo block, ck, blo blockckage, blockaded, blockading, blockages, blocked, blocking, and blocks. Related words are embargo, force, intervention, penetration, and sanctions. “Blockage” itself appears relatively infrequently, but “blockade” is used 69  times. In the case of the Gulf War, blockage is seen in the form of an economic embargo that is a major alternative strategic option. “Embargo”

and its cognates appear 260 times. Opposite terms, such as “unblockage” or its di dist stan antt co cogn gnat ates es “l “lib iber erat atio ion” n” an and d “f “fre ree, e,”” ap appe pear ar 167 time times. s. “Pe “Penet netraration” is the opposite of blockage age in penetration another anoth er dimension. dimension When liberation liberation relieves or dissolves theblock blockage, pierces. it. “Penetration” was used infrequently, but the notion of “intervene” was used  374 times. “Intervention,” like “blockage,” is a standard means of foreign policy and is densely connected in the theory and practice of international relations. Center-periphery has wide play in international political economy. “Center” emerges as the key term in this dyad, appearing  37  times compared with “periphery’s”  3. “Center” evokes a very clear circular spatial grid. Indeed, as Sen. Steven Symms (R-Idaho) used “center,” he conjured up an image of a spider – Saddam Hussein – sitting at the center of a web of domestic power: “The Iraqi dictator sits at the center of a web of  state, party, military, and secret police organizations” (CR, S-380). When thee we th web b sp sprrea eads ds ou outw twar ard d be beyo yond nd th thee na nati tion onal al bo boun unda dari ries es of Ir Iraq aq,, it en enta tanngles an ever-growing number of participants, including the international world of terrorism. However, as in a real web, control always remains at the center. Indeed, as Sen. Orrin Hatch (R-Utah) suggested: “We all know that the world’s most vicious terrorists have taken up residence in

 

Concepts

 

109

Baghdad . . . Terrorists are on the move, and weapons, and equipment are  being put into place. Iraq stands at the center of three actions, providing the crucial support – false passports, sophisticated equipment, vast sums of  mone mo ney y – th that at onl nly y a sta tate te spo pon nso sorr of te terrror ha hass av avaail ilab able le”” (C (CR, R, S-385). Se Sennator Daniel Akaka (D-Hawaii) invoked the schema of center to describe contemplated actions against Saddam Hussein: “Following Iraq’s illegal takeover of Kuwait, the United Nations has adopted  12  resolutions over the past months in an attempt to resolve the crisis peacefully and without the use of force. The centerpiece of the UN initiatives was an agreement to apply economic sanctions against Iraq, which would result in its peaceful withdrawal from Kuwait” (CR, S-396). “Periphery” is an opposite of “center” and its textual use illustrates another important dimension of bodily orientation. For example, Sen. Paul Sarbanes (D-Maryland) distinguishes between vital (or central) and periph ri pher eral al co comp mpon onen ents ts of th thee na nati tion onal al in inte terres est: t: “O “Off co cour urse se,, we ha have ve in inte terres ests ts in the Gulf. But it is essential to distinguish between peripheral interests and an d vi vita tall in inte terres ests ts.. Vit ital al in inte tere rest stss ex exis istt wh when en ou ourr na natio tiona nall se secu curi rity ty is tr trul uly y at risk. Vital interests interests are those you kill and die for” (CR, S-154). In much the same way, way, peripheral elements of the human body – such as skin or even limbs – may be sacrificed in order to maintain the “center” of the body – the life essence, or “soul.” “Col “C olle lect ctio ion” n” sh show owss up mo most st fr freq eque uent ntly ly as “c “col olle lect ctiv ive” e” an and d is us used ed to referr to th fe thee in indi divi vidu dual al le lead ader er or na nati tion on an and d co comm mmon on ac acti tivi vitie tiess ta take ken n in co conc ncer ertt  by democratic societies. The American president president acts as the collective reprepresentative of a democratic people, mobilizing and applying their united forces and strength. As Rep. Frank McCloskey (D-Indiana) pointed out, “Such power ultimately is not up to one man, but the collective wisdom of the people through their elected representative” (CR, H- 152). Demo-

cratic judgment, conscience, and decision are collective. As Rep. Patricia Schr Sc hroe oede derr (D (D-C -Col olor orad ado) o) po poin inte ted d ou out: t: “T “Tha hatt is wh what at th this is de demo mocr crat atic ic pr prin inci ci-ple is all about. This is not a country where we recognize that one person has all the wisdom. Every one of us has feet of clay, clay, and the best judgment we ca can n ha have ve is a lo lott of co coll llec ecti tive ve ju judg dgme ment nt in th this is wo wond nder erfu full Re Repu publ blic ic”” (C (CR, R, H-153). Ano Anothe therr re repr prese esenta ntativ tive, e, Rep Rep.. Hen Henry ry Waxm axman an (D(D-Cal Califo iforni rnia), a), not noted ed that America has a leadership role and “bears the brunt of our collective security burden” (CR, H-156). “Compulsion” is used as a frame to distinguish between free and slave societies, the free American Self and the enslaved Other. One of the marks dist di stin ingu guis ishi hing ng th thee op oppr pres essi sive ve re regim gimee of Sa Sadd ddam am Hu Huss ssei ein n is th thee us usee of co commpulsory labor. labor. The theme of compulsion also enters the democratic debate in Co Cong ngrres ess. s. Se Sena nato torr Jo Jose seph ph Li Libe berm rman ann n (D (D-C -Con onne nect ctic icut ut)) ma made de it cle clear ar th that at he did not wish to create an unseemly compulsion of the president to go to war. Rather, he wanted Congress to share the collective responsibility for the actions that must be taken: “I make my choice today to support the

 

110

 

Embodiment and Cognitive Science

President of the United States, to give him not a compulsion to go to war, war,  but an authorization to commit our troops to battle should he determine it necessary to protect our national security” (CR, S-376). The network of political life is constructed through contact. The world of international relations is bound together through contact. As Sen. Trent Trent Lott (R-Mississippi) noted: “The world is united – and I have to commend President Bush for the effort he has made through the United Nations and with wi th pe pers rson onal al di dipl plom omac acy y, pe pers rson onal al co cont ntac acts ts wi with th le lead ader erss al alll ov over er th thee wo worl rld d to bring the world together, unite against the aggression of one man, really, Saddam Hussein” (CR, S-376). The great issues of war and peace also depend on contact. During the Gulf War debate, Sen. Patrick Moynihan (D-New York) noted that “No battle plan ever survived contact with the enemy” (CR, S-394). Likewise, Sen. Hank Brown (R-Colorado) evinced  belief in contacts by stating his hope “that economic sanctions and diplomatic contacts would convince Saddam to restore Kuwaiti sovereignty” (CR, S-396). “Container” obviously translates into “containment,” one of the major orienting terms of postwar international relations. The Gulf War debate evokes new nuances. Iraq is a container that contains Kuwait, and Kuwait is a container that contains significant territory and oil. As Sen. Symms reported: “According to maps distributed to Iraqi embassies, this territorial enclave consists of Kuwait’s Northern Province which contains approximately one-third of Kuwait’s territory, and one-fifth of its oil” (CR, S-380). “Containment,” in contrast to “container,” suggests a state of being rather than an actual object. As Sen. Kerry (D-Massachusetts) pointed out: “We sustained our fight against the Soviets for  40  years after Stalin took over Eastern Europe. We We contained Stalinism, and in time, an isolated and decaying Soviet Union has been going through a process of caving in” (CR, S-249). The proposed strategy of containment took on a more economic flavor. flavor. As Senator Sarbanes spoke of the Iraqi case, the assumption of those who supported a sanctions policy was that over time “as the bite

of these economic sanctions were felt and the punitive containment the embargo, the blockage, the use of force to make the sanctions effective through the blockade – as that bite (became) stronger and stronger with thee pa th pass ssag agee of ti time me,, it wo woul uld d ov over er ti time me le lead ad to hi hiss de depa part rtur uree fr from om Ku Kuwa wait it”” (CR, S-151). Finally, the New the  New York Times talked Times  talked of the wider political and military containment when mentioning what would happen if economic cont co ntai ainm nmen entt wa wass no nott ef effe fect ctiv ive: e: “t “the he co confl nflic ictt wo woul uld d th then en be beco come me re regio giona nall lly y destabilizing, on a scale that is difficult precisely to define but that could  become also impossible to contain” (CR, S-155). This discussion demonstrates how many key political concepts can be trac tr aced ed ba back ck to bo bodi dily ly re refe ferren ents ts.. At Attr trac acti tion on is co conn nnec ecte ted d to al allia lianc nce, e, ba bala lanc ncee to the balance of power, physical blockage to blockade, center-periphery

 

Concepts

 

11 111 1

to core and marginal interests, collection to collective interests as well as collective defense and security, compulsion to the use of force and coercion, contact to diplomatic discourse and military friction, and container to containment. Mathematical Concepts

Mathematics is reputed to be the ideal case of disembodied thought. On the surface, mathematics seems to reflect highly abstract, transcendental ideas. But recent studies suggest that mathematical concepts are formed  by two fundamental types of embodied metaphors: grounding metaphors and an d lin linkin king g me meta taph phor orss (L (Lak akof offf & Nu Nune nez, z, 2000). Gr Grou ound ndin ing g me meta taph phor orss si sittuate mathematical ideas in everyday embodied experience. For instance, grounding metaphors allow us to conceptualize arithmetic operations in terms of forming collections, constructing objects, or moving through space. These metaphor preserve image-schematic structures so that inferences about collecting, constructing, and moving are mapped onto the abstract domain of arithmetic. Some of the most basic grounding metaphors are the following: ARITHMETIC IS OBJECT COLLECTION –   numbers are are collections of physical physical objects of uniform uniform size –   the mathematical agent is the collector of objects –   the results results of an arithmetic operation is a collection of objects –   th thee si size ze of th thee nu numb mber er is th thee ph phys ysic ical al si size ze gg ggvo volu lume me)) of th thee co coll llec ecti tion on –   equations are are scales weighing collections that balance –   addition is putting collections collections together to form larger collections –   subtractionistakingsmallercollectionsfromlargeronestoformother collections –   mu mult ltip ipli lica cati tion on isof theetimes th rep epea eate ted d ad addi diti tion on of co coll llec ecti tion onss of th thee sa same me si size ze a given number –   di divi visi sion on is th thee rep epea eate ted d di divi vidi ding ng up of a gi give ven n co coll llec ecti tion on in into to as ma many ny smaller collections of a given size as possible –   zero is an empty collection ARITHMETIC IS OBJECT CONSTRUCTION –   numbers are physical objects objects –   arithmetic operations are acts of object construction construction

–   the results results of an arithmetic operation is a constructed object object –   the size of the number is the size of the object –   equations are are scales weighing objects that balance –   addition is putting objects together with other objects to form larger larger objects –   su subt btra ract ctio ion n is ta taki king ng sm smal alle lerr ob obje ject ctss fr from om la larg rger er ob obje ject ctss to fo form rm ot othe herr objects

 

112

 

Embodiment and Cognitive Science

repeated addition of objects of the same size a –   multiplication is the repeated given number of times –   division is the repeated repeated segmentation of a given object into as many objects of a given smaller size as possible –   zero is the absence absence of any object Everyday talk about arithmetic reveals the constraining presence of  these different governing metaphors, including “A trillion is a big num ber.”  ber .” “How many  5’s are there in 20?” “There are four  5s in  23, and 3 left over.” “How many times does  2  go into  10 ?” “If  10  10  is on one side of the equati equ ation on and 7 is on the the other other,, what do you have to add to 7 to balance the equation?” Different linguistic examples distinguish the Object Collection metaphor from the Object Construction metaphor. Object Collection: “How many more than 5 is 8?” “8 is 3 more than 5.” Object Construction: “If you put 2 and 2 together, it makes 4.” “What is the product of  5  5 and 7?” “Two is a small fraction of  248  248.” A different grounding metaphor is ARITHMETIC IS MOTION –   numbers are are located on on a path –   the mathematical agent is a traveler along that path path –   arithmetic operations are acts of moving along the the path –   the result result of an arithmetic operation is a location location on the path –   zero is the origin (starting point) –   the smallest whole number (one) (one) is a step forward forward from the origin –   the size of the number of the length length of the trajectory from from the origin to the location –   equations are are routes routes to the same location –   addit addition ion of a given quantity quantity is taking steps a given distance distance to the right (or forward) –   su subt btra ract ctio ion n of a gi give ven n qu quan anti tity ty is ta taki king ng st step epss a gi give ven n di dirrec ecti tion on to th thee left (or backward) –   multiplication is the repeated repeated addition of quantities of the same size a given number of times –   div ivis isio ion n is th thee rep epea eate ted d seg egme ment ntat atio ion n of a pa path th of a gi give ven n le len ngt gth h in into to

as many smaller paths of a given length as possible Once more, everyday language illustrates the ARITHMETIC IS MOTION metaphor, such as “How close are these two numbers?” “ 37  is far

 

Concepts

 

113

away from 189,712,” “ 4.9 is almost  5,” “The result is around 40,” “Count up to   20  without skipping any numbers,” “Count backward from   20,” “Count to 100, starting at 20,” and “Name all the numbers from  2  to 20.” These examples show how embodied metaphors underlie systematic talk of arithmetic and arithmetic operations. Zero is not the same kind of thing as a number in the Object Collection and Object Construction metaphors,  but is a number under the Motion metaphor metaphor.. Thus, for the Object Collection and an d Co Cons nstr truc ucti tion on me meta taph phor ors, s, ze zerro re repr pres esen ents ts th thee ab abse senc ncee of at attr trib ibut utes es,, bu butt under the Motion metaphor zero is a specific location in space. Lakoff and Nunez argue that the Collection and Construction metaphors are so basic that it took a long time for zero to be included i ncluded as a number. number. There are two related experiences that serve as the metaphorical basis for set theory: (1) grouping objects with conceptual containers, and ( 2) comparing the number of objects in two groupings. The source domain of  thee me th meta taph phor or em empl ploy oyss a co cont ntai aine nerr sc sche hema ma th that at sp spec ecifi ifies es a bo boun unde ded d reg egio ion n of space, with an interior interior,, a boundary boundary,, and an exterior. exterior. Objects within the  boundary are in the container. container. Sets in mathematics are conceptualized as cont co ntai aine nerr sc sche hema mass an and d th thee nu numb mber erss of th thee se sets ts ar aree vi view ewed ed as ob obje ject ctss in insi side de the container: THE SETS-AS-CONTAINER-SCHEMA METAPHOR –   a set is a container schema schema –   a member of a set is an object in a container container schema –   a subset of a set is a container-schema within a container-schema This metaphor can easily be extended to metaphorically define unions, intersections, and complements. A different metaphor, the SETS ARE OBJECTS metaphor, makes it possible for objects to be set numbers. Alth Al thou ough gh th this is me meta taph phor or sh show owss ho how w se sets ts ca can n be me memb mber erss of ot othe herr sets, combining this metaphor with the SETS-AS-CONT SETS-AS-CONTAINER-SCHEMA AINER-SCHEMA metaphor illustrates why a set cannot be a member of itself. The reason is that the container-schema can not be inside itself. i tself. My di disc scus ussi sion on he herre is lim limit ited ed to ar arith ithme meti tica call me meta taph phor ors. s. La Lako koff  ff  and Nunez (2000)   provide more extensive analyses of how embodied meta me taph phor orss un unde derl rlie ie ma many ny as aspe pect ctss of co comp mple lex x ma math them emat atic ics, s, in incl clud udin ing g to toppicss in lo ic logic gic,, tr tran ansfi sfini nite te nu numb mber ers, s, po poin ints ts at in infin finity ity,, in infin finite itesi sima mals ls,, an and d so on on.. They Th ey sh show ow ho how w th thee ba basi sicc co cogn gnit itiv ivee me mech chan anis isms ms fo foun und d th thro roug ugho hout ut hu huma man n conceptual systems (i.e., image schemas, conceptual metaphors, conceptual blends) are also part of where mathematics comes from. As they note, “Mathematics is not built into the universe. The portrait of mathematics has a human face.” More importantly, mathematics comes into being because of the kinds of brains and bodies we possess. These are provocative, and certainly speculative ideas, on what is traditionally viewed as one of the most disembodied abstract concepts. Yet these analyses are quite

 

114

 

Embodiment and Cognitive Science

consistent with the evidence described in this chapter on the embodied character of many concrete and abstract concepts. Questions about Image Schemas

The fact that one can talk about different kinds of image schemas and different ways in which these can be transformed certainly suggests that image schemas are definable mental representations. But how are image sche sc hema mass re repr pres esen ente ted, d, gi give ven n th thei eirr cr cros osss-mo moda dall ch char arac acte ter? r? Wh Wher eree mig might ht im im-age schemas be represented in the brain, given that they arise from recurring bodily experiences that cut across vision, audition, kinesthetic movement, and so on (i.e., are the SOURCE-PATH-GOAL and MOMENTUM schemas encoded in visual cortex or some other part of the brain)? The abstract, yet still definable, character of image schemas does not provide easy answers to these questions. At this point, linguists and psychologists should be cautious in making concrete claims about how and where image schemas might be mentally represented. Image schemas may best be understood as experiential gestalts that do not necessarily get encoded as explicit mental repr representations. esentations. Asdifferent question concerns which aspects recurring bodily ences ence nece ne cess ssar arily ily gi give ve ri rise se to ima image ge sche sc hema mas. s. Fo Forrofin inst stan ance ce,, lik like e mo most stexperipeop pe ople le and animals, I regularly scratch itches on my skin. Does this mean that I must have a SCRATCH image schema? I do not believe that people have SCRATCH image schemas, primarily because these are not part of our cross-modal experience. However, However, scratching behavior helps restore bodily equilibrium and thus is part of the varied bodily experiences that give rise ri se to a BA BALA LANC NCE E sc sche hema ma.. Mo Most st ge gene nera rall lly y, th thee bo bodi dily ly be beha havi vior orss th that at wi will ll give gi ve ri rise se to im imag agee sc sche hema mass ar aree th thos osee th that at ar aree re recu curr rrin ing g an and d he help lp so solv lvee ad adap ap-tive problems. Schemas such as BALANCE, SOURCE-PATH-GOAL, and MOME MO MENT NTUM UM co cont ntri ribu bute te to ov over eral alll bo body dy sc sche hema mass th that at he help lp en ensu sure re hu huma man n survival. The emergent nature of image schemas as in-the-moment embodied simulations is best understood theoretically in terms of the complex interplay of brain, body, and world. Image schemas may be described as emergent properties, a kind of structural coupling between brain, body, and world, that arise from different “cycles of operation” constituting a person’s life. Image schemas reflect a form of stability within cognitive systems. According to self-organization theory, order in a system arises around attractors that help create and hold stable patterns within the system. Attractors are preferred patterns, such that if the system is started from one state it will evolve until it arrives at the attractors and will stay there in the absence of other factors. An attractor can be a point (e.g., the center of a bowl containing a rolling ball), a regular path (e.g., a planetary orbit), a complex series of states (e.g., the metabolism of a cell), or an

 

Concepts

 

115

infinite sequence (called a strange attractor). A complex system will have many attractors and the study of self-organizing systems is focused on investigating the forms and dynamics of these attractors. My suggestion is that image schemas are attractors within human self-organizing systems. Attractors, such as BALANCE, SOURCE-P SOURCE-PA ATHGOAL, RESIS RESIST TANCE, VER VERTICALI TICALITY TY,, and PATH refl reflect ect emer emerging ging points pointsof  of  stab st abili ility ty in a sy syst stem em as it en enga gage gess in re real al-w -wor orld ld in inte tera ract ctio ion. n. Ne New w, su surp rpri risi sing ng patterns encountered in the environment throw a system into momentary chao ch aoss (e (e.g .g., ., th thee sy syst stem em go goes es ou outt of BA BALA LANC NCE) E),, un unti till th thee sy syst stem em,, th thrrou ough gh it itss self-a sel f-asse ssembl mbly y pr proce ocess, ss, re reor organ ganize izess and re reach aches es a new sta stabil bility ity (e. (e.g., g., re reach aches es a ne new w st stat atee of eq equi uili libr briu ium m or BA BALA LANC NCE) E).. Th Thee im impo port rtan antt po poin intt he herre is th that at attractors are not localized representations, but emerging patterns of entirre sy ti syst stem emss in ac acti tion on (i (i.e .e., ., in inte terp rpla lay y of br brai ain, n, bo body dy,, an and d wo worl rld) d).. In th this is wa way y, the stable properties of image schemas (e.g., the topographic structure of  something like SOURCE-P SOURCE-PA ATH-GOAL) are not separate from sensorimotorr ac to acti tivi vity ty.. Ima Image ge sc sche hema mass sh shou ould ld no nott be re redu duce ced d to se sens nsor orimo imoto torr ac acti tivi vity ty,,  but it is a mistake to view image schemas as mental repre representations sentations that are abstracted away from experience. One implication of this dynamical view is that each construal of an image schema will have a different profile depend pe ndin ing g on th thee ov over eral alll st stat atee of th thee or orga gani nism sm in invo volv lved ed in so some me ac acti tivi vity ty an and d past basins of attractions created within the system (i.e., past simulations of a particular behavioral modes such as BALANCE). Question about Conceptual Metaphors

The cognitive linguistic research on conceptual metaphor has provided significant evidence on the embodied grounding of abstract thought. But there are several problems with the theory. First, conceptual metaphors appear to differ in the way they are experientially grounded (Grady,  1997, 1999). For instance, consider the well-known conceptual metaphor MORE IS UP (e.g., “Inflation is up this year”). It is easy to correlate having more of some objects or substance (i.e., quantity) with seeing the level of those objects or substance rise (i.e., verticality). But many conceptual metaphors do not sug sugges gestt suc such h str straig aightf htforw orwar ard d exp experi erient ential ial cor corre relat lation ions. s. For ins instan tance, ce, the well-known conceptual metaphors THEORIES ARE BUILDINGS and LOVE IS A JOURNEY do not seem to have the same kind of correlation in experience as seen in MORE IS UP. Thus, actual travel has little to do with the progress of relationships, and theories are not closely tied to the  buildings in which people generate, generate, discuss, and dismantle these ideas. A related problem with conceptual metaphor theory is that it does not explain why certain source-to-target domain mappings are not likely to occur (Grady (Grady,, 1997, 1999). For instance, a common way for people to think about the concepts of “theory” is in terms of the conceptual metaphor THEORIES ARE BUILDINGS. This conceptual metaphor motivates many

 

116

 

Embodiment and Cognitive Science

meaningful linguistic expressions such as “The theory needs to be but-

tressed” or “The foundation for your theory is shaky.” But some aspects of buildings are clearly not mapped onto the domain of theories, which is one reason why it sounds odd to say “The theory has no windows.” One problem that metaphor theorists have long struggled with is why some portions of a source domain get mapped onto a target domain, but nott ot no othe hers rs.. Co Cons nsid ider er th thee ex expr pres essi sion on “T “Thi hiss bo book ok is ha harrd to di dige gest st.” .” We rea eaddily recognize this expression as conveying something about the domain of  thinking in terms of the t he domain of eating. Yet Yet only certain aspects of what we know about eating get mapped into our understanding of thinking. Thus, we rarely hear people talk about their mouths in conventional expressions about thinking, but we do hear people say things such as “The author tried to bite off more than he could chew.” chew.” What accounts for some expressions being acceptable and others not? What explains the “gaps” in metaphorical mapping processes? Another problem for metaphor theory concerns how different conceptual metaphors relate to one another. Conceptual metaphors differ considerably.. Some metaphors have greater detail and complexity in the kind siderably of inferences they evoke. For instance, the conceptual metaphor MORE IS UP relates quantity to vertical elevation, as in “Gas prices are up this year” or “Inflation has stayed down over the last four years.” This mapping of vertical elevation into quantity leads to a straightforward set of  meta me taph phor oric ical al ma mapp ppin ings gs.. Yet co cons nsid ider er th thee co conc ncep eptu tual al me meta taph phor or LO LOVE VE IS A  JOURNEY,, as in “Our relationship  JOURNEY relationship is at a cross-roads” cross-roads” or “My marriage is on the rocks.” In fact, LOVE IS A JOURNEY inherits the structure of, and elaborates on, the more general metaphorical idea of LONG-TERM PURPOSEFUL ACTIVITIES ARE JOURNEYS (Lakoff,  1993). Although these two metaphorical mappings seem logically related, it is difficult to determine which conceptual metaphors can be elaborated and are susceptible to being inherited. A further challenge for metaphor theory is that some metaphors are equall equ ally y app appro ropri priate ate to des descri cribe be dif differ ferent ent con concep ceptua tuall dom domain ains. s. For ins instan tance, ce, the word “feed” can be used to describe a professor’s teaching style, as in “The professor spoon-feeds his students,” as well as to talk about a completely different domain, as in “The minor leagues feed players into major league baseball.” How may theorists account for this wide range of  metaphorical uses to talk about abstract concepts? A New View of Embodied Metaphor

An in inte tere rest stin ing g so solu luti tion on to th thes esee pr prob oble lems ms su sugg gges ests ts th that at co conc ncep eptu tual al metaphors are not the most basic level at which metaphorical mappings exist in human thought and experience. Grady (1997)   argued that the strong correlation in everyday embodied experience leads to the creation

 

Concepts

 

117

of “primary” metaphors. Some of the most prominent primary metaphors are the following: INTIMACY IS CLOSENESS C LOSENESS (e.g., “We “We have a close relationship”) DIFFICULTIES DIFFICUL TIES ARE BURDENS (e.g., “She’s weighed down by respon-

sibilities ) AFFECTION IS WARMTH (e.g., “They greeted me warmly”) IMPORTANT IS BIG (e.g., “Tomorrow is a big day”) MORE IS UP (e.g., “Prices are high”) SIMILARITY IS CLOSENESS (e.g., “Those colors aren’t the same, but they’re close”) ORGANIZATION IS PHYSICAL STRUCTURE (e.g., “How do theories ORGANIZATION fit together?”) HELP IS SUPPORT (e.g., “Support your local charities”) TIME IS MOTION (e.g., “Time flies”) STATES ARE LOCATIONS (e.g., “I’m close to being in a depression”) CHANGE IS MOTION (e.g., “My health has gone from bad to worse”) PURPOSES ARE DESTINATIONS (e.g., “He’ll be successful, but isn’t there yet”) CAUS CA USES ES AR ARE E PH PHYS YSICA ICAL L FO FORCE RCES S (e (e.g .g., ., “T “The hey y pu push shed ed th thee bi bill ll th thrrou ough gh Congress”) KNOWING IS SEEING (e.g., “I see what you mean”) UNDE UN DERS RST TAN ANDI DING NG IS GR GRAS ASPI PING NG (e (e.g .g., ., “I “I’v ’vee ne neve verr be been en ab able le to gr gras asp p complex math”) These metaphorical correlations arise out of our embodied functioning in the world. In each case, the source domain of the metaphor comes from the body’s sensorimotor system. A primary metaphor is metaphorical mapping for which there is an independent and direct experiential  basis and independent linguistic evidence. A “compound” or “complex” metaphor, on the other hand, is a self-consistent metaphorical complex composed of more than one primary metaphor. Complex metaphors are crea cr eate ted d by bl blen endi ding ng pr prim imar ary y me meta taph phor orss an and d th ther ereb eby y fit fittin ting g to toge geth ther er sm smal alll metaphorical pieces into larger metaphorical wholes. For instance, consider the following three primitive metaphors: PERSISTING IS REMAINING ERECT, STRUCTURE IS PHYSICAL STRUCTURE, and INTERRELATED IS INTERWOVEN. These three primitives can be combined in different ways to give rise to compound metaphors that have traditionally been seen as conceptual metaphors. But the com bination of these primitives allows metaphorical concepts without gaps. Thus, Thu s, com combin bining ing PER PERSIS SISTIN TING G IS REM REMAIN AINING ING ERE ERECT CT wit with h STR STRUCT UCTURE URE IS PHYSICAL STRUCTURE provides for a compound THEORIES ARE BUILDINGS that nicely motivates the metaphorical inferences that theories need support and can collapse, and so on, without any mappings such as that theories need windows. In a similar way, the combination

 

118

 

Embodiment and Cognitive Science

of STRUCTURE IS PHYSICAL STRUCTURE and INTERRELATED IS INTERWOVEN INTERWOV EN gives rise to a different metaphorical compound for theories, namely, THEORIES ARE FABRICS. This compound metaphor gives rise ri se to th thee rea easo sona nabl blee in infe ferren ence cess th that at th theo eori ries es ca can n un unra rave vell or ma may y be wo wove ven n together, without generating less likely entailments such as that theories are colorful in the way that some fabrics have colors. This view of the embodied basis for metaphorical thought and language solves the “poverty of mapping” problem often noted for conceptual metaphor and other theories of metaphor (Grady,  1997). There is no

need to posit specific mechanisms that override parts of source-to-targetdomain primitive metaphors because thetarget positive correlation inmappings embodiedinexperience between the sourceofand domains (Grady,  1997). Moreover, the correlation between source and target domains may possibly be instantiated in the body via neural connections (Lakoff & Johnson, 1999). Under this view view,, neural connections in the brain may reflect how inferences from the sensorimotor source domain (i.e., verticality) are projected onto the subjective target (i.e., quantity). Quite generally, metaphor is an extended consequence of topographic mappings, As in all topographic mappings, the structure of the source domain is preserved in the target domain because the neurons of the former map to (i.e., stimulate) the latter through reentrant signaling (or its equivalent). How do these connections form? Consider the MORE IS UP meta me taph phor or.. In th this is ma mapp ppin ing, g, th thee ab abst stra ract ct do doma main in of qu quan anti tity ty or va valu luee co corr rreelatess wi late with th th thee rel elat ativ ivee ch chan ange gess al alon ong g th thee ve vert rtic ical al ax axis is of th thee sp spat atia iall do doma main in,, as in “my stocks skyrocketed” and “his productivity is way up.” These correlations occur by the neural networks characterizing each of these domains are coactivated in everyday experience, as when we pile more  books on the desk and their height goes up or we add water to a container.. If the mappings are connected by reentrant pathways, coactivation tainer will strengthen the connections. Once those connections have formed, the relations in the source domain of verticality will be preserved by the mapping and therefore can form the basis of inferences in the target domain of quantity. If something shoots up, it is propelled quickly upward and in a very short time is much higher than before. Hence, the phrase “her famee sky fam skyro rocke cketed ted”” ind indica icates tes a sud sudden den and sub substa stanti ntial al inc incre rease ase in cel celebr ebrity ity.. Meta Me taph phor or,, th then en,, is a ne neur ural al me mech chan anis isms ms th that at en enab able less ne netw twor orks ks us used ed in se sennsorimotor activity also to serve as the substrates hat make abstract reason possible. Is Cognitive Linguistic Evidence Relevant to the Study of Cognition?

As is th thee ca case se wi with th al alll sc scie ient ntifi ificc me meth thod ods, s, th ther eree ar aree li limit mitat atio ions ns to th thee st stra rate tegy gy of trying to infer something about conceptual structure from from a systematic anal an alys ysis is of lin lingu guis istic tic st stru ruct ctur uree an and d be beha havio viorr. Th Thee pr prim imar ary y lim limit itat atio ion n is th that at

 

Concepts

 

119

shared by most linguistic research, namely the problem of making conclusions about phenomena based on the individual analyst’s own intuitions. Many cognitive scientists believe that trying to infer aspects of conceptual knowledge from an analysis of systematic patterns of linguistic structure makes these theories appear post hoc. For instance, the claim that the systematicity in expressions such as “He’s wasting our time,” “I save an hour doing my paper on the computer computer,” ,” and “I can no longer invest that much energy into my marriage” is due to the presence of an independent, preexis ex istin ting g co conc ncep eptu tual al me meta taph phor or TI TIME ME IS MO MONE NEY Y pr prov ovid ides es on only ly a mo moti tiva vate ted d expl ex plan anat ation ion fo forr lin lingu guis isti ticc be beha havi vior or.. Cog Cogni nitiv tivee sc scie ient ntis ists ts wi wish sh to pr pred edic ictt be be-havior hav ior in adv advanc ancee acc accor ordin ding g to the hyp hypoth otheti eticoco-ded deduct uctive ive met method hod of sci scienentifi tific cceptua infe in fere renc nce. e. Wh What atdge they th ey seek se ek iswem empi piri rica cal, l,the obje ob ject ctive ivence evid ev iden ceferent that th at pe peop ople le’s ’sconcep con tual l kno knowle wledge someho som ehow predi pr edicts cts existe exi stence ofence differ dif ent linguis ling uis-

tic be beha havi vior or,, no nott th that at pe peop ople le s li ling ngui uist stic ic be beha havi vior or ca can n be ex expl plai aine ned d by po posi sitting theoretical entities such as conceptual metaphor. As argued by psychologist Sam Glucksberg, in his review of Koveceses’ book Metaphor book Metaphor and Emotion ( Emotion  (2000), “From the perspective of cognitive science, the cognitive lingu lin guis isti tics cs pr prog ogra ram, m, as ex exem empl plifie ified d by sc scho hola lars rs su such ch as Ko Kove vecs cses es,, is ov over erly ly (and, in my mind, unnecessarily) limited. It would benefit greatly from a deployment of convergent operations – that is, multiple methods to validate inferences drawn from linguistic evidence” (Glucksberg, 2002: 765). This Th is sk skep eptic ticis ism m ab abou outt th thee th theo eore reti tica call cl clai aims ms of co cogn gnit itive ive lin lingu guis ists ts ab abou outt the embodied, metaphorical nature of abstract concepts is surely reasonable. All scientific theories require the support of evidence obtained from convergent operations. But I strongly reject the implicit assumption that cognitive linguistic research stands outside of “cognitive science,” which presumably possesses appropriate appropriate scientific methods to satisfactorily test important hypotheses about the human mind. First, the detailed analyses of how peo eop ple ta talk lk ab abou outt ab abst strrac actt co conc nceept pts, s, on only ly som omee of wh whic ich h hav avee be been en described in this chapter, must be explained, and not merely dismissed out of hand, as is done by many psychologists and philosophers. The vast majority of studies in experimental psycholinguistics on metaphor compreh pr ehen ensi sion on fo focu cuss on si simp mple le res esem embl blan ance ce,, or “A is li like ke B, B,”” me meta taph phor ors, s, su such ch as “M “My y la lawy wyer er is a sh shar ark, k,”” an and d ig igno nore re sy syst stem emat atic ic co conv nven entio tiona nall me meta taph phor ors, s, such as those described above. However, However, cognitive linguistics li nguistics has demonstrated the pervasiveness of conventional metaphors of varying types in speech and writing. This linguistic evidence must be explained by psycholinguists who propose general models of metaphor understanding. Psychologists sometimes defend their neglect of embodied, conventional metaphors by arguing that some cognitive linguistic analyses are contrary to their own intuitions. Thus, psychologists voice skepticism about the intuitive, introspectionist methods of cognitive linguistics, but then justify their neglect of conventional metaphors because of their own intuitions! It would be far better, better, in my view view,, for psychologists and others

 

120

 

Embodiment and Cognitive Science

to ex expl plic icit itly ly st stud udy y em embo bodi died ed me meta taph phor orss fo forr ab abst stra ract ct co conc ncep epts ts ac acco cord rdin ing g to acce ac cept pted ed em empi piri rica call me meth thod odss an and d ma make ke de deci cisi sion onss ab abou outt co cogn gnit itiv ivee li ling ngui uist stic ic claims based on these studies, and not simply to dismiss this work out of  hand. Chapter  6  offers a more detailed discussion of some of this work, much of which is supportive of cognitive linguistic claims on embodied metaphors. Apart from concerns about whether cognitive linguistic claims about abstract concepts are “psychologically real,” there are some issues about cogn co gnit itive ive lin lingu guis isti ticc re rese sear arch ch me meth thod odss th that at sh shou ould ld be fu furt rthe herr ex exam amin ined ed.. Fo Forr example, exactly what constitutes sufficient “systematicity” among conventional expressions to properly infer that these are motivated by some underlying conceptual metaphor? How does one determine the appropriat ate e le leve vell of ge gene nera rali lity ty when wh en id iden enti tify fyin ing g co conc ncep eptu tual al A meta me taph phor ors? s? Th Thus us,, is th thee appropriate source domain for the classic LIFE IS JOURNEY metaphor really journeys, and not some other idea, such as travel, movement from onee pl on plac acee to an anot othe herr, or ph phys ysic ical al mo move veme ment nt of an any y so sort rt?? Fu Furt rthe herm rmor ore, e, ho how w do conceptual metaphors relate to one another within human conceptual

system syst ems? s? Co Cogn gnit itiv ivee lin lingu guis ists ts ha have ve va vari riou ouss res espo pons nses es to so some me of th thes esee qu ques es tions (see Gibbs,  1994;  Grady,  1999;  Kovecses,  2000 b  b;;  Lakoff,  1990,  1993; Yu, 1999). But cognitive linguists li nguists need to provide more explicit criteria for identifying conceptual metaphors that scholars in other disciplines may employ in conducting their own empirical investigations on the possible embodied metaphorical nature of certain abstract concepts. Conceptual metaphors as prototypical representations are assumed to exist as enduring knowledge structures in long-term memory and are essentia sen tiall to the con conten tentt of eve everyd ryday ay,, abs abstra tract ct con concep cepts. ts. Yet cog cognit nitive ive psy psycho cholologist gi stss ar aree of ofte ten n sk skep epti tica call of th this is cl claim aim,, pr prim imar arily ily be beca caus usee th they ey do doub ubtt wh whet ethe herr linguistic lingui stic evidence alone can rev reveal eal much about human conceptual conceptual systems (Murphy,  1996). One instance of this skepticism about metaphors as conceptual prototypes is seen in the problem of multiple metaphors (Murphy, 1996; Gibbs, 1996). According to cognitive linguistics analyses, the concept of love, for example, can be understood through several different metaphors (e.g., LOVE IS A JOURNEY, LOVE IS INSANITY, LOVE IS AN OPPONENT, LOVE IS A VALUABLE COMMODITY). The entailments of these different metaphors vary in certain respects. Thus, LOVE IS A JOURNEY refers to the structure of a love relationship over time, wher wh erea eass LO LOVE VE IS AN OP OPPO PONE NENT NT pe pers rson onifi ifies es lo love ve as an op oppo pone nent nt ag again ainst st wh whom om we of ofte ten n st stru rugg ggle le.. Th Thes eseeand difffe di fer ent t me meta taph phor orss to appe ap pear ar,, at such time ti mes, s,inconto be inconsistent with one another it risen unclear how resolve sistencies in the mental representation for our concept of love. This argument preserves a view of concepts as monolithic entities that should be internally consistent. But the so-called problem of multiple meta me taph phor orss fo forr co conc ncep epts ts ca can n be ea easi sily ly ha hand ndle led d if we vi view ew th thes esee pr prot otot otyp ypic ical al concepts not as fixed, static structures, but as temporary representations

 

Concepts

 

121

that are dynamic and context-dependent. The LOVE IS A JOURNEY metaphor may better reflect a particular embodied conceptualization of  love in certain situations, whereas LOVE IS AN OPPONENT may arise in forming a concept for love in other situations. These alternative ways of  thin th inki king ng ab abou outt hu huma man n co conc ncep epts ts al allo low w, an and d ev even en en enco cour urag age, e, th thee us usee of mu mulltiple metaphors to access different aspect of our rich, body-based knowledge ed ge ab abou outt lo love ve to dif diffe ferren enti tial ally ly co conc ncep eptu tual aliz izee of th thes esee ex expe peri rien ence cess at va vari ri-ouss mo ou mome ment ntss of ou ourr ex expe peri rien ence ce.. Ea Each ch me meta taph phor oric ic co cons nstr trua uall of a co conc ncep eptt in some context results in a concept that is independent as a temporary represe re sent ntat atio ion n ap apar artt fr from om em embo bodi died ed so sour urce ce do doma main in in info form rmat atio ion n in lo long ng-t -ter erm m memory.. My suggestion, then, is that conceptual metaphors may not prememory exist in the sense of continually structuring specific conceptual domains. But conceptual metaphors may be used to access different knowledge on different differ ent occ occasi asions ons as peo people ple imme immedia diatel tely y con concep ceptua tualize lize som somee abs abstra tract ct tar tar-gett do ge doma main in gi give ven n a pa part rticu icular lar ta task sk.. Co Conc ncep eptu tual al me meta taph phor orss ma may y al also so si simp mply ly emerge as the product of conceptualizing processes, rather than serve as the underlying cause of these processes (Gibbs, 1999 b  b). ). Chapter 6 reports the findings of some new experiments that should be especi esp eciall ally y com comfor fortin ting g to cog cognit nitive ive psy psycho cholog logist istss see seekin king g mor moree obj object ective ive eviden id ence ce on th thee ro role le of em embo bodie died d co conc ncep eptu tual al me meta taph phor or in ev ever eryd yday ay la lang ngua uage ge use. us e. Th This is ne new w wo work rk pr prov ovid ides es a co comp mple leme ment ntar ary y wa way y of do doin ing g co cogn gnit itiv ivee li linnguis gu isti tics cs th that at pa part rtia ially lly he help lpss to el elim imin inat atee th thee st stri rict ct re relia lianc ncee on th thee in indi divi vidu dual al

analysts’ own intuitions in assessing different kinds of linguistic phenomena. It also shows how motivated explanations of linguistic structure can  be used to predict predict people’s linguistic behavior in experimental experimental situations. Conclusion

Studies from cognitive psychology and cognitive linguistics paint a new view of concepts, one that is contrary to the traditional position in cognitivee sci tiv scienc encee in whi which ch con concep cepts ts ar aree abs abstra tract, ct, dis disemb embodi odied, ed, dec decont ontext extual ualize ized, d, enduring mental representations. Both concrete and abstract concepts are temporary, dynamic, embodied, and situated representations. Moreover, concepts arise from acts of perceptual/embodied simulation and are not merely accessed as static representations in long-term memory. This em bodied perspective explains why concepts are flexible, multimodal, and productive and give rise to explicit inferences as they are tuned to realworld contexts. My ofcess embodied simulation innsor the creation ofreconcepts incontex cont exttadvocacy does do es no nott ne nece ssar aril ily y imp imply ly th that at th thee se sens orim imot otor or na natu tur of co conc ncep eptuall pr tua proce ocessi ssing ng is inh inher erent ently ly non nonre repr prese esenta ntatio tional nal.. Aft After er all, sim simula ulation tion pr proocesses operating to create specific concepts in context use various kinds of  know kn owle ledg dge, e, in incl clud udin ing g th that at ab abou outt th thee bo body dy,, th that at is re repr pres esen ente ted. d. Th Thes esee si simu mu-latio la tions ns ar aree no nott id iden entic tical al to th thee ne neur ural al st stat ates es th that at un unde derl rlie ie pe perrce cept ptio ion, n, ac acti tion on,,

 

122

 

Embodiment and Cognitive Science

and cognition. But conceptual simulations surely involve brain processes in cooperation with the entire nervous system and body to create imaginati na tive ve un unde ders rsta tand ndin ings gs of ev even ents ts,, bo both th wh when en en envi viro ronm nmen enta tall in info form rmat atio ion n is present, and when it is not. Metaphor is fundamental to conceptual processing. Abstract concepts are partly created from the metaphorical mapping of embodied source domains onto various target domains. In fact, abstract concepts would not exist in the ways that they do in ordinary cognition without body based metaphor metaphor.. Metaphor is not a way of accessing previously articulated abstract knowledge, but is inherent in the creation and maintenance of  abstra abs tract ct con constr strual ualss in dif differ ferent ent sit situat uation ions. s. Thi Thiss pos positio ition n sug sugges gests, ts, the then, n, tha thatt human conceptual processing is deeply grounded in embodied metaphor, metaphor, especially in regard to abstract understandings of experience.

 

5 Imagery, Memory, and Reasoning

The history of science reveals many notable examples of the power of  embodied thought in creativity and imagination. Scientists frequently acknowledge that their great discoveries are brought about not through formal, ma l, pu purrel ely y an anal alyt ytic ic rea easo soni ning ng,, bu butt by “g “gut ut fe feel elin ings gs”” th that at ta take ke sh shap apee in th thee form of rich sensory images and bodily sensations. Albert Einstein, who always recognized his weakness in mathematics, described his creative process in the following way: The words of the language, as they are written or spoken, do not seem to play any role in my mechanisms of thought. The psychical entities which seem to serve as the elements in thought are certain signs and more or less clear images which can  be voluntarily reproduced and combined. . . . The above mental entities are, in my case, of visual and some of muscular type. (Hadamard,  1945: 142–3)

Einstein’s embodied thought processes took particular shape in one of his famous thought experiments where he pretended to be a photon moving

at the speed of light. He first imagined what he saw and how he felt, and then became a second photon and imagined what he now experienced of  the first photon. Many scientists, like Einstein, have conceded that formal mathematics was useful for communicating their scientific discoveries, but that the locus of their original ideas was rooted in embodied possibilities. Another scientist, Cyril Stanley Smith, purposefully studied graphic arts to better develop his sense of the structure of metals. When he was developing alloys lo ys,, Sm Smit ith h wr wrot otee “I ce cerrta tain inly ly ca came me to hav avee a ve verry str tron ong g fe feeeli lin ng of nat atur ural al understanding, a feeling of how I would behave if I were a certain alloy, a sense of hardness and softness and conductivity and fusability and deformability and brittleness – all in a curious internal and quite literally sensuous way” (Smith, 1981: 359). These embodied images were not incident de ntal al to Sm Smit ith’ h’ss cr crea eati tive ve wo work rk,, fo forr hi hiss res esea earc rch h de depe pend nded ed up upon on “a “aes esth thet etic ic 123

 

124

 

Embodiment and Cognitive Science

feeling for a balanced structure and a muscular feeling of the interfaces pulling against one another” (Smith, 1981: 359). Simila Sim ilarly rly to sci scient entific ific thi thinki nking, ng, art artist istic ic cr creat eativi ivity ty als also o inv involv olves es ima imagin gined ed  bodily sensations. One commentator commentator on creativity in the arts suggested, To the pianist and sculptor, the instrumentalist, dancer, surgeon, and manual artisans, they [ideas] burst upon awareness in a kinesthetic form, feeling their way into varying types of muscular experience. Fingers “itch” to play, music “flows” from the hands, ideas “flow” from the pen. Movement expresses the “idea” of the dancer or orche orchestra stra conductor; conductor; the almost sensuous sensuous desir desiree to model plastic form  becomes compulsive in sculpture. (Hutchinson, 1959: 142)

Cognitive Cognit ive sci scient entist istss rar rarely ely ack acknow nowled ledge ge the emb embodi odied ed nat natur uree of hig higher her-order cognition. Gardner (1983)  persuasively argued for the concept of  kine ki nest sthe heti ticc th thin inki king ng as on onee of th thee se seve ven n fo form rmss of mu mult ltip iple le in inte tell llig igen ence ce.. Bu Butt the tendency in cognitive science has been to view kinesthetic intelligence as a separate module of mind that does not necessarily interact with other aspect aspe ctss of mi mind nd an and d la lang ngua uage ge.. Yet th ther eree is an em emer ergi ging ng li lite tera ratu ture re in se seve vera rall area ar eass of co cogn gniti itive ve sc scie ienc ncee th that at ex expl plici icitl tly y de demo mons nstr trat ates es di dire rect ct lin links ks be betw twee een n higher-order cognition and embodied action, such that embodiment is essent se ntia iall to va vari riou ouss co cogn gnit itive ive fu func nctio tions ns.. Th This is ch chap apte terr de desc scri ribe bess th this is wo work rk an and d its implications for theories of higher-order cognition. Mental Imagery

The vast majority of scholarship on mental imagery within psychology, as well as many other disciplines, ignores the role that embodiment (e.g., people’s subjective felt experiences of their bodies in action) may play in mental imagery activities. For instance, the classic empirical work on mental men tal ima imager gery y inv invest estiga igates tes pos possib sible le cor corre respo sponde ndence ncess bet betwee ween n men mental tal imagery and visual perception (e.g., Finke, 1989). Following this trend, most contemporary cognitive psychology textbooks talk about mental imagery i magery only in terms of visual perception (and, to a much lesser extent, audition). Although there are numerous studies examining people’s kinesthetic and motor imagery, imagery, few scholars, until recently recently,, searched for explicit links be-

tween kinesthetic activity and mental imagery. The recent work suggests that many aspects of visual and motor imagery share a common representational, and possibly neuropsychological, substrate. As Paivio (1986: 72) oncee not onc noted, ed, “al “alll men mental taltra transf nsform ormati ations onseng engage agemot motor or pr proce ocesse ssess tha thatt der derive ive originally from active manipulation of the referent objects.”  Imagining Human Movement 

The embodied approach to mental imagery suggests that the long-noted equivalence between mental imagery and visual perception is not inaccurate, as long as one recognizes that visual perception is shaped by

 

Imagery, Memory, and Reasoning

 

125

kinesthetic activity (see Chapter  3 ). My view of mental imagery is quite  broad. Following Newton (1996), I use the term “image” to refer to any imag im agin inar ary y in inst stan ance ce wh wher eree on onee co cons nsid ider erss wh what at it ma may y be li like ke to mo move ve on one’ e’ss  body in a certain way or to think what it may be like to manipulate an object different ort what itlymust feel like to upon in y a certai cer tain nin manner man ner,, allways withou wit hout actual act ually physic phy sically ally doing doi ngact what wha t wean areeobject ar curre cur rentl ntly thinking (Gibbs & Berg, 2002). For example, to entertain in our conscious minds the possible feeling that arises when we bend over to grab our left foot shows how people form proprioceptive mental images of an action. Thes Th esee me ment ntal al im imag ages es ar aree no nott me mere rely ly pe perc rcep eptu tual al,, bu butt ki kine nest sthe hetic tic in th thee se sens nsee of entertaining what it is like to move our bodies in particular ways. There is considerable research showing that our ability to imagine ourselves moving in certain ways subsequently influences our actual performance of those movements (a few recent studies include Corriss & Kose, 1998;  Hanrahan, Tetreau, & Sarrazin,  1995;  Hardy & Callow,  1999; Murphy, 1990; Smy Smyth th & Walle allerr, 1998). A rec ecen entt st stud udy y, fo forr ex exam ampl ple, e, sh show owed ed that when participants had to copy figures, engaging in visual imagery  benefited the drawing of overall form, whereas engaging in kinesthetic imagery facilitated fine-tuned movements of the two hands (Fery,  2003). Alth Al thou ough gh th ther eree is mu much ch de deba bate te ov over er wh whic ich h im imag agin inin ing g te tech chni niqu ques es ha have ve th thee mostt mea mos measur surabl ablee influ influenc encee on lea learni rning ng and per perfor forman mance ce (se (seee Ahs Ahsen, en, 1995), these empirical studies show, at the very least, that imagining our bodies moving has some relationship to subsequent real-life human action. Ideo Id eomo moto torr ac acti tion on ref efer erss to th thee fa fact ct th that at ju just st th thin inki king ng ab abou outt an ac acti tion on ca can n make people perform the action without any special influence of the will. will . Arnold (1946) found that the more vividly a person imaged a movement, the more it occurred. For instance, standing still and imagining falling over, by thinking about both what it would look like and what it would feel fe el li like ke,, pr prod oduc uces es mo morre te teet eter erin ing g th than an do does es th thin inki king ng ab abou outt ei eith ther er th thee lo look ok or alone. people to imagine bending their activity arms, without actuallyfeel doing so,Asking provokes movement-r movement-relevant elevant electrical in the arms’  biceps-brachial muscles (Jacobson, 1932). On Once ce mo morre, th thin inkin king g ab abou outt ac actin ting g can produce movement without the feeling of doing. Merely thinking about a kind of person can induce ideomotor mimicry of that person’s behavior (Bargh, Chen, & Burrows,  1996). In one study, college students completed a scrambled-sentence task in which some words repeatedly mentioned the idea of aging (e.g., sentences containing the words “wrinkled,” “gray,” “retired,” “wise,” and “old”). Afterward,

each participant s gait was secretly measured as he or she left the experiment room. People who earlier read words referring to the elderly in the scrambled-sentence task actually walked out of the room more slowly than did participants who were not presented with words referring to senior citizens. Postexperiment interviews suggested that participants were not consciously aware ofhaving been exposed to ideas about the elderly,

 

126

 

Embodiment and Cognitive Science

or having walked slowly out of the experiment room. But reading words referring to characteristics of the elderly unconsciously prompted people to walk slowly. Interestingly, in a different experiment, when participants were told beforehand that the words mentioned in the sentences were expect pe cted ed by th thee ex expe peri rime ment nter er to in influ fluen ence ce th thei eirr be beha havi vior or,, th they ey di did d no nott ex exhi hibi bitt the slowness in to walking afterward. In this way, conscious the cognitive encesame on action seems occur outside of participants’ will.influMany Ma ny ot othe herr st stud udie iess ha have ve re repl plic icat ated ed an and d ex exte tend nded ed th thes esee or orig igin inal al findings. One study asked college students to think about professors (Dijksterhuis & van Knippenberg,  1998). Afterward, the students gave morre co mo corr rrec ectt an answ swer erss to qu ques estio tions ns fr from om th thee ga game me “T “Tri rivi vial al Pu Purs rsui uit” t” th than an di did d participants who did not first think about professors. On the other hand, when participants were first asked to think about soccer hooligans (in a study stu dy con conduc ducted ted in Hol Hollan land), d), the they y sub subseq sequen uently tly wer weree inf inferi erior or at ans answer wering ing “Trivial Pursuit” questions compared to control subjects. Related studies have also demonstrated that getting college students to think about old age can facilitate some loss of memory (Dijksterhuis, Bargh, & Miedema, 2001). In ea each ch of th thee ab abov ovee st stud udie ies, s, th ther eree is no me ment ntio ion n of im imag ager ery y in de desc scri ribi bing ng the locus of the experimental findings. Researchers simply assume that some kind of abstract knowledge is activated (e.g., when a person reads words such as “wrinkled,” “gray,” “wise,” and “old”) and integrated via symbolic mental processes. Yet participants may actually be creating rich ment me ntal al im imag ages es wh while ile en enga gage ged d in ta task skss su such ch as un unsc scra ramb mblin ling g wo worrds to fo form rm grammatical sentences. These mental images are not simply pictorial, but reflect complex interplay between image, somatic response, and meaning. Much Mu ch of th thee wo work rk on im imag ager ery y fo focu cuse sess on pe peop ople le’s ’s de delib liber erat atee im imag agin inin ings gs ofphysicalevents.Imaginingthelocationofobjectsaroundusalsodepends on kinesthetic action. For example, one study asked people to memorize thee lo th loca cati tion on of ob obje ject ctss in a roo oom m (P (Prres esso son n & Mo Mont ntel ello lo,, 1994). Af Afte terw rwar ard, d, th thee pa part rtic ipan ants tsand werreaccur we blin bl indf dfol olde ded ddoing and an d as aske ked dBut to po poin intt toparticipants spec sp ecifi ific c ob obje ject cts. s. Pe Peop ople le were wer eicip quick accurate ate in this. when partici pants were then asked to imagine rotating 90 degrees and to point to specific objects again, they were slow and inaccurate. When participants were asked to actually rotate  90 degrees, when blindfolded, and to point to specific objects, they were just as fast and accurate as they were before rotating. A similar set of results has been reported with children (Rieser & Rider,  1991). Five- and  9-year-old children were tested on their ability to imagine (when at home) their classrooms and to point to objects from various perspectives. When the perspective changes were accompanied by actually changing positions, 5-year-olds were correct in 100% of the trials, and the   9-year-olds   98% of the time. Yet when the children only imag-

ined changing perspectives, the 5-year-olds were correct only  29% of the time and the  9-year-olds 27% of the time. A comparison group of adults

 

Imagery, Memory, and Reasoning

 

127

showed tha showed thatt whe when n act actual ually ly cha changi nging ng pos positi itions ons,, 100% of th thee th thei eirr res espo pons nses es required less than  2  seconds, whereas when they imagined the perspectivee ch tiv chan ange ges, s, on only ly 29% of th thee res espo pons nses es req equi uirred le less ss th than an 2 sec second onds. s. Onc Oncee more, important aspects of the imagination are shaped by bodily actions. Peop Pe ople le al also so ha have ve th thee ab abili ility ty to im imag agin inee th thee en envi viro ronm nmen enta tall co cons nseq eque uenc nces es of their actions. During tool use, people often change position, for example, when they use a wrench to turn a bolt. One set of studies specifically examined the idea that hand movements can facilitate imagery for object rotations but that the facilitation depends on people’s model of the tool in some situation (Schwartz & Holton,  2000). Physically turning a block without vision reduced mental rotation times compared with imagining the same rotations without bodily movement. A second study showed that pulling a string from a spool facilitated participants’ mental rotation of an obj object ect sit sittin ting g on the spo spool. ol. Ove Overal rall, l, peo people ple’s ’s ima imagis gistic tic tra transf nsform ormati ations ons are not dependent on the objective, geometric characteristics of an action. But people’s imagistic abilities are dependent on their subjective modeling of the tools that mediate motor action and the environmental consequen qu ence cess of th that at ac actio tion, n, an and d ho how w th they ey ca can n tr tran ansf sfer er th that at un unde ders rsta tand ndin ing g to ne new w situations. Thes Th esee rep epre rese sent ntat ativ ivee fin findi ding ngss ill illus ustr trat atee th thee im impo port rtan ance ce of em embo bodi died ed ac ac-tion ti on in ho how w pe peop ople le le lear arn n ov over er ti time me to me ment ntal ally ly im imag agin inee th thei eirr lo loca cati tion ons, s, an and d the locations of objects, in the world around them. Thus, imagining the location of objects in space, a task that many cognitive psychologists view as purely cognitive and divorced from the body, body, is strongly influenced by  body movements. In general, these data support the idea that mental imagery for different real-world events incorporates embodied information. Several studies have suggested that the mental representations of overt and covert action actionss are to a larg largee degr degree ee “functionally “functionally equiva equivalent” lent” (Hall, Bernoties, & Schmidt,  1995;  Vogt,  1995). For instance, there is a close relationship between the time needed to mentally rotate a hand or fist in line with the same orientation as a target picture, and the time needed to physically perform the same rotations (Parsons,  b,,  1994are  b ). Moreover, imagined representations of the human body 1987 in motion limited by the same biomechanical factors that constrain real movements (Kourtzi & Shiffrar, 1999). Planning movements, and not just moving covertly or overtly, is the common element underlying embodied action and mental imagery performance (Salway & Logie, 1995). For example, Johnson (2000) reports the findings from a series of studies showing that motor imagery, imagery, or mentally simulated actions, is essential in people’s prospective judgments of the awkwar awk wardne dness ss in pr prehe ehensi nsion on (e. (e.g., g., peo people ple’s ’s jud judgme gments nts for exe execut cuting ing dif differ fer-ent hand movements with a dowel). People appear to think ahead about thei th eirr em embo bodi died ed mo move veme ment ntss no nott by ac acti tiva vati ting ng a co comp mple lete ted d mo moto torr pl plan an,, bu butt  by planning the simulated action.

 

128

 

Embodiment and Cognitive Science

People often overestimate their physical abilities when they mentally simulate their possible future actions (Landau, Libkuman, & Wildman, 2002). In one study, a group of participants mentally simulated lifting a heavy object (a refrigerator) and then estimated how much weight they could lift. A different group of people estimated the weight they could lift without first engaging in the mental simulation exercise. People who mentally simulated lifting a heavy object reported being able to lift more weight than did participants in the no-simulation condition. Follow-up studies showed that people who mentally simulated an event, such as lifting a 100-lb. weight, many times beforehand reported that they could lift more weight than did participants who completed fewer simulations. Furthermore, people who simulated lifting a larger amount of weight estimated that they could lift more weight than did people who imagined lifting a smaller amount of weight. These findings nicely illustrate how even brief mental simulations of embodied activities shape people’s abilities to predict future physical performance. Predicting the outcome of a future action seems to require some ability to in inte tern rnal ally ly rep eprres esen entt a mo mode dell of th thee si situ tuat atio ion n an and d th then en dr draw aw a co conc nclu lusi sion on on the basis of these representational structures. One study investigated whether people’s verbal reports about imagined actions can be explained withou wit houtt app appeal eal to re repr prese esenta ntatio tional nal str struct uctur ures es (va (van n Roo Rooij, ij, Bon Bonger gers, s, & Haselager,  2002). While standing in one spot, participants were handed rod odss of di difffe ferren entt le leng ngth thss th that at th they ey th then en he held ld at an up upwa warrd 45-degr -degree ee angle angle.. Thee pa Th parrti ticcip ipan ants ts’’ ta tassk wa wass si simp mply ly to sa say y wh whet eth her or no nott th they ey co cou uld use th thee rod to touch a distant object. Across the series of trials, the rods presented presented to participants either increased in length and then decreased, decreased decreased in length and then increased, or were of random lengths. Determining whether a rod can reach an object involves assessing information on the rod’s length and one’s own bodily abilities (i.e., one’s posture, ability to lean forward with feet planted on one spot, arm length, and so on). A traditional representational account would argue that particip tic ipan ants ts mu must st ca calc lcul ulat atee vi viaa so some me in inte tern rnal al st stan anda dard rd a co comp mpar aris ison on be betw twee een n the representations of the rod’s length, the postural possibilities, and the estimated distance to the object. Successful imagined action, therefore, is  based on mental calculations that transform these different, separate representations. A dynamical systems account, however, maintains that a person’s behavior is best described at the level of the whole embodied system, as a self-o sel f-org rgani anizin zing g pat patter tern, n, eme emerg rging ing fr from om the int intera eracti ction on amo among ng sub subsys system tems. s. As wi with th al alll dy dyna nami mica call ac acco coun unts ts of hu huma man n pe perf rfor orma manc nce, e, th thee em emph phas asis is he herre is on the temporal dynamics of the participants’ behaviors across the different trials of the experiment. Differential equations are used to show how different potential functions capture the long-term dynamics underlying the participants’ performance. These potential functions describe an

 

Imagery, Memory, and Reasoning

 

129

attr trac acto torr la land ndsc scap ape, e, wh whic ich, h, at di diff ffer eren entt ti time mess fo forr a pa part rtic icip ipan ant, t, refl eflec ects ts re rela la-at tivel tiv ely y st stab able le an and d un unst stab able le st stat ates es of be beha havi vior or.. Van Ro Rooi oijj et al al.. te test sted ed a sp spec ecifi ificc twotw o-at attr trac acto torr sp spac acee mo mode dell th that at ma made de sp spec ecifi ificc pr pred edic ictio tions ns ab abou outt th thee re relat lativ ivee freq fr eque uenc ncie iess of di diff ffer eren entt dy dyna nami mica call pa patte ttern rns, s, on once ce mo more re,, ac acrros osss di diff ffer eren entt se se-quences of trials in the experiment (i.e., presentation of rods from shortest to longest length, longest to shortest length, and rods of varying lengths presented randomly). In fact, the results of van Rooij et al. (2002) sh show owed ed th that at se seve vera rall dy dyna namic mic patterns explained the participants’ performance. Participants tended to give the same categorical responses in the random sequence condition. This assimilative effect is consistent with the dynamical view that the system tends to cling to the state it resides in. Second, there was an inverse relationship between rod length and probability of “yes” responses. This Th is co cont ntra rast stiv ivee ef effe fect ct wa wass en enha hanc nced ed wh when en th thee co coup uple led d se sequ quen ence ce ra ran n fr from om shor sh orte terr to lo long nger er ro rods ds ra rath ther er th than an th thee op oppo posi site te,, ex exac actl tly y wh what at is ex expe pect cted ed be be-caus ca usee th thee mu multi ltist stab able le reg egio ion n is rel elat ativ ivel ely y la larrge he herre. Fin Final ally ly,, th thrree dy dyna namic mic patter pat terns ns of hys hyster teresi esis, s, cri critic tical al bou bounda ndary ry,, and enh enhanc anced ed con contra trast st wer weree all observed, to different extents, in each participant’s behavior throughout the tria tr ials ls in th thee ex expe peri rime ment nt.. Mo Most st ge gene nera rally lly,, th thes esee da data ta ar aree co cons nsis iste tent nt wi with th a dy dy-nami na mica call ac acco coun untt in wh whic ich h th thee pa part rtic icip ipan ants ts’’ ima imagi gine ned d ac actio tions ns ar aris isee fr from om th thee interplay between a control parameter (a parameter that leads the system through various dynamical patterns) and a collective variable governing the ent entir iree sys system tem.. The These se par parame ameter terss ar aree not re repr prese esente nted d int intern ernally ally,, but pr proovide an “imagining landscape” that is an emergent property of the entire embodied system. Van Van Rooij et al. argue that it is difficult to imagine how a tr trad adit itio iona nall rep epre rese sent ntat atio iona nall th theo eory ry co coul uld d ex expl plai ain n th thee dy dyna nami mica call pa patte ttern rnss observ obs erved ed in par partic ticipa ipants nts’’ tas task-b k-beha ehavio viors, rs, giv given en the com comple plexit xity y of hav having ing to integr int egrate ate dif differ ferent ent int intern ernal al mec mechan hanism ismss tha thatt ar aree usu usuall ally y pos postul tulate ated d for eac each h finding (i.e., the problem of integrating hysteresis and enhanced contrast within a single mechanism). Nonetheless, the complex patterns in participants’ performance can be explained within a more general dynamical model of self-organizing behavior. behavior. For the present purposes, a dynamical model also gives proper acknowledgment to the body’s role in cognitive  behavior, such as that used to momentarily imagine differ  behavior, different ent human actions tio ns.. Th This is wo work rk,, mo morre ge gene nera rall lly y, de demo mons nstr trat ates es ho how w a dy dyna namic mical al mo mode dell of  human behavior can scale up to explain higher-orde higher-orderr cognitive behavior. behavior.  Motoric Processes in Mental Imagery It may not be terribly surprisingly that people are able to mentally imagine their bodies in action, and sometimes use their specific embodied experiences when imagining physical events in the real world. But some very interesting studies have more directly explored the links between motor processes and visual mental imagery abilities. For example, the transformation of mental images, as done in the classic Cooper and

 

130

 

Embodiment and Cognitive Science

Shepard (1982)   task, depends on motor processes (Wexler et al.,  1998; Wohlschlager & Wohlschlager, 1998). “Visuomotor anticipation is the en-

gine that drive mental rotation (Wexler et al.,  1998:   79). Similar mech anismss may underlie both visual image transformation anism transformation and the production/control of embodied movements. Wexler et al. (1998) examined the relationship between mental rotation and an d mo moto torr pr proc oces esse sess by as aski king ng pa part rtic icip ipan ants ts to rot otat atee a ha hand nd-h -hel eld d jo joys ystic tick k in a direction either in congruence or in opposition to the direction of simultaneous rotation of a mental image. Prior to the main experimental task, partic part icip ipan ants ts pr prac acti tice ced d th thee jo joys ystic tick k rot otat atio ion n ta task sk.. A vi visu sual al tu tunn nnel el pr prev even ente ted d part pa rtic icip ipan ants ts fr from om se seei eing ng th thei eirr ha hand ndss as th they ey ma mani nipu pula late ted d th thee jo joys ysti tick ck.. Pa Parrticipants ticipa nts practiced practiced rot rotating ating the joyst joystick ick at one of two specific speeds (45 or 90 degrees/second) in both clockwise and counterclockwise directions until they were adept at the task. During the main experiment, participants simultaneously performed  both a mental imagery rotation task and the motor rotation task. The mental rotation task used two-dimensional block drawings. One figure was presented at the top of a display for 5 seconds. Immediately afterward, an arrow was briefly displayed indicating where a second figure would appear pe ar.. Th Thee se seco cond nd fig figur uree th then en ap appe pear ared ed an and d wa wass ei eith ther er a rot otat atio ion n (o (off va vary ryin ing g degrees) of the original figure or a rotation of a mirror reflection (flipped 180 degrees on its vertical axis) of the original figure. The participants had to in indi dica cate te wh whet ethe herr th thee se seco cond nd fig figur uree wa wass id iden enti tica call to th thee fir first st (a (and nd si simp mply ly rotated) or a mirror image of the first figure. When performing the motor task ta sk,, pa part rtici icipa pant ntss we werre in inst stru ruct cted ed to be begi gin n ro rota tatin ting g th thee jo joys ysti tick ck (i (in n th thee sp spec ec-ifie fied d direction and at the proper speed) at the same time as the onset of the initial figure in the mental rotation task. The joysti joystick ck rotation continued continued until the participant made a response in the mental rotation task. The main finding in this study was that “clockwise motor rotation facilitates clockwise mental rotation and hinders counterclockwise mental rotation, and vice-versa for counterclockwise motor rotation” (Wexler et al.,  1998:  86 ). Mental rotation was faster when it was in the same direction as the motor rotation than when the two rotations were in opposite directions. The speed of the motor rotation also influenced the speed of  the mental rotation. People typically perform faster with practice across trials in a mental rotation task. Yet Yet in this study, study, people who completed a firsst se fir sess ssio ion n of tr tria ials ls wi with th a fas astt mo moto torr rota tati tio on spe peed ed fo foll llow oweed by a se seco con nd session with a slow motor rotation speed did not perform in this manner manner.. Mental rotation speed decreased slightly for these participants in the second session compared to the first, indicating a tight link between mental and an d mo moto torr rot otat atio ion n sp spee eeds ds.. In ge gene nera ral, l, Wex exle lerr et al al.’ .’ss (1998) res results ults suppo support rt the idea of a tight, dynamic relation between mental and motor rotation. Anot An othe herr so sour urce ce of ev evid iden ence ce on th thee li link nk be betw twee een n me ment ntal al im imag ager ery y and motoric processes comes from studies showing that spontaneous

 

Imagery, Memory, and Reasoning

 

131

eye-mo vement entss occ occur ur dur during ing vis visual ual imag imagery ery tha thatt clo closel sely y re reflec flectt the con conten tentt eye-movem and an d sp spat atia iall ar arra rang ngem emen entt of th thee or orig igin inal al vi visu sual al sc scen enee (B (Bra rand ndtt & St Star ark, k, 1997; Laeng & Teodorescu, 2002). Eye movements appear to have a significant, functional role in the process of mental imagery and may specifically be important in activating and arranging parts of a complex scene into their proper location. One reason that mental imagery must involve aspects of kinesthetic

experience is the fact that the congenitally blind are quite capable of forming imagistic representations (Zimler & Keenan,  1983). Different experimental studies indicate that congenitally blind participants exhibit typical mental rotation, mental scanning, and size/inspection time effects (Carpenter & Eisenberg,  1978;  Marmor & Zaback,  1976). The strength of  these effects is somewhat diminished and they are slower overall than those obtained with sighted individuals, but the pattern of results is similar for both sighted and congenitally blind people. This array of empirical findings for congenitally blind participants is clearly due to these individuals’ tactile/kinesthetic, or haptic, imagery, given that blind participants are unable to complete purely visual tasks (Arditi, Holtzman, & Kosslyn, 1988). Blind persons’ haptic understanding of objects and spatial relations arises from their active, exploratory physical movements. Most generally, generally,  both blind and sighted individuals’ haptic abilities are constrained by a complex coordination between tactile senses, proprioception, and the involvement volve ment ofrejects motorthe cortex cortex. range of imager imagery y evide evidence nce for the  blind clearly idea. The that large mental images are necessarily amodally visual or amodally spatial (Intos-Peterson & Roskos-Ewoldsen, 1984). These findings suggest that there is no reason to believe a visual repr repreesentation is necessary for mental imagery imagery.. There may be two anatomically distinct distin ct corti cortical cal syste systems ms for deali dealing ng with visual rep repres resentat entations ions (one involved in representing the appearance of objects, the other to represent the location of objects in space) (Farah et al.,  1988). One neurological case study showed that a patient with brain damage from an automobile accident suffered from several deficits in visual recognition, but performed normally on most spatial mental imagery tasks (Farah et al., 1988). Many Man y neu neuro ropsy psycho cholog logica icall stu studie diess sup suppor portt the cla claim im tha thatt bas basic ic pr proce ocesse ssess underlying embodied action are activated even in the absence of physical movement. For instance, brain-imaging studies, especially using positron emissi emi ssion on tom tomogr ograph aphy y (PE (PET), T), dem demons onstra trate te tha thatt the act activa ivatio tion n of sen sensor sorimo imo-tor cortex occurs even in the absence of physical movement when people are engaged in a variety of mental tasks, ranging from judging the meaningfulness of imaginal movements to using different mnemonic strategies (Dec (D ecet ety y et al al., ., 1994). Ot Othe herr PE PET T st stud udie iess fo foun und d th that at pa part rtic icul ular ar ar area eass of co cort rtex ex are activated not only when people imagine themselves making different  body movements, but also when people speak the name of a tool (Martin et al.,  1996). Visual discrimination between mirror-reversed forms, such

 

132

 

Embodiment and Cognitive Science

as a pair of hands, results in the strong activation of frontal motor cortex,  both when people imagine the hand movements and when they physicall ca lly y pe perf rfor orm m th them em (P (Par arso sons ns et al al., ., 1995). Fr Fron onta tall mo moto torr co cort rtex ex is ev eveen ac acti ti-vate va ted d wh when en so some meon onee ob obse serv rves es an anot othe herr pe pers rson on mo move ve hi hiss or he herr ow own n ha hand ndss (Rizzolatti, (Rizzo latti, Fogassi, & Gales Galese, e,  1997;  DiPelligrino et al.,  1992). Lesions in parietal cortex severely interfere with people being able to anticipate the outcome outco me of a motor action, action, as well as the ability to engage in mental imagery tasks (Georgopoulos et al., 1989). Neuro Neu roima imagin ging g stu studie diess hav havee sub subseq sequen uently tly re revea vealed led tha thatt pos poster terioioparietal areas of the brain are activated when people mentally rotate pictures tur es of mul multiar tiarmed med blo blockck-like like obj object ectss (Ko (Kossl sslyn, yn, DiGi DiGiro rolam lamo, o, & Tho Thomps mpson, on,

1998). Other brain areas (e.g., parts of the motor cortex) are also activated when wh en pe peop ople le me ment ntal ally ly ro rota tate te pi pict ctur ures es of ha hand ndss (K (Kos ossl slyn yn et al al., ., 1998). Th Thes esee

finding findi ngss su sugg gges estt th that at pe peop ople le ma may y rot otat atee ob obje ject ctss in me ment ntal al im imag ages es ei eith ther er by imagining the consequences of an external force or by imagining the consequences of manipulating an object with their own hands. Thus, mental rotation involves motor processes only when the forces involved are endogenous, and not when the rotating force is exogenous. Most Mo st sc scho hola lars rs no now w ag agrree th that at th thee mo moto torr pr proc oces esse sess ac acti tiva vate ted d du duri ring ng pe perrcept ce ptio ion n an and d im imag agin inat atio ion n ar aree al alwa ways ys a lim limit ited ed su subs bset et of th thos osee ac acti tiva vate ted d du durring overt movement (Ellis,  1995;  Ramachandran & Hirstein,  1997). More genera gen erally lly,, tho though ugh,, the var variou iouss beh behavi aviora orall and neu neuro roima imager gery y find finding ingss hig highhlight that motoric elements are recruited whenever the perceived perceived or imagined objectmodels is conceptualized action-oriented terms. Several have beenin proposed to account for people’s ability to imagine, often correctly, how their bodies move in space and, more generally,, the effect of motion on the behavior of objects. The most discussed erally mode mo dell is “m “mot otor or im imag ager ery y th theo eory ry”” (J (Jea eann nner erod od,, 1994, 1995). Un Unde derr th this is vi view ew,, imagery arises as the conscious experiences of an ongoing nonconscious premotor plan. As noted above, there is some dispute as to whether complete motor plans necessarily must exist for people to experience mental imagery (see Ito, 1999). The alternative “imagery as planning hypothesis” proposes that mental imagery is invoked when a person mentally transforms somatosensory representations in order to anticipate the result of  upcoming movement in advance of its execution (Rosenbaum, 1991). A “dynamic (depiction) model of imagery” acknowledges the importanc ta ncee of fo forrce dy dyna nami mics cs in so solv lvin ing g ph phys ysic ical al im imag ager ery y ta task skss (S (Sch chwa wart rtz, z, 1999). Research supporting this view includes tasks studying self-motion (Parsons,  1987a,  1994), biological motion and friction (Hubbard,  1995), and momentum (Freyd & Johnson,  1987)  and judging the behavior of water in tilting glasses (Schwartz, 1999). One computational instantiation of this view vie w on how peo people ple re repr prese esent nt con contex text-s t-spec pecific ific dyn dynamic amic inf inform ormati ation on inc incor or-porates rate-based representation of physical properties such as friction, elasticity,, and balance (Schwartz & Black, 1996), each of which we clearly elasticity

 

Imagery, Memory, and Reasoning

 

133

expe rien ence ce wi with th ou ourr bo bodi dies es.. Th This is mo mode dell ca capt ptur ures es,, am amon ong g ot othe herr th thin ings gs,, ho how w experi peop pe ople le me ment ntal ally ly im imag agin inee th thee sp spee eed d an and d di dirrec ecti tion on wi with th wh whic ich h on onee ob obje ject ct reacts to the movement of another. Not surprisingly, people conceptualize how ho w ob obje ject ctss rea eact ct wh when en mo movi ving ng ag agai ains nstt on onee an anot othe herr in ve very ry hu huma man n te term rms. s. In his classic studies, Michotte (1963) observed that people described the coll co llis isio ion n of obj bjec ects ts in te terrms of how hu huma man ns act wh when en th theey com omee in into to phy hyssical contacts (e.g., “A gave B a kick in the pants and sent him flying”). As will be argued below, many mental imagery phenomena, including how people imagine the movement of nonhuman objects, are understood directly via people’s recurring embodied experiences. One proposal that explicitly aims to explain the sensory-motor aspects of mental imagery is “perceptual activity theory” (Thomas, 1999). Under this view, mental imagery is not assumed to be the end product of perception ceptio n (i.e., no inner picture picture or depiction of some stimuli is spec specifically ifically created). However, mental imagery is intimately tied to the ongoing ac-

tivity of perceptual/motor exploration of the environment. People have the phenomenological experience of having a mental image whenever a schema that is not directly relevant to the exploration of the present environment momentarily takes control of the body’s exploratory apparatus. Perceptual activity theory explains various traditional mental imagery findingss (Thom finding (Thomas, as, 1999). Me Ment ntal al sc scan anni ning ng pa para rall llel elss rea eall-wo worl rld d vi visu sual al sc scan an-ning in that it takes longer to scan through a larger visual angle than a smal sm alle lerr on one. e. Wh When en ha hand nd an and d ey eyee mo move veme ment ntss ar aree su supp pprres esse sed, d, th thee sc sche hema ma’s ’s failed attempt attempt to initiate and contr control ol these movements movements will still result in aoccur timesuch course similar to real-world Size/inspection time to effects that smaller details in ascanning. visual scene take longer than pick out in larger ones, in part because people must narrow their attentional focus or move closer to the target object to perceive smaller details. When people attempt to find smaller details in a mental image, the additional time to covertly move closer to the object or to narrow one’s visual focus on the small details will also increase processing effort and time. Mental rotation effects, as suggested above, are tightly linked to motion processes so th that at our me ment ntal al rot otat atio ion n of an ob obje ject ct is si simi mila larr to phy hyssic ical ally ly tu turrni ning ng it in our hands (Kosslyn, 1994). Fina Fi nall lly y, so some me sc scho holar larss co cont nten end d th that at co cons nsci ciou ouss me ment ntal al im imag ager ery y en enha hanc nces es peop pe ople le’s ’s pe perf rfor orma manc ncee in a va vari riet ety y of pe perc rcep eptu tual al-m -mot otor or an and d co cogn gnit itiv ivee ta task skss (Marks, 1999). In this view, mental imagery is again not the end product of a sp spec ecifi ificc ki kind nd of co cogn gnit itiv ivee pr proc oces essi sing ng,, bu butt pr prov ovid ides es th thee bu buil ildi ding ng bl bloc ocks ks for thinking, problem solving, memory memory,, and imagination, i magination, especially in regard to how people plan action. For instance, Newton ( 1996) claimed that intentional mental states are images of goal-directed action episodes that are created in response to external stimuli. Much of our conscious experience of sensorimotor imagery reflects these underlying mental states. Conscious imagery is consequently essential for the planning of human

 

134

 

Embodiment and Cognitive Science

action (Marks rks,, 1999), an and d em embo bodi died ed mo move veme ment nt pr prov ovid ides es th thee roo oots ts fo forr co connaction (Ma scious experience (Sheets-Johnstone, 1998, see Chapter 9). For this reason, some scholars go so far as to suggest that people’s conscious preparation of action and their imagery of this action are very difficult to distinguish (Rizzolati, 1994). In ge gene nera ral, l, an in incr crea easi sing ng nu numb mber er of re rese sear arch cher erss no now w ra rais isee the possibility that motor imagery may not be separable from visual and auditory imagery (Klatzky, 1994). Do Mental Images Arise from the Brain?  The previous The  previous section mentioned section  mentioned neuroimagining studies show that the motor cortex is activated during many mental rotation tasks. Correlating distinct brain regions with performance on different cognitive tasks may poin po intt to th thee ne neur ural al me mech chan anis isms ms th that at un unde derl rlie ie,, fo forr in inst stan ance ce,, me ment ntal al im imag ager ery y. In fact, Kosslyn (1994) claims that neuroscientific data resolve many of the trad tr aditi ition onal al de deba bate tess on th thee na natu ture re an and d fu func ncti tion on of me ment ntal al im imag ager ery y. Ko Koss ssly lyn, n, Thompson, Wraga, and Alpert (2001) recently published the results of anothe ot herr ne neur uroi oima magi ging ng st stud udy y te test stin ing g th thee id idea ea th that at di dist stin inct ct ne neur ural al me mech chan anis isms ms underlie different ways of imagining object rotation. Participants in this study either first viewed an electric motor rotate an angular object, or ac-

tually ro tually rotat tated ed the obj object ect man manual ually ly.. Aft Afterw erwar ard, d, neu neuro roima images ges wer weree re recor corded ded as participants performed a rotation task in i n which they compared pairs of  objects in different orientations. Participants were specifically instructed to imagine the object rotations as they had just seen the objects rotated (done by the electric motor or by their own hands). The results showed that the motor cortex was only activated in the condition in which participantss im pant imag agin ined ed th thee rot otat atio ion n as a co cons nseq eque uenc ncee of th thei eirr ow own n ma manu nual al ac acti tivi vity ty.. Kosslyn et al. (2001)  argued that these findings support the existence of  qualitatively different ways of imagining object rotation, where each one can be adopted voluntarily depending on the task. Howeve How everr, I re resis sistt dra drawin wing g the these se con conclu clusio sions, ns, esp especi eciall ally y in sim simply ply re reduc duc-ing imagery to brain states and ignoring the rest of the body and bodily action act ion.. Fir First, st, re resea searc rcher herss pos posit it the exi existe stence nce of dif differ ferent ent neu neural ral mec mechan hanism ismss given their interest in particular hypotheses, yet ignore other possible influences on people’s successful performance on some experimental task. For example, Kosslyn et al. (2001)  found activation of motor cortex only when participants first rotated an object, but not when they observed an electric motor rotate that same object. But they also found activations in many ma ny ot othe herr br brai ain n ar area eass fo forr th thes esee di diff ffer eren entt ex expe peri rime ment ntal al co cond nditi ition ons, s, in incl clud ud-ing mos mostt con conser servat vative ively ly ar areas eas 4, 6, 7,and 9 (a (ass we well ll as si sign gnifi ifica cant nt ac acti tiva vati tion on,, acco ac cord rdin ing g to mo more re lib liber eral al st stat atis istic tical al te test sts, s, in ar area eass 18, 19, 37, 45/47, and 47). Why is activation in these areas not considered as evidence for multiple mechanisms operating during mental rotation tasks? Researchers too often focus their attention on data that seemingly support their hypotheses (e.g., the idea that there are distinct neural mechanisms responsible for

 

Imagery, Memory, and Reasoning

 

135

mental rotation) and disregard alternative ways of explaining their data (e.g., (e.g ., mi migh ghtt th ther eree be as ma many ny “m “mec echa hani nism sms” s” in invo volv lved ed in me ment ntal al rot otat atio ion n as ther th eree ar aree ar area eass of br brai ain n ac acti tivi vity ty?) ?).. To th thei eirr cr cred edit it,, Ko Koss ssly lyn n et al al.. (2001) comment me nt “W “Wee re reco cogn gniz izee th that at so some me in inve vest stig igat ator orss ma may y ha have ve sp spec ecifi ificc hy hypo poth thes eses es rel elat ated ed to reg egio ions ns ot othe herr th than an th thee on ones es fo forr wh whic ich h we ex expe pect cted ed di difffe ferren ence cess in activation” (p. 2522). Yet to claim that motoric mechanisms function only when wh en pe peop ople le fir first st ma manu nual ally ly ro rota tate te ob obje ject ctss gr grea eatly tly si simp mpli lifie fiess th thee co comp mple lexi xity ty of how brains, bodies, and the world interact during mental imagery imagery.. More generally, there are several general and specific reasons to question whether mental imagery arises simply as the output of neural processes. First, the correlation between activation of certain brain areas and  behavior on differen differentt experimental tasks should not be interpreted as evidence for distinct neural, or even cognitive, “mechanisms” of mind. Psychologists too often quickly rush to postulate the existence of distinct “mechanisms” based on different patterns of behavioral and brainrecording recor ding data. Yet Yet different patterns of behavioral and brain data can almost as easily be understood as arising from distributed mental processes (Rumelhart & McClelland, 1986), holonomic processes (Pribram, 1991), or nonlinear dynamical interactions (Kelso, 1995; Port & van Gelder, 1995). The problem with traditional imagery research in cognitive and neuropsychology is that it primarily conceives of mental images as internal mental representations that, perhaps, are rooted in distinct areas of the  brain. But activating activating neural structures structures does not generate the experience of 

mental ment al im imag ager ery y, ev even en fo forr th thee re relat lativ ivel ely y im impo pove veri rish shed ed me ment ntal al im imag ages es st stud ud-ied ie d by mo most st co cogn gniti itive ve ps psyc ycho holo logi gist sts. s. Me Ment ntal al ima image gess ha have ve a pa part rtic icul ular ar “f “fee eel” l” to them that is only meaningful in the context of kinesthetic experiences (Damasio, 1999). The search for the “holy grail” of mind in brain mechanismsconcealsasubtleformofdualismwherethebodybecomestheinstrument for executing the brain’s instructions. But as many scholars now argue, the particular brains we have, with their distinct neural organization, are shaped by the bodies we have and the actions we continue to perform in th thee rea eall wo worl rld d (F (Frree eema man, n, 2000; Ke Kels lso, o, 1995; La Lako koff ff & Jo John hnso son, n, 1999). Underr th de this is vi view ew,, br brai ains ns ar aree no nott th thee so sole le un unde derl rlyi ying ng ca caus usal al ag agen ents ts fo forr im imag agis isti ticc experience. Mental imagery arises from neural and somatic activity that is understood by ongoing actions of the whole person. The idea that mental imagery is a whole-person activity paints a different picture of the “mental simulation” perspective of mental imagery adopted adopt ed by many cogni cognitive tive and neur neuropsyc opsycholog hologists. ists. Follow Following ing Berth Berthoz oz (2000), I specifically argue that “simulation” should be replaced by the idea of “simulator.” Consider, for a moment, a meteorologist creating a comp co mput uter er si simu mula lati tion on of th thee pa path th of a hu hurr rric ican anee al alon ong g th thee ea east ster ern n se seab aboa oarrd of the United States. This simulation nicely maps various topological relationshipsofthisweathersystemandmayevenbeusedtopredictthebehavior of real hurricanes. To To some extent, computer simulations of cognition,

 

136

 

Embodiment and Cognitive Science

includ ing men mental tal ima imager gery y pr proce ocesse sses, s, ar aree sim simila ilarr to the met meteor eorolo ologis gist’s t’s sim sim-including ulations of the behavior of hurricanes – they capture relevant information about the formal characteristics of a set of operations that are carried out on particular repr representations. esentations. A bet ette terr not otio ion n of sim imul ulat atio ion n is th thaat of an ac actu tual al “s “sim imul ulat ator or”” (Berthoz,  2000). Unlike a computer simulation of behavior (e.g., neural networks or any symbolic computing device), a simulator provides something close to what it i t actually feels like in a full-bodied manner to, say say,, fly an ai airrcr craf aft. t. Me Ment ntal al im imag ager ery y, in ou ourr vi view ew,, is a ki kind nd of si simu mula lato torr of ac acti tion on th that at is ba base sed d on rea eall-li life fe ac acti tion onss an and d po pote tent ntia iall ac acti tion onss th that at a pe pers rson on ma may y en enga gage ge in. As a simulator, mental imagery provides a kinesthetic feel that is not simply the output of some abstract computational machine, but provides some so meth thin ing g of th thee fu fullll-bo bodi died ed ex expe peri rien ence cess th that at ha have ve te text xtur ures es an and d a fe felt lt se sens nsee of three-dimensional depth. The traditional focus in cognitive psychology on visu vi sual al ima image gery , to the th negl ne ect t of othe ot her ratse sens nsor ory y an and dager kine ki nest tic c do doma main s, some so meti time mes s fo fool ols sry, peop pe ople leein into toglec thin th inki king ng that th ment me ntal al imag im ery ysthe isheti comp co mple lete tely lyins, in the “mind’s eye” without much engagement from the rest of the body. But as Damasio (1994, 1999, 2003) has long argued, we have an ongoing awareness of our somatosensory systems. Noting that the brain continually receives feedback signals from the body’s autonomic processes, Damasio suggests that this feedback provides us with a constant background awareness of our own bodies’ somatosensory systems. This low level of awareness is akin to mood that colors our ordinary consciousness: “The background body sense is continuous, although one may hardly notice it, since it represents not a specific part of anything in the body but an overall state of most everything in it” (1994:  152 ). Although Damasio contends in his “somatic marker hypothesis” that people mark certain

somatosensory sensations as positive or negative, depending on the context, these sensations are more complex and often attributed as properties of the object that may be the focus of our imagination. Our “markings” of  these somatosensory sensations allow us to refer to something external to the body (e.g., as when comparing two objects in differe different nt orientations in a mental rotation task) and to refer to something that is embodied within us (as if the object has been introjected into our bodies). Mental imagery experiences often retain both these objective and subjective components. Grush (2004) argued that simulation alone though motor planning, or mere activation of motor cortex, is insufficient to explain imagery. He advo ad voca cate tess an “e “emu mula lati tion on th theo eory ry”” in wh whic ich h em emul ulat atio ion n of th thee mu musc scul ulos oske kele le-tal system is employed and imagery that is produced with the efferent motor center drives this emulation. By analogy, motor imagery is like a pilot sitting in a flight simulator, and the pilot’s efferent commands (hand and foot movements) are translated into faux sensory information (instrument readings) by the flight simulator, which is essentially an emulation of an aircraft. This emulation theory of imagery, therefore, gives much

 

Imagery, Memory, and Reasoning

 

137

grea terr em emph phas asis is to pr prop opri rioc ocep epti tion on an and d kin kines esth thet etic ic ac acti tion on in th thee co cont ntex extt of  greate the entire body than does the simulation theory. The emulation represents obje ob ject ctss an and d th thee en envi virron onme ment nt as th thin ings gs en enga gage ged d wi with th in ce cert rtai ain n wa ways ys,, as op op-posed to how they are considered apart from their role in the organism’s environmental engagements. How Ho w is it po poss ssib ible le to de desc scri ribe be th this is in inte tera ract ctio ion n of br brai ains ns,, bo bodi dies es,, an and d rea eallworld wor ld obj object ects/e s/even vents ts dur during ing men mental tal ima imagis gistic tic exp experi erienc ences es as sug sugges gested ted by emulation theory? My view of mental imagery is that it can best be understood within an “enactive” approach to cognitive science. Once more, the most mo st imp impor orta tant nt fe feat atur uree of dy dyna nami mica call sy syst stem emss is th that at me ment ntal al ima image gery ry ar aris ises es as a generic feature of “emergence” in complex, self-organizing, systems. The nervous system, the body, and the environment are highly structured dynamical systems, coupled to each other at multiple levels. Emergence through self-organization has two directions (Thompson & Varela, 2001). First, there is local-to-global determination, or upward causation, as a result which processes unique imagistic experiences) emerge thatt of tha have hav e the their irnovel own dis distin tincti ctive ve (e.g., featur fea tures, es, life lifeline lines, s, and int intera eracti ction on wit with h oth other er aspects of thought and language. Second, there is global-to-local determinate, or downward causation, whereby global characteristics of a system gove go vern rn lo loca call ne neur ural al in inte tera ract ctio ions ns.. Co Conc ncei eivi ving ng of me ment ntal al im imag ager ery y as an em emer er-gent phenomenon suggests that it may have causal effects on substrate large-scale neural assemblies. This two-way causal interaction has not yet  been demonstrated empirically for mental imagery per se, but various studies reveal similar recipro reciprocal cal relationships between differ different ent conscious acts and neural events (Freeman,  2001;   Thompson & Varela,  2001;   see Chapter 8). The relationship between neural dynamics and conscious mental imagery experience can be described in terms of the participation of neural processes in the “cycles of operation” that constitute a person’s life. Three kinds of cycles are most relevant (Thompson & Varela,  2001): cycles of  organismic regulation of the entire body, body, cycles of sensorimotor coupling

 between organism organism and environment, and cycles of intersubjective interac tion, involving the recognition of the intentional meaning of actions and linguistic communication. Freeman (2000) nicely describes how these differren fe entt cy cycl cles es co coop oper erat atee to pr prod oduc ucee “t “the he bi biol olog ogy y of me mean anin ing. g.”” He no note ted, d, “A mean me anin ingf gful ul st stat atee is an ac acti tivi vity ty pa patt tter ern n of th thee ne nerv rvou ouss sy syst stem em an and d bo body dy th that at has a particular focus in the state space of the organism, not in the physical space of the brain. As meaning changes, the focus changes, forming a trajectory that jumps, bobs, and weaves like the course of a firefly on a summer night. The elements of each dynamic state consist of the pulses and waves in the brain, the contraction of the muscles, the joint angles of the skeletal system, and the secretions of cells in the autonomic and neuroendocrine neuroendocr ine system. Meaning emerges from the whole of the synaptic connections among the neurons of the neuropil, the sensitivities of their

 

138

 

Embodiment and Cognitive Science

trigger zones, determined by the neuromodulators, and to less extents the grow gr owth th,, fo form rm,, an and d ad adap apta tati tion onss of th thee res estt of th thee bo body dy.. Th Thee sk skil ills ls of at athl hlet etes es,, dancers, and musicians live not only in their synapses but also in their limbs, fingers, and torsos” (Freeman, 2000: 1  115 15). My emphasis on a dynamical account of mental imagery is primarily motivated by a desire not simply to reduce mental imagery to states of  neur ne ural al ac acti tiva vati tion on.. Th This is do does es no nott de deny ny,, of co cour urse se,, th thee cr crit itic ical al rol olee th that at br brai ains ns play in mental imagery imagery,, as well in other cognitive functions. But cognitive scie sc ient ntis ists ts ne need ed to rec ecog ogni nize ze ho how w im imag ager ery y is ac acco comp mpan anie ied d by se sens nsor orimo imoto torr sensations, or whole “body-loops” (Damasio, 1994), which give imagistic experi exp erienc encee its ric rich h phe phenom nomena enall qua quality lity.. Dyn Dynami amical cal acc accoun ounts ts ar aree sug sugges gestive tive of how people engaging in various actions may momentarily experience these meaningful, and indeed adaptive, imagistic states that are clearly embodied and not just locked within the brain.  Image Schemas and Mental Imagery

How the recent empiricalactivity work and theories on the similarities betweendo motoric and imagistic relate to broader views on human thought and language? In this section, I advance the idea i dea that recent findings and theories on the kinesthetic nature of mental imagery fit in nicely with wi th a ra rath ther er di diff ffer eren entt se sett of de deve velo lopm pmen ents ts on th thee em embo bodi died ed fo foun unda dati tion on fo forr though tho ughtt and lan langua guage ge (se (seee Cha Chapte pterr 4). I de desc scri ribe be th thes esee id idea eass an and d sh show ow ho how w they th ey sp spec ecifi ifica call lly y ap appl ply y to tr trad adit itio iona nall me ment ntal al im imag ager ery y fin findi ding ngss an and d po poin intt to a more comprehensive view of the embodied grounding of higher-order cognition. The possible relevance of cognitive psychology research on mental imagerytoimageschemaswasfirstnotedbyJohnson( 1987) andLakoff(1987). They both described several studies on mental imagery that supported the idea that embodied image schemas and their transformations play an important role in cognitive functioning. Johnson (1987) suggested that the data from several studies on selective interference in mental imagery (Brooks, 1968; Segal & Fusella, 1970) provide evidence for image schemas. Thus Th us,, pe peop ople le se seem emed ed ab able le to ac acce cess ss ce cert rtai ain n mo mode dess of co cogn gnit itio ion, n, ei eith ther er re reca call ll of verbal information or visual imagery, through multiple channels, such as ki kine nest sthe hetic tic or ve verb rbal al re repo port rt.. Mo Morreo eove verr, bo both th La Lako koff ff an and d Jo John hnso son n cl claim aimed ed that classic studies on mental rotation (e.g., Cooper & Shepard, 1982) also

provide evidence in support of image schemas and their transformations. As Johnson (1987:  25 ) concluded from his discussion of the mental rotation data, “we can perform mental operations on image schemata that are analogs of spatial operations” (i.e., we rotate things quickly in our imagination because of our bodily experience of rotating things with our eyes, hands, and other body-parts). In other words, data that th at im imag age e sc sche hema mas s ha have ve a ki kine nest sthe heti ticc ch char arac acte terrthe , as empirical they th ey ar aree no not t ti tied edsuggest to an any y single perceptual modality.

 

Imagery, Memory, and Reasoning

 

139

Does our ability to mentally rotate images truly reflect the operation of  imag im agee sc sche hema mas? s? To an answ swer er th this is qu ques esti tion on,, on onee mu must st be ve very ry cl clea earr ab abou outt th thee diff di ffer eren ence cess be betw twee een n me ment ntal al ima image gery ry as ty typi pica call lly y st stud udie ied d by co cogn gnit itiv ivee ps psyychologists and the idea of image schemas. Image schemas are presumably morre ab mo abst stra ract ct th than an or ordi dina nary ry im imag ages es an and d co cons nsis istt of dy dyna namic mic sp spat atia iall pa patt tter erns ns that underlie the spatial relations and movement found in actual concrete images. Mental images are traditionally viewed as temporary representations, whereas image schemas are permanent properties of embodied experience. Finally, image schemas are emergent properties of subjective felt bodily experience, whereas mental images are the result of more effort fo rtfu full co cogn gnit itiv ivee pr proc oces esse ses. s. Fo Forr ex exam ampl ple, e, re rese sear arch ch sh show owss th that at so some me me ment ntal al images can be generated by assembling the parts of the image i mage one part at a time (see Finke, 1989). Despit Des pitee the these se dif differ ferenc ences, es, the there re ar aree int inter erest esting ing sim simila ilarit rities ies bet betwee ween n men men-tal images and image schemas that make the study of mental imagery especially relevant to our quest for the embodied foundations of mental imagery. One body of research that quite specifically points to the role of image schemas and their transformations in The mental functioning comes from studies on representational momentum. term representational momentum (RM) was coined by Freyd and Finke (1984)  to refer to an internalized representation of physical momentum. A variety of experiments have studied different different aspects of RM. The typica typicall parad paradigm igm used to investigate RM consists of the presentation of a sequence of three static images, referred to as the inducing stimuli, of an object (usually a simple geographic shape or a dot) that appears to be moving linearly or rotating in one direction. A final target position of the image is then presented and participants are asked to determine whether this target image’s position is the same as the third static image of the object. People’s participation in a RM task involves their ability to follow in their imagination the path of a moving object and then focus on the point where it will come to rest (an example of the Path-focus to end-point focus image schema transformation). Thee cl Th clas assi sicc fin findi ding ng fr from om RM st stud udie iess is th that at pa part rtic icip ipan ants ts’’ me memo mory ry fo forr th thee final position of an object undergoing implied motion is shifted toward the direction direction of the motion. For exam example, ple, if partic participant ipantss watch an image of an object that appears to be rotating, and then have to remember the final position of the object, they will typically report that the object’s final position was further along in the rotation than it actually was. The same sort of effect holds for linearly moving objects. The effect was first discovered for rotating objects (Freyd & Finke,  1984) and was later extended to

linearl linea rly y mo movi ving ng ob obje ject ctss (F (Fin inke ke & Fr Frey eyd, d, 1985; Hu Hubb bbar ard d & Bh Bhar arac acha ha,, 1988), centripetal force and curvilinear impetus (Hubbard,  1996a), and spiral  1994 paths (Freyd &be Jones, ).ong Ifgparticipants an ofrean object that th at ap appe pear arss to movi mo ving ng al alon a li lin nea earr pa path th,watch , and th then enimage hav ha ve to memb me mbeer

 

140

 

Embodiment and Cognitive Science

this object’s final position, they will report that the final position was further along the path than it actually was. The RM effect is not due to apparent motion because increasing the amount of time between the presentations of the static images up to2 seconds still results in RM (Finke & Freyd,  1985). RM presumably “reflects the internalizatio internalization n in the visual system of the principles principles of physical momentum” (Kelly & Freyd, 1987: 369). Indeed, many characteristics of realworld physical momentum have been found in RM. For instance, the apparent velocity of the inducing stimuli affects RM (Freyd & Finke,  1985; Finke, Freyd & Shyi,  1986). Participants’ memory for the final position of  a quickly moving object is displaced further along in its path than if the obje ob ject ct is mo movi ving ng sl slow owly ly.. Ap Appa parren entt ac acce cele lera rati tion on of th thee in indu duci cing ng st stimu imuli li als also o affects RM in that objects that appear to be accelerating will produce a larger memory displacement (Finke, Freyd, & Shyi, 1986). Also, displacements that go beyond what one would expect in real-world momentum do no nott prod oduc ucee RM RM,, (Fi Fin nke & Fr Frey eyd d, 1985). If the target im imaage of the obje jecct is in a position that corresponds to the “next” position in the sequence of  inducing images, or is even further along in the path or rotation than the “next” position, the RM effect goes away. Furthermore, memory displacement is greater for horizontal than for vertical motion (Hubbard & Bharacha, 1988). This may be a result of the predominance of horizontal motion in our environment. Gravity also affect fe ctss RM (H (Hub ubba bard rd & Bh Bhar arac acha ha,, 1988). Ob Obje ject ctss mo movi ving ng do down wnwa warrd ar aree di dissplac pl aced ed mo more re alo along ng th thei eirr di dire rect ctio ion n of mo moti tion on th than an ob obje ject ctss mo movi ving ng up upwa ward rd.. If  an object is moving horizontally and then disappears, participants consistent te ntly ly ma mark rk it itss va vani nish shin ing g po poin intt lo lowe werr th than an it ac actu tual ally ly wa was. s. Th Thee sa same me res esul ultt occurs with ascending oblique motion. Interestingly, descending oblique motion usually produces displacement above the actual vanishing point. These results suggest internalized environmental constraints on momentum. What goes up must come down, what comes down comes down faster than what goes up, things moving linearly usually drop toward the grou gr ound nd,, an and d th that at wh whic ich h dr drop opss at an an angl glee us usua uall lly y en ends ds up mo movi ving ng ho hori rizo zonntally along the ground. It appears that RM is something more complicated than a simple representation of what an object’s motion is like given that it has momentum. Fina Fi nall lly y, an and d im impo port rtan antl tly y, RM ef effe fect ctss ha have ve be been en fo foun und d no nott on only ly fo forr vi visu sual al stimuli, but for auditory stimuli as well (Kelly (Kell y & Freyd, 1987; Freyd, Kelly Kelly,, & DeKay DeKay,, 1990). St Stud udie iess wit with h mu musi sica call pi pitc tch h ha have ve de demo mons nstr trat ated ed th that at a se seri ries es of inducing tones either rising or falling in pitch, followed by a target tone either higher or lower in pitch than the third inducing tone, produces the same sa me RM ef effe fect ctss as wi with th th thee st stud udie iess us usin ing g vi visu sual al st stim imul uli. i. Th This is au audi dito tory ry RM appears not to be due simply to a correlation with visual RM, but rather seems only abstractly related to visual RM (Kelly & Freyd, 1987).

 

Imagery, Memory, and Reasoning

 

141

Many aspects of the data on visual and auditory RM can be explained in terms of image schemas and their transformations. First, the SOURCEPATH-GOAL schema must underlie critical aspects of RM as a person observes an object move from a starting position along some path toward an imagined goal. The SOURCE-PATH-GOAL schema must be one of the most basic image schemas that arise from our bodily experience and percept ce ptua uall in inte tera ract ctio ions ns wi with th th thee wo worl rld d (i (i.e .e., ., no note te al alll th thee ac acti tion onss wh wher eree an any y pa part rt of a bod ody y mo move vess to reac ach h som omee phy hyssic icaal ob obje ject ct or lo loca cati tion on)) (s (see ee Ch Chaapt pter er 4). Besi Be side dess th thee sc sche hema ma of SO SOUR URCE CE-P -PA ATH TH-G -GOA OAL, L, th ther eree ma may y al also so be a sp spec ecifi ificc schema for MOMENTUM. When we encounter the inducing stimuli in a RM task, either visual or auditory, a stored representation for momentum tu m is no nott ac acti tiva vate ted. d. In Inst stea ead, d, we us usee th thee im imag agee sc sche hema ma fo forr MO MOME MENT NTUM UM,, derived jointly by our minds, bodies and our environment, to expect the next stimuli to be further along in the path, rotation, or musical scale. Such an expectation would not occur using only the PA PATH image schema or FOLLONG-A-TRAJECTORY transformation. These may provide the direction that a moving or rotating object is about to traverse, but they cannot account for an expectation about the distance that the object will travel given that it i t has momentum. Yet a MOMENTUM schema accounts for specific, quantitative aspects of visual RM. Thus, our experience tells us that the faster something is moving, the more momentum it will have and thus the more distance it will travel when a stopping force is applied to it. Moreover, the notion of momentum as image schema also explains the cross modal aspects of RM. We abstract away from our experiences of seeing momentum, hearing momentum and feeling momentum those aspects that are shared or which are similar to one another. Thus, we get thee sa th same me kin inds ds of ef effe fect ctss in au audi dito tory ry RM as in vi visu suaal RM eve ven n th thou ough gh th theey are not always correlated in the environment (Kelly & Freyd, 1987). The research on visual and auditory RM may also be used to speculate on how momentum can be created by image schema transformation, such as LANDMARK, PATH, PATH, BLOCKAGE, REMOVAL REMOVAL OF BLOCKAGE, and GOAL (see Johnson,  1987  for more discussion of the embodied evidence for these schemas). Image schema transformations like these would fun fu nct ctio ion n in RM in th thee fo foll llow owin ing g wa way y. Fi Firrst st,, we in inv vok okee th thee la lan ndm dmar ark k im imag agee sche sc hema ma wh when en we im imme medi diat atel ely y at atte tend nd to an ob obje ject ct.. As th this is ob obje ject ct mo move ves, s, we transform the landmark image schema into the path image schema in that our attention is now additionally focused upon the path of the landmark. This is known as the LANDMARK-PATH image schema transformation. We then invoke the BLOCKAGE image schema when the moving object disappears. This image schema is transformed into the REMOVAL OF BLOCKAGE image schema when the target stimulus appears. This transformation is known as the BLOCKAGE-REMOVAL image schema transformation. Finally, to determine the endpoint of the moving object given

 

142

 

Embodiment and Cognitive Science

that it was a landmark moving along a path that encountered blockage, which was subsequently removed, we transform the PATH image schema into a MOMENTUM image schema, and then that t hat into an endpoint focus or goal image schema. This gives us information about the likely position of the object given that it had not encountered any blockage. People use the position provided by image schema transformations to compare to the target stimuli in a RM task. If there is a match between our exp expect ected ed pos positio ition n giv given en by dif differ ferent entima image ge schema sch ema transf tra nsform ormati ations ons and the target stimuli, we respond affirmatively affirmatively. . As the RM literature has shown, howe ho weve verr, we ar aree fr freq eque uent ntly ly mis mista take ken n in sa sayi ying ng th that at ta tarrge gett po posi sitio tions ns th that at ar aree furt fu rthe herr al alon ong g th thee pa path th co corr rrec ectly tly in indi dica cate te th thee po posi sitio tion n th thee ob obje ject ct wo woul uld d ha have ve.. Thiss mis Thi mistak takee is pr produ oduced ced by the PATHTH-END END-PO -POINT INT FOC FOCUS US ima image ge sch schema ema tran tr ansf sfor orma mati tion on.. Th This is tr tran ansf sfor orma mati tion on giv gives es us in info form rmat atio ion n ab abou outt wh wher eree th thee object should be given that it was moving at a certain speed, in a certain direction, and encountered blockage that was then removed. If we were inst in stea ead d re relyi lying ng on only ly up upon on th thee in info form rmat atio ion n in me memo mory ry on th thee ac actu tual al po posi sitio tion n of the most recent image of the object, we would not make these errors. In summary, summary, although there are significant differe differences nces between mental imagery and image schemas, there is good evidence that spatial, kinesthetic, the tic, and vis visual ual re repr prese esenta ntatio tions ns exi exist st for men mental tal ima imager gery y. Thi Thiss con conclu clusio sion n is qu quit itee co cons nsis iste tent nt wi with th th thee id idea ea th that at di diff ffer eren entt mo mode dess of pe perc rcep eptu tual al/b /bod odily ily expe ex peri rien ence ce gi give ve ri rise se to co cogn gniti itive ve sc sche heme mess th that at ha have ve an anal alog og-l -lik ikee pr prop oper erti ties es.. To the extent, then, that people’s mental images reflect the operation of  various modalities and kinesthetic properties of the body, body, the experimental findings on mental imagery support the idea that image schemas play a significant role in certain aspects of perception and cognition. Memory

Memory has traditionally been studied as if it constituted a functionally sepa se para rate te st stor orag agee de devic vicee of mi mind nd co cont ntain ainin ing g in info form rmat ation ion th that at is mo most stly ly re reppresented in abstract, symbolic terms. There may be different memory systems, such as short-term and long-term memory, and varying contents, such su ch as se sema mant ntic ic an and d ep epis isod odic ic in info form rmat atio ion. n. Bu Butt wi with th th thee ex exce cept ptio ion n of so some me information that is “procedural” (e.g., knowledge about how to tie one’s shoe sh oes) s),, th thee tr trad aditi ition onal al vi view ew is th that at me memo mory ry is co cons nsti titu tute ted d by di dise semb mbod odie ied, d, abstract symbols.  Memory as Embodied Action Much recent work suggests that memory and the process of remem bering are partly based on embodied activity activity.. Imagine yourself in your kitchen gathering the ingredients needed to bake a cake (Cole, Hood, & McDermott, 1997). You need not remember exactly where each ingredient presumably is kept in the cabinets around you because you can simply go

 

Imagery, Memory, and Reasoning

 

143

to the cabinets and move through the contents inside until each necessary ingredient is located. The external world (i.e., the cabinet) may take the plac pl acee of a fu full lly y de deta tail iled ed in inte tern rnal ally ly rep eprres esen ente ted d me memo mory ry co code de fo forr th thee in ingr greedients and where they are located. The cabinet may even allow you not to retrieve in memory a detailed list of all the ingredients needed to make a cake ca ke.. We kn know ow th thaat wh what atev eveer is nee eede ded d wi will ll li like kely ly be fo foun und d in th thee ca cab bin inet et.. It is not just the existence of the cabinet that allows these environmental structures to partly take over significant information processing. Rather, our movement through the cabinet, as to webake pushthe some enables us to figure out what is needed cake.ingredients aside, Of co cour urse se,, th thee ki kitc tche hen n ca cabi bine nett ma may y on only ly se serv rvee as an ex exte tern rnal al ai aid d to me memmory or y, bu butt no nott be pa part rt of th thee co cogn gnit itiv ivee pr proc oces esss of rem emem embe beri ring ng.. Yet dr draw awin ing ga rigid distinction between what is external and what is internal to remem bering is less compelling, compelling, given that memory is composed of both internal representations repr esentations and manipulation of environmental structures. Cons Co nsid ider er th thee ca case se of th thee ex expe pert rt ba bart rten ende derr. Fa Face ced d wi with th mu mult ltip iple le dr drin ink k or or-ders de rs in a no nois isy y an and d cr crow owde ded d en envi virron onme ment nt,, th thee ex expe pert rt mi mixe xess an and d di disp spen ense sess drinks with amazing skill and accuracy. What is the basis of this expert perf pe rfor orma manc nce? e? Do Does es it al alll st stem em fr from om fin finel ely y tu tune ned d me memo mory ry an and d mo moto torr sk skil ills ls?? Studies comparing novice and expert bartenders show that expert skill involves a delicate interplay between internal and environmental factors (Beach, 1988). The experts select and array distinctively shaped glasses at the time of ordering. They then use these persistent cues to help recall and sequen seq uence ce the spe specifi cificc or order ders. s. Exp Expert ert per perfor forman mance ce thu thuss plu plumme mmets ts in tes tests ts involving uniform glassware, whereas novice performance is unaffected by any such manipulations. The expert has learned to transform the working environment to simplify the task that confronts the body in action. Ourr em Ou embo bodi died ed ex expe peri rien ence ce cl clea earl rly y af affo forrds ea ease se of rem emem embe beri ring ng.. On Onee th theeory or y pr prop opos oses es th that at me memo mory ry’s ’s pr prim imar ary y fu func ncti tion on is to me mesh sh th thee em embo bodi died ed co connceptualization of projectible properties of the environment (e.g., a path or a cup cup)) wit with h emb embodi odied ed exp expre ressi ssions ons tha thatt pr provi ovide de non nonpr proje ojecti ctible ble per percep ceptio tions ns (Glenberg, 1997). This meshed conceptualization serves to control action in a three-dimensional environment. For example, people recalled objects in reoom onver the th bas asis isdofth the th e ph phys ical pro pr oxi ximi mity oft in theete th obje ob ject cts s to ach hmant oth ot her as ath the obs bser erve remo move ved thr rou ough ghysic the th eal roo oom, m, and an dtyno not ter rms of theeeac th sema se ntic ic relatedness of the objects (Brewer,  1998). A diff differe erent nt study had climbers reproduce on a scale model the locations and orientations of  23  23  holds of  a climbing wall that each had just climbed (Boschker et al., 2002). Expert climbers correctly recalled more of the holds than did the novices. But the experts also focused on the functional aspects of the wall (i.e., its affordanc da nces es), ), co comp mpar ared ed to no novi vice cess wh who o ex excl clus usiv ivel ely y rep epor orte ted d th thee st stru ruct ctur ural al,, bu butt nott fu no func ncti tion onal al,, as aspe pect ctss of th thee wa wall. ll. Pe Peop ople le’s ’s me memo mory ry fo forr pl plac aces es is gr grou ound nded ed in their embodied experience, as a perceptual symbolic form, and not in some abstract, amodal, schematic representation.

 

144

 

Embodiment and Cognitive Science

Another set of studies demonstrated this point by having participants read and memorize spatial layouts corresponding to scenarios viewed from fr om particu particular lar perspec perspective tivess (e. (e.g., g., in a hot hotel el scene, scene, “T “To o you yourr lef leftt . . . you seee a sh se shim imme meri ring ng in indo door or fo foun unta tain in”) ”) (B (Bry ryan antt & Wri righ ght, t, 1999). Ob Obje ject ctss we werre

located above, below, in front of, in back of, to the left of, and to the right of th thee pa part rtic icip ipan antt in th thee im imag agin ined ed sc scen ene. e. Af Afte terr a sc scen enar ario io wa wass im imag agin ined ed,, th thee time it took participants to locate a particular object was measured. Given that the objects were equally well memorized, one might expect retrieval time to be independent of location. On the other hand, the time needed to locate anmentally object might to the mental that search has to do to find correspond the object. But thedegree resultsofshowed theone fastest responses were given to objects located on the head/foot axis, followed  by the front/back axis, followed by the left/right axis. Participants used a “spatial framework” that was sensitive to environmental asymmetries (such as gravity) and bodily asymmetries (we generally look and attend to things in front of us). Retrieval processes are are clearly shaped, in part, by constraints on embodied experiences. Of course, environmental information can be suppressed so that concept ce ptua uali liza zati tion on is gu guid ided ed by pr prev evio ious us ex expe peri rien ence ces, s, wh whic ich h is a co cons nsci ciou ouss an and d effo ef fort rtfu full us usee of me memo mory ry.. Cl Clos osin ing g on one’ e’ss ey eyes es or lo look okin ing g to towa warrd a bl blue ue sk sky y is an ac acti tion on th that at he help lpss to su supp ppre ress ss th thee en envir viron onme ment nt by el elimi imina natin ting g pr proj ojec ectib tible le prope pr operti rties es tha thatt nor normal mally ly int interf erfer eree wit with h tho though ughtt pr proce ocesse sses. s. Res Resear earch ch sho shows ws that people avert gaze when working on moderately difficult recollection task ta skss (b (but ut no nott ea easy sy on ones es)) an and d th that at th this is be beha havi vior or en enha hanc nces es ac accu cura rate te rem emem em- bering (Glenberg, Schroeder, Schroeder, & Robertson, 1995). The ability to suppress environmental information contributes to prediction, the experience of remembering, and language comprehension. Working Memory Wor orki king ng me memo mory ry is a ty type pe of sh shor ortt-te term rm me memo mory ry th that at fu func ncti tion onss as te temp mpoorary storage for the information needed to accomplish particular tasks, including reasoning, problem solving, and language understanding. A classic model claims that there are three components to working memory: a phonological loop, a visuospatial sketchpad, and a central executive  1986 (Baddeley, ;  Baddeley & Hitch, ). But several recent proposals suggest that working memory also  1974 reflects different embodied abilities (Carlson, 1997; Glenberg, 1997; Wilson, 2001). Speaking is a bodily activity that, in different situations, may either faci fa cili lita tate te or hi hind nder er me memo mory ry of ve verb rbal al ma mate teri rial al.. Fo Forr in inst stan ance ce,, se sear arch ch tasks involving speech disrupt short-term memory for verbal information (Badd (Baddeley eley,, 1986). Of co cour ursse, th thee se seri rial al nat atur uree of sp spea eak kin ing g is on onee rea easo son n that th at ve verb rbal al re rehe hear arsa sal, l, ei eith ther er ov over ertt or co cove vert rt,, is pa part rtic icul ular arly ly im impo port rtan antt wh when en peop pe ople le mu must st re reca call ll ite items ms in se seri rial al or orde derr (H (Hea eale ley y, 1982). Ot Othe herr st stud udie iess rev evea eall that th at sh shor ortt-te term rm me memo mory ry sp span an fo forr ve verb rbal al it item emss de depe pend ndss on th thee ti time me ne need eded ed

 

Imagery, Memory, and Reasoning

 

145

to pronounce the items. Thus, Chinese and English speakers have differentt me en memo mory ry sp span ans, s, be beca caus usee of th thee di difffe ferren entt ti time mess ne need eded ed to pr pron onou ounc ncee th thee same items in these different languages (Stigler, Lee, & Stevenson, 1986). Cros Cr osss-li ling ngui uist stic ic di diff ffer eren ence cess ar aree als also o fo foun und d wi with thin in in indi divi vidu dual als. s. In a st stud udy y of  Wel elsh sh-E -Eng nglis lish h bi bilin lingu gual als, s, me memo mory ry sp span an wa wass fo foun und d to be la larg rger er fo forr En Engl glis ish h than th an fo forr Wel elsh sh wi with thin in th thee sa same me pa part rtic icip ipan antt sa samp mple le (E (Ell llis is & He Henn nnel elly ly,, 1980). Differencess in articulation rates can also account for the fairly extensive Difference differences in memory span between sign and spoken languages. Deaf 

participants, who use American Sign Language, typically have memory spans of around four items, in contrast to the approximately seven items for oral speakers (Wilson & Emmory,  1997; Wilson, Iverson, & Emmorey, 2000). The difference in span appears to reflect differences in articulatory time for speakers and signers (Marschark, 1994). Working memory for sequences of body postures and for sequences of  spatial locomotion displays some of the same effects that characterize immediate memory for language. Thus, sensorimotor rehearsal in working memory is not restricted to spoken language. Experimental participants can ca n en enha hanc ncee th thei eirr sh shor ortt-te term rm me memo mory ry sp span an fo forr di digi gits ts wh when en th they ey ar aree ta taug ught ht to serially tap the appropriate fingers corresponding to different digits at the same time that they verbally rehearse the digits they hear (Reisberg, Rappaport, & O’Shaughnessy O’Shaughnessy,,  1984). Other studies show that eye movement me ntss du duri ring ng a ta task sk of co copy pyin ing g bl bloc ock k pa patt tter erns ns red educ uces es th thee lo load ad on wo work rkin ing g memory mem ory and fac facilit ilitate atess fas fast, t, acc accura urate te per perfor forman mance ce (Ba (Ballar llard, d, Hay Hayhoe hoe,, Poo Pook, k, & Rao, 1997). Many studies have explored the relation between overt movement and memory for information placed along an imaginary path. For example, Baddeley and Lieberman (1980) had participants imagine a four-by-four square matrix pattern. Participants were then required to imagine placing consecutive numbers in a series of adjacent squares following a path arou ar ound nd th thee ma matr trix ix.. Af Afte terr th thee “n “num umbe berr pa path th”” ha had d be been en de desc scri ribe bed, d, th thee pa part rtic ic-ipants were then asked to recall verbally the sequence of imagined movements required to reproduce reproduce the imagined path. In one experimental condition the participants were blindfolded and were given a flashlight with which they had to follow the motion of a swinging pendulum. A tone signaled nal ed whe whethe therr the theflas flashli hlight ght was shi shinin ning g on or of offf the pen pendul dulum. um. The There refor fore, e, participants were performing two distinct tasks concurrently – generating the mental image of a path and moving their arms back and forth in time with wi th a me metr tron onom ome. e. Th Thee me ment ntal al im imag ager ery y ta task sk in invo volv lved ed on only ly au audi dito tory ry in inpu putt and an d vo voca call re reca call ll,, wh wher erea eass th thee mo move veme ment nt ta task skss in invo volv lved ed au audi dito tory ry fe feed edba back ck from the tone and controlled tracking movement of the hand and arm. Neither task involved visual input. Under these circumstances, participants’ recall of the matrix paths was significantly impaired relative to that for performing the imagery task without concurrent movement. This interference seemed to be specific to

 

146

 

Embodiment and Cognitive Science

the combination of the imagined path task and concurrent movement. When the imagined path task was concurrent with visual discrimination of patches of light, recall performance was unimpaired. In other words, the cognitive processes involved in mentally imaging a sequence of location ti onss al alon ong g a pa path th ap appe pear ar to ov over erla lap p wi with th th thee co cogn gnit itiv ivee pr proc oces esse sess in invo volv lved ed in controlling arm movement. The fact that t hat participants were blindfolded in the movement condition indicates that this overlap in processing resources is linked to spatial representations and movement control rather than relying solely on the visual system. Other studies showed that recall of an imagined path was disrupted  by concurrent arm movement (Quinn, (Quinn, 1994). For example, participants in one condition tapped areas on a table top in a random fashion. In a dif-

ferent condition, the experimenter held the participant s hand and moved it across the table top in a random fashion (i.e., the participants had no control over their movements). The results showed that random movement me nt ge gene nera rate ted d by th thee pa part rtic icip ipan antt res esul ulte ted d in di disr srup upti tion on of im imag agin ined ed pa path th recall, but not when the experimenter generated the movement. When the experimenter held the participant’s hand and moved it in a regular and an d pr pred edic icta tabl blee pa patt tter ern, n, th thee du dual al ta task sk dis disru rupt ptio ion n rea eapp ppea eare red. d. In su summ mmar ary y, participants had difficulty recalling the imagined matrix path when they controlled their movements or could predict where their hand was going next. However, However, the participants had little difficulty recalling the imagined matrix path when they did not controlled their own movements. These results demonstrate how working memory may be disrupted by planning movements as well as by executing them. There is also evidence that motor activity influences how people recall info in form rmat atio ion n fr from om co cons nstr truc ucte ted d me ment ntal al im imag ages es (K (Kos ossl slyn yn et al al., ., 1988). Pa Part rtic iciipant pa ntss we werre pr pres esen ente ted d wi with th a se sequ quen ence ce of fo four ur co cont ntig iguo uous us se segm gmen ents ts of a vi vi-sual su al ma matr trix ix.. Ea Each ch se segm gmen entt co cont ntain ained ed an ar arro row w in indi dica catin ting g ho how w th thee se segm gmen entt should be drawn. The participants’ task was to combine the segments by physically drawing them as a single shape. Afterward, participants were shown individual segments of the matrix, and judged if each presented elem el emen entt wa wass pa part rt of th thee ov over eral alll fig figur ure. e. Re Resp spon onse se ti time mess to ma make ke th thes esee ju judg dg-ments from memory increased with serial position in the drawing. Thus, peoplee took about  50 % longer to respond “yes” to a given segment if it peopl was the last segment in the matrix than if it was the first segment. Motor activity to influence the way people recallinvisual stimuli. Thereappears may, however, may, however , be functional differences working memory for purely visual and movement-based information. Various work supports the idea that temporary memory for visual information may be distinct from temporary memory for paths between objects or targeted movement sequences. Logie and Marchetti (1991) examined two contrasting memory task ta skss to te test st th this is po poss ssib ibili ility ty.. On Onee ta task sk in invo volv lved ed pr pres esen enti ting ng pa part rtic icip ipan ants ts wi with th a sequence of squares appearing one after another in different random

 

Imagery, Memory, and Reasoning

 

147

locati ons on a com comput puter er scr screen een.. Par Partic ticipa ipants nts’’ re recog cognit nition ion mem memory ory was the then n locations measured for the sequence tested after a retention interval of  10  seconds during which the screen was blank. A second task presented participants with wi th an ar arra ray y of sq squa uarres ea each ch in a di difffe ferren entt hu huee of th thee sa same me ba basi sicc co colo lorr (e (e.g .g., ., shades of blue). During the retention interval, participants either tapped outt a re ou regu gula larr pa patt tter ern, n, or si simp mply ly vi view ewed ed ra rand ndom om se seque quenc nces es of li line ne dr draw awin ings gs of objects in the same location. Analysis of the recognition scores showed that th at pr pres esen enta tati tion on of th thee lin linee dr draw awin ings gs di disr srup upte ted d re rete tent ntio ion n of th thee co colo lorr hu hues es,,  but did not disrupt retention of the sequence of squares. In contrast, tapping out a pattern disrupted memory for the sequence of squares in different locations, but did not affect memory for color hues. These data suggest thatt a sep tha separa aratio tion n exis exists ts bet betwee ween n a visu visual al tem tempor porary ary mem memory ory sys system tem and a spatial movement-based memory system. Some scholars argue that visuospatial working memory processes are part pa rt of a dy dyna namic mic mo moto torr sy syst stem em,, du dubb bbed ed th thee “i “inn nner er sc scri ribe be,” ,” wh whic ich h is lin linke ked d to a static visual store, called the “visual cache” (Logie,  1995;  Logie &

Pearson, 1997). In this view, the inner scribe is capable of redrawing the contents conte nts of the visual cache, which permits visual and spatia spatiall reh rehearsa earsal, l, manipu man ipulati lation, on, and tra transf nsform ormati ation on of inf inform ormati ation on wit within hin wor workin king g mem memory ory.. Some So me ne neur urop opsy sych chol olog ogic ical al ev evid iden ence ce is co cons nsis iste tent nt wi with th th this is id idea ea.. Fa Fara rah h et al al.. (1988)  described a patient who had great difficulty with mental imagery task ta skss th that at in invo volv lved ed ju judg dgme ment ntss ab abou outt vi visu sual al ap appe pear aran ance ce,, su such ch as “W “Whi hich ch is darker blue, the sky or the sea?” However, the same patient had no difficult cu lty y wi with th im imag ager ery y ta task skss th that at in invo volv lved ed me ment ntal al ac acti tion onss su such ch as im imag agin ing g an and d reca re call llin ing g a pa path th be betw twee een n ta targ rget ets. s. On Onee co comp mput utat atio iona nall mo mode dell of vi visu sual al me memmory appears to capture some of these very performative characteristics of  visuospatial working memory (Kosslyn, 1987). The research described in this section all points to the conclusion that working memory is at the very least partly related to motoric activity. Various bodily actions, including private speech, covert looking, and motoric activity, enhance short-term memory. Some of the evidence suggests a strong isomorphism between short-term memory performance and em bodied action. Findings such as this reveal reveal how working memory is more more a co coll llec ecti tion on of pe perf rfor orma mati tive ve st stra rate tegi gies es (t (tou ouch chin ing, g, sp spea eaki king ng)) th than an it is a se sepparate, architectural component of mind. A sensorimotor account of working memory posits a rapid cycling of information i nformation between perceptual perceptual and motoric forms of coding (Wilson,  2001). The automaticity of translation  between the two forms of coding suggests that isomorphism between perceptual and motoric repr representations esentations confers processing advantages, even in the absence of overt perceptual and motoric activity. activity.  Imagination Inflation Ther Th eree is gr grow owin ing g in inte tere rest st am amon ong g me memo mory ry re rese sear arch cher erss in ho how w im imag agin inin ing g an even ev entt al alte ters rs au auto tobi biog ogra raph phic ical al be belie lief. f. Fo Forr in inst stan ance ce,, wh when en pe peop ople le ar aree as aske ked d to

 

148

 

Embodiment and Cognitive Science

rate thee li th like keli hood od th that ata th they ey ha have ve these expe ex peri rien ence ced d athey sett of se even ev ents ts,, an and d th then en late la terr rateasked are toliho imagine subset of events, will more likely report late la terr on th that at th they ey ha had d ac actu tual ally ly ex expe peri rien ence ced d th thee ev even ents ts pr prev evio ious usly ly ima imagi gine ned d (Garry & Polaschik, 2000). This false memory finding is known as “imagination inflation.” Various studies have demonstrated that the more times someone imagines an event, the more likely he or she will be to say that he or sh shee ha hass in inde deed ed ex expe peri rien ence ced d th thee ev even entt (G (Gof offf & Ro Roed edig iger er,, 1998), al alth thou ough gh this result is not simply due to an increase in familiarity with an event. Of  specific interest here are findings that visually imagining an event from a first-person (own) perspective leads to greater imagination inflation than imagining from a third-person (observer) perspective (Libby, 2003). Most Mo st rec ecen entl tly y, st stud udie iess ha have ve sh show own n th that at fa fals lsee me memo mori ries es we were re mo more re li like kely ly in cases when individuals were instructed to imagine events in great sensory detail (Thomas, Bulevich, & Loftus, 2003). In one study, study, participants sat at a table filled with numerous objects. They heard a series of statements (e.g., “flip a coin”) and then had to perform or imagine performing the cal called led-fo -forr act action ions. s. Sev Severa erall day dayss lat later er,, whe when n par partic ticipa ipants nts cam camee bac back, k, the they y simply had to imagine performing various actions on objects, but without any an y ob obje ject ctss in fr fron ontt of th them em.. Fi Fina nally lly,, pa part rtic icip ipan ants ts’’ me memo mori ries es we werre te test sted ed fo forr what they did on the first day of the experiment. After even a few imaginings in gs,, pe peop ople le so some meti time mess rem emem embe berred pe perf rfor ormi ming ng ac acti tion onss th that at th they ey ha had d no nott

performed. Not only did they falsely claim to have engaged in common acti ac tion onss (e (e.g .g., ., “r “rol olll th thee di dice ce”) ”),, bu butt al also so sa said id th that at th they ey ha had d do done ne ra rath ther er bi biza zarr rree or unusual actions (e.g., “kiss a plastic frog”). These different results suggest that people will have greater confusion  between what they actually have experienced and what they merely imagined the more that they engaged in embodied simulations of the events during the imagination phase of the experiment. One possibility is that planting false memories is more successful to the extent that people engage ga ge in ki kine nest sthe heti ticc im imag agin inat atio ion n of th thes esee ev even ents ts.. Th This is id idea ea is ri ripe pe fo forr fu furt rthe herr empirical study.  Memory for Language A well-known finding from experimental psychology is that people rememb me mber er wo worrds bet ette terr wh wheen th theey rea ead d an and d say th theem alo loud ud th than an whe hen n th they ey only read them (Slamecka & Graf,  1978). This “generation effect” demonstrates that engaging the body, through speaking, leads to more durable memo me mory ry fo forr wo worrds ds.. Re Rese sear arch ch al also so sh show owss th that at pe peop ople le in inci cide dent ntal ally ly re reca call ll th thee

statements in a scripted dialogue better if participants enacted them than if they simply read them (Jarvella & Collas, 1974). Learning to consciously rec ecal alll la lang ngua uage ge ma may y al also so be en enha hanc nced ed if pe peop ople le sp spea eak, k, ra rath ther er th than an rea ead, d, th thee material. The great acting coach Stanislavski (1936/1982) maintained that placing oneself into the situation of the character, and creating a history for a fictitious role, allows actors to give a more believable performance.

 

Imagery, Memory, and Reasoning

 

149

Actors are taught to envision themselves in the particular situations that th thei eirr ch char arac acte ters rs arealso ar e fa faci cing ng an and imagin ima gineepersonal them th emse selv lves es res espo pond ndin ing g evoke as th thee emochar ch ar-acters. Actors are told todidentify memories that tions similar to those supposedly experienced by their characters. One study provided support for the importance of enactment in memory for discourse (Scott, Harris, & Rothe,  2001). Participants were told to read a 5-minute monologue and try to learn as much as they could about the character (they were not told to memorize the text). All participants then engaged in one of five differe different nt 30-minute activities. These were (1) a read-only activity where participants simply performed an unrelated distrac tr acto torr ta task sk,, (2) a wr writ itin ing g ta tassk wh wheere ea each ch pe perrso son n wr wro ote ou outt an ansswe werrs to fiv fivee questions about the character voicing the monologue, ( 3) a collab collaborativ orativee discussion task where groups of participants discussed the five character questions, (4) an independent discussion task where only one person at a time responded to the same set of character questions, and (5) an improvisation task where the participants dramatized in small groups their reactions to the five questions. After the 30-minute activities, the participants recalled the monologue they originally read. Participants in the improvisation group exhibited memory superior (based on the gist of their recalls) to that of people in any other condition. This finding demonstrates the value of embodied activity, where participants directly dramatized the action, in memory for dramatic monologues. Asking people to actively experience a character as a full person (with appropriate cognitive, affective, and emotional dimensions) leads to better memory than in conditions that still provoked

high-level cognitive, but less embodied, involvement. Performing activities thatlanguage. are compatible the meanings of words is important in remembering Peopletowho were induced to nod while incidentally reading positive and negative adjectives were later on more likely to recognize positive adjectives, but participants who were indu in duce ced d to sh shak akee th thei eirr he head adss we were re mo more re lik likel ely y to re reco cogn gniz izee ne nega gativ tivee wo worrds (Foster & Strack,  1996). Moreover, when people moved their heads in a manner that was compatible with the adjectives they read (e.g., nodding for positive adjectives), they were better able to perform a secondary task than th an wh when en th thee wo word rdss an and d he head ad mo move veme ment ntss we werre in inco comp mpat atib ible le.. On Once ce mo more re,, performing incompatible motor and cognitive task concurrently requires more cognitive capacity, which appears to hinder memory for words. A different source of embodiment in memory for language is seen in recent studies on brain imagining. For instance, one PET study showed that rem that emem embe beri ring ng th that at vi vivi vid d wo worrds ha had d be been en pa pair ired ed wi with th so soun unds ds at en enco codding activated auditory brain regions that were engaged during encoding (Nyberg et al., 2000). This finding suggests that memory for language includ cl udes es mo moda dalit lityy-sp spec ecific ific in info form rmat atio ion, n, an and d do does es no nott si simp mply ly in invo volv lvee ac acce cess ss-ing an amodal representation for words in memory.

 

150

 

Embodiment and Cognitive Science

In a related study, using fMRI, participants were first presented with word-sound or word-picture pairs (Wheeler, Peterson, & Buckner,  2000). Afterward, participants engaged different tasks, including recalling the asso as soci ciat ated ed so soun und d or pi pict ctur ure e gi give ven n in a wo word rd,, or re reme memb mber erin ing g wh whet ethe herr a wo wor rd wass pa wa pair ired ed wi with th a so soun und d or pi pict ctur ure. e. Br Brai ain n im imag agin ing g du duri ring ng th thee me memo mory ry po porrtion of the experiment revealed that regions of visual and auditory cortex weree act wer activa ivated ted dif differ ferent ential ially ly dur during ing re retri trieva evall of pic pictur tures es and wor words, ds, re respe specctivel tiv ely y. Fu Furt rthe herm rmor ore, e, th thee reg egio ions ns ac acti tiva vate ted d du duri ring ng re retr trie ieva vall we werre a su subs bset et of  those activated during a separate perception task in which people viewed pictures or heard sounds. These data support the idea that retrieval or visual and auditory information associated with word learning, reactivates some of the same sensory region that was active during learning. Once again, embodied processes engaged in learning language appear to be encoded as part of the representation of language in memory. Finally, many studies have investigated the benefits and costs associated with imagining embodied action when remembering linguistic statement me nts. s. Fo Forr in inst stan ance ce,, re rese sear arch ch sh show owss th that at pe peop ople le’s ’s me memo mory ry fo forr ve verb rbal al st stat ateements about embodied actions, such as “peeling potatoes” or “lighting a cigarette,” are facilitated when people pretended to perform these actions without using real objects (Engelkamp,  1998). People also recall phrases that th at th they ey ha have ve en enac acte ted d be bett tter er th than an th they ey do ph phra rase sess th that at th they ey ha have ve wa watc tche hed d other people enact (Hornstein & Mulligan,  2001). Moreover, phrases requiring quir ing int intera eracti ctions ons wit with h obj object ectss ar aree bet better ter re recal called led tha than n ena enacte cted d sta statem tement entss without objects. Why are enacted statements most memorable? People who enact statement me ntss wi with th an ob obje ject ct ar aree pr prov ovid ided ed mo morre de deta tail iled ed se sens nsor ory y in info form rmat atio ion n th than an are people who only pretend to perform the same action. During its execution, a person receives feedback on his or her own action. Ifrmareal obje ob ject ctss ar are e in invo volv lved ed in th theevisual task ta sk,, th then en ad addi diti tion onal al vi visu sual al an and d ta tact ctil ilee in info form a-

tion is provided. Engelkamp and Zimmer (1984) demonstrated that when participants rate the similarity of pairs of action phrases, when they enact thee fir th first st ac acti tion on ph phra rase se,, ra rath ther er th than an ju just st he hear ar it it,, th thei eirr ju judg dgme ment ntss of si simi mila lari rity ty to the second action phrase are faster than when both action phrases are only presented verbally. Thus, the enactment of the first phrase activates or primes the motoric information necessary for the subsequent comparison so n mo morre ef effe fect ctiv ivel ely y th than an do does es th thee ve verb rbal al de desc scri ript ptio ion n of th thee ac acti tion on.. Si Simp mply ly imag im agin inin ing g th thee fir first st ac acti tion on wa wass no nott su sufffic ficie ient nt,, ho howe weve verr, to ac achi hiev evee th thee pr prim im-ing effect on similarity judgments (Engelkamp, 1998). Thes Th esee va vari riou ouss fin findi ding ngss ad addr dres esss se seve vera rall imp impor orta tant nt is issu sues es ab abou outt me memo mory ry and language (Engelkamp,  1998). First, three kinds of information contribute to the recall of action phrases, namely the visual sensory information provided provided by obse observing rving one’s own body movement movement req require uired d for the action, the visual sensory information from observing the physical objects involved in the actions, and the motoric or kinesthetic information from

 

Imagery, Memory, and Reasoning

 

151

the movement. Second, visual sensory information from perceiving real objects improves memory, but it is not crucial for the enactment effect. Third, visual information from observing other people’s actions does not enhance memory for action statements compared to enactment by the rememberer. Motoric or kinesthetic information appears to play the most critical role in remembering linguistic statements. Reasoning

Assessing human intelligence often focuses on how people reason, make deci de cisi sion ons, s, an and d so solv lvee pr prob oble lems ms.. As is th thee ca case se fo forr me ment ntal al im imag ager ery y an and d me memmory, most cognitive scientists view human reasoning as a computational skill that requires altering abstract, symbolic representations from initial problem states to reach different goal states (e.g., means-ends analysis). People presumably solve problems via general reasoning strategies and domain-specific knowledge. But consider the following reasoning problem. Imagine running across a large room from one wall to another (Schwartz, 1999). Now imagine doing the same activity, this time in waist-deep water. Although the spatial relations are identical in both cases, our imaginary feeling of the two situations is quite different. In fact, people take longer to imagine i magine how long it takes to walk 30 yards carrying a heavy backpack than it does to walk the same distance without carrying anything (Decety, (Decety, Jeannerod, & Problanc, 1989). Although this finding may simply be due to people’s beliefs about the effect of carrying weight on their walking speed, other work shows that people always take longer to respond to the  30 -yard question when told they are wearing a backpack, no matter what they actually believed (Finke & Freyd,  1985). In fact,   5-year-olds correctly modified the force ne need edeed to push pu shdespite an obj bjeetheir ct offincorrect a ta tab ble atbeliefs dif iffe ferreabout nt he heig igh hts to ma make it la land nd on a specific target where theke object would land la nd wh when en th they ey we werre as aske ked d be befo fore reha hand nd (K (Kri rist st,, Fi Fieb eber erg, g, & Wil ilke keni ning ng,, 1993). It is difficult to account for these findings in terms of abstract reasoning strate str ategie giess tha thatt do not ack acknow nowled ledge ge the imp import ortanc ancee of emb embodi odied ed sim simula ulatio tion n

in ho how w pe peop ople le so solv lvee pr prob oble lems ms.. In fa fact ct,, st stud udie iess sh show ow th that at pe peop ople le ca can n so solv lvee a physical problem by simulated doing, which involves an imagined action that facilitates correct inferences even when people are unable to verbally arti ar ticu cula late te wh what at th they ey ar aree do doin ing g (S (Sch chwa wart rtzz & Bl Blac ack, k, 1999). Fo Forr in inst stan ance ce,, pe peooplee im pl imag agin ined ed th that at tw two o be beak aker erss of wa wate terr wi with th id iden enti tica call he heig ight htss bu butt di difffe ferren entt widths were tilted at various angles. When people were explicitly asked about when the water would reach the top of both beakers, they did so incorrectly. However, when people closed their eyes and tilted the glasses until the imagined water reached the top, they did so correctly for both types of beaker. beaker. Simulated doing through imagery includes motor as well as vis visua uall co comp mpon onen ents ts,, an and d th that at mo moto tori ricc in info form rmat atio ion n is no nott al alwa ways ys av avail ailab able le

 

152

 

Embodiment and Cognitive Science

to visual awareness or consciousness when solving real-world problems (Schwartz & Black, 1999). How people move their bodies may even influence creative problem solving. A recent series of studies showed how arm flexion elicits a systematic processing strategy that facilitates creative insight (e.g., the ability to engage in contextual set-breaking, restructuring, and mental search),  but arm extension impairs insight processes (Friedman & Foster Foster,,  2000). Furthermore, data from the same studies revealed that people solve more analogy problems when flexing their arms as opposed to extending them. Thes Th esee em empi piri rica call fin findi ding ngss ar aree no nott du duee to pa part rtic icip ipan ants ts’’ ow own n af affe fect ctiv ivee st stat ates es or moods that may arise from the activity of moving their arms in particular ways wa ys.. In Inst stea ead, d, mo moto torr ac actio tions ns,, su such ch as mo movi ving ng yo your ur ar arms ms in pa part rtic icul ular ar wa ways ys,, influence cognitive processes associated with creative insight and problem le m so solv lvin ing. g. On Onee po poss ssib ibil ilit ity y is th that at ap appr proa oach ch an and d av avoi oida danc ncee mo moto torr ac acti tion onss not only elicit bodily signals that trigger differential processing processing strategies,  but may also differen differentially tially activate the brain-based motivational system (Lang, 1995). People often rely on environmental resources in an embodied manner when wh en so solv lvin ing g di diff ffer eren entt pr prob oble lems ms.. Pa Part rtic icip ipan ants ts wi with th go good od ima imagis gistic tic ca capa paci ci-tiees we ti werre as ask ked in one st stud udy y to ob obse serv rvee an and d rec ecal alll a dra raw win ing g of an amb mbig iguuous picture (duck/rabbit) (Chambers & Reisberg, 1992). The drawing was flipp fli ppab able le in th that at it co coul uld d be vi view ewed ed as ei eith ther er of tw two o di difffe ferren entt th thin ings gs,, th thou ough gh not at once. The participants had not seen the duck/rabbit picture before  but were trained on related examples (Necker cubes, face/vase pictures) to ensure that they were familiar with the phenomenon in question. They were briefly shown the duck/rabbit, told to form a mental picture so that they could draw it later, and then asked to consult their mental image to seek an alternative interpretation for the two pictures. Participants were given hints that they should try to shift their visual fixation (e.g., from lo lowe r le left ft to up uppe perrinterpretations righ ri ght) t).. Fi Fina nall lly y, pa part icip ipan ants ts dr drew ew th thei eirr im imag ages es an and d tr trie ied d tower find alternative of rtic their drawing. Despite the fact that some of the participants were vivid imagers, none of the participants tested were able to recognize the alternative image of  the stimuli (e.g., from duck to rabbit). In sharp contrast, all participants were able to find the alternate construal after they had made their own drawings. This pattern of findings shows how people’s problem-solving capacities are significantly extended by the simple device of externalizing

inform info rmat atio ion n vi viaa bo bodi dily ly ac activ tivit ity y (e (e.g .g., ., dr draw awin ing g th thee im imag agee fr from om me memo mory ry)) an and d then confronting the external trace using on-line visual perception. People employ various “complementary strategies” to alter the environment to enhance their reasoning abilities. These strategies include em bodied actions, such as using one’s hands to manipulate Scrabble pieces or to do complex arithmetic using pencil and paper, that help people improve their thinking and memory to solve problems. For example, when

 

Imagery, Memory, and Reasoning

 

153

view ing g an up upsi side de-d -dow own n ph phot otog ogra raph ph,, pe peop ople le do no nott tr try y to me ment ntal ally ly rot otat atee viewin thee im th imag agee in th thei eirr he head ads, s, bu butt ph phys ysic icall ally y tu turn rn th thee pi pict ctur uree ri righ ghtt si side de up up.. Th Thus us,, people physically alter the environment, rather than altering their mental abilities, to better identify what are looking at. Complementary strategies clearly enhance people’s problem-solving abilities. For instance, in one study, participants were shown two sets of  30  coins (i.e., different quarters, dimes, and nickels) and asked to calculate the amount in dollars and cents (Kirsh,  1995). People were faster and more accurate in determining the sum when they were allowed to touch the coins than when they were not allowed to use their hands. Touching thee co th coin inss ap appe pear arss to he help lp pe peop ople le rem emem embe berr in inte term rmed edia iate te su sums ms,, in th thee sa same me way that writing down the intermediate sums facilitates solving complex multiplication problems. A different study asked participants to find the sum of digits presented on dice-like markers (Cary & Carlson,  1999). When participants were permitted to handle the markers, every participant actually did so. But when participants were prohibited from touching the markers, they talked aloud far more than in the other condition. Talking provided a kind of environmental support for working memory when other resources such as touching the dice were unavailable. In a related way, way, abacus experts often solve mental arithmetic problems employing specific routines of physically manipulating the abacus (Hatano & Osawa,  1983; Miller & Stigler,  1991; Stigler, 1984). Finally, children who are allowed to gesture count more accurately than when they are prohibited from doing so (Alibali & DiRusso,  1999). Active gesture helps children both to keep track and to coordinate tagging the items and saying the number words. Obse Ob serv rver erss wh who o ac acti tive vely ly ro rota tate ted d th thrree ee-d -dim imen ensi sion onal al no nove vell ob obje ject ctss on a co commputer screen screen later showed faster visua visuall rec recogniti ognition on of these objects than did di d pe peop ople le wh who o pa pass ssiv ivel ely y ob obse serv rved ed th thee sa same me se sequ quen ence cess of im imag ages es of th thes esee visual vis ual obj object ectss (Ha (Harma rman, n, Hum Humphr phrey ey,, & Goo Goodal dale, e, 1999). Pe Peop ople le wh who o en enga gage ge in active rotation also were subsequently faster in a mental rotation task involving the studied objects (James, Humphrey, Humphrey, & Goodale, 2001). People continuously use perceptual information from the environment to guide their actions and reduce cognitive effort in reasoning. One study examined this idea in the context of people’s performance playing the computer game Tetris (Kirsh & Maglio,  1994). In Tetris, pieces, or zoids, enter the board from the top and the players have to decide whether to move them right or left or to rotate them. Physically rotating a piece can save considerable cognitive effort in placing the zoid in their appropriate place over that needed to do the rotation mentally. Overall, people took about 150 ms msec ec to ph phys ysic ical ally ly ro rota tate te a zo zoid id,, wh wher erea eass do doin ing g so me ment ntal ally ly ta take kess

 between 700  and  1500  msec. People clearly off load the internal compu tations needed to place their zoid by physically doing the same transformation in the environment. Thus, people use their bodies and the world

 

154

 

Embodiment and Cognitive Science

to save internal computations. In high-speed tasks, such as playing Tetris, the person and environment can be so tightly coupled that it is better to conceive of the two as constituting a single conceptual system rather than two independent systems (Kirsh, 1995). People also create and employ tools when solving everyday tasks to explore the possible variations in their ideas when seen in the real world. Having Havin g to tool olss av avail ailab able le fo forr do doin ing g th this is,, an and d ex exte tenu nuat atin ing g in inte tern rnal al rep epre rese sent ntaations, makes this easy to do. The most intelligent use of space is to try out possibilities that are difficult to imagine using only internalized thought (Kirsh, 1995). Co Cons nsid ider er ho how w on onee pl play ayss th thee bo boar ard d ga game me of Sc Scra rabb bble le.. Fi Find ndin ing g the best set of words that can be created through combining the letters is easiest to do when the letters are shuffled around. Making a complex recipe often involves the same sort of physical explor pl orat atio ion. n. Wh When en pa part rtic icip ipan ants ts we were re as aske ked d to pr prep epar aree a re reci cipe pe in incl clud udin ing g 3/4 of  2  2/3 of a cup of cottage cheese, most people engaged in some embodied action using external resources, rather than figuring out the exact amount using arithmetic (Brown, Collins, & Duguid,  1989). For instance, one participant took 2/3 cup of the cheese, flattened it out into a uniformly thick circular disk on a utility board, and drew a cross on it with his finger to see the desired amount by discarding the quarter quarter.. Like most participants, this person never verified the procedure algorithmically through arithmetic (e.g., 3 /4 × 2/3 = 1/2). Other studies show that when people employ pl oy ex exte tern rnal al res esou ourc rces es du duri ring ng pr prob oble lem m so solv lvin ing g (e (e.g .g., ., en enga gagi ging ng in co comp mple lex x paper-folding tasks), they actively leave traces of the work, such as written notations, that facilitate their retracing their work later on (Shirouzu, Miyake, & Masukawa,  2002). In these ways, formal reasoning skills are not independent of people’s ability to physically manipulate the external environment. All these studies point to the possibility that the dynamic relations between embodied agents and the environment form a complex computational system (Hutchins,  1995;  Suchman,  1987;  Wilson,  2004). Embodied agents can inventively exploit facts about the physical environment to avoid explicit representations and reasoning (Agre & Chapman, 1987; Brooks,  1991). For example, Hutchins (1995) demonstrated how the cognitive processe processess involved in flying a plane or piloting a ship do not take place in the pilot’s head, but are distributed throughout throughout the cockpit, in the members of the crew, crew, the control panel, and the manuals. Anothe Ano therr eme emerg rging ing lite literat ratur uree on emb embodi odied ed re reaso asonin ning g con concer cerns ns stu studen dents’ ts’ learning of physics. Solving physics problems generally requires that students understand and correctly apply abstract physical laws. But novice physics physi cs stude students’ nts’ knowledge knowledge about physical physical pheno phenomena mena is not a tightl tightly y connected, logically organized structure that might be properly called a theory. Instead, studies shows that physical knowledge is a set of loosely connected ideas about the world, abstracted from concrete experiences,

 

Imagery, Memory, and Reasoning

 

155

that can be used to generate explanations in particular situations and in response to particular questions or cues (diSessa,  1993). These ideas, called “p-prims,” are “phenomenological primitives,” because they are self-evident to the holder holder,, requiring no further explanation. For example, students do not try to explain why you get more results when wh en yo you u ex expe pend nd mo morre ef effo fort rt pu push shin ing g a bi big g roc ock, k, be beca caus usee th ther eree is no noth thin ing g puzzling about this phenomenon. Students’ preconceptual understanding of something like impetus is simply an invention particular to a small class of problems. Thus, when explaining the physics of tossing a ball, the action of one’s hand on the ball is never described because it is entirely unproblematic from the student’s point of view. A p-prim of “force as mover” describes and explains precisely the situation of tossing. In contras tr ast, t, af afte terr th thee ba ball ll de deta tach ches es fr from om th thee ha hand nd,, th thee si situ tuat atio ion n is pr prob oble lema mati tic. c. Bu Butt why does the ball keep going upward until the peak of the trajectory even though gravity acts on the ball to go down? At the peak, the ball appears to stop, and it looks as if the forces are balancing. What is balancing the force of gravity? DiSessa claims that this kind of problem presents a conflict that forces students to invent a concept such as impetus to explain the continuous force that keeps the ball moving and balancing at the peak of  its trajectory. In this way, naive theories are consistent ad hoc explanations situatedly invented, rather thanphysical a product of some theory or “representation” “repr esentation” in the mind. In everyday life, people often communicate with each other about motion and force, not to explain these phenomena,  but to coordinate their situated actions for collaboration. A person may say “Pus “P ush h it ha hard rder er”” or “K “Kee eep p yo your ur st stre reng ngth th fo forr pu pull llin ing, g,”” fo forr ex exam ampl ple, e, wi with thou outt needing to articulate the meaning of force and motion. To take another case, the question of why it is hotter in the summer may activate for students a p-prim connecting proximity and intensity. intensity. As students tacitly know from their own experiences, one is more strongly influenced by something the closer one is to it. Candles are hotter and  brighter the closer you get to them, music is louder the closer you are to the speaker, the smell of garlic is more intense the closer you bring it to your nose. Simple activation of the idea “closer means stronger” with the p-pris p-p rism m “co “conne nnecti cting ng pr proxi oximit mity y to int intens ensity ity”” allo allows ws stu studen dents ts to und unders erstan tand d why it is hotter in the summer than in the winter, winter, because the sun is closer to the earth in summer. Another problem students face concerns the idea that gravity holds a rolling ball onto the ground. One student explained this first by saying “gravity holds the ball right onto the surface,” which indicates a misconception of gravity as a constraint, “holding the ball at a certain distance.” Onee ma On may y ch char arac acte teri rize ze th thee st stud uden ent’ t’ss mi mist stak aken en id idea ea as a pi piec ecee of kn know owle ledg dgee he could incorrectly apply in many situations. But the student’s idea may alternative alter natively ly be seen as involving the activa activation tion of one or mor moree p-pri p-prims ms from what diSessa called the “constraint cluster,” cluster,” including “supporting,”

 

156

 

Embodiment and Cognitive Science

“guiding,” and “clamping.” Thus, the student’s explanation is specific to the situation, and reflects an imaginative line of reasoning, because in the situation, the student’s idea was not inconsistent with Newtonian reasoning. In this way, the student’s primitive, embodied reasoning actually indicates the seeds of a more complex Newtonian understanding. These few examples serve to illustrate the power of embodied thought, in th thee fo form rm of pp-pr pris isms ms,, in ho how w st stud uden ents ts le lear arn n to so solv lvee el elem emen enta tary ry ph phys ysic icss problems. Learning to solve these physics problems eventually refines pprims, but does not replace them (diSessa, 1993). Thus, embodied knowledge constitutes an enduring part of people’s advanced reasoning about physical events. Somee cog Som cognit nitive ive sci scient entist istss ar argue gue tha thatt exp experi erienc encee wit with h the phy physic sical al world sometimes confuses students in reasoning about physics problems (Clement, 1987; McCloskey & Kohl, 1983). Yet teachers of young children often oft en ack acknow nowled ledge ge the imp import ortanc ancee of all allowi owing ng chi childr ldren en to phy physic sically ally int inter er-act with materials when learning mathematics and science concepts. One case study clearly demonstrated how one child’s direct kinesthetic experience supported her learning about the mathematics of motion (Wright, 2001). Karen is a third/fourth grader in an urban public school in the Boston area. Her teacher claimed that students’ first-hand experiences of  motion (e.g., running, walking, etc.) mathematical concepts of motion. Analyses of helped Karen’sthem workunderstand with a teacher shows thrree ex th exam ampl ples es of ho how w sh shee us used ed he herr bo body dy as a res esou ourrce fo forr le lear arni ning ng ma math theematics. First, Karen used her body to produce different different positive outcomes of a two-person race. Second, Karen used her hands in place of dropped objects to compare patterns of speed. In these two instances, Karen came to understand that any one motion (e.g., “run” or “walk”) can have a variety of speeds, and that these differences (e.g., walk vs. walk slowly vs. run fast vs. moderately fast) predictably affect the outcomes of a race. Finally, Karen’s enacting different motion types facilitated her correct interpretation of a data table. She recognized how distance affected varying speeds when time was held constant. All three cases showed how Karen used us ed he herr bo body dy to en enac actt mo motio tion n ty type pess to co conc ncep eptu tual aliz izee ne new w re repr pres esen enta tati tion onss about time and space. One mathematics and physics teacher reports that his students more easily learned ideas about space, time, and speed by his havi ha ving ng th them em mo move ve th thei eirr bo bodi dies es in va vari riou ouss wa ways ys (L (Lil ilje jeda dahl hl,, 2001). Do Doin ing g so enab en able led d hi hiss st stud uden ents ts to co conn nnec ectt st stat atic ic re repr pres esen enta tati tion onss of mo moti tion on di dirrec ectl tly y to their own embodied actions. Children are not the only people to use embodied activity in thinking about math and science concepts (see Chapter  7). A study of how physicists interpret graphs revealed that “scientists engaged in collaborative interpretive activity to transport themselves through talk and gesture into constructed visual representations through through which they journey with their wordsandbodies”(Ochs,Jacoby,&Gonzales, 1994: 168).Ingeneral,studies

 

Imagery, Memory, and Reasoning

 

157

like these show how physical enactment of different sorts is a critical re-

source in scientific reasoning. Summary

Traditional views of higher-order cognition as computational processing on sy symb mbol olic ic rep epre rese sent ntat atio ions ns fa fail il to ca capt ptur uree th thee imp impor orta tanc ncee of em embo bodi dime ment nt in human thought. People’s previous and current embodied actions serve as the grounding for various aspects of imagination, memory memory,, and reasoning. in g. On On-li -line ne em embo bodi died ed pr proc oces esse sess em emph phas asiz izee ov over ertt se sens nsor orim imot otor or ac acti tivi vity ty to assi as sist st wi with th co cogn gnit itiv ivee ta task skss th that at in inte tera ract ct wi with th th thee im imme medi diat atee wo worl rld. d. Of Offf-lin linee embodiment occurs when sensorimotor processes run covertly to assist with the representation and manipulation of information in the temporary absence of task relevant input or output. Both of these aspects of  embodiment work to create an embodied model of mind that is not internall to pe na peop ople le’s ’s he head ads, s, bu butt is di dist stri ribu bute ted d as a “c “cog ogni niti tive ve we web” b” ac acrros osss br brai ains ns,,  bodies, and world. This distributed, embodied view of cognition offers a vision of human thought that is far less internally computational and far more bodily extended into the real world of action than is traditionally understood in cognitive science. Once again, this claim does not imply that cognition never relies on disembodied, computational processes. Yet Yet there is an ever-growing literature support a view of imagery, memory, and reasoning as intimately tied totobodily activity activity, , such that higher-order cognitive process are situated, embedded, and embodied.

 

6 Language Languag e and Communic Communication ation

Communicating with others mostly requires that we move our bodies. When I speak to a friend, I move my lips, tongue, and vocal apparatus, along with vario various us body parts not dire directly ctly associated associated with speech, such as my eyes, hands, head, and torso. In some situations, I can effectively communicate some idea or belief simply by nodding my head or blinking my eyes. Even the absence of overt body movement can communicate, as when you stare blankly at me after I ask you a question. We interpret the lack of body movement as meaningful precisely because bodily motion normally communicates. The traditional belief among many cognitive scientists is that meaning is an abstract entity divorced from bodily experience. Understanding language is assumed to require breaking down the physical information (e.g., speech sounds) into a language-independent medium that constitutes the “language of thought.” The meaning of any sentence presumably can be represented as a complex proposition consisting of a pred pr edic icat atee wi with th se seve vera rall ar argu gume ment nts. s. Lo Long nger er te text xtss ar aree rep epre rese sent nted ed in as asso soci ciaativee ne tiv netw twor orks ks of pr prop opos osit itio ions ns (p (prred edica icate te-a -arrgu gume ment nt sc sche hema mas) s) or as ab abst stra ract ct mental models (Fletcher,  1994;  Fletcher, van den Broek, & Arthur,  1996; Kintsch,  1988). More recent approaches to the semantics of words and sentences include powerful quantitative tools, such as hyperspace analog to language (HAL) or latent semantic analysis (LSA), both of which reduce the problem of meaning to a simple matter of computing word cooccurrence (Burgess, 2000; Burgess & Lund, 2000; Kintsch, 1998). The resulting high-dimensional semantic space of words can be used to predict several psychological effects, such as the acquisition of vocabulary in children, word categorization, sentence coherence, priming effects, meaning similarity,, and learning difficulty of texts. similarity The primary problem with these different views of language, similar to those with the other topics covered in this book, is that they conceive of meaning, and human cognition more generally, in terms of abstract 158

 

Language and Communication

 

159

and disembodied symbols. Traditional views of language and communication ignore the fundamental problem of how meaning is grounded in ordinary experience (the “symbol grounding problem”) (Harnad, 1990; Searle,  1980), especially in regard to how meaningful symbols relate to embodi emb odimen mentt and re realal-wor world ld re refer ferent entss (Jo (Johns hnson, on, 1987). Al Alth thou ough gh co cogn gniti itive ve neuro neu roling linguis uists ts exa examin minee the neu neural ral bas basis is for hum human an lin lingui guisti sticc abi abiliti lities, es, mos mostt rese re sear arch ch on th thee lin links ks be betw twee een n la lang ngua uage ge an and d br brai ain n fu func ncti tion onss do no nott pr prop oper erly ly acknow ack nowled ledge ge the imp import ortanc ancee of peo people ple’s ’s or ordin dinary ary kin kinest esthet hetic ic exp experi erienc ences. es. This neglect has seriously undermined scientific understanding of the relation lat ionss bet betwee ween n min mind d and bod body y, and and,, mor moree spe specifi cifical cally ly,, lin lingui guisti sticc mea meanin ning, g,

communication, and embodiment. Myaiminthischapteristoarguethecaseforembodimentinpsychologicall th ca theo eori ries es of la lang ngua uage ge an and d co comm mmun unic icat atio ion. n. I co cons nsid ider er a ra rang ngee of hy hypo poth theesess on th se thee po poss ssib ible le in influ fluen ence ce of em embo bodi died ed ac acti tivi vity ty in la lang ngua uage ge us use, e, ra rang ngin ing g from speech perception and the evolution of language to word meaning and discourse comprehension. The chapter also describes the significant work wo rk in co cogn gnit itiv ivee sc scie ienc ncee on ge gest stur uree an and d it itss cr crit itic ical al rol olee in co comm mmun unic icat atio ion n and cognition. Time Course of Linguistic Communication

Embodied experience may influence linguistic communication at several levels. Each of these levels reflects how the body shapes communication at different time scales, ranging from slow-moving linguistic evolution to fast-moving aspects of immediate, online language production and comprehension. Consider the following hypotheses: (1)  Embodiment plays a role in the development of and change in the mean me anin ings gs of wo word rdss an and d ex expr pres essi sion onss ov over er tim time, e, bu butt do does es no nott mo moti tiva vate te contemporary speakers’ use and understanding of language. l anguage. (2)  Embodiment motivates the linguistic meanings that have currency with wi thin in lin lingu guis isti ticc co comm mmun unit itie ies, s, or ma may y ha have ve so some me role in an id idea ealiz lized ed speakers/hearers’ speakers/hear ers’ understanding of language. But embodied experience does not play any part in speakers’ abilities to make sense of  or process language. (3)   Embodiment motivates contemporary speakers’ use and understanding of why various words and expressions mean what they do, but does not play any role in people’s ordinary on-line production or comprehension of everyday language. (4)  Embodiment functions automatically and interactively in people’s on-line use and understanding of linguistic meaning. These hypotheses reflect a hierarchy of possibilities about the interaction of embodied experience with different aspects of language use

 

160

 

Embodiment and Cognitive Science

and understanding. Because they relate to different time scales on which linguistic meaning occurs, each hypothesis requires appropriate methods to empirically study it, with certain disciplines being better able to provide evidence in support of these different possibilities. My essential claim is that debates about embodied language and communication will be best served by looking specifically for its effects on human performance at these differing levels, and not just arguing a priori that em bodied experience plays little litt le or no role in speaking, understanding, and communication. Language Change

Embodi Embo dime ment nt ha hass a cl clea earr ro role le in ho how w la lang ngua uage gess ch chan ange ge,, es espe peci cial ally ly through the use of metaphoric reasoning. Many types of word mean-

ings become extended from body concepts to conceive, and talk of, ideas from dissimilar domains, such as space and time (Geeraerts,  1997;  Traugott & Dasher, 2002). For instance, Sweetser (1990) has shown how many polysemous words in Indo-European languages acquired their nonphysical meanings via metaphorical extensions from earlier acquired concrete, physical meanings, so that VISION/HEARING/TACTILE ACTS gets ge ts ma mapp pped ed on onto to id idea eass ab abou outt IN INTE TELL LLEC ECT T. Th Thus us,, me meta taph phor oric ical al ma mapp ppin ings gs from the idea of visually seeing things to the idea of intellectually understanding things defines a pathway for semantic change. The presence of co conc ncep eptu tual al me meta taph phor orss su such ch as UN UNDE DERS RST TAN ANDI DING NG IS SE SEEI EING NG no nott on only ly explai expl ains ns ho how w wo worrds ch chan ange ge th thei eirr me mean anin ings gs hi hist stor oric ical ally ly (i (i.e .e., ., wh why y th thee ph phys ys-ical sense of “see” regularly gets extended via metaphor at a later l ater point to have ha ve a no nonp nphy hysi sica call me mean anin ing) g),, bu butt al also so mo motiv tivat ates es fo forr co cont ntem empo pora rary ry sp spea eakkers just why it is that polysemous words have the specific meanings they do (e.g., why it just makes sense to us to talk about understanding ideas using expressions such as “I clearly see the point you’re making in this essay”). With few exceptions, words in Indo-European languages meaning “see “s ee”” reg egul ular arly ly ac acqu quir iree th thee me mean anin ing g “k “kno now” w” at wi wide dely ly sc scat atte tere red d tim times es an and d places (see Andrews, 1995 for similar evidence from Russian). A different instance of how bodily ideas/experiences drive semantic change through metaphor is seen in the development of modal verbs (Sweetser, 1990). Modal verbs, such as “must,” “may,” “may,” and “can,” pertain to experiences of actuality, possibility, and necessity. Thus, we often represent se nt ou ourr ex expe peri rien ence ce of th thin ings gs,, ev even ent, t, an and d rel elat atio ions ns as be bein ing g ac actu tual al,, po poss ssib ible le,, or necessary. We often feel ourselves able to act in certain ways (“can”), perm pe rmit itte ted d to pe perf rfor orm m ac acti tion onss of ou ourr ow own n ch choo oosi sing ng (“ (“ma may” y”), ), an and d co comp mpel elle led d  by forces beyond our control (“must”). A pervasive embodied metaphor of THE MIND IS THE BODY drives the use of physical experiences to conceive of mental processes of reasoning as involving forces and barriers

 

Language and Communication

 

161

anal alog ogou ouss to ph phys ysica icall an and d so soci cial al fo forc rces es an and d ob obst stac acle les. s. Co Cons nsid ider er th thee fo follo llowwan ing examples (Gibbs, 1994: 160): “You must move your foot, or the car will crush it.” “You (Physical necessity necessity.) .) “Sally can reach the fried eel for you.” (She is physically capable of reaching it.) “Paul must get a job now now.” .” (Paul is forced to get a job, although the compulsion is not physical.) “You may now kiss the bride” “You (No social barrier prevents you from kissing the bride)

Thedifferentrootmeaningsofthesemodalverbsinvolvenotionsofforce and obligation that are metaphorically extended from the bodily sense to describe more abstract ideas about mental processes. The historical evidence demonstrates that semantic change for many words, as seen here for modal verbs, is driven by embodied metaphor. Most generally, there aree ext ar extens ensive ive map mappin pings gs fr from om per percep ceptua tual/k l/kine inesth stheti eticc exp experi erienc ences es ont onto o cog cog--

nitive processes, and later on ideas about verbal expression, that appear to drive semantic change in a wide variety of languages across different times and places (Koivisto-Alanko, 1998). Speech Perception

How do listeners link speech and meaning? How do listeners learn to extract phonetic information from a complex speech array to identify individual words? Categorizing speech sounds is an enormously complex task. Environmental noises, including that from other speech, often interferre wi fe with th th thee sp spee eech ch si sign gnal al.. Ot Othe herr fa fact ctor ors, s, su such ch as sp spea eaki king ng ra rate tess an and d vo voic icee of the speaker(s) (i.e., high vs. low pitch), also complicate stable speech identification. The traditional approach acknowledges the invariance problem by focusing on invariant acoustic functions used in phoneme identification (Steve (St evens ns & Blu Blumst mstein ein,, 1981). Th Thee ph phon onet etic ic in info form rmat atio ion n in th thee sp spee eech ch si sign gnal al is th then en co comp mpar ared ed wi with th th thee id iden enti tica call ab abst stra ract ct rep epre rese sent ntat atio ions ns fo forr ph phon onem emes es in the mental lexicon. For example, the word “keep” is composed of three phon ph onet etic ic se segm gmen ents ts:: an in init itia iall co cons nson onan antt (/ (/k/ k/), ), a me medi dial al vo vowe well (/ (/i/ i/), ), an and da final fin al co cons nson onan antt (/ (/p/ p/). ). Ea Each ch ph phon onet etic ic se segm gmen entt ma may y be de desc scri ribe bed d in te term rmss of  a sm smal alll se sett of di dist stin inct ctive ive fe feat atur ures es th that at re reco comb mbin inee to fo form rm th thee se sett of se segm gmen ents ts in a given language. Butt th Bu ther eree ar aree pr prob oble lems ms wi with th th this is vi view ew.. Pr Prim imar aril ily y, th ther eree is no nott a co comp mple lete te set of invar invariant iant properties properties that can unambiguously unambiguously identify all phon phonetic etic segm se gmen ents ts in th thee sp spee eech ch si sign gnal al (K (Kla latt tt,, 1989). Ma Many ny fe feat atur ures es co cont ntri ribu bute te to th this is

 

162

 

Embodiment and Cognitive Science

comp lexi xity ty in ma mapp ppin ing. g. Fo Forr ex exam ampl ple, e, wh when en sp spea eake kers rs ta talk lk,, th they ey do no nott pr proocomple ducee the pho duc phonet netic ic seg segmen ments ts of a wor word d seq sequen uentia tially lly,, bec becaus ausee the art articu iculat latory ory gestures for certain segments overlap with others. Coarticulation enables speakers to rapidly produce sequences of phonetic segments, but it complicates listeners’ mappings of acoustic signals and phonetic structures. Onee so On solu luti tion on to th thee pr prob oble lem m of a la lack ck of in inva vari rian ance ce in th thee ac acou oust stic ic si sign gnal al is to su sugg gges estt th that at sp spee eech ch pe perrce cept ptio ion n ma may y be al alig igne ned d wi with th em embo bodi died ed ar arti ticu cu-latio la tion n pr proc oces esse sess (L (Lib iber erma man n et al al., ., 1967). Alt Althou hough gh the re relati lations onship hip bet betwee ween n acoustic stimuli and perception is quite complex, the link between articulati la tion on an and d pe perrce cept ptio ion n is mo morre di dirrec ect. t. So Soun unds ds pr prod oduc uced ed in si simi mila larr wa ways ys bu butt with varying acoustic representations are perceived in similar ways. For example, the acoustic cues for a [d] in a syllable onset differ depending on thee vo th vowe well th that at fo follo llows ws,, ye yett an ar artic ticula ulato tory ry de desc scri ript ption ion of [d [d]] as an “a “alv lveo eola larr constriction” compatible in each environment. decoding results in an abstractisphoneme-based representation via Speech a listener’s own knowledgee of the ef edg effec fects ts of coa coarti rticul culati ation on on his his/he /herr own pr produ oductio ctions. ns. Ane Anecdo cdotal tal eviden evi dence ce sug sugges gests ts tha thatt pr produ oducin cing g spe speech ech sil silent ently ly fac facilit ilitate atess stu studen dents’ ts’ lea leanning of new speech sounds. Trying Trying to say new sounds encourages learners to at atte tend nd to su subt btle le mo moto torr pr proc oces esse sess th that at mig might ht ot othe herw rwis isee be ov over ersh shad adow owed ed  by auditory sensations. sensations. This motor theory of speech speech perception perception maintains that th at lis liste tene ners rs us usee th thei eirr ow own n in inte tern rnali alize zed d ar arti ticu cula lato torr mo moto torr pa patt tter erns ns,, en enca cappsulated in a “specialized speech module,” to interpret spoken language

(Liberman, 1970; Liberman & Mattingly, 1989). The motor theory was devised to explain how disparate acoustic stimuli could represent the same phoneme. The discontinuity in the acoustic stimulusfor“got”and“gaze”providesastrikingexampleinwhichthepercepts are in accord with the articulatory maneuvers that make the sounds, rath ra ther er th than an th thee ac acou oust stic icss of th thee so soun unds ds em emit itte ted. d. An Anot othe herr ex exam ampl plee is fo foun und d in the interposition of brief silences in a spoken sentence. If a tape recording is made of the phrase “please say stop” and a silence of about  5  ms is inserted after “say,” the phrase becomes “please say chop” (Dorman, Studdert-Kennedy,, & Raphael, 1977). This makes sense given the articuStuddert-Kennedy latory gestures required to say “please say chop.” Producing the sudden  burst of sound for “chop” requires requires closing the airway briefly, briefly, thereby crecreating a brief silence. Considerable evidence shows that perception of phonemes is accomplis pl ishe hed d no nott si simp mply ly by an analy alysi siss of th thee ph phys ysic ical al ac acou oust stic ic pa patt tter erns ns bu butt th thrrou ough gh thei th eirr ar arti ticu cula lato tory ry ev even ents ts,, su such ch as mo move veme ment ntss of th thee li lips ps,, to tong ngue ue,, an and d so on  1996; Liberman  2000and (Fowler,  Liberman,findings & Whalen, ). For instance, Fowler ( 1994 1987); analyzed from speech production perception tasks tas ks and andpr propo oposed sedtha thatt lis listen teners ers“fo “focus cus on aco acoust ustic ic cha change nge,, bec becaus ausee cha changnging regions of the sound spectrum best reveal the gestural constituency of  the tal talker ker’s ’s utt uttera erance nces.” s.” Not sur surpri prisin singly gly,, sig sign n lan langua guage ge use users rs als also o foc focus us on opti op tica call ch chan ange ge (i (i.e .e., ., mo move veme ment nt)) be beca caus usee ch chan angi ging ng reg egio ions ns of th thee ob obse serv rver’s er’s

 

Language and Communication

 

163

visu sual al fie field ld be best st in indi dica cate te th thee ge gest stur ural al co cons nsti titu tuen ency cy of th thee si sign gner er’s ’s ut utte tera ranc nce. e. vi There is also evidence that children’s own attempts at vocal articulation are governed to some extent by their visual observations of articulations patterns of others (Studdert-Kennedy (Studdert-Kennedy,, 1981). The child recognizes speech sounds as patterns of gestures and in attempting to represent them, often fails to produce the correct sound because of an error in timing. Othe Ot herr ev evid iden ence ce su sugg gges estin ting g th that at sp spee eech ch is pr proc oces esse sed d in te term rmss of ar artic ticula ula-tory gestures is found in resear research ch on immediate serial recall. For example, the recency effect refers to people’s superior performance in recalling the last few spoken items in a list of words. The suffix effect refers to the decrea cr easse in rec ecal alll fo forr th thee la lasst fe few w it iteems wh when en th thee li list st is fo foll llow owed ed by a sin ingl glee ir ir-rel elev evan antt sy syll llab able le or wo worrd (t (thu hus, s, a red educ ucti tion on in th thee rec ecen ency cy ef effe fect ct)) (N (Nai airn rnee & Walters, 1983). Bo Both th th thee rec ecen ency cy an and d su sufffix ef effe fect ctss ar aree si sign gnifi ifica cant ntly ly red educ uced ed forr no fo nons nspe peec ech h st stim imul uli, i, su such ch as wh when en pe peop ople le rea ead d pr prin inte ted d wo worrds (k (kno nown wn as the modality effect). theories claim that modality deffects  because the last few Traditional items are briefly held in an unprocessed unprocesse form arise (i.e., their acoustic properties) that is then disrupted by hearing additional material (Surprenant, Pitt, & Crowder,  1993). But studies show that the recency and suffix effects occur for human speech stimuli, but are nonexistent for nonlinguistic sounds (Greene & Samuel,  1986). In fact, when people hear a list of words followed by an ambiguous suffix, and they think it is the sound of a trumpet, then the suffix effect disappears, but not when told that this same sound was the syllable “wa” (Ayres et al., 1979). These findings suggest that it is not sensory modality (hearing vs. reading) per se that guides speech perception. Instead, speech perception is organized around the fact that the auditory information was spoken

speech, or even thought to be human speech, with articulatory gestures. Anothe Ano therr phe phenom nomeno enon n dem demons onstra tratin ting g how pho phonem nemee per percep ceptio tion n is shaped sha ped by re recog cognit nition ion of the their ir art articu iculati lation on is the “Mc “McGur Gurk k ef effec fect” t” (Ma (Massa ssaro ro,, 1987;  McGurk & MacDonald,  1976). Watching a speaker’s mouth saying a syllable that conflicts with a heard syllable often changes what syllable is heard. For example, when participants hear the bilabial stop “ba” but see a speaker uttering the velar consonant “ga,” they often report hearing thee al th alv veo eola larr st stop op “d “da” a” th that at reta tain inss som omee of th thee ph pho one neti ticc fe feat atur ures es of th thee tw two o soun so unds ds.. Pa Part rtic icip ipan ants ts in th thes esee st stud udie iess ar aree no nott aw awar aree of th this is co confl nflic ictt be betw twee een n thee tw th two o so sour urce cess of in info form rmat atio ion. n. Th Thee Mc McGu Gurk rk ef effe fect ct ill illus ustr trat ates es ho how w lis liste tene ners rs proc pr oces esss ph phon onem emes es no nott st stri rict ctly ly in te term rmss of ac acou oust stic ic pr prop oper ertie ties, s, bu butt by ar arti ticu cu-latory gestures as well. Listeners appear to use information about the way a sound is produced from both auditory and visual modes in the process of speech Lachs Pisoni (2004 ) proposed, in fact, visu vi sual al an and dperception. audi au dito tory ry di disp spla lays ysand of sp spee eech ch ar are e or orga gani nize zed d by th the e sa same methat laws la wsboth un-un derlying articulatory events. Cross-modal matching between speech and vision (i.e., face and mouth movements) is achieved, under this view, by comparison to a common source of information about vocal tract activity. activity.

 

164

 

Embodiment and Cognitive Science

The correspondence between sight and sound is also made by infants. If an 18- to 20-week-old infant is simultaneously shown two videos of the same face saying two different syllables, and hears one of the syllables through a loudspeaker placed directly between the two video screens, the child gazes longer at the video face corresponding to the audible signal (Kuhl & Meltzoff, 1987). Within a broad perspective, each of the above findings is consistent with the idea that speech perception is linked with the articulatory gestures of speaking. The motor theory, more specifically, fails, nonetheless, to explain certain aspects of speech perception. First, there seem to be as many motor manifestations of a given consonant as there are acoustic signals (MacNeilage, 1975). Second, listeners are still able to perceive speech when they suffer from speech production (i.e., motor) difficulties. Third, very young infants appear able to recognize speech sounds easily easily,, despite  being poor speech articulators articulators (Jusczyk, 1995). Finally, early studies show that nonhumans can learn to discriminate different speech sounds, an unlikely possibility if human speech production processes are involved in speech perception (Kuhl & Miller, 1975). Beca Be caus usee of th thes esee pr prob oble lems ms,, a re revi vise sed d mo moto torr th theo eory ry em emph phas asiz ized ed ab abst stra ract ct,, as op oppo pose sed d to rea eal, l, ph phon onet etic ic ge gest stur ures es th that at ca can n be de dete tect cted ed th thrrou ough gh sp spec ecia iallized processes (Liberman & Mattingly, 1985). Phonetic gestures, including movements such as rounding of the lips, raising the jaw jaw,, and so on, are invariant motor commands sent from the brain to the vocal tract. Under this view, the conversion from acoustic signals to intended phonetic gestures is ra rapi pidl dly y an and d au auto toma mati tica call lly y pr proc oces esse sed d by a ph phon onet etic ic mo modu dule le.. Th This is re revis vised ed view suggests that the important thing may be the biological capacity for spee sp eech ch,, ev even en in th thee ab abse senc ncee of ex expe peri rien ence ce wi with th a fu func ncti tion onin ing g vo voca call ap appa para ra-tus. tu s. Th This is rel elat atio ions nshi hip p wo would uld no nott ne nece cess ssar aril ily y be di disr srup upte ted d by da dama mage ge to th thee vocal tract. It is interesting, nonetheless, that electrical stimulation of certain ta in sit itees in th thee hum uman an br brai ain n le lead adss to th thee fin final al ges estu turre of sp speeec ech h an and d al alsso affects the perception of phonemes (Ojemann & Mateer,  1979). Thus, elec-

trical stimulation in the same location of the periSylvan area of the left hemisphere produces deficits in phoneme perception and oral gesture repetition. This revision of the motor theory is, unfortunately, difficult to experimentally test, because the concept of gesture remains elusive, at both the underlying and surface levels of production. Some scholars question whether the evidence that phonetic gestures are processed necessarily implies that speech perception requires access to a speaker’s motor system. te m. Fo Forr in inst stan ance ce,, Fo Fowl wler er’s ’s (1986; Fow Fowler ler & Ros Rosenb enblum lum,, 1991) direct-realist theory thatperception, a unit of motor organization (i.e., the gesture) fundamental claims to speech but that phonetic gestures are distalis events, and that speech perception involves recovery of distal events from proximall st ma stim imul ulat atio ion. n. Fi Fina nally lly,, so some me sc scho hola lars rs cl clai aim m th that at sp spee eech ch pe perrce cept ptio ion n is no nott

 

Language and Communication

 

165

special, but shares underlying pattern recognition processes involved in all types of perception (Massaro, (Massaro, 1987). But more recent research research attempts to demonstrate how phonetic primitives are gestural, and not abstract features (Browman & Goldstein, 1995). Articulatory gestures are unified primitives characterizing phonological patt pa tter erns ns,, in ad addi diti tion on to ca capt ptur urin ing g so some meth thin ing g ab abou outt th thee ac acti tivi vity ty of th thee vo voca call tract articulators. More specifically, the lexicon is composed of dynamically specified gestures, where lexical items differ from one another in terms of these gestures and their organization. Once again, gestures are not the movements themselves, but abstract characteristics of the movements. Several dynamical models have been developed that implement phonetic gestures as fundamental couplings among the articulators and the gestural laryngeal characteristics of coarticulation (Kelso, 1995). These modeli mod eling ng ef effor forts ts mos mostt gen genera erally lly illu illustr strate ate how pho phonet netic ic ges gestur tures es can str strucucture acoustic signals directly. There is another body of literature that highlights the embodied nature of speech processing. Many studies now suggest that the speech signal includes not only phonetic and prosodic information, but also nonlinguistic or indexical information. Listeners can readily identify talkers, thei th eirr ph phys ysic ical al an and d em emot otio iona nall st stat ates es,, th thei eirr se sexe xes, s, th thei eirr re regio giona nall di dial alec ects ts,, an and d other qualities associated with speaking rates and the dynamics of articulatio ula tion n (Ny (Nygaa gaard rd,, Som Sommer mers, s, Mit Mitche chell, ll, & Pis Pisoni oni,, 1994). Stu Studie diess dem demons onstra trate te that the mental lexicon may be an episodic or exemplar memory system in which each occurrence of a word leaves a detailed trace in memory (Goldinger,  1998; Nygaard, Sommers, Mitchell, & Pisoni,  1994;  Remez, Fellowes, & Rubin, 1997). This indexical information, including information ti on ab abou outt th thee pr prod oduc ucti tion on of so soun unds ds,, is en enco code ded d in me memo mory ry al alon ong g wi with th th thee phonetic properties of speech. This recent research is consistent with the theory of perceptual symbols (Barsalou, 1999a) in that high-level symbols are represented represented in memory in terms of their perceptual properties and not in terms of their amodal form. Gesture and Speech

People move their bodies when they speak. These movements are not

accidental, but are often tightly linked to the communicative messages spea sp eake kers rs wi wish sh to co conv nvey ey.. Li List sten ener erss ta take ke no noti tice ce of th thes esee ge gest stur ural al mo move veme ment ntss and an d ma may y in infe ferr di diff ffer eren entt th thin ings gs ab abou outt sp spea eake kers rs an and d th thei eirr me mess ssag ages es as a res esult ult of what they see. There are several views of the relationship between speech and gesture (Iverson & Thelen,  1999). One position maintains that speech and gesture are separate communicative systems, which occasionally become linked due to the cognitive demands associated with speech production (Butterworth & Beattie,  1978;  Butterworth & Hadar,  1989;  Hadar,  1989;

 

166

 

Embodiment and Cognitive Science

Hadar, Wenkert-Olenik, Krauss, & Soroker,  1998). Gesture serves to support speech production activities, for example, by compensating when speech is temporarily disrupted (e.g., by coughing) or when speakers are unable to put their thoughts into words. But gestures do not influence underlying speech production processes. A second view holds that there are deep cognitive linkages between speech and gesture, presumably located at the physiological encoding stage (i.e., the stage at which words forms must be accessed from lexicall me ca memo mory ry)) (K (Kra raus uss, s, 1998; Kr Krau auss ss & Ha Hada darr, 1999). Ge Gest stur ures es ar aree es espe peci cial ally ly usef us eful ul,, on th this is vi view ew,, wh when en sp spea eake kers rs ex expe peri rien ence ce di diffi fficu cult lty y re retr trie ievi ving ng wo worrds ds,,  because the production of gesture activates relevant spatio-dynamic features of the concept (i.e., thought) that a speaker had in mind. The link  between speech and gesture is limited, then, to a particular stage of speech production. A third view of speech and gesture assumes that these communicative activities are grounded in common thought processes (Iverson & Thelen, 1999; McNeil, 1992). Speech and gesture have a strong reciprocal reciprocal relationship th ship thro roug ugh h th thee en enti tire re pr proc oces esss of sp spee eech ch pr prod oduc ucti tion on,, ra rang ngin ing g fr from om ph phon onoological encoding up through producing syntax, semantics, and discourse. Even though speech and gesture may communicate different aspects of  people’s thoughts, the tight coupling of these activities suggests that any disr di srup upti tion on in on onee (e (e.g .g., ., ge gest stur ure) e) wi will ll ha have ve ne nega gativ tivee ef effe fect ctss on th thee ot othe herr (e (e.g .g., ., speech). Several kinds of empirical evidence support the third view that speech and gesture are fundamentally grounded in the same underlying cognitive processes. First, when speakers momentarily hesitate, or stutter, their gestures tend to be held motionless until speech continues (Mayberry &  Jacques, 2000). Th Ther eree ar aree tw two o hy hypo poth thes eses es to ex expl plai ain n th this is fa fact ct.. Th Thee le lexi xica call retrieval hypothesis claims that gesture plays an active role in lexical access, particularly for words with spatial content (Butterworth & Hadar,  1989; Krauss, 1998). An al alte tern rnat ativ ivee vi view ew,, th thee in info form rmat atio ion n pa pack ckag agin ing g hy hypo poth thes esis is,, suggests that gesture is involved in the conceptual planning of messages. Specifi Spe cifical cally ly,, ges gestur turee hel helps ps spe speake akers rs to pac packag kagee spa spatial tial inf inform ormati ation on int into o ver ver- balizable entities. Thus, gesture plays a role in speech production production because it plays a role in the process of conceptualization. One study comparing these two hypotheses had   5-year-olds engage in one of two tasks that required comparable lexical access but different information packaging (Alibali, Kita, & Young,  2000). In the explication task, children answered whether two items did or did not have the same

quantity (Piagetian conservation). For the description task, children described how two items looked different. These two tasks elicited similar utterances (e.g., “This one is lower, and this one is higher”), but made different dem ferent demand andss on chi childr ldren’ en’ss con concep ceptua tuall pac packag kaging ing of spa spatia tiall inf inform ormati ation. on. Thus, children in the explication task must consider multiple perceptual

 

Language and Communication

 

167

dime mens nsio ions ns in ju just stif ifyin ying g th thei eirr jud judgm gmen ents ts,, so some meth thin ing g th that at is no nott re requ quir ired ed in di the description task. Not surprisingly surprisingly,, children presented more substitute gestures (e.g., right hand moves from back to top of the glass), and used morre no mo nonr nred edun unda dant nt ge gest stur ures es in th thee ex expl plic icat atio ion n ta task sk th than an in th thee de desc scri ript ptio ion n task. This finding is most consistent with the information packaging hypothesis. may facilitate lexical retrieval, finding is not the Although only placegestures where gestures are involved in speechword production. Gesture appears to play a critical role in the conceptualizing and planning of messages. Neur Ne urop opsy sych chol olog ogic ical al re rese sear arch ch su supp ppor orts ts th thee id idea ea th that at ge gest stur uree an and d sp spee eech ch aree ti ar tigh ghtl tly y li link nked ed.. Fi Firs rst, t, ha hand nd an and d ar arm m mo move veme ment ntss ar aree rep epre rese sent nted ed in br brai ain n sites closely related to those responsible for movement of the vocal tract. Seco Se cond nd,, co comm mmon on br brai ain n me mech chan anis isms ms ex exis istt fo forr la lang ngua uage ge an and d se sequ quen entia tiall mo mo-tor functions, specifically in the lateral periSylvan cortex of the dominant hemispher hemis pheree (Ojem (Ojemann, ann, 1994). St Stimu imula latio tion n of th this is br brai ain n reg egio ion n di disr srup upts ts bo both th oral facial movements and speech production (e.g., naming in reading). These findings raise the possibility that the tight temporal link between speech and gesture may be achieved because of spreading activation from the brain region responsible for speech production to that associated with hand and arm movements, and vice versa. Stud St udie iess al also so sh show ow th that at hi high gh le leve vels ls of EE EEG G ac acti tivi vity ty ar aree fo foun und d in mo moto torr ar ar-eass of th ea thee br brai ain n wh when en pe peop ople le ar aree as aske ked d to rea ead d si sile lent ntly ly,, es espe peci cial ally ly wh when en th thee words are verbs (Pulvermueller, 1999). A PET study showed high activity in the left premotor cortex when people had to retrieve words for tools,  but not words for other conceptual categories (e.g., animals) (Grafton et al., 1997). Verbs Verbs and tool names may exhibit the strongest brain activity in the motor cortex because people encode the motoric functions of these words as part of their semantic representations, which become activated when speakers attempt to retrieve their words from memory. These data suggest that there may be a strong connection between the cerebellum and traditional language areas of the brain such as Broca’s area. A different set of brain studies, employing fMRI, showed that there is some overlap between brain areas activated during language and motor tasks (Loring et al.,  2000). In this experiment, right-handed participants performed several motor movements (e.g., random finger tapping, toe movement, complex finger tapping, copying displayed hand shapes) and a verb generation task. The results showed that there was significant acti ac tiva vati tion on th thro roug ugho hout ut Br Broc oca’ a’ss ar area ea du duri ring ng bo both th th thee lan langu guag agee ta task sk an and d ea each ch motor task, especially those that required hand movements. Most surprisingly,, perhaps, is that evidence also suggests that Broca’s area is activated ingly when people just think about moving their hands (Tanaka & Inui, 2002). Recent functional brain imaging studies, more generally, have reported Broca’s area activation during tasks outside the linguistic domain, includ-

ing motor execution (Iacoboni, Woods, & Mazziotta,  1998), perception of 

 

168

 

Embodiment and Cognitive Science

others’ actions (Decety et al., 1997), and mental simulation (Grafton et al., 1996). Contrary to the popular belief that Broca’s area is only associated with certain aspects of language production, this area of the brain appears to be widely involved with any coherent sequence of body movements (Grezes & Decety Decety,, 2001; Rizzolatti & Arbib, 1996). Arbib (1998) specifically argue arg ue that Bro Broca’s ca’s are areaa funct functions ions includ includee rep repres resentat entational ionalcapac capacities ities rel related ated to action/recognition of coupling oro-facialbetween and brachio-manusual basedalso behavior. behavior Evidence of the tight speech and gesture comes. from research showing that impairment in some motor functions, such as the ability to move sequentially, also hurts language performance. For instance, a study of right-hemisphere damaged patients showed that they did significantly more poorly than left-hemisphere damaged patients in a copying hand movement task (e.g., closed fist, thump sideways on the table, slap palms on a table), in a task demonstrating the use of common objects, and in a task where people had to produce familiar gestures given a ve verb rbal al co comm mman and d (e (e.g .g., ., sh show ow ho how w to wa wave ve go good od-b -bye ye)) (K (Kim imur ura, a, 1973). Th This is pattern of interferences suggests an overlap in cerebral representation of  speaking and certain manual activities. Many researchers now agree that speech and gesture originated from the same neural system (Corballis, 1994). These different lines of evidence from neuropsychology are generally very consistent with the idea that speech and gesture constitute a tightly coupled cognitive system (Iverson & Thelen, 1999). One proposal on the linkage between speech and gesture claims that the origins for this tight coupli cou pling ng occ occur ur in ear early ly dev develo elopme pment nt of mot motor or and han hand d coo coord rdina ination tion(Iv (Iver er-son & Thelen,  1999). A great deal of hand-mouth contact occurs during infancy.. Research with infants 9–15 months old shows that there are sysinfancy tematic relationships between types of hand actions and oral activity in face-t fac e-to-f o-face ace int intera eractio ction n (Fo (Fogel gel & Han Hannan nan,, 1985). Al Alth thou ough gh th thee ma manu nual al an and d vocal voc al sys system tem dev develo elop p ind indepe epende ndentl ntly y, the they y app appear ear to influ influenc encee eac each h oth other er to a si sign gnifi ifica cant nt de degr gree ee,, es espe peci cial ally ly in th thee pr prod oduc ucti tion on of rhy hyth thmic mical al mo move veme ment ntss (Butcher & Goldin-Meadow, ). Infants’speech. first gestures tend to be produced without speech or with 2000 meaningless Later on, when speech and gesture begin to occur, they are not tightly linked in time, with one preceding the other. Finally, speech-gesture synchrony emerges quite dramatically when infants simultaneously combine meaningful words with gesture. Iverson and Thelen (1999) claim that speech and gesture momentarily activate and entrain one another as a coupled oscillator. At first, an infant’s manual activity takes precedence, but through rhythmical activity, and later through gesture, manual behavior gradually entrains the speech prod pr oduc uctio tion n sy syst stem em.. Th Thee in init itial ial ba basi siss to mo move ve ha hand nd an and d mo mout uth h to toge geth ther er ca casscade with a single coupled connected system where the mental thought is manifested as movement. This initial activation increases as infants learn

 

Language and Communication

 

169

to vocally communicate through words and phrases, leading to a tight synchrony of speech and gesture. Eventually, every communicative act, either by speech or gesture, is remember remembered ed as an ensemble, including the prop pr opri rioc ocep eptiv tivee co cons nseq eque uenc ncee of th that at mo move veme ment nt.. Th This is li link nkag agee of sp spee eech ch an and d gesture provides another example of the sensorimotor origins of thought and of the continual importance of embodied action in mental life. A fin final al,, ve very ry di diff ffer eren ent, t, ex exam ampl plee of th thee in inte tera ract ction ion of sp spee eech ch an and d ge gest stur uree is se seen en in aStates stud st udy y(Farnell, of th thee me mean anin ingf gful ul ge gest stur ures es us used ed by th thee Pl Plai ains ns In Indi dian anss of th thee United  1995). Among the Assinboine or Nakota people of  northern Montana, storytellers use hand gestures that constitute a unique sign si gn sy syst stem em,, on onee th that at ca can n be us used ed in inde depe pend nden entt of sp spee eech ch.. Th This is si sign gn sy syst stem em,, called Plains Sign Talk (PST), has for centuries been an intertribal lingua franca for Plains Indians speaking different languages. Although fluent sign talkers are not nearly as common today as they were one hundred year ye arss ag ago, o, va vari riou ouss el elde ders rs wh who o le lear arne ned d PS PST T wh when en yo youn ung, g, de deaf af fa fami mili lies es,, an and d participants in ceremonial rituals have kept PST alive. For the Assinboine, PST is not merely a dramatic enhancement of the speaker’s narrative, because all speech acts are simultaneously voiced and manual, with both the verbal and signed elements being considered aspects of “talking.” As one teacher of PST commented when asked for the spoken equivalent of a signed utterance, “Like I just showed you.” The Ass Assinb inboin oinee phi philos losoph ophy y of “be “being ing-in -in-th -the-w e-worl orld” d” mak makes es bod body y mov moveement me nt fu fund ndam amen enta tall as a wa way y of kn know owin ing. g. Fo Forr th thee As Assi sinb nboi oine ne,, ph phys ysica icall be bein ing g is es esse sent ntia iall to th thee at atta tain inme ment nt of po powe werr. Fo Forr ex exam ampl ple, e, pr pray ayer er is a hi high ghly ly em em- bodied activity where bodily suffering in the hot steam of a sweat lodge and an d fr from om fa fast stin ing g an and d pe peri riod odss of is isol olat atio ion n pr prov ovid ides es a si sign gnifi ifica cant nt pa path thwa way y to seek se ekin ing g an and d gi givi ving ng po powe werr. Su Suff ffer erin ing g is mo most st in inte tens nsee wh when en pa part rtic icip ipan ants ts fa fast st,, dance, and endure the sun’s heat during the sun dance ceremony ceremony.. People only on ly ow own n th thei eirr bo bodi dies es an and d th this is ma make kess bo bodi dily ly sa sacr crifi ifice ce th thee mo most st me mean anin ingf gful ul way to obtain spiritual guidance and personal power. The act of dancing in the sun dance ceremony is itself prayer and not merely action that accompanies spoken prayer. PSTreferring contraststowith American Signand Language (ASL)ininASL significant ways. Signs thoughts, minds, intelligence center on the head, and signs for emotion and feelings are enacted near the heart and chest. ASL signs reflect folk ideas about the spatial locations of the entities and an d po powe wers rs in th thee bo body dy wi with thin in Am Amer eric ican an so soci ciet ety y. Bu Butt PS PSL L si sign gn fo forr co conc ncep epts ts succh as “k su “kn now ow”” an and d “t “thi hink nk”” ar aree ena nact cteed ar arou oun nd th thee he hear art, t, an and d th thee si sign gn fo forr “doubt” is literally “being of two hearts.” To To say that someone has a good mind, a Nakota speaker will move a pinched index finger from the heart away from the body with the finger pointing straight ahead, followed  by the sign for “good.” The movement from the heart is important here,  because the heart is not simply the location for the mind. Of course, hearts have little relevance to minds in most Western societies.

 

170

 

Embodiment and Cognitive Science

Seeing Seei ng is a po powe werf rful ul me meta taph phor or fo forr th thin inki king ng in En Engl glis ish, h, es espe peci cial ally ly in ref ef-eren er ence ce to th thee en endd-pr prod oduc uctt of th thou ough ght. t. To sa say y “I se seee wh what at yo you u me mean an”” refl eflec ects ts thee me th meta taph phor oric ical al id idea ea th that at to pe perrce ceive ive so some meth thin ing g is to co corr rrec ectly tly un unde ders rsta tand nd it. However, in PST, the active process part of the visual metaphor is emphasized rather than its end-product. Thus, talk of thinking refers to the action of looking, specifically looking from the heart. Unlike in English, where there are significant differences differences between thinking and feeling, PST incorporates the Assinboine folk idea that to know something is to “to know in one’s heart,” an idea that reduces the distinction between what one personally experiences and what one might objectively know to be true. Even though the Assinboine primarily view thinking as an activity of  looking from the heart, the head still plays a significant role in PST related to th thee se sens nsor ory y ab abili ilitie tiess of he hear arin ing, g, lis liste teni ning ng,, se seei eing ng,, lo look okin ing, g, ta tast stin ing, g, ea eatin ting, g, drinking, and smelling. The individual signs for these center around the face and head near the respective organs. Beyond this, many names for individuals and tribes are performed near, or touching parts of, the face and an d ot othe herr reg egio ions ns of th thee he head ad.. Fo Forr in inst stan ance ce,, BL BLOO OOD D (a pa part rt of th thee Bl Blac ackf kfee eett confederacy) is referred to in PST by a circular action of the fist on the cheek, SIOUX or ASSINBOINE is described by a flat hand moving across the throat referring to “cut throats,” NEZ PERC E´ are signed as “pierced noses,” CREE by fingers drawn along the side of the face, FLATHEAD FLATHEAD by a flattened head, and KIOWA by a head cut at one side. These head- and face fa ce-o -ori rien ente ted d si sign gnss ar aree hi high ghly ly me meta taph phor oric ical al an and d wo work rk as th thee pr prim imar ary y wa ways ys of identifying strangers. This case study shows how a unique sign system, which can work independent of speech, reveals people’s embodied understandings of the world and human events. The PST system also illustrates the importance of embodied metaphor in how people effectively conceptualize abstract ideas and events. Body Movements and Discourse

Discourse analysts longisargued embodied activity, such as eye gaze, gestures, andhave posture, centralthat in establishing speakers’ goals in conversation. As Goodwin (1981: 125) noted, “Emergent displays thus integrate the bodies of the participants into the production of talk, and are important constitutive features of the conversation.” There are many examples of how speakers and listeners coordinate their body positions to express different different kinds of meaning. For example, one study demonstrated how students use gestures and bodily position vis-`a-vis a-vis one another (and the teacher) in the construction of silence (Leander,  2002). One conversation in this analysis took place in a high school class in American history. The teacher, Sid, at the front of the room, began to review material from

 

Language and Communication

 

171

law, and then asked the class a the previous day’s class on Constitutional law, ques qu esti tion on.. Th Thee st stud uden ents ts we werre se seat ated ed ar arou ound nd th thee roo oom m wi with th fo four ur wh whit itee gi girl rls, s, including Chelle, seated in the back, and four African American students,

including Shameen, Rod, and Trent, grouped in the front of the room. sid:   “Do women have full equal rights?” rod:   “Y “Yes, es, the.” sid:   “And wh-what guaranteed them full and equal rights” shameen:   “19th Amendment?”

“The 19th Amendment gave them the right to vote. What guaranteed them full and equal rights?” chelle:   (from the back of the room) “No we don’t have equal rights” robert:   “She says – we got somebody somebody back here that says they don’t have equal right.” sid:   “They don’t have equal rights?” trent:   “Y “You’re ou’re kidding. There was an amendment that did it.” robert:   “That’s what what she said, ‘No, we don’t’ don’t’ ” (facing forward, forward, pointing toward Chelle with thumb of left l eft hand, over shoulder) chelle:   “W “Wee don’t.” don’t.” (Chelle smiles as Robert looks back back at her) sid:   “What, what are the laws that give women full and equal rights, from the Amendment to the Constitution?” (Kareena enters, sits in desk behind Robert, sitting up on knees) shameen:   “13th, 14th, and 15th?” (Ian gets up from seat, walks to back of room) sid:   “13th says all people are – no – 13th says you can’t own a person, no slavery, as you’ll find out. 14th says – what’s the 14th? (holding hands up, palms pressed together) “Raise your hand.” (Kareena moves from seat in front to seat in back) shameen:   “Ah-hah, I said it though.” sid:   “14th says?” shameen:   “14th says that all people are created equal.” sid:   “No.” sid:

 

The conversation reveals several ways that people reproduce reproduce the social structures through their routine embodied movements. For example, in the midst of Sid and Shameen’s exchange, Robert reproduced Chelle’s speech with “She says – we got somebody back here that says they don’t havee equ hav equal al rig rights hts.” .” By dra drawin wing g att attent ention ion to Che Chelle lle’s ’s bac backch kchann annel el com commen ment, t, Robert brings himself and others into the exchange for several seconds, following which Sid and Shameen continue their questions and answers. Butt Ro Bu Robe bert rt’s ’s rem emar ark k ab abou outt Ch Chel elle le’s ’s co comm mmen ent, t, al alth thou ough gh br brin ingi ging ng it up fr fron ontt for class discussion, also marks an embodied alignment with the “we” of  the cla class, ss, bec becaus ausee he mai mainta ntaine ined d a fac face-f e-forw orwar ard d pos positi ition, on, and poi pointi nting ng ove overr

 

172

 

Embodiment and Cognitive Science

his shoulder with his thumb at Chelle and then looking back at her only  briefly.. The embodied positioning aligns Robert firmly with the normative  briefly ongoing class discussion, yet destabilizes the conversation by physically pointing to a dissenting voice. Inviting dissension while turning away from fr om it is an im impo port rtan antt pa part rt of th thee si sile lenc ncin ing g pr proc oces ess. s. Th Thus us,, on once ce ca can n si sile lenc ncee someone not just by closing off a particular speaker, speaker, but also by creating a silenced position in regard to a dominant one.

During Robert and Chelle s interaction, Kareena s entrance into the classroom and taking a seat illustrates how students negotiate their em bodied positions with regar regard d to how they view their social alignments in context. For instance, temporarily shifting where one sits can be related to shifting topics, the ideologies called up by them, and students’ relations with one another. Kareena had originally entered the classroom and left her books in a seat in front of Trent. Yet when she re-entered the room during discussion, Kareena took a position in the back of the classroom, sitting on her knees at her desk, and observed the interaction. Eighteen seco se cond ndss la late terr, Ka Karree eena na mo move ved d to th thee fr fron ontt of th thee roo oom, m, ret etri riev eved ed he herr bo book oks, s, and took a seat behind Robert. This new seating allowed Kareena to align with the group of students in the back of the class, while avoiding the particular conversational focus of the entire classroom. In fact, Kareena demonstrates her new alignment with Chelle by “repositioning” herself  in the class. Thus, students take positions with their bodies as well as in their speech. People’s bodies are thus a valuable resource in marking their ideological views. Much research in psycholinguistics demonstrates the importance of  “common ground” between speakers and listeners in successful communication (Clark,  1996;  Gibbs,  1999a). There are three primary sources for common ground (Clark, 1996). The first source is “linguistic co-presence,” where the listener takes as common ground all of the conversation up to and including the utterance currently being interpreted. A second source for common ground is “physical co-presence,” where the listener takes as common ground what he or she and the speaker are currently experiencing in terms of their immediate physical environment, including the actions and positioning of their own bodies. The final source of evidence is community membership. This includes information that is universally known in a community and can be represented represented by mental structures such as scripts (Schank & Abelson, 1976) or schemata (Rumelhart, 1980). Moreoverr, it als ove also o cov covers ers mut mutual ually ly kno known wn con conven ventio tions ns gov govern erning ing the pho phonol nology ogy,, syntax, and semantics of the sentence uttered. Normally, mutual knowledge is established by some combination of  physicalorlinguisticco-presenceandmutualknowledgebasedoncommunity ni ty me memb mber ersh ship ip.. Bu Butt ph phys ysic ical al co co-p -prres esen ence ce (i. (i.e. e.,, si sigh ght, t, so soun und, d, to touc uch) h) alo alone ne provides multiple resources for conversational grounding. One study of 

 

  173 Language and Communication collab labora orativ tivee pr probl oblemem-sol solvin ving g (e. (e.g., g., re repai pairin ring g a bic bicycl ycle) e) sho showed wed tha thatt phy physscol ical co-presence has several independent sources of visual information: (1) participants’ hands and faces, (2) parti participan cipants’ ts’ beha behavior vior and action actions, s, (3) focused task objects, and (4) the work environment in context (Kraut, Fussell, & Siegel,  2003). When people work side by side, they have all four sources of visual information available. Thus, participants can monitor each other’s facial expressions and body orientations vis-`a-vis a-vis task objects (e.g., the bicycle, its parts, tools for repairing it). Facial expressions and visible actions directed toward the task provide evidence of whether someone understands an instruction. In fact, research shows that when people work side by side, they perform the task of fixing a bicycle faster

and using fewer utterances than when they are remotely linked by video and audio, or audio alone (Kraut et al., 2003). There are major differences between visual co-presen co-presence, ce, or just sharing a joint view, and full physical co-presence, in which spatial relations between people and task objects are monitored. For example, seeing another person’s upper body allows a remote partner to observe that he or she is pointing at something. Yet having a spatially consistent view of both participants and task objects is needed to understand the precise target of  a pointing gesture, as well as the targets of the partner’s eye gaze. Several systems have been developed that provide participants with multiple cues as to, for example, where a person at a meeting is looking at any moment in time, or which object a hand gesture is aimed at (Luff et al., 2003; Stiefehagen, Yang, & Waibel, 2002). Computer interface designers have attempted to build embodied interface agents to provide a higher bandwidth of communication than would otherwise be possible with a less embodied system. Unfortunately, many new interface agents do not provide much bodily information other than ornamental ornam ental things such as red redundan undantt point pointing ing gestures, gestures, a few facial expres pr essi sion ons, s, a co cock cked ed he head ad,, an and d an ex exte tern rnal al wa warrdr drob obe. e. Bu Butt on onee mo mode dell of em em- bodied conversation aims to exploit the afford affordances ances of the body to facilitate meaningful dialogue between interface agents and human users (Cassell et al.,  2001). REA is an embodied, multimodal, real-time conversational agen ag entt th that at ac acts ts as a rea eall es esta tate te sa sale lesp sper erso son n wh whil ilee sh show owin ing g us user erss ar arou ound nd vi virrtual houses. REA has a fully articulated graphical body, body, can sense the user passively through cameras and audio input, and is capable of speech with intonation, facial display, display, hand and eye movements, and gestural output. Thee sy Th syst stem em co cons nsis ists ts of a la larrge pr proj ojec ecti tion on sc scrree een n in wh whic ich h RE REA A is di disp spla laye yed, d, and an d wh whic ich h th thee us user er st stan ands ds in fr fron ontt of of.. Var ario ious us mic micro roph phon ones es ca capt ptur uree sp spee eech ch input. Two cameras mounted on the top of the projector screen track the user’s hand and head positions in space. One computer runs the graphics and conversation of REA, while a second manages the speech recognition and image processing.

 

174

 

Embodiment and Cognitive Science

The following provides an actual conversational interaction between a user, Tim, and REA (Cassell et al.,  2001: 60): Tim approaches REA. REA notices and looks toward Tim and smiles. Tim says, “Hello.” REA responds, “Hello, how can I help you?” with a hand wave. Tim says, “I’m looking to buy a place near MIT.” REA RE A sa says ys,, “I ha havea vea ho hous usee . . . ” wi witha tha be beat at ge gest stur uree toemp toempha hasi size zeth thee ne new w in info form rmat atio ion n “house.” Tim interrupts by beginning to gesture. REA finishes her current utterance by saying “in Cambridge” and then gives up her turn. Tim refines his house request. REA finishes the house description and then continues.

Although REA is somewhat awkward in her conversational interac-

tions, she converses as well as she can because of the system’s ability to track the user’s hand and head movements, infer the specific speech acts the user conveys, and use voice information to recognize when the other speaker’s turn begins and ends. A future goal of this project is to allow REA RE A to en entr trai ain, n, or in incr crea easi sing ngly ly ad adap apt, t, he herr be beha havi vior orss in sy sync nchr hron ony y wi with th th thee user. Word Meaning

Words are traditionally defined in terms of semantic features that are usually abstract and are thought to reflect different conceptual relations. Yet in recent years, scholars have argued that some aspects of word meanings arise ari se fr from, om, and ar aree men mental tally ly re repr prese esente nted d in ter terms ms of, per percep ceptua tual/e l/embo mbodie died d experience. Consider the word “stand” in the following sentences: “Please stand at attention.” “He wouldn’t stand for such treatment.” “The clock stands on the mantle.” “The law still stands.” “He stands six foot five.” “The part stands for the whole.” “She had a one-night stand with a stranger.” stranger.” These sentences represent represent just a few of the many senses of “stand” that are common in everyday speech and writing. Some of these senses refer to the physical act of standing (e.g., “Please stand at attention,” “The clock stands on the mantle,” “He stands six foot five”), whereas others have nonphysical, perhaps figurative, interpretations (e.g., “We “We stood accused of the crime,” “The part stands for the whole,” “He wouldn’t stand

 

Language and Communication

 

175

the physical meanings of  for such treatment”). What are thewhat principles polysemous words? For instance, relatesthat the relate different and nonphysical senses of “stand” in the examples noted above? Some So me li ling ngui uist stss in rec ecen entt ye year arss ha have ve ar argu gued ed th that at ma many ny po poly lyse se-mous mo us wo word rdss re resi sist st be bein ing g de defin fined ed by a ge gene nera ral, l, ab abst stra ract ct,, co corre se sens nsee (Brugman & Lakoff, 1988; Fillmore, 1982; Geeraerts, 1993; Sweet  Sweetser ser,, 1986). Cognitive linguists have suggested that the meanings of polysemous words can be characterized by metaphor, metonymy, metonymy, and differen differentt kinds ki nds of im imag agee sc sche hema mass (L (Lak akof off, f, 1987; Johnson, 1987; Sweetser, 1990). Und ndeer th this is view, the lexical organization of polysemous words is not a repository of random, idiosyncratic information, but is structured by general cognitive principles that are systematic and recurrent throughout the lexicon. Most important, perhaps, is the claim that these principles arise from our phenom phe nomeno enolog logica ical, l, emb embodi odied ed exp experi erienc ence. e. One pos possib sibilit ility y is tha thatt bod bodily ily experience partly motivates people’s intuitions as to why different senses of  “stand” have the meanings they do. Gibbs et al. (1994) attempted to experimentally show that the different senses of the polysemous word “stand” are motivated by different image schemas that arise from our bodily experience of standing. Their general aim was to empirically demonstrate that the meanings of the polysemous

word “stand” are not arbitrary for native speakers, but are motivated by people’s recurring bodily experiences in the real world. As a first step toward understanding how image schemas partly motivate the meanings of the polysemous word “stand,” a preliminary experiment sought to determine which image schemas best reflect people’s recurring bodilyaexperiences of standing. A group participants were guided through brief set of bodily exercises to get of them to consciously thin th ink k ab abou outt th thei eirr ow own n ph phys ysic ical al ex expe peri rien ence ce of st stan andi ding ng.. Fo Forr in inst stan ance ce,, pa part rtic ic-ipan ip ants ts we werre as aske ked d to st stan and d up up,, to mo move ve ar arou ound nd,, be bend nd ov over er,, to cr crun unch ch,, an and d to stretch out on their tiptoes. Having people actually engage in these bodily experiences facilitates participants’ intuitive understanding of how their experience of standing related to many different possible image schemas. After this brief standing exercise, participants then read brief descriptions of  12  12 different image schemas that might possibly have some relationship to the experience of physical standing (e.g., VERTICALITY, BALANCE, RESIST RESIS TANCE, ENABL ENABLEMEN EMENT T, CENTE CENTER-PER R-PERIPHER IPHERY Y, LINKA LINKAGE). GE). Finall na lly y, th thee pa part rtic icip ipan ants ts ra rate ted d th thee de degr gree ee of re rela late tedn dnes esss of ea each ch ima image ge sc sche hema ma to their own embodied experiences of standing. The results of this first study showed that five image schemas are primary to people’s bodily experiences of standing (i.e., BALANCE, VERTICALITY, CENTERPERIPHERY, RESISTANCE, and LINKAGE). A second experiment investigated people’s judgments of similarity for different senses of “stand.” The participants sorted  35  different senses of  “stand” into five groups based on their similarity of meaning. An analysis

 

176

 

Embodiment and Cognitive Science

of these groups revealed that participants did not categorize physical senses of “stand” separately from the nonphysical or figurative senses. For example, the physical idea of standing in “to stand at attention” was often grouped with the metaphorical senses of “stand” in “let the issue stand” and “to stand the test of time.” The third experiment in this series examined the relationship between thee fiv th fivee im imag agee sc sche hema mass fo forr th thee ph phys ysic ical al ex expe peri rien ence ce of st stan andi ding ng an and d th thee va varrious senses of “stand” studied in Experiment 2 . Once again, participants were first asked to stand up and focus on different aspects of their bodily experience of standing. As they did this, the participants were presented withverbaldescriptionsofthefiveimageschemasBALANCE,VERTICALITY,, CENTE ITY CENTER-PER R-PERIPHER IPHERY Y, RESIS RESIST TANCE, and LINKAG LINKAGE. E. After Afterwar wards, ds, the participants were given a list of  32  32 senses of “stand” and asked to rate the degree of relatedness between each sense and the five image schemas. The rating data from this third study allowed Gibbs et al. (1994) to construct an image schema profile for each of the  32 uses of “stand.” Several interesting similarities emerged in the image schema profiles for some of  the 32 se sens nses es of “s “sta tand nd.” .” Fo Forr ex exam ampl ple, e, “i “itt st stan ands ds to rea easo son” n” an and d “a “ass th thee ma mattter now stands” both have the same image schema profile (in their rankorde or derr of im impo port rtan ance ce)) of LI LINK NKAG AGE E – BA BALA LANC NCE E – CE CENT NTER ER/P /PER ERIP IPHE HER RY – RESISTANCE – VERTICALITY. The expressions “don’t stand for such treatment” and “to stand against great odds” are both characterized  by the image schema profile RESIST RESISTANCE ANCE – CENTER/PERIPHERY – LINKAGE – BALANCE – VERTICALITY.

The pr The prim imar ary y go goal al of th this is st stud udy y, th thou ough gh,, wa wass to as asse sess ss wh whet ethe herr th thee se sens nses es of “s “sta tand nd”” se seen en as be bein ing g si simi mila larr in me mean anin ing g in th thee se seco cond nd ex expe peri rime ment nt we werre reliably predictable from the image schema obtained in the third experiment. Statistical analyses showed thatprofiles knowing the image schema profiles for different senses of “stand” allowed us to predict   79% of all the groupings of “stand” in Experiment 2. These data provide very strong support for the hypothesis that people’s understandings of the meanings of “stand” are partly motivated by image schemas that arise from their  bodily experiences of standing. standing. A fo four urth th st stud udy y sh show owed ed th that at pa part rtic icip ipan ants ts’’ so sort rtin ings gs of “s “sta tand nd”” in dif diffe ferren entt groups cannot be explained simply in terms of their understanding of  the contexts in which these words appeared. Thus, people did not sort phrases, such as “don’t stand for such treatment” and “to stand against great odds,” because these phrases refer to the same types of situations. Instead, it appears that people’s similarity judgments are best attributed to their tacit understanding of how different patterns of image schemas motivate different uses of the polysemous word “stand.” These studies demonstrate that people make sense of different uses of  “stand” because of their tacit understanding of several image schemas that arise partly from the ordinary bodily experience of standing. These

 

Language and Communication

 

177

image schemas, the most important of which are BALANCE, VERTICALITY, CENTER-PERIPHERY, RESISTANCE and LINKAGE, not only produce the grounding for many physical senses of “stand” (e.g., “he stands six- foot five,” “stand in the way,” and “stand at attention”), but also underlie people’s understanding of complex, metaphorical uses (e.g., “the part stands for the whole,” “as the matter now stands,” and “the engine can’ ca n’tt st stan and d th thee co cons nsta tant nt we wear ar”) ”).. Pe Peop ople le pe perrce ceiv ivee di diff ffer eren entt se sens nses es of “s “sta tand nd”” as similar in meaning partly on the basis of the underlying image schema profile for each use of the word in context. Similar work shows the em bodied basis of people’s understandings of the various meanings of the preposition “on” (Beitel, Gibbs, & Sanders, 2000). My argument about the meanings of several polysemous words does nott im no impl ply y th that at pe peop ople le ju judg dgee si simi mila lari rity ty of me mean anin ing g be betw twee een n tw two o se sens nses es of a word only on the basis of image schemas. Many aspects of word meaning that have little to do directly with image schemas certainly play some role in people’s understanding of word meaning and their judgments of similarity of meaning for different senses of a polysemous word. At the same time, this experimental research does not imply that people automatically access some specific pattern of image schemas each time they encounter a particular use of a word. The main conclusion, though, from the experimental work on “stand” and “on” is that people tacitly recognize some connection between these schematic bodily experiences and different aspects of lingu linguistic istic meaning, including meanings meanings that are highly abstract abstract and/or metaphorical. Theworkonimageschemasandwordmeaningprovidessupportforthe ideaa tha ide thatt som somee asp aspect ectss of mea meanin ning g ar aree gr groun ounded ded in con contem tempor porary ary spe speake akers’ rs’ embodied experience, which they can tacitly recognize under the right

experimental conditions. But other psycholinguistic studies suggest that people may automatically infer perceptual/embodied characteristics of  wordmeaningwhilereadingorlistening.Someresearchhasfoundpriming  between words words that refer refer to objects objects with similar perceptual perceptual characteristics characteristics (Schreuder, Flores d’Arcais, & Glazenborg, 1984). For instance, the words “orange” and “ball” are perceptually similar, similar, because the objects have the same shape. The words “skipping rope” and “ball,” on the other hand, are conceptually related, because the objects are both toys. Other word pairs are both perceptually and conceptually similar similar,, such as “butter” and “ball.” Finally, some word pairs are both perceptually and conceptually unrelated, such as “hoe” and “ball.” Several studies, employing tasks where participants either simply pronoun no unce ce or gi give ve le lexi xica call de deci cisi sion onss on ta targ rget et it item emss (e (e.g .g., ., “b “bal all” l”), ), de demo mons nstr trat ated ed significant perceptual and conceptual priming effects. These priming effects were even stronger when the word pairs were both perceptually and conceptually related (e.g., “butter” and “ball”) (Schreuder et al.,  1987). More recent studies, however, suggest that perceptual priming is found

 

178

 

Embodiment and Cognitive Science

only when participants are first alerted to an object’s perceptual characteristics (e.g., does the word “ball” refer to an oblong object?) (Pecher, Zoelen Zoe lenber berg, g, & Raa Raajam jamado adon, n, 1998). Th Thes esee st stud udie iess on pe perrce cept ptua uall pr prim imin ing g in word identification tasks generally suggest that the perceptual characteristi is tics cs of ob obje ject cts, s, in incl clud udin ing g th thos osee th that at ob obje ject ctss af affo ford rd,, mi migh ghtt be au auto toma mati tica cally lly acti ac tiva vate ted d in me memo mory ry,, bu butt pe perh rhap apss no nott fo forr al alll ca case ses, s, wh when en co conc ncrret etee no noun unss ar aree read. One set of studies examined the idea that perceptual symbols are used in on-line language comprehension (Stanfield & Zwaan, 2001). Perceptual symb sy mbol olss ar aree th thee res esidu idues es of pe perc rcep eptu tual al ex expe peri rien ence ce,, st stor ored ed as pa patt tter erns ns of ac ac-tivation, in the brain. Unlike amodal representations, representations, perceptual symbols  bear an analogue relationship with their real-world refer references. ences. Participants in these studies were presented with sentences such as “He hammered the nail into the wall” and “He hammered the nail into the floor.” After reading a specific sentence, participants saw a picture depicting the object mentioned in the sentence (e.g., the nail). This picture presented the object either in a horizontal or in a vertical orientation, thus creating a match or mismatch with the orientation of the object implied by the sentence. In fact fa ct,, re resp spon onse sess we were re si sign gnifi ifica cant ntly ly qu quic icke kerr wh when en th ther eree wa wass a ma matc tch h be betw twee een n thee im th impl plie ied d or orie ient ntat atio ion n an and d th thee pi pict ctur uree th than an wh when en th thes esee we were re mi mism smat atch ched ed.. Thes Th esee re resu sult ltss su supp ppor ortt th thee id idea ea th that at pe peop ople le ac activ tivat atee an and d ma mani nipu pula late te pe perrce cepptual symbols when understanding the context-specific meanings of words during utterance interpretation. A follow-up set of studies extended the previous findings to the representation of an object’s shape in sentence comprehension (Zwaan, Stanfield, & Yaley,  2002). For example, participants saw the sentence “The ranger saw the eagle in the sky” followed by a picture of an eagle with either folded or outstretched wings. Not surprisingly, people gave faster recognition judgments to the eagle when the picture matched the shape implied by the sentence. A second study showed the same findings using a naming task that did not involve people matching the picture with the previous sentence. Once more, the results support the hypothesis that

people peop le ac activ tivat atee pe perc rcep eptu tual al sy symb mbol olss of ref efer eren ents ts fo forr wo word rdss du duri ring ng la lang ngua uage ge interpretation. Research shows that people translate words and sentences into a flow of events comparable to normal perceptual experiences. For instance, resear se arch cher erss ar argu guee th that at th thee wo worrds in wh whic ich h ev even ents ts ar aree re repr pres esen ente ted d in la lang ngua uage ge reflect their chronological order (i.e., the iconicity assumption; Dowty, 1986). Un Unde ders rsta tand ndin ing g th thee wo word rd “c “cro ross ssin ing” g” in “c “crros ossi sing ng th thee ri rive ver” r” de dema mand ndss that th at rea eade ders rs tr trac ack k th thee sp spat atia iall ev eval alua uati tion on ov over er ti time me of so some me ta tarrge get, t, or tr traj ajec ec-tor, which starts at one side and ends up at the other. Because the trajector cannot be represented at both sides at one time, readers must create a dynamic representation that captures the temporal perceptual character of  the phrase’s meaning.

 

Language and Communication

 

179

Experiments show that people create dynamic, temporal representations as part of their understanding of word meaning. Zwaan (1996) demo de mons nstr trat ated ed th that at ti time me sh shif ifts ts in na narr rrat ativ ives es in incr crea ease se pr proc oces essi sing ng tim time. e. Th Thus us,, peoplee rea peopl reading ding the phras phrasee “An hour later later”” after some event take longe longerr to process this phrase than when a minor time shift is implied, such as with the phrase “A moment later.” later.” These findings are consistent with the “iconicity assumption,” not only in that events are assumed to occur in chronological order, but also occur contiguously. Other data indicates that continuing actions in sentences are more activated in memory than are events not continuing. Thus, people are faster to say that “walked” is a word after reading the pair of sentences “Teresa “Teresa walked onto the stage. A moment later she collapsed” than they did having first read the sentences “Ter “T eres esaa wa walk lked ed on onto to th thee st stag age. e. An ho hour ur la late terr sh shee co coll llap apse sed. d.”” Re Rela late ted d st stud ud-ies show that embodied actions that continue remain more activated than events that have been discontinued. Thus, people were slower to judge that “kicking” was a word after reading “Steve stopped kicking the soccer  ball” after reading “Steve was kicking the soccer ball” (Carreiras etsed al., 1997)than . Th Thes ese e fin findi ding ngs, s, ag agai ain, n, sh show ow ho how w pe peop ople le’s ’s cons co nstr trua uall of even ev ents ts,, ba base d on their embodied understandings, play an important role in the processing and representation in memory of words in linguistic expressions. Very recent work indicates that schematic images are recruited during immediate processing of verbs (Richardson et al.,  2003). A norming study first showed that participants were generally consistent in pairing four different pictures that reflect various schematic images (e.g., a circle, a square, an arrow looking up, down, left, or right) with different concrete and abs abstra tract ct ver verbs bs (e. (e.g., g., “pu “push, sh,”” “lif “lift,” t,” “ar “argue gue,” ,” “r “resp espect ect”). ”). A sec second ond nor normming study had participants create their own schematic images for verbs in a simple computer-based drawing environment. Once more, there was good consistency in the spatial shapes people thought best described the meanin mean ings gs of th thee di diff ffer eren entt ve verb rbs. s. Th Thes esee fin findi ding ngss sh show ow th that at pe peop ople le ha have ve re reggularr int ula intuit uition ionss abo about ut the spa spatia tiall re repr prese esenta ntatio tions ns und underl erlyin ying g dif differ ferent ent ver verbs, bs, even abstract ones. Additional studies showed that verbs activate underlying spatial representations during online language comprehension. For instance, in one study, participants heard a sentence (e.g., “The girl hopes for a pony”) with two pictures presented sequentially in the center of the computer

screen. The two pictures reflected different different images of the main and object nouns in either vertical or horizontal position. Afterwards, participants were tested on their memory for the pictures in a speeded recognition task. As predicted, people recognized the pictures faster when they were oriented along the same axis of the associated verb. Verb comprehension appears to activate schematic images that act as scaffolds for visual memory of the pictures. The pictures that were encoded as oriented similarly to the verbs’ meanings were identified faster during the memory tests.

 

180

 

Embodiment and Cognitive Science

These results suggest that verb meanings are actively linked with perceptual mechanisms that influence online comprehension and memory. One possib pos sibilit ility y is tha thatt dif differ ferent ent per percep ceptua tuall and mot motor or exp experi erienc ences es bec become ome ass assoociat ci ated ed wi with th ve verb rbs, s, wh whic ich h ar aree ac acti tiva vate ted d as pa part rt of pe peop ople le’s ’s pe perc rcep eptu tual al-m -mot otor or simulations of the sentence during understanding (Barsalou, 2001). Differ Dif ferent ent exp experi erimen ments ts dem demons onstra trate te tha thatt emb embodi odied ed act action ion infl influen uences ces immediate med iate sym symbol bolic, ic, or sem semant antic, ic, jud judgme gments nts for sim simple ple lin lingui guisti sticc sta statem tement ents. s. In these studies, participants were first asked to make hand shapes corresponding to verbal descriptions such as “pinch” and “clench” (Klatzky et al., 1989). Following this, the participants made speeded judgments on thee sen th enssib ible len nes esss of phr hraase sess suc uch h as “a “aim im a da darrt” (s (sen ensi sibl ble) e) or “cl clo ose a na nail il”” (not (n ot se sens nsib ible le). ). Em Embo bodi died ed ac acti tion on rel elev evan antt to th thee ph phra rase sess fa faci cilit litat ated ed pe peop ople le’s ’s speeded verifications of these phrases. For instance, the hand shape for “pinch” speeded the sensibleness judgments for “throw a dart” but not “throw a punch.” Interestingly, when participants were asked to make verbal responses (but not hand shapes) to the nonverbal prime (e.g., the word “pinch” when shown the nonverbal signal for pinch), the priming ef effec fectt was elimin eli minate d. Sen Sensib siblen leness ess jud judgme gments nts, , like onl online inemotoric compr com prehe ehensi nsion, on, require a type of ated. mental simulation using an embodied, medium. Image Schemas and Utterance Interpretation

Image schemas are cognitive representations that arise from people’s recurrin cur ring g emb embodi odied ed exp experi erienc ences es (se (seee Cha Chapte pterr 4). Th Thee em empi piri rica call fin findi ding ngss fr from om cognitive psychology (see Chapter   5)  correspond to some of the inference en cess pe peop ople le ap appe pear ar to dr draw aw wh when en un unde ders rsta tand ndin ing g di diff ffer eren entt se sent nten ence cess th that at metaphorically refer to momentum. Consider the following utterances. “I was bowled over by that idea.” “Wee have too much momentum to withdraw from the election race.” “W “I got carried away by Iewas doing.” “We “W e bet ette terr qu quit it argu ar guin ing gwhat bef efor ore it pic icks ks up to too o mu mucch mo mome men ntu tum m an and d we can’t stop.” “Once he gets rolling, you’ll never be able to stop him talking.” These utt These uttera erance ncess re reflec flectt how the ima image ge sch schema ema for MOM MOMENT ENTUM UM all allows ows disc di scus ussi sion on of ve very ry ab abst stra ract ct do doma main inss of co cogn gnit itio ion, n, su such ch as po poli litic tical al su supp ppor ort, t, control, arguments, and talking in terms of physical objects moving with mome mo ment ntum um.. We ma may y be ab able le to pr pred edic ictt im impo port rtan antt as aspe pect ctss of th thee in infe ferren ence cess people draw when understanding these sentences, given what is known about representational momentum from cognitive psychological research

(see Chapter 5). One of the findings from representational momentum research is that people behave as if an apparently moving object continues to move even after encountering an obstacle. Essentially, the moving object appears to carry the obstacle along with it rather than deflecting off it or stopping.

 

Language and Communication

 

181

When understanding the sentence “I was bowled over by that idea,” people should infer that the idea was important and that the speaker was conv co nvin ince ced d by th thee id idea ea.. Th This is fo foll llow owss fr from om on onee of th thee ch char arac acte teri rist stic icss of mo movving objects – the bigger objects are, the more momentum they have when moving. Accordingly, a big object encountering an obstacle should result in that obstacle being carried along with the big object. Applying the conceptual metaphor IDEAS ARE OBJECTS, one should infer when reading or hearing “I was bowled over by that idea” that the person encountering an important (big) idea would be convinced (carried along) by that idea. Anot An othe herr res esul ultt fr from om th thee re rese sear arch ch on re repr pres esen enta tati tion onal al mo mome ment ntum um is th that at obje ob ject ctss mo movi ving ng wi with th mo mome ment ntum um ar aree pe perrce ceiv ived ed as be bein ing g un unab able le to st stop op im im-mediately. Even if a force is applied to stop the object, it will continue for some distance before coming to rest. One might infer from this situation that if reaching a particular destination is desired, then the more momentum tu m an obj bjeect ha hass th thee bet ette terr ar aree th thee ch chan ancces for th thee ob obje ject ct to rea each ch th thee de dest stiinati na tion on.. We ca can n ap appl ply y th this is kn know owle ledg dge, e, alo along ng wi with th th thee co conc ncep eptu tual al me meta taph phor or ACCO AC COMP MPLI LISH SHME MENT NTS S AR ARE E MO MOVE VEME MENT NTS, S, to th thee se sent nten ence ce “W “Wee ha have ve to too o mu much ch mo mome um to wi with thdr draw aw fr from theechance th elec el ecti tion on race ra ce””momentum) to in infe ferr th that at th the canndidate in ment thentum election race has a om good (much toe ca win the election, and therefore should not attempt to withdraw (stop). A related finding from representational representational momentum research is that an object with unchecked momentum will move a long distance, perhaps even overshooting overshooting some desir desired ed destination. destination. This situation gives rise to the inferences drawn when comprehending “I got carried away by what I was doing.” Specifically, a person doing something without monitoring the time involved or the resources devoted to doing it (an object moving with unchecked momentum) might devote too much time or too many resources resour ces to the task (overshoot the desired destination). A di diff ffer eren entt as aspe pect ct of th thee rep epre rese sent ntat atio iona nall mo mome ment ntum um re rese sear arch ch co conc ncer erns ns thee ap th appa parren entt sp spee eed d an and d ac acce cele lera ratio tion n of th thee mo movi ving ng ob obje ject ct.. Th This is fa fact ctor or af affe fect ctss thee pe th perc rcei eive ved d am amou ount nt of mo mome ment ntum um th that at an ob obje ject ct wi will ll ha have ve.. Ap Appl plyi ying ng th this is finding to the sentence “Once he gets rolling, you’ll never get him to stop talk ta lkin ing” g” le lead adss to th thee in infe ferren ence ce th that at in inte terr rrup upti ting ng (s (sto topp ppin ing) g) th thee pe pers rson on ea earl rly y in the conve conversati rsation on (when speed is low) will be easie easierr than interruptin interrupting g him hi m la late terr (w (whe hen n sp spee eed d is hi high gh). ). Th This is res esul ultt al also so ap appl plie iess to th thee se sent nten ence ce “Y “You ou had ha d be bett tteer sto top p th thee ar argu gume ment nt now bef efor oree it pi pick ckss up to too o mu much ch mo mome men ntu tum m and we can’t stop it.” The inference here might be that arguments start off  fairly innocuously (with low speed), but as they progress, things may be said sa id th that at ar aree un unrret etra ract ctab able le (h (hig igh h sp spee eed) d).. Fo Forr bo both th se sent nten ence ces, s, we un unde ders rsta tand nd that the talking or argument should be stopped as early as possible. There is no experimental evidence to support these speculative ideas. But this discussion illustrates important possibilities about how image schemas may underlie various aspects of the rather subtle inference patterns associated with the meanings of utterances.

 

182

 

Embodiment and Cognitive Science

Embodied Metaphor in Figurative Language Interpretation

Many experiments, however, illustrate how image schemas serve as the source domains in different metaphorical concepts and partly account for the rich meanings of a wide range of linguistic phenomena, including idioms,, conve idioms convention ntional al expr expressio essions, ns, and nove novell metap metaphors hors (Gibb (Gibbs, s, 2002).Conside si derr th thee id idio iom m “s “spi pill ll th thee be bean ans. s.”” Try to fo form rm a me ment ntal al im imag agee fo forr th this is ph phra rase se and then ask yourself the following questions. Where are the beans before they are spilled? How big is the container? Are the beans cooked or uncooked? Is the spilling accidental or intentional? Where are the beans oncce th on they ey’v ’vee be been en spi pill lled ed?? Are th thee bea eans ns in a nic ice, e, ne neaat pi pile le?? Wh Wher eree ar aree th thee  beans supposed to be? After the beans are spilled, are they easy to retrieve? Most Mo st pe peop ople le ha have ve de defin finite ite re resp spon onse sess to th thes esee qu ques esti tion onss ab abou outt th thei eirr me menntal images for idioms (Gibbs & O’Brien, 1990). They generally say that the  beans were were in some pot that is about the size of a person’s head, head, the beans beans are uncooked, the spilling of the beans is accidental, and the spilled beans are all over a floor and are difficult to retrieve. This consistency in people’s intuitions about their mental images is quite puzzling if one assumes that the meanings of idioms are arbitrarily determined. People’s descriptions about their mental images for idioms some the em bodied metaphorical knowledge that motivates thereveal meanings ofof idiomatic phra ph rase ses. s. On Onee st stud udy y ex exam amin ined ed pe peop ople le’s ’s me ment ntal al ima image gess fo forr gr grou oups ps of id idio ioms ms with similar figurative meanings, such as about revelation (e.g., “spill the  beans,” “let the cat out of the bag,” “blow the lid off”), anger (e.g., “blow your stack,” “hit the ceiling,” “flip your lid”), insanity (e.g., “go off your rocker,” “lose your marbles,” “bounce off the walls”), secretiveness (e.g., “keep it under your hat,” “button your lips,” “keep in the dark”), and exerti er ting ng co cont ntrrol (e (e.g .g., ., “c “cra rack ck th thee wh whip ip,” ,” “l “lay ay do down wn th thee la law w,” “c “cal alll th thee sh shot ots” s”)) (Gibbs & O’Brien, 1990). Participants were asked to describe their mental images for these idioms and to answer questions about the causes, intentionality,, and manner of actions in their mental images for these phrases. tionality Overall, participants’ descriptions of their mental images were remarkably consistent for different idioms with similar figurative meanings. The general schemas underlying people’s images were not simply repr representaesentativee of th tiv thee id idio ioms ms’’ fig figur urat ative ive me mean anin ings gs,, bu butt ca capt ptur ured ed mo more re sp spec ecifi ificc as aspe pect ctss of the kinesthetic events with the images. For example, the anger idioms such as “flip your lid” and “hit the ceiling” all refer to the concept of “getting angry,” angry,” but participants specifically imagined for these phrases some force causing a container to release pressure in a violent manner. There is nothing in the surface forms of these different idioms to tightly constrain the images participants reported. After all, lids can be flipped and ceilings can be hit in a wide variety of ways, caused by many different circumstances. But the participants’ protocols in this study revealed little variation in the general events that took place in their images for idioms with similar meanings.  

Language and Communication

 

183

Participants’ responses to the questions about the causes and consequences of the actions described in their images were also highly consistent. Consider the most frequent responses to the probe questions for the anger idioms (e.g., “blow your stack,” “flip your lid,” “hit the ceiling”). When imagining anger idioms, people reported that pressure (i.e., stress or frustration) causes the action, that one has little control over the pressure once it builds, that its violent release is done unintentionally (e.g., the  blowing of the stack) and that once the release has taken place (i.e., once the ceiling has been hit, the lid flipped, the stack blown), it is difficult to reverse the action. Simila Sim ilarr find finding ingss hav havee bee been n fou found nd for men mental tal ima imager gery y of pr prove overbs rbs (Gi (Gibbs bbs,, Strom, Str om, & Spi Spivey vey-Kn -Knowl owlton ton,, 1997). Fo Forr in inst stan ance ce,, pe peop ople le im imag agin inee th thee ph phra rase se “A rolling stone gathers no moss” in particular ways, in part because of  the metaphorical idea that LIFE IS A JOURNEY, JOURNEY, which is grounded in the image schema of SOURCE-PATH-GOAL. Thus, people are limited in the kindsofimagestheycreateforidiomsandproverbsbecauseofveryspecific embodied knowledge that helps structure their metaphorical understanding in g of va vari riou ouss co conc ncep epts ts (G (Gib ibbs bs & O’ O’Br Brie ien, n, 1990). Fo Forr ex exam ampl ple, e, pe peop ople le’s ’s im im-ages ag es fo forr th thee an ange gerr id idio ioms ms ar aree ba base sed d on fo folk lk co conc ncep epti tion onss of ce cert rtai ain n ph phys ysic ical al events. That is, people use their embodied knowledge about the behavior of he heat ated ed flu fluid id in co cont ntai aine ners rs (e (e.g .g., ., th thee bo bodi dies es as co cont ntai aine ners rs an and d bo bodi dily ly flu fluid idss within them) and map this knowledge onto the target domain of anger to help he lp th them em co conc ncep eptu tual aliz izee in mo morre co conc ncrret etee te term rmss wh what at is un unde ders rsto tood od ab abou outt thee co th conc ncep eptt of an ange gerr. Var ario ious us sp spec ecifi ificc en enta tailm ilmen ents ts res esult ult fr from om th thes esee ge gene nera rall metaphorical mappings, ones that provide specific insight into people’s consistent responses about the causes, intentionality, manner, and consequen qu ence cess of th thee ac acti tivit vitie iess de desc scri ribe bed d by st stac acks ks bl blow owin ing, g, li lids ds flip flippi ping ng,, ce ceili iling ngss  being hit and so on. It appears, then, that the embodied metaphorical ways in which people partially conceptualize experiences actually provide part of th thee mo moti tiva vati tion on fo forr wh why y sp spea eake kers rs ha have ve co cons nsis iste tent nt me ment ntal al ima image gess an and d sp speecific knowledge about these images for idioms and proverbs with similar figurative meanings. Embodied Action in Metaphor Processing

A new line of research investigated the possible influence of bodily action on people’s speeded processing of simple metaphoric phrases (i.e., hypothesis 4  above). Phrases such as “stamp out a feeling,” “push an issue,” “sniff out the truth,” and “cough up a secret” all denote physical actions upon abstract items. Wilson and Gibbs (2005) hypothesized that if  abstract concepts are indeed understood as items that can be acted upon  by the body body,, then performing a related action should facilitate making a sensibleness judgment for a figurative phrase that mentions this action. For example, if participants first move their leg as if to kick something, and then read “kick around the idea,” they should verify that this phrase

 

184

 

Embodiment and Cognitive Science

is meaningful faster than when they first performed an unrelated body action. Participants were first taught to perform various specific bodily actions given different nonlinguistic cues. The sixteen bodily actions were throw, throw, stamp, tear, push, swallow, sniff out, cough, spit out, poke nose, grasp, shake off, put finger, chew, stand, stretch, and shake. The participants lear le arne ned d th thes esee ac acti tion onss by wa watc tchi hing ng a vi vide deot otap apee of an ac acto torr pe perf rfor ormin ming g th thes esee actions after showing a distinct icon before each event. Participants then had to demon demonstrat stratee perfe perfect ct memor memory y for the diff differe erent nt actio actions ns given their respective cues. Following this, participants were individually seated in front of a computer screen. The experiment consisted of a series of trials wher wh eree an ic icon on fla flash shed ed on th thee sc scrree een, n, pr prom ompt ptin ing g th thee pa part rtic icip ipan antt to pe perf rfor orm m the appropriate bodily action. After this was done, a string of words appeared on the screen and participants had to judge as quickly as possible whether that word string was “sensible.” Half of the word strings were sensible and half were not. The sensible phrase phr asess wer weree all con conven ventio tional nalmet metaph aphori oricc phr phrase asess re refer ferrin ring g to an emb embodi odied ed action on some abstract concept. In the experiment, some of the bodily actions participants first performed were relevant to the following verbal phrases (e.g., the motor action kick was followed by “kick around the idea”), and some were not (e.g., the motor action chew was followed by “kick around the idea”). A third type of trial involved no prime at all (i.e., participants did not perform any bodily action before seeing the word string). Part Pa rtic icip ipan ants ts ma made de se sens nsib ible lene ness ss ju judg dgme ment ntss mo morre qu quic ickl kly y to th thee metaph met aphori orical cal phr phrase asess tha thatt mat matche ched d the pr prece ecedin ding g act action ion tha than n to the phr phrase asess that did not match the earlier movement. People were also faster in respon sp ondi ding ng to th thee me meta taph phor or ph phra rase sess af afte terr ha havi ving ng pe perf rfor orme med d a re rele leva vant nt bo body dy moment than when they did di d not move at all. A control study, study, where people simply had to provide a word that best described each action, showed that these priming effects were not just due to prior lexical associations  between the primes and targets. In short, performing an action facilitates understanding of a figurative phrase containing that action word, just as it does for literal phrases. People do not understand the nonliteral meanings of these figurative phrases as a matter of convention. Instead, people actually understand “toss out a plan,” for instance, in terms of physically toss to ssin ing g so some meth thin ing g (i (i.e .e., ., th thee pl plan an is vi view ewed ed as a ph phys ysic ical al ob obje ject ct). ). In th this is wa way y, processing metaphoric meaning involves some imaginative understanding of the body’s role in structuring abstract concepts. Desire as Hunger: A Case Study in Embodied Metaphor

A different research project on embodied action in metaphorical meaning looked at people’s interpretations of metaphorical expressions about

 

Language and Communication

 

185

human desires (Gibbs, Lima, & Francuzo,   2004). Consider the following last lines from the book titled   Holy Holy Hu Hung nger er:: A Me Memo moir ir of De Desi sire re (Bullitt-Jones,  1999), in which the author summarizes her spiritual journey after the death of her father:

In my case, I hungered, I yearned for something – or Someone – that would really fill me up, fill up my life, give me something to live for, something larger than the ordinary of everyday but found there, nevertheless, in the turning of the days and the seasons, the rising and the setting of the sun, in the sheer gift of being alive. By the time my father died, I knew I was on my way. I had set my course. I knew that whatever my life was about, it was about desire, the desire beyond all desire, the desire for God. It was about learning to listen to my deepest hunger and to let this hunger guide me, as a ship steers at night by the stars.

These few lines poetically describe how even the most abstract desires, such su ch as th thee ne need ed fo forr sp spir irit itua uall fu fulfi lfill llme ment nt,, ar aree of ofte ten n co conc ncep eptu tual aliz ized ed in te term rmss of felt embodied experiences, such as those associated with hunger. The meta me taph phor oric ical al ma mapp ppin ing g of hu hung nger er on onto to de desi sirre is fr freq eque uent ntly ly fo foun und d in ta talk lk of  various kinds of desires, including lust and the desires for both concrete objects and abstract ideas/events. Thus, American English speakers often talk of abstract desires in terms of hunger: He hungers for recognition. He hungers for adventure. He had a hunger for power. He hungers for revenge.

Asserting this metaphorical relationship is not just a conventional or arbitrary way of speaking about desire, because there appear to be rich, systematic correspondences between feeling hunger and feeling differe different nt aspe as pect ctss of de desi sirre. Gi Gibb bbss et al al.. (i (in n pr pres ess) s) in inve vest stig igat ated ed wh whet ethe herr un univ iver ersi sity ty st stuudents in two cultures, the United States and Brazil, metaphorically understand different desires in terms of their embodied experiences of hunger. They first examined people’s embodied experiences of hunger, apart from their understanding of hunger in talk of desire. Some bodily experiences of hunger should be far more prominent than others across both American English and Brazilian Portuguese speakers. If hunger and desire are high hi ghly ly co corr rrel elat ated ed,, an and d if pe peop ople le me meta taph phor orica ically lly ma make ke se sens nsee of th thei eirr de desi sire ress part pa rtly ly in te term rmss of hu hung nger er,, th then en th thes esee mo morre pr prom omin inen entt pa part rtss of th thei eirr hu hung nger er experiences should be invariantly mapped onto their different concepts for desire. Thus, people should subsequently view certain ways of talking abou ab outt de desi sirres in te term rmss of sp spec ecific ific hu hung nger er ex expe peri rien ence cess mo more re ac acce cept ptab able le th than an less prominent aspects of feeling hunger. hunger. A first study presented American and Brazilian college students with three types of symptoms that may possibly result from a person being hungry (these were translated into Brazilian Portuguese for the Brazilian

 

186

 

Embodiment and Cognitive Science

participants). “Local” symptoms referred to specific parts of the body, “gener “ge neral” al” sym sympto ptoms ms re refer ferre red d to who whole le bod body y exp experi erienc ences, es, and “be “behav havior ioral” al” symptoms referred referred to various behaviors that may result as a consequence of a person being hungry. Each of these three symptoms included items that th at we pr pres esum umed ed ma may y be cl clos osel ely y rel elat ated ed to th thee ex expe peri rien ence ce of be bein ing g hu hung ngry ry,, items possibly related, and items not at all related to hunger. An analysis of these ratings showed that English and Portuguese speakers gave similar ratings to the different items. For example, the two groups of participants agreed that strong effects of hunger on the human body include the

stomach grumbling, thought of food making one’s mouth water, water, having a stomac sto machac hache, he, andbecoming having hav ing a hea headac dache he (lo (local cal sympto sym ptoms) ms); feeling ling dis discom fort, t,  becoming weak, dizzy, dizzy , getting annoyed, and; fee having an comfor appetite (gen (g ener eral al sy symp mpto toms ms); ); an and d th thee pe pers rson on fe feel elin ing g ou outt of ba bala lanc nce, e, be beco comi ming ng em emootiona tio nall lly y fr frag agile ile,, an and d be beco comi ming ng ve very ry an anxi xiou ouss (b (beh ehav avio iorr sy symp mpto toms ms). ). Th Thee tw two o groups of participants also agreed on those items that were not related to their hunger experiences. Examples of these items include the following: the knees swell, the feet hurt, the hands itch, and the fingers snap (local symptoms); one wants to run, does not wish to see anyone, becomes talkative, and gets a fever (general symptoms); and the person behaves normally and the person can work well (behavior symptoms). Overall, these findings indicate significant regularities in people’s embodied experien ri ence cess of hu hung nger er,, at le leas astt as su sugg gges este ted d by sp spea eake kers rs fr from om th thes esee tw two o di diff ffer eren entt cultures. A second study examined whether people’s folk knowledge about hunger hun ger is cor corre relat lated ed wit with h the their ir und unders erstan tandin dings gs of dif differ ferenc encee exp experi erienc ences es of  desire. To To do this, t his, English and Portuguese speakers from the same populations sampled in the first study were asked to give their intuitions about two types of questions. The first set of questions focused on how people’s  bodies felt when experiencing experiencing three three types of desire: desire: love, lust, and the desire for things other than human beings, such as fame, adventure, money, money, and so on (the “other” category). Participants were asked to read each question and then rate the relevance of various bodily experiences (e.g.,  becomes dizzy, dizzy, weak, annoyed, talkative) when that person was in love, in lust, or experiencing some other desire. Thee se Th seco cond nd se sett of qu ques esti tion onss fo focu cuse sed d on pe peop ople le’s ’s in intu tuit itio ions ns ab abou outt th thee ac ac-ceptability different half ways linguistically expressing desire. Similarly to the bodyof questions, ofof the items were constructed from bodily experiences strongly (or highly) rated for hunger as shown in the first study, study, with wi th th thee ot othe herr ha half lf co comi ming ng fr from om we weak akly ly (o (orr lo low) w) ra rate ted d hu hung nger er it item ems. s. Th Thes esee lingu lin guis isti ticc qu ques estio tions ns we were re po pose sed d fo forr th thrree ty type pess of de desi sirre (i (i.e .e., ., lo love ve,, lus lust, t, an and d other), as was the case for the body questions. The participants’ task was simply to read each statement (e.g., “My whole body aches for you,” “I have a strong headache for knowledge,” “My hands are itching for you,”

 

Language and Communication

 

187

“My knees ache for information about my ancestry”) and rate whether it was an acceptable way of talking in their respective language. An analysis of the mean ratings showed that the findings for both the Body and Linguistic questions are generally consistent across English and Portuguese for the three types of symptoms for the three types of desire (love, lust, other). For instance, in regard to students’ ratings of the acceptability of different linguistic expressions, both the American and Braz Br azil ilia ian n st stud uden ents ts vi view ewed ed “I ha have ve a gr grea eatt ap appe peti tite te fo forr mo mone ney” y” an and d “I ha have ve a st stom omac ach h pa pain in fo forr my ol old d wa way y of lif life” e” as be bein ing g rea easo sona nabl ble, e, ac acce cept ptab able le wa ways ys of talking about different desires. But they also rated expressions such as “I became talkative for adventure” and “My knees swell for information about my being ways ofgtalking about desire. Over Ov eral all, l,ancestry” then th en,, th thee as findi fin ding ngss unacceptable show sh owed ed ho how w kn know owin ing some so meth thin ing g ab abou outt pe peoo-

ple’s embodied experiences of hunger allows scholars to empirically predict which aspects of desire will and will not be thought of, and talked about, in terms of our complex embodied understandings of hunger. This evidence is generally consistent across two different languages and cultural communities. People use their knowledge of their bodily experiences/actions as the primary source of metaphorical meaning and understanding. Understanding Time Expressions

One of the topics that has generated significant debate in discussions of  embodied metaphor is language about time. Time, like most abstract domains, can be described by more than one metaphor. In English, two separate metaphors are used to sequence events in time (Lakoff & Johnson, 1980). Th Thee fir first st is th thee eg egoo-mo movi ving ng me meta tap pho horr in wh whic ich h th thee ego or ob obse serv rveer’s movement progresses along the time-line toward the future (e.g., “We’re coming up on Christmas”). The second is the time-moving metaphor in which a person is standing and the time-line is conceived as a river or a conv co nvey eyer er-b -bel eltt in wh whic ich h ev even ents ts ar aree mo movi ving ng fr from om th thee fu futu turre to th thee pa past st (e (e.g .g., ., “Christmas is coming up”). These two metaphors lend to different assignments of the front and  back in the time line. In the ego-moving metaphor metaphor,, front is assigned to a future or later event (e.g., “The revolution is before us”; the revolution is ath later orrver future is tion said been before because itves issfurther the e ob obse serv er’s ’s di dirrevent ecti ec tion onand of mo moti on). ).to When Wh an ob obse serv rver er mo move alon al ong g aalong path pa th,, objects are ordered according to the direction of motion of the observer. In thee ti th time me-m -mov ovin ing g me meta taph phor or,, fr fron ontt is as assi sign gned ed to a pa past st or ea earl rlie ierr ev even entt (e (e.g .g., ., “The revolution was over before breakfast”; the revolution is the earlier event, and is said to be before because it is further along in the direction of motion of time). Once again, an analo analogous gous system exists for ord ordinary inary

 

188

 

Embodiment and Cognitive Science

obje ctss in sp spac ace. e. Wh When en ob obje ject ctss wi with thou outt in intr trin insi sicc fr fron onts ts ar aree mo movi ving ng,, th they ey ar aree object assigned fronts based on the direction of motion. Aree ego Ar ego-mo -movin ving g and tim time-m e-movi oving ng exp expre ressi ssions ons und unders erstoo tood d thr throug ough h dif dif-ferent conceptual schemes? A common argument in the debate on time is that linguistic metaphors play no causal role in shaping abstract domains (Murphy,  1996). One study that tested this idea provided participants with a block of temporal statements that either were consistent with one scheme, or switched between ego-moving and time-moving schemes (Gentner, Imai, & Boroditsky, 2002). For each statement (e.g., “Christmas is si six x da days ys be befo forre Ne New w Yea ear’ r’ss Da Day” y”), ), pa part rtic icip ipan ants ts we werre gi give ven n a ti time me-l -lin inee of  events eve nts (e. (e.g., g., pas pastt . . . New Year ear’s ’s Day Day . . . fut futur ure), e), and had to plac placee an eve event nt (e.g.,, Christ (e.g. Christmas) mas) on the time-line. Participants Participants took more time to do this when the temporal statements switched between the two metaphors. A different study on understanding time expressions asked people at an airport (Chicago O’Hare) a priming question in either the ego-moving form (e.g., “Is Boston ahead or behind in i n time?”) or the time-moving form (e.g., “Is it earlier or later in Boston than it is here?”) (Gentner et al.,

2002). After answering, the participants were asked the target question

“So should I turn my watch forward or back?” which was consistent with the ego-moving form. The experimenter measured response times to the target question with a stopwatch disguised as a wristwatch. Once again, response times for consistently primed questions were shorter than for inconsistently primed questions. Switching schemas caused an increase in proce pr ocessi ssing ng time time.. The These se re resul sults ts sug sugges gestt tha thatt two dis distin tinct ct con concep ceptua tuall sch scheme emess are involved in sequencing events in time. Additional evidence on embodied metaphor in understanding temporal statements comes from a study in which participants answered  blocks of questions about days of the week phrased in either the egomoving metaphor (e.g., “We passed the deadline two days ago”) or the time-moving metaphor (e.g., “The deadline passed two days ago”) (McGlo (Mc Glone ne & Har Hardin ding, g, 1998). Fo Forr ea each ch st stat atem emen ent, t, pa part rtic icip ipan ants ts in indi dica cate ted d th thee day of the week on which the event in question has occurred or would occur. At the end of each block, participants read an ambiguous temporal statement such as “The meeting originally scheduled for next Wednesday Wednesday has been moved forward two days” and were asked to perform the same task. The “move-forward” statement is ambiguous because it could be interp te rprret eted ed us usin ing g on onee or th thee ot othe herr sc sche hema ma to yi yiel eld d di difffe ferren entt an answ swer ers. s. Pa Part rtic iciipants in the ego-moving tended to disambiguate the “moved forward” statement in ancondition ego-moving-consistent manner (thought the meeting was on Friday), whereas participants in the time-moving condition tio n te tend nded ed to di disa samb mbig igua uate te in a ti time me-m -mov ovin ingg-co cons nsis iste tent nt ma mann nner er (t (tho houg ught ht the meeting was on Monday). These studies provide strong evidence for the psychological reality of two distinct globally consistent schemas for sequencing events in time.

 

Language and Communication

 

189

Some scholars suggest that even if languages differ in the metaphors they use to describe abstract domains, speakers of these languages should not differ in their mental representations of these domains (Murphy, 1996). Recent evidence suggests that this is not the case (Boroditsky, 2001). English and Mandarin speakers talk tal k about time differently. differently. English speakers use predominately horizontal terms to talk about time, whereas Mand Ma ndar arin in sp spea eake kers rs us usee bo both th ho hori rizo zont ntal al an and d ve vert rtic ical al te term rms. s. A me meta taph phor oric ical al stru st ruct ctur urin ing g ac acco coun untt wo woul uld d pr pred edict ict th that at Ma Mand ndar arin in sp spea eake kers rs wo woul uld d be mo morre likely to rely on vertical spatial schemas when thinking about time than English speakers. This is indeed what was observed. When answering true/false questions tio ns ab abou outt ti time me (e (e.g .g., ., “M “Mar arch ch co come mess ea earl rlie ierr th than an Ap Apri ril” l”), ), Ma Mand ndar arin in sp spea eakkers were faster after vertical spatial primes than after horizontal spatial primes. This result implies that Mandarin speakers were relying on vertical representations of time to answer the time questions. The reverse was true for the English speakers. English speakers were faster answering the ques qu estio tions ns af afte terr ho hori rizo zont ntal al sp spat atia iall pr prime imess th than an af afte terr ve vert rtic ical al sp spat atial ial pr prim imes es.. This difference is particularly striking because both groups performed the task in English, and all of the Mandarin speakers had had at least 10 years of sp spea eaki king ng En Engl glis ish. h. Fu Furt rthe herr, En Engli glish sh sp spea eake kers rs wh who o we werre br brie iefly fly tr trai aine ned d to talk ta lk ab abou outt tim timee us usin ing g ve vert rtic ical al me meta taph phor orss pr prod oduc uced ed re resu sult ltss th that at we werre st stat atis is--

tically indistinguishable from those of Mandarin speakers. This is strong evidence that embodied metaphorical concepts play an important role in shaping abstract thought. People’s understanding of time is not necessarily based on online sensorimotor activity, activity, but rather on people’s repr representations esentations of and thoughts about their past and present spatial experiences. Support for this claim come co mess fr from om se seve vera rall ex expe peri rime ment ntss in wh whic ich h pe peop ople le in di diff ffer eren entt se sett ttin ings gs we were re asked a question about time (Boroditsky & Ramscar, 2002). For instance, students waiting in line at a cafe were given the statement “Next WednesWednesday’s meeting has been moved forward two days” and then asked “What day da y is th thee me meet etin ing g th that at ha hass be been en res esch ched edul uled ed?” ?” St Stud uden ents ts wh who o we werre fa fart rthe herr along in line (i.e., who had thus experienced more forward spatial motion) were more likely to say that the meeting had been moved to Friday. Similarly,, people riding a train were presented with the same ambiguous Similarly statement and question about the rescheduled meeting. Passengers who were at the end of their journeys reported that the meeting was moved to Friday significantly more than did people in the middle of their journeys. Although bothofgroups werethey experiencing the same physical experience sittingof inpassengers a moving train, thought differently about their journeys and consequently responded differently differently to the rescheduled meet me etin ing g qu ques esti tion on.. Th Thes esee fin findi ding ngs, s, al alon ong g wi with th ot othe hers rs,, su sugg gges estt th that at it is ho how w people think about spatial motion, and not the physical experience itself, that influences their thoughts about temporal events.

 

190

 

Embodiment and Cognitive Science

How people think about motion through space also influences their comprehension of fictive motion sentences, such as “The road runs along the coast.” Although fictive motion expressions communicate no explicit moti mo tion on (i (i.e .e., ., ro road adss do no nott lit liter eral ally ly ru run) n),, pe peop ople le ma may y un unde ders rsta tand nd th thes esee st stat ateements in terms of implicit, imaginary sensations of movement (Talmy, 1996). Experimental studies show this is true (Matlock, 2004). Participants read stories about protagonists traveling through spatial regions and then made speeded decisions about whether a fictive motion statement related to the story. Reading times in different experiments were faster when sentences followed stories depicting short distances, fast motion, and uncluttered terrains. People gave faster positive decisions to statements containing in g fa fast st ve verb rbs, s, su such ch as “T “The he roa oad d ru runs ns al alon ong g th thee co coas ast, t,”” th than an to ex expr pres essi sion onss with wi th sl slow ower er ve verb rbs, s, su such ch as “T “The he roa oad d me mean ande derred al alon ong g th thee co coas ast. t.”” No None ne of  these differences were found in control studies examining comprehension of nonfictive motion spatial sentences, suggesting that the fictive motion effects are not due to lexical priming. Over Ov eral all, l, th thes esee st stud udie iess im impl ply y th that at em embo bodi died ed si simu mula lati tion on is cr crit itic ical al to pr proocessing fictive motion. Thus, people interpret the meanings of fictive motion statements by “replaying” the movement, reconstructing a mental enactment of movement implicit in the sentence. People are not aware of  these simulations, and so fictive motion processing is not dependent on deliberate thought about motion. These psycholinguistic studies provide additional support for the general claim that language is closely tied to imagination that is grounded in ordinary embodied action. In general, the experimental work on understanding linguistic state-

ments about time shows how people conceive of time in embodied, meta me taph phor oric ical al wa ways ys.. Th Thes esee da data ta dir direc ectly tly co cont ntra radi dict ct th thee vi view ew th that at me meta taph phor or-ical ic al ta talk lk of ti time me em emer erge gess fr from om ab abst stra ract ct si simi mila lari riti ties es be betw twee een n sp spac acee an and d ti time me (Jackendoff & Aron, 1991). Embodied Metaphors in American Sign Language

Another example of contemporary research on how bodily action shapes utte ut tera ranc ncee in inte terp rprret etat atio ion n co come mess fr from om th thee st stud udy y of Am Amer eric ican an Si Sign gn La Lang ngua uage ge (ASL (A SL). ). A co comm mmon on co conc ncep eptu tual al me meta taph phor or wi with th em embo bodi died ed ro root otss in Am Amer eric ican an English and other languages is COMMUNICATION IS SENDING AND RECE RE CEIV IVIN ING G OB OBJE JECT CTS. S. Th This is co conc ncep eptu tual al me meta taph phor or un unde derl rlie iess sp spea eake kers rs’’ us usee and an d un unde ders rsta tand ndin ing g of li ling ngui uist stic ic ex expr pres essi sion onss su such ch as “W “Wee to toss ssed ed so some me id idea eass  back and forth” and “His meaning went right over my head.” In each case, ideas correspond to objects and the act of communicating corresponds to the sending and receiving of these objects. Recent work on ASL demonstrates that a similar conceptual metaphor underlies ideas about communication in ASL (Taub,  2001; Wilcox, 2001). ASL sig signer nerss gen genera erally lly exp exploi loitt sig signin ning g spa space ce to sch schema ematic tically ally re repr preesent spatial relations, time, order, and aspects of conceptual structure

 

Language and Communication

 

191

(Emmorey, 2002). When signers describe spatial relations, there is a structural analogy between the form of a construction and aspects of the described Specifically, elements inveme ASL (the map to phys ph ysic ical alscene. elem el emen ents ts wi with thin in th theephysical scen sc enee (o (obj bjec ects ts), ), mo move ment nts s of hands) thee ha th hand nds s ma map p to the motion of referent objects, and locations in signing space map to physic phy sical al loc locati ations ons with within in the sce scene. ne. Thr Throug ough h met metaph aphori orical cal map mappin pings, gs, sig signners can extend the use of classifier constructions and signing space to describe abstract concepts and relations. However,, the conceptual metaphors in ASL are different from those in However spoken language, because they involve a double mapping (Taub,  2001). First, there is a metaphorical mapping from a concrete, embodied source domain to an abstract target domain (e.g., objects that can be grasped and passed pas sedto to oth others ersar aree map mapped pedont onto o ide ideas/ as/tho though ughts/ ts/con concep cepts) ts).. Sec Second ond,, the there re is an iconic mapping mapping fro from m the concrete concrete domain to the linguistic linguistic domain (e.g., cylindrical objects map to cylindrical cyli ndrical handshapes). Forr ex Fo exam ampl ple, e, si simil milar arly ly to En Engl glis ish h sp spea eake kers rs,, AS ASL L si sign gner erss us usee th thee communicating-as-sending metaphor. For both speakers and signers, the discou dis course rsess of com commun munica icatin ting g ide ideas as and thr throwi owing ng obj object ectss ar aree lin linked ked,, whereby an idea corresponds to an object, and telling or explaining the idea corresponds to throwing the object to someone. But unlike spoken English, ASL has an additional iconic mapping between the concrete domain ma in (o (obj bjec ects ts)) an and d th thee ar artic ticula ulato tors rs (t (the he ha hand nds) s).. Co Cons nsid ider er th thee En Engl glis ish h st stat ateement “I didn’t get through to him” in reference to a speaker trying to get a listener to understand some idea or belief. In ASL, the equivalent sign (parap (pa raphra hrased sed as THI THINKNK-BOU BOUNCE NCE)) ind indica icates tes a fai failur luree to com commun munica icate te and consists of an iconic depiction of a projectile bouncing off a wall (the dominant handshape moves from the head and bounces off the nondominant han ha nd) d).. Th Thus us,, AS ASL L has tw two o le leve vels ls of hum uman an mo move veme men nt in th thee sig ign n ref efeerr rrin ing g to failure to communicate. ASL AS L ex exhi hibi bits ts do doub uble le ma mapp ppin ings gs in ma many ny co conc ncep eptu tual al do doma main ins. s. Fo Forr ex exam am--

ple, the metaphorical mapping between power and height (e.g., power is up) has an additional iconic mapping between height and signing space. Thus, authority figures are associated with higher locations in signing space, and less powerful people are associated with lower locations. This nicely illustrates how bodily action helps to articulate abstract ideas such as our conceptualization of people in authority. Similarly, signers use the “intimacy is proximity” metaphor to associate known or preferred objects/people with locations near the body and less preferred ob jects/people with locations away from from the body. body. Taub also argues that some signs (e.g., THINK-PENETRATE) are fully motiva moti vate ted d by a si sing ngle le me meta taph phor or,, wh wher erea eass ot othe herr si sign gnss ar aree on only ly pa part rtia iall lly y mo mo-tivated, motivated by several metaphors simultaneously, simultaneously, or are motivated  by both metaphorical and pure iconicity iconicity.. For example, the sign for SAD consists of a downward motion of both spread fingered hands, palms in, in fr fro ont of th thee fa facce. Th Thee si sign gner er bu buil ilds ds on th thee ma mapp ppin ing g of th thee up up-d -do own sca cale le

 

192

 

Embodiment and Cognitive Science

onto emot otio ion, n, wh wher eree ne nega gati tive ve em emot otio ions ns ha have ve do down wnwa ward rd mo move veme ment ntss (e (e.g .g., ., onto em “I’m feeling down today”). Thee si Th sign gn fo forr TH THRI RILL LL em empl ploy oyss bo both th th thee me meta taph phor orss in sa sad d an and d ha happ ppy y an and d adds a third. Thus, the of upward movement for HAPPY ARE UP begins at the center t he chest, the where THE LOCUSEMOTIONS OF EMOTION IS THE CHEST, and includes the open-8 handshape, which is motivated by the metaphor FEELING IS TOUCHING. Finally,, the sign for EXCITE incorporates both metaphor and iconicity. Finally iconicity. EXCITE uses the same three metaphors as THRILL but differs in meaning from THRILL in that rather than having both hands move upward in a single,, long rapid str single stroke, oke, the two hands alter alternate nate making short upwar upward d movements at the chest. Thus, whereas the sign for THRILL represents a  brief, rapid experience, the the sign for EXCITED represents represents the experience of  of  an ongoing state. Another metaphor that is structured in terms of movement around the  body is time (W (Wilcox, ilcox, 2001).TimeinASLisexpressedinrelationtothe“time line,” with each time sign running along an imaginary line through the  body.. The signer’s body repr  body represents esents present time, and areas in the front and  back of the body repr represent esent future and past, respectively (see Emmorey Emmorey,, 2002 forr a di fo difffe ferrent mo mode dell of ti time me in AS ASL) L).. Tim imee sig igns ns su such ch as NO NOW W, WI WILL LL,, an and d ONE-DAY-PAST have relative locations on the line that agree with the temporal message, even though their specific locations are not to scale. Most generally, ASL represents time as a perceptual experience in terms of spatial path and temporal unidirectionality. Thus, time is perceived as running from past (back) to future (front). The front and back of a human  body correspon correspond d to the body’s daily movements of running ahead into the future and stepping back into the past. As in spoken language, the container schema is also prevalent in ASL. For example, being knowledgeable expressed by sign si gne ers by th thee the usee idea us of th theeofCahperson ands an dsha hap p e at th the e fr fron ontt of th thee fo forris ehe head ad.. Th Thee si sign gn demo de mons nstr trat ates es th that at th thee mi mind nd co coul uld d be vi visu sual aliz ized ed as a fu full ll co cont ntai aine nerr. A si sign gner er can convey the idea of incomplete understanding, or a momentary lapse in thought, by collapsing the handshape. But the container metaphor in ASL AS L is mo morre th than an a si simp mple le on onto tolo logi gica call me meta taph phor or de desc scri ribi bing ng an ab abst stra ract ct en en--

tity. Understanding metaphorical mappings convey abstract connections withintheinteriorofthecontainer.Thesemappingsareorganizedbydifferent ima image ge sch schema ema,, suc such h as SOU SOURCE RCE-P -PA ATHTH-GOA GOAL, L, LIN LINK, K, PAR ART-W T-WHOL HOLE, E, CENTER CEN TER-PE -PERIP RIPHER HERY Y, and FRO FRONTNT-BAC BACK. K. The These se ima image ge sch schema emass und underl erlie ie thee so th sour urce ce do doma main inss in ma many ny co comp mple lex x me meta taph phor ors. s. Fo Forr in inst stan ance ce,, th thee FR FRON ONTTBACK schemas refer to ideas such as “It’s in the back of my mind somewhere.” Deaf people know that the brain’s activities are specialized in different regions, but they typically use the forehead area in making signs different such as remember, understand, memorize, think, imagine, idea, puzzle, and hypothesize, among many others.

 

Language and Communication

 

193

One example of taking ideas out of one’s mind concerns a signer who wass at wa atte temp mpti ting ng to wr writ itee a bo book ok de deta tail ilin ing g al alll th thee jo joke kess an and d fo folk lklo lorre he ha hass rememb me mber ered ed fr from om ta talki lking ng wi with th de deaf af pe pers rson onss al alll ov over er th thee wo worl rld d (W (Wilc ilcox ox,, 2001). Thee si Th sign gn fo forr “p “poo ooll-id idea eass-in into to-b -boo ook” k” be bega gan n wi with th th thee si sign gner er ha havi ving ng bo both th fis fists ts close to his hislap forehead andfingers then throwing them outward and downward toward area, with splayed out, into the place where a book might be written or read. Thus, the ideas are taken from the mind as a container and put into a differe different nt container or book. ASL AS L ha hass th thee me meta taph phor or ID IDEA EAS S IN EX EXIS ISTE TENC NCE E AR ARE E ST STRA RAIGH IGHT T. En Entit titie iess in the world that are erect and straight tend to be objects that persist. Living objects that are alive have integrity and stand tall, whereas dead trees tr ees,, flow flowers ers,, and eve even n peo people ple top topple ple ove overr. The These se exp experi erient ential ial eve events nts ser serve ve as the source domain for understanding abstract mental processes. Thus, ideas, thoughts, or understanding can be metaphorically viewed as living li ving things. When referring to the process of abstract thoughts, and coherent ideas, ASL has the G classifier handshape in which the index finger is extended upright and pointed near the forehead. This icon of the straight finger serves as a metaphor for physical life or existence. Not surprisingly, there is an equally pervasive countermetaphorical mapping that IDEAS NOT FULLY IN EXISTENCE ARE BENT. This metaphor is based on the experience of entities not in i n existence being difficult to see. Thus, when an idea metaphorically disappears from view, view, by signers bending or flexing their index finger, finger, it is permanently gone. ASL signs that evoke IDEAS I DEAS NOT FULLY FULLY IN EXISTENCE ARE BENT include weak-m wea k-mind inded, ed, dr dream eams, s, and mul mull-o l-over ver,, wit with h eac each h sig sign n art articu iculat lated ed wit with h ben bentt fingers. Most generally, generally, thought corresponds to our experiences of watching living things come into being or crumbling away – with the bending or straightening of a finger mirroring life. Fina Fi nall lly y, th thee co conc ncep eptu tual al me meta taph phor or ID IDEA EAS S AR ARE E OB OBJE JECT CTS S TO BE GRASPED represented byThus, a handshape with fist simultaneous and metaphoricisrepresentations. repre sentations. a fully closed handshapeiconic is used in the sign to represent someone reaching out to grab objects as in “Ryan scooped up the jewels with one hand.” This same classifier is employed in cases like “I will take grandmother with me,” even if we don’t literally grab people in taking them somewhere. Not surprisingly, the same sign can be used metaphorically as when signing “Hang onto that idea.” This particular handshape maps the grasping of an object in such a way that it cannot escape onto the intellectual process of permanently holding onto

an idea in memory. Ling Li ngui uist stic ic an analy alyse sess su such ch as th thes esee sh show ow ho how w sp spok oken en an and d si sign gned ed la lang ngua uage ge share many of the same schematic mappings between embodied experiences/ enc es/act action ionss and mor moree abs abstra tract ct con concep ceptua tuall dom domain ains. s. The There re is no emp empiri irical cal evidence as yet demonstrating that embodied metaphors are accessed, or acti ac tiva vate ted, d, du duri ring ng co comp mpre rehe hens nsio ion n of AS ASL. L. Bu Butt th thee ab abov ovee lin lingu guis istic tic an anal alys ysis is

 

194

 

Embodiment and Cognitive Science

suggests that this is a likely possibility, and should be a topic for future experimental work. Neural Theory of Language

The emp empiri irical cal wor work k illu illustr strati ating ng the imp import ortanc ancee of emb embodi odied ed exp experi eri-ence in linguistic understanding has focused mostly on behavioral and neuropsychological evidence. But there are also recent developments in cognitive science on the neural basis for language interpretation. The neural the theory ory of lan langua guage ge (NT (NTL) L) is an int inter erdis discip ciplina linary ry pr proje oject ct at the Uni Univer versit sity y of California, Berkeley devoted to understanding how neural structures of the human brain shape thought and language and influence language learning and understanding (Feldman & Narayanan,  2004). These studies attempt to define the representations and computations used to link  brain functions, including those related to emotion and social cognition, with language use. A general assumption, based on neuroscientific work, is that there should not be brain are areas as that are specialized specialized for langu language, age, and that language processing should not be confined to only a few select regions of the brain. An early NTL project provided a neural model for learning spatial relations terms in the world’s languages (Regier,  1996). For example, English has several concrete and abstract spatial terms, such as prepositions that can ca n be us used ed to ex expr pres esss bo both th sp spat atia iall an and d no nons nspa pati tial al me mean anin ings gs,, as in “I “I’m ’m in a depression,” “Prices went up,” and “He’s beside himself.” These nonspatial uses of spatial relations arise from systematic conceptual metaphors that preserve the spatial logic of the source domain (see Chapter  4). Previous linguistics work showed that many elementary spatial relations are topographical in nature (Talmy, 2000). To To build a neural model for spatial relations, Regier adopted ideas from several aspects of cognitive neuroscie sc ienc nce. e. Fi Firs rst, t, to topo pogr grap aphi hica call ma maps ps of th thee vi visu sual al fie field ld we werre us used ed to co comp mput utee image schemaswere that also topological. Second, orientation-sensitive cell assemblies usedwere to compute the orientational aspects of spatial concep con cepts ts tha thatt re rely ly on bod bodily ily ori orient entatio ation n (e. (e.g., g., abo above) ve).. Thi Third rd,, cen center ter-se -sensi nsitiv tivee receptor fields were employed to characterize concepts such as contact. Finally,, a filling-in architectur Finally architecturee was used to deal with notions such as containment. Regier’s model was tested in the following way. A few simple figures (squares, circles, triangles) were presented with various spatial relations,  both static and moving (e.g., “in,” “on,” “through,” “above”), in a simple comp co mput uter er mo mode dell of a ret etin inaa (n × m pi pixe xels ls). ). On Onee fig figur uree se serv rved ed as La Land ndma mark rk and an d th thee ot othe herr as Tra raje ject ctor or (e (e.g .g., ., if th thee ci cirrcl clee is un unde derr th thee sq squa uarre, th thee sq squa uarre is thee La th Land ndma mark rk an and d th thee ci cirrcl clee is Tra raje ject ctor or). ). Th Thee mo mode del’ l’ss ta task sk wa wass to le lear arn n th thee

spatial-relations system of a language and the spatial-relations terms, so that th at th thee sy syst stem em co coul uld d pr prov ovid idee th thee ri righ ghtt na name me fo forr ne new w sp spat atia iall co confi nfigu gura rati tion onss presented on the computer screen. A difficult challenge here was to learn

 

Language and Communication

 

195

thes esee sp spat atia iall-re rela latio tions ns te term rmss wi with thou outt an any y fe feed edba back ck ab abou outt wh when en th thee sy syst stem em th was incorrect. In fact, Regier’s model was extremely good at learning these spatial terms, and even accurately displayed prototype effects without being trained on prototypes. One implication of the work is that conceptual and linguistic categories may be formed using perceptual apparatus from the visual system. Thus, conceptual categories of spatial relation are created  based on the brain’s structure and our bodily experience of spatial relations. A di difffe ferren entt mo mode dell wi with thin in th thee NT NTL L fr fram amew ewor ork, k, ca call lled ed KA KARM RMA A (knowledge-based action representations for metaphor and aspect), characterized metaphoric reasoning about events (Narayanan,  1997). Many narratives describe abstract plans and events in terms of spatial motion and an d ma mani nipu pula latio tions ns.. Fo Forr ex exam ampl ple, e, rea ead d th thee fo follo llowi wing ng br brie ieff ne news wspa pape perr st stor ory y about European economics (Narayanan, 1997: 1): Britain was deep in recession while Germany was flourishing three years ago. France kept moving steadily long after Germany had fallen into recession. But now France is plunging deeper while the German economy continues to struggle. Britain has been taking small steps toward stimulating its economy by cutting interest rates, and has finally started to emerge from recession.

Narayanan Naraya nan’s ’s bas basic ic hyp hypoth othesi esiss was tha thatt peo people ple und unders erstan tand d thi thiss nar narrat rative ive from their knowledge of embodied metaphors, including those related to moving steadily, falling, taking small steps, plunging deeper, struggling, and starting to emerge. These embodied metaphors function to project features of spatial motion and manipulation onto abstract plans and processes. Representational structures, called “x-schemas,” encode embodied metaphors in a way that retains their dynamic and highly responsive realtime nature to reason about abstract events. Narayanan’s computational model included a detailed account of how any motor schema may be modeled in the form of Petri nets that are reducible to structured connectionist neural networks. The x-schema representation reflects low-level motor synergies that perform motor control and connects motor actions to produce complex motor sequences. These motor events, actions, and processes are invariantly projected onto more abst ab stra ract ct do doma main inss to li link nk ph phys ysic ical al an and d ec econ onom omic ic do doma main ins, s, su such ch as AC ACTI TION ON IS MO MOTI TIO ON, A REC ECES ESS SIO ION N IS A HO HOLE LE,, an and d MO MORE RE IS UP UP.. Ne Neur uraal mo mod del elss usee th us thee ph phys ysic ical al la lang ngua uage ge in th thee ne news ws st stor ory y to ac acti tiva vate te a me ment ntal al si simu mula lati tion on of physical action, using control structures (with actual motor action assumed to be inhibited). The results showed that this computational model drew dr ew th thee sa same me in infe ferren ence cess pe peop ople le do wh when en rea eadi ding ng a wi wide de va vari riet ety y of ne news ws-paper stories about economics. For instance, the system drew inference related to goals (their accomplishment, modification, subsystem, concordance, or thwarting), aspect (temporal structure of events), frame-based inferences, perspectival inferences, and inferences about communicative intent. In general, Narayanan’s system shows how the same structured

 

196

 

Embodiment and Cognitive Science

neural network used to control high-level motor schemas also operates during abstract reasoning about economic events. A second part of Narayanan’s work focused on aspect, or the linguistic devic de vices es th that at sp spea eake kers rs us usee to di dire rect ct li list sten ener ers’ s’ at atte tent ntio ion n to th thee in inte tern rnal al te temp mpooral characteristics of a situation. For instance, the verb “tap” is inherently iterative and suggests repeated actions, “pick up” has a purpose and a final state, “run” has no inherent final state, “slip” is nonvolitional, and “walk” “wal k” is du dura rati tive ve (i (itt ta take kess an ex exte tent nt of ti time me). ). En Engl glis ish h ha hass di difffe ferren entt le lexi xica cal, l, morphological, and grammatical means of specifying aspect. The English progressive progr essive construction (“be + V-ing”), for instance, enables speakers to focu fo cuss on th thee on ongo goin ing g na natu ture re of an un unde derl rlyi ying ng pr proc oces esss th that at ha hass be been en st star arte ted, d, yett no ye nott co comp mple lete ted. d. Th Thee pa past st-p -per erfe fect ct co cons nstr truc uctio tion n (“ (“ha hass VV-ed ed”) ”) giv gives es sp spea eakkerss th er thee ab abil ilit ity y to sp spec ecif ify y co cons nseq eque uenc nces es of a si situ tuat atio ion, n, as wh when en so some meon onee ha hass completed some action. Different aspectual verbs, such as “tap,” “walk,” and “run,” also denote temporal characteristics of situations. Narayanan (1997) built a computer model to illustrate the semantics of  aspects aspec ts in terms of embod embodied ied action within a neura neurall syste system. m. This model showed how aspectual expressions in language are linked to schematic processes that recur in sensorimotor control, such as inception, interruption termination, iteration, enabling, completion, force, and effort. Conside si derr “J “Jac ack k wa walk lked ed to th thee st stor ore. e.”” Wal alki king ng in invo volv lves es sp spec ecifi ificc en enab abli ling ng co cond ndiitions tio ns (e (e.g .g., ., an up upri righ ghtt po posi siti tion on,, a vi visu sual al/k /kin ines esth thet etic ic te test st in indi dica cati ting ng a st stea eady dy ground) and specific resources (e.g., energy) and may have a specific goal (e.g (e .g., ., ge gett ttin ing g to th thee st stor ore) e).. Th Thes esee fe feat atur ures es in inte tera ract ct wi with th th thee co cont ntro roll ller er,, wh whic ich h results in specific meaning inferences. Thus, when a listener hears “John walked to the store,” he or she is likely to infer that Jack got to the store. The statement “John was walking to the store” does not imply that John actually arrived at the store. Most theories of aspects are unable to deal with wi th th this is “i “imp mper erfe fect ct pa para rado dox, x,”” bu butt Na Nara raya yana nan’ n’ss mo mode dell ill illus ustr trat ates es ho how w th thee difference between these two sentences arises from the constraint that in one case (e.g., “John walked to the store”) the result is obtained only if  the goal (of reaching the store) is reached. In the case of “John was walking to the store,” no such constraint exists and the result occurs only after  John gets to the store. In a similar way way,, Narayanan’s model gives statements such su ch as “H “Hee is ru rubb bbin ing g oi oint ntme ment nt”” an and d “H “Hee is co coug ughi hing ng,” ,” ite itera rati tive ve rea eadi ding ngs, s,  because of the inherently iterative nature of activities like “rubbing” and “coughing.” The dynamic, highly responsive nature of x-schemas enables Naraya Nar ayanan nan’s ’s com comput putati ationa onall mod model el to mak makee re realal-time time def defeas easibl iblee inf infer erenc ences es associated with understanding aspect. One other neural model was developed for learning the verbs of hand motion (Bailey,   1998). English has a large number of verbs associated with different hand motions, including “push,” “pull,” “shove,” “wave,” “pinch,” “yank,” “slap,” “clutch,” “hold,” rub,” “squeeze,” and “fling.” Each verb denotes a slightly different embodied action. Other languages have ha ve th thei eirr ow own n sp spec ecia iall co coll llec ecti tion on of ve verb rbss fo forr ha hand nd mo moti tion on.. Fo Forr in inst stan ance ce,, in

 

Language and Communication

 

197

Farsi “zadan” denotes many different kinds of object manipulation with quick motion. Tamil’s “thallu” and “ilu” are equivalent to “push” and “pull” in English, but refer to ballistic actions, rather than to a smooth continuous movement. In Cantonese “meet” refers to both pinching and tearing, and suggest forceful manipulation by two fingers, but is also approp pr opri riat atee to us usee wh when en ref efer erri ring ng to te tear arin ing g la larrge it item emss wh when en tw two o fu full ll gr gras asps ps aree em ar empl ploy oyed ed.. Sp Span anis ish h ha hass th thre reee se sepa para rate te wo word rdss fo forr “p “pus ush, h,”” wi with th “p “pul ulsa sar” r” referring to poking or using a single finger to push a button, “presionar” referring to applying pressure to something, and “empujar” meaning to push open a door, door, or push another person, usually employing two hands. Human bodies provide the conceptual basis for defining the range of  meanings associated with each verb for hand motion in any language. People do not have direct access to complex neural networks that coordinate na te th thei eirr ac acti tion ons. s. Bu Butt ho how w do does es th this is wo work rk?? Ba Bail iley ey cr crea eate ted d a co comp mput utat atio iona nall model to simulate the acquisition of verb semantics for hand action. The first step was to adapt a computational model of the body, body, named “Jack,” that th at co coul uld d ca carr rry y ou outt ar arm m mo move veme ment ntss co corr rrec ectl tly y. Th This is wa wass do done ne by cr crea eati ting ng a sett of mo se modi difie fied d Pe Petr trii ne nets ts th that at we were re ma mapp pped ed on onto to a st stru ruct ctur ured ed co conn nnec ecti tion onst st network. The resulting neural model embodied a collection of motor synergi er gies es wh which ich ar aree lo loww-le leve vell se self lf-c -con ontr trol olle led d mo moto torr ac acti tion onss su such ch as tig tight hten enin ing g a grip, extending a finger, releasing a grip, pivoting a wrist, and so on. These synergies were coordinated to provide a mechanism for executing motor schemas in real time, given different environmental conditions. Another Anoth er compu computation tational al mecha mechanism nism enabled Bailey to model the rela rela-tionship between low-level motor synergies and the feature structures for different verbs. For example, suppose that the program learned that the word “shove” involves using the slide-executing schema with high force and short duration. This information, in which schemas and parameters define the word, serves as part of its stored definition. Bailey trained the model in English and several other languages by presenting the system with  165  labeled examples of actions corresponding to  15  English verbs and 18 word senses. The model posse possessed ssed suf sufficien ficientt compu computation tational-toal-to-neur neural al local mappi mappings ngs to co corr rrec ectl tly y le lear arn n ve verb rbss fo forr ha hand nd mo moti tion on so th that at th thee sy syst stem em co coul uld d rec ecog ogni nize ze the action, name it correctly, and perform the appropriate hand motions give gi ven n a sp speeci cific fic ve verb rb (f (fo or Eng ngli lish sh,, an and d to a le less sser er ex exte ten nt He Hebr brew ew,, Fa Farrsi si,, an and d Russian). Although the model was not perfect (it performed at about the 80% le leve vell on bo both th ta task sks) s),, Ba Bail iley ey’s ’s mo mode dell de demo mons nstr trat ates es ho how w th thee co conc ncep eptu tual al role ro le of di dist stin ingu guis ishi hing ng be betw twee een n di diff ffer eren entt ve verb rbss fo forr ha hand nd mo moti tion on is gr grou ound nded ed in the sensorimotor system. Thes Th esee res esul ults ts ar aree in intr trig igui uing ng,, al alth thou ough gh th thee re rese sear arch cher erss in invo volv lved ed wi with th th this is projectnotethatthefindingsarepresentedas“existenceproofs”ratherthan as models of real neural structures operating in real time. Yet as existence proo pr oofs fs,, th thes esee st stud udie iess su sugg gges estt so some me of th thee co comp mple lex x wa ways ys th that at br brain ains, s, bo bodi dies es,, and world interact to provide the grounding for linguistic meaning.

 

198

 

Embodied Construction Grammar

Embodiment and Cognitive Science

Embodied construction grammar (Bergan & Chang, in Chang,  in press) press) follows the  basic tenets of mainstream construction grammar (Goldberg, 1995;   Kay & Fillmore, 1999), and cognitive grammar (Langacker,  1991) in assuming that all linguistic knowledge, at all levels, can be characterized as pairings of form and meaning, called “constructions.” Understanding utterances, quite broadly, involves internal activation of “embodied schemes,” along with the mental simulation of these representations in context to generate a rich set of inferences. Constructions are important in this account  because they provide the interface between phonological and conceptual knowledge, thereby evoking embodied semantic structures. Consider the expression “Mary tossed me a drink.” A constructional analysis presumes that the active ditransitive argument structure imposes an in inte terp rpre reta tati tion on in wh whic ich h on onee en enti tity ty ta take kess so some me ac acti tion on th that at ca caus uses es an anot othe herr enti en tity ty to rec ecei eive ve so some meth thin ing. g. Al Alth thou ough gh th thee ve verb rb “t “tos oss” s” ma may y be se seen en in ma many ny argument structures, its appearance in “Mary tossed me a drink” is permiss mi ssib ible le on only ly if it itss me mean anin ing g ca can n be rec ecog ogni nize zed d as co cont ntri ribu buti ting ng to a tr tran ansf sfer er event. The word “tossed” evokes a specific physical action that also denote no tess te tens nsee an and d as aspe pect ct in info form rmat atio ion n rel elat ated ed to th thee la larrge gerr ev even entt in wh whic ich h it is involved. Prototypically, Prototypically, the TOSS schema represents knowledge about a lo loww-en ener ergy gy ha hand ndTOSS actio ac tion nschema that th at ca caus uses es anprofile enti en tity ty to move ve th thro ugh hForce-Motion thee ai th airr. Mo Morre specifically, the helps themo role ofroug the sche sc hema ma wi with thin in th thee ac acti tive ve di ditr tran ansi siti tive ve co cons nstr truc ucti tion on.. Th Thus us,, th thee ac acti tion on of to toss ss-ing profiles a forceful action on an entity that causes its resulting motion, including a tosser (agent) and the tossed object (the object). Constructions aree no ar nott de dete term rmin inis isti tic, c, bu butt fit a sp spec ecifi ificc ut utte tera ranc ncee an and d co cont ntex extt to so some me de degr gree ee,, with the result of processing being the best-fitting set of constructions. This brief overview of constructional analysis provides the first step in determining the meaning of the utterance “Mary tossed me a drink.” The meaning arises from the simulation of grounded semantic structures as characterized by the constructional analysis. First, executing schemas, or x-schemas, are used for executing and perceiving an action and brought to bear in understanding larger abstract action. For example, the TOSS schema evoke schema evoked d by “toss “tossed” ed” enter enterss the tossin tossing-exe g-executio cution n sche schema, ma, which is the explicit, grounded representation of the semantic pattern used by an agent (or tosser) to perform a tossing action. This scheme specifically capt ca ptur ures es a se sequ quen ence ce of ac acti tion onss th that at ar aree re relat lated ed to to toss ssin ing g an ob obje ject ct,, in incl clud udin ing g possibly preparatory actions (e.g., grasping the object and moving it into a suita suitable ble starting position) position) and the arm movement necessary necessary to launc launch h the object. Subsidiary actions that move the object along a suitable path with low force are also included. This execution scheme for tossing may also specify other conditions that possibly hold at different stages of the event, such as that the tossed object must be in the agent’s hand before the

 

Language and Communication

 

199

action takes place, and that the object will be flying toward some target afterward. In general, constructions, such as the active ditransitive, enable listenerss to ac er acce cess ss de deta tail iled ed dy dyna nami micc kn know owle ledg dgee th that at ch char arac acte teri rize ze ri rich ch em embo bodi died ed structures merely by specifying a limited set of parameters. One important result of this analysis is that x-schemas provide significant inferential

power to evoke detailed meanings for any utterance. For instance, part of  thee in th infe fere renc nces es as asso soci ciat ated ed wi with th un unde ders rsta tand ndin ing g “M “Mar ary y to toss ssed ed me a dr drin ink” k” re re-ferr to as fe aspe pect ctss of th thee ch chrron onol olog ogic ical al st stag ages es in inhe here rent nt in th thee ev even entt-le leve vell tr tran ansf sfer er scheme and the action-level tossing scheme, such as the following: SPEAKER does not have drink Mary exerts force via TOSS DRINK in reach of Mary DRINK in hand of Mary Mary launched DRINK toward SPEAKER Mary exerts energy (force-amount = low) DRINK flying toward SPEAKER DRINK not in hand of Mary Mary causes SPEAKER to receive DRINK SPEAKER has received DRINK

This analysis of embodied action schemas can also specify a rich set of inferential meaning evoked by metaphorical utterances such as “Mary tossed the Enquirer the  Enquirer a  a juicy tidbit.” This expr expressio ession n empha emphasizes sizes the same cons co nstr truc ucti tion on as in “M “Mar ary y to toss ssed ed me a dr drin ink, k,”” in incl clud udin ing g th thee ac acti tive ve di ditr tran ansi si-tive construction. But for the metaphorical expression, “the Enquirer “the Enquirer”” cannot be a literal recipient within the TRANSFER schema. A solution to this problem is to construct a metaphorical map that allows a target domain involving Communication to be structured in terms of a corresponding source domain of Object-Transfer, which enables “the Enquirer “the  Enquirer”” to be construed as a suitable recipient. Once this mapping occurs, this links to the inference that the object transfer may be metaphorical as well, belonging to the domain of information, and not food, interpretations of “juicy” and “tidbit.” Moreover, Moreover, both the overall event and the means by which it take plac pl aces es ca can n be un unde ders rsto tood od as a ve verb rbal al ac actt of tr tran ansf sfer er,, as op oppo pose sed d to ph phys ysic ical al.. Mostt gen Mos genera erally lly,, emb embodi odied ed con constr struct uction ion gra gramma mmarr is a sim simula ulatio tionn based model of language understanding. Critical to this perspective is the idea that motor action may be simulated and applied to understandi sta nding ng of var variou iouss asp aspect ectss of lan langua guage ge (Be (Berg rgan an & Cha Chang, ng,   in press; press; Feldman & Narayanan, 2004). Embodied Text Understanding

Readerss use the Reader their ir emb embodi odied ed abi abilit lities ies to imm immedi ediate ately ly cr creat eatee con constr strual ualss of the differentt perspectives, and shifts of perspective, of the objects and actions differen

 

200

 

Embodiment and Cognitive Science

described language ge (MacW (MacWhinne hinney y, 1998). Co Cons nsid ider er th thee se sent nten ence ce “A “Ass fa farr described by langua as the eye could see, stalks of corn were bending as waves under the battering force of a surging curtain of rain” (MacWhinney,  1998). How might readers construct a meaningful interpretation of this sentence? One could argue that readers comprehend comprehend this sentence by simply creating a picture of a heavy rain pouring down on a large corn field. But this characterization underdetermines the embodied richness of what people normally understand from this sentence. A more embodied view of understanding claims that readers adopt different perspectives to make sense of the complex actions described in the sentence. Readers might first adopt the

perspective of the eye and imagine scanning the scene from the fore ground to the horizon. This spatial perspective provides an interpretation of the phrase “As far as the eye can see.” The spatial perspective required to imagine “stalks of corn” necessitates a shift from our point of view, as readers understand the corn stalks to be a distributed figure located across the vast ground. Readers next view the stalks as bending, which arises from the secondary spatial perspective suggested by “under the battering force,” and then elaborate upon the shift of perspective to the “surging curtains of rain.” Each perspective shift, therefore, is guided by specific words, such as “as far as” and “under.” In general, people’s embodied comprehensio comprehension n of this sente sentence nce req require uiress a shift across four perspectives: “eyes,” “stalks of corn,” “battering force,” and “curtains of rain.” Note that the syntactic form of the sentence, emphasizing “corn” aster the responding external force, shapes the dynami na mic c ch char arac acte r ofsubject sent se nten ence ce proc pr oces essi sing ngtoasan a se seri ries es of em embo bodi died ed pe pers rspe pect ctiv ivee shifts. Cons Co nsid ider er no now w a di diff ffer eren entt se sent nten ence ce,, th this is on onee wi with th a me meta taph phor oric ical al co cont nten ent: t: “Casting furtive glances at the seamstress, he wormed his way into her heart” (MacWhinney, 1998). Readers first adopt the perspective of the implied subject and imagine him, or her, casting glances at the seamstress. After this, readers shift to the subject’s embodied action of worming (i.e., moving as if a worm), yet soon recognize that the action here is not literal but metaphorical in the sense of the subject trying to place himself  inside the seamstress’s heart. Even here, readers soon comprehend that  by metaphorically inserting himself into the seamstress’ heart, he has really placed himself closer to the seamstress’ emotions and affections. The embodied action of worming, again, suggests a slow, deliberate process of becoming emotionally closer to the seamstress, which she implicitly accepts, allowing the suitor to enter into her affections. The different spatial perspectives readers readers adopt as they comprehend the sentence give rise to a rich, embodied interpretation. This approach to language understanding argues that people create meaningful construals by simulating how the objects and actions depicted in language relate to embodied possibilities. Thus, people use their

 

Language and Communication

 

201

embo bodi died ed ex expe peri rien ence cess to “s “sof oftt-as asse semb mble le”” me mean anin ing, g, ra rath ther er th than an me mere rely ly ac ac-em tivate pre-existing abstract, conceptual representations. Empiri Emp irical cal re resea searc rch h ind indica icates tes tha thatt re reade aders rs use spa spatia tiall per perspe specti ctives ves to con con-struct mental models of narrative texts. One study asked participants to first memorize the layout of unnamed rooms in a building, along with objects in the rooms (Morrow, Bower, & Greenspan,  1989). Afterwards, part pa rtic icip ipan ants ts rea ead d a st stor ory y de desc scri ribi bing ng a pe pers rson on’s ’s mo move veme ment ntss th thro roug ugho hout ut th thee  building. At various points when reading the story story,, participants were asked to ju judg dgee th thee lo loca cati tion on of sp spec ecifi ificc ob obje ject cts. s. Th Thee res esul ults ts sh show owed ed th that at pe peop ople le we werre quicker to make these judgments when the objects were located in rooms visited by the protagonist. Thus, participants constructed a spatial model of th thee na narr rrat ativ ivee by ad adop opti ting ng th thee em embo bodi died ed pe pers rspe pect ctiv ivee of th thee pe pers rson on in th thee story,, and not by simply creating an objective sketch of the rooms and the story objects in them. Other work provides evidence of spatial indexing and motor participa-

tion in a va tion vari riet ety y of lan langu guag agee co comp mprreh ehen ensi sion on an and d me memo mory ry ta task sk (R (Ric icha hard rdso son n & Spivey, 2000; Spivey & Geng, 2002). For example, when participants are faci fa cing ng a bl blan ank k sc scrree een n (o (orr ha have ve th thei eirr ey eyes es cl clos osed ed)) an and d ar aree li list sten enin ing g to a sc scen enee descri des cripti ption on tha thatt con contai tains ns spa spatiot tiotemp empora orall dyn dynami amics cs in one par partic ticula ularr dir direcection,theireyemovementstendtowardthatdirectionmoresothatanyother (Spi (S pive vey y & Ge Geng ng,, 2001). Fu Furt rthe herm rmor ore, e, wh when en pe peop ople le at atte temp mptt to re reca call ll se sema manntic properties of a visual object or of a spoken factoid that was previously pres pr esen ente ted d in a pa part rtic icul ular ar lo loca cati tion on of th thee co comp mput uter er sc scre reen en,, th they ey te tend nd to fix fixat atee on that (now empty) location in the display (Richardson & Spivey,  2000). Spatia Spat iall in info form rmat atio ion n se seem emss cl clos osel ely y ti tied ed to a ra rang ngee of me ment ntal al rep epre rese sent ntat atio ions ns,, even ev en th thos osee fo forr wh whic ich h th thee sp spat atia iall pr prop oper erti ties es ar aree ar arbi bitr trar ary y or ir irre rele leva vant nt,, su such ch as a random factoid delivered by a talking head in a particular corner of  the computer screen. Readers construct mental models for narrative by adopting the perspective of the protagonist. Participants in one study read texts describing a protagonist and a target item, such as a jogger and sweatshirt (Glenberg, Meyer, & Lindem,  1987). The protagonist and target item in one experimental condition were spatially linked (e.g., the jogger put on the sweatshirt before jogging), whereas in a different condition, the two were dissociated (e.g., the jogger took off the sweatshirt before jogging). After reading the main part of the story, participants judged whether they earlier read the word “sweatshirt.” People were faster to make this  judgment in the linked, or associated, condition than when the protagonist and target were dissociated. Thus, readers appear to create mental models for narrative in which spatial information (e.g., the location of the sweatshirt) is tied to the story’s characters and their embodied actions. Research also shows that readers construct fairly elaborate, embodied microworlds when comprehending literary stories (Zwaan, Magliano, &

 

202

 

Embodiment and Cognitive Science

Graesser,  1995). One study had college students read short literary storiess to det rie determ ermine ine whe whethe therr re reade aders rs aut automa omatic tically ally ima imagin ginee dif differ ferent ent pos possib sible le dimensions of the microworld, including information about the characters, temporality, spatiality, causality, and intentionality (i.e., the characters’ goals and plans). Examination of the time needed to read portions of  these stories revealed increased reading times for sentences when a new character entered the microworld, when there there was a significant gap in the story’s timeline (e.g., flashforwards and flashbacks), when the spatial setting changed, when a story action was not causally related to the prior context, and when a character generated a new plan or goal. These data supporttheideathatreadersactually“fleshout”importantembodiedcharacters of the stories they read, which takes varying amounts of cognitive effort. Other empirical findings also show that people assume the perspective of th thee pr prot otag agon onis istt wh when en rea eadi ding ng na narr rrat ativ ives es.. Th Thus us,, pa part rtic icip ipan ants ts in on onee st stud udy y were faster to pronounce the single word “sat” having just read “After standing through the three-hour debate, the tired speaker walked over to his chair” than when they had just read “The tired speaker moved the chair that was in his way and walked to the podium to continue the t he three-

hour debate” (Keefe & McDaniel, 1993; O’Brien & Albrecht, 1992). Notice, again, aga in, how dif differ ferent ent asp aspect ectss of an obj object ect’s ’s (e. (e.g., g., cha chair’ ir’s) s) af affor fordan dances ces bec become ome prom pr omin inen entt de depe pend ndin ing g on th thee ty type pe of bo body dy ac acti tion on th thee pr prot otag agon onis istt wa wass lik likel ely y to perform (i.e., sitting). Readers’ creation of embodied representations nott on no only ly in influ fluen ence cess th thei eirr un unde ders rsta tand ndin ing g of a pr prot otag agon onis ist’ t’ss ac acti tion ons, s, bu butt als also o shapes their understanding of the orientation of objects (Sanford & Moxy, Moxy, 1995).

Finally, a specific type of embodied knowledge that is useful in text processing is “scripts” (Schank & Abelson,  1976). Scripts consist of welllearned scenarios describing structured embodied situations in everyday life.. Man life Many y stu studie diess sho show w tha thatt re reade aders rs aut automa omatica tically lly inf infer er app appro ropri priate ate scr script ipt-rela re late ted d ac actio tions ns wh when en th thes esee ar aree no nott ex expl plic icitl itly y st stat ated ed (A (Abb bbot ott, t, Bl Blac ack, k, & Sm Smit ith, h, 1985; Bower, Black, & Turner, 1979; Gibbs & Tenney, 1980; Graesser et al., 1980). Oth Other er exp experi erimen ments ts re revea veall tha thatt pri prior or act activa ivatio tion n of scr script ipt-ba -based sed kno knowlwledge provides readers with a highly available set of causal connections that can facilitate sentence-by-sentence integration (Bower et al.,  1979; Garrod & Sanford,  1985;  Sanford & Garrod,  1981; Seifert, Robertson, & Black, 1985; Sharkey & Sharkey, 1987). One difficulty with the idea of script-based narrative understanding is that these categorizations are often too rigid to accommodate variations from fr om wh what at ma may y be ty typi pica call lly y ex expe pect cted ed.. Fo Forr ex exam ampl ple, e, we do no nott us usua uall lly y ha have ve a “w “wal alki king ng in na natu ture re an and d ha havi ving ng a pe pers rson onal al rev evel elat atio ion” n” sc scri ript pt.. A so solut lutio ion n to this th is pr prob oble lem m as assu sume mess th that at sc scri ript ptss “d “do o no nott ex exis istt in me memo mory ry as pr prec ecom ompi pile led d chunks” (Schank, 1982: 16). Instead, the different parts of a script may be reconstructed reconstruc ted depending on the context.

 

Language and Communication

 

203

There are two kinds of high-level processing mechanisms that allow us to create the right script at the right time during text understanding: memory organization packets (MOPs) and thematic organization packets (T (TOPs OPs)) (Do (Downi wning, ng, 2000; Schank, 1982). MO MOPs Ps ar aree pr proc oces essi sing ng st stru ruct ctur ures es that allow people to relate new information with existing expectations to generate reasonable predictions about future events. TOPs are related to MOPs, but are specifically abstractions that allow people to establish connect ne ctio ions ns be betw twee een n di diff ffer eren entt ev even ents ts an and d di disc scov over er si simil milar ariti ities es be betw twee een n th them em.. Thus Th us,, rea eadi ding ng “W “Wes estt Si Side de St Stor ory” y” ma may y rem emin ind d us of “R “Rom omeo eo an and d Ju Juli liet et”” be be-caus ca usee of th thee si simi mila lari rity ty be betw twee een n th thei eirr go goal alss (e (e.g .g., ., mu mutu tual al go goal al pu purs rsui uit) t),, co connditions (e.g., outside opposition), and specific features (e.g., young lovers, false report of death). In this way, TOPs are not static memory representations of abstract prototypical categories, but are processing capabilities that allow readers to be creative in their understandings of events, such as those encountered in literary texts. One analysis of the American novel   Catch-22 (Heller, Catch-22  (Heller,   1961)  demonstrates the power of MOPs and TOPs to provide coherence to disparate texts (Downing,  2000).  Catch-22  Catch-22 is  is the story of an American bombardier squadron during World War II on an imaginary island, Pianosa, off the coast of Italy Italy.. The novel describes, often quite humorously, humorously, the contradictions and absurdities of war and America’s military-industrial power in the twentieth century century.. Consider one excerpt from the novel:

Sharingatentwithamanwhowascrazywasn’teasy,butNatelydidn’tcare.Hewas crazy,, too, and had gone every crazy e very free day to work on the officers’ club that Yossarian had not helped h elped build. Actually Actually,, there were many officers’ clubs that tha t Yossarian Yossarian had not helped build, but hehis was proudest of the one on .Pianosa. It a sturdy and complex comple x monume monument nt to powers of determ determination ination. Yossaria ossarian n was never went there to help until it was finished – then he went there often, so pleased was he with the large, fine, rambling, shingled building. It was truly a splendid structure, and Yossarian throbbed with a mighty sense of accomplishment each time he gazed at it and reflected that none of the work that had gone into it was his. (Heller, 1964: 28)

This excerpt illustrates an incompatibility between two MOPS: “help  build officer club” and “refuse to cooperate.” But Nately simultaneously holds both these conflicting beliefs and does so with great pride. This contradiction differs from our usual expectation that people feel pride when they actually do something to achieve a goal. But readers resolve this contradiction by creating a meta-MOP “Pride” where the prototype and the expe ex pect ctat atio ions ns us usua ually lly as asso soci ciat ated ed wi with th it ar aree no nott fu fulfi lfille lled. d. In th this is wa way y, re read ader erss “soft-assemble” a new concept with its own prototypical structure. This Th is an anal alys ysis is do does es no not, t, ho howe weve verr, ex expl plai ain n wh why y rea eade ders rs pe perrce ceiv ivee th this is pa passsage from Heller’s Heller’s nove novell as funny and informative. informative. But TOPs are useful for this purpose, because they establish connections between apparently

 

204

 

Embodiment and Cognitive Science

unconnected schemas. Thus, readers of the above passage must draw a novel connection between two war situations: one where an officers’ club is bu built ilt,, an anot othe herr wh wher eree co coop oper erat atio ion n wi with th th thee en enem emy y ta take kess pl plac ace. e. Re Read ader erss do this by creating an analogy between competing with the enemy and cooperating in the building of an officer’s club. Both events are then perceived as th them emat atic ical ally ly rel elat ated ed an and d ne nega gativ tive. e. Th This is an analo alogy gy be betw twee een n th thee en enem emy y an and d thee hi th high gher er of offic ficer erss refl eflec ects ts a pa para rall llel elis ism m th that at is rec ecur urrren entt in th thee no nove vell an and d is explicitly pointed out by Yossarian Yossarian when he says “The enemy is anybody who is going to get you killed, no matter what side he’s on.” The incongruity of this analogy helps account for the humorous nature of Yoss ossari arian’ an’ss pr predi edicam cament ent.. Alt Althou hough gh coo cooper perati ating ng wit with h the ene enemy my is a ver very y seri se riou ouss ma matt tter er,, or to pu putt it in ot othe herr wo worrds ds,, it is so some meth thin ing g im impo port rtan ant; t; bu buil ildding in g an of offic ficer ers’ s’ cl club ub,, by co comp mpar aris ison on,, is a tr triv ivia iall ma matt tter er.. Un Unde ders rsta tand ndin ing g th this is specific theme is representative of a broader theme repeated throughout theThis novel, where trivial reveal a more dramatic background. discussion of onesituations portion of the novel Catch-22 novel  (offered by Down Catch-22 (offered ing,  2000)  demonstrates the adaptive character of many aspects of text understanding. Skilled readers do not comprehend texts by simply activating pre-existing prototypes in the form of scripts. Instead, prototypical understandings arise as the products of dynamic and embodied meaning construction processes (in this case through the interaction of MOPs and TOPs). There are other reasons to doubt whether people activate pre-stored prototypical knowledge, such as scripts, when understanding texts. Most prototype event sequences (e.g., the events that occur at a restaurant) are typically part of common knowledge within a given culture. However, routin ro utinize ized d seq sequen uences ces of eve events nts can als also o be qui quite te idi idiosy osyncr ncratic atic.. For exa exampl mple, e,

my fr frie iend nd Jo John hn reg egul ular arly ly ge gets ts up at 5:30 a. a.m. m.,, dr drin inks ks a gl glas asss of to toma mato to ju juic ice, e, puts pu ts ou outt th thee ca cat, t, and goe oess jo jogg ggin ing. g. Whe hen n he fin finis ishe hess runn nnin ing g, Jo Joh hn put utss on the coffee, shaves, brushes his teeth, and sits down to read  17th-century Engli En glish sh po poet etry ry.. Al Alth thou ough gh Jo John hn en enga gage gess in th this is se sequ quen ence ce of ev even ents ts ea each ch da day y, the sequence is unlikely to be performed by anyone but him (Colcombe & Wyer, 2001). Do people form person-specific prototypes that are then used to comprehend new experiences that exemplify them? One may believe that prototypes for the self’s and well-known others’ actions may be the most enduring scripts, given that they are continually embodied in everyday life. Butt as ar Bu argu gued ed in Ch Chap apte terr 4, pr proto ototyp types es nee need d not nec necess essari arily ly be pr pre-e e-exis xistin ting g mental representations that get activated to facilitate processing of related events. In fact, research shows that when individuals read descriptions of  events that pertain to themselves or to a familiar other (i.e., a parent or roommate), they do not activate a prototypical representation of the behavior in interpreting these events (Colcomb & Wyer,  2001). This is true regardless of whether the sequence is similar to one that they personally

 

Language and Communication

 

205

experience or observe on a daily basis, or whether it exemplifies a more general prototype of the events that occur in a particular type of situation (e.g., cashing a check). Thus, even if prototypical sequences of embodied actions (i.e., a script) exist in memory and are used to comprehend the  behavior of unfamiliar persons, individuals do not apply these when compreh pr ehen endi ding ng ev even ents ts th that at pe pert rtai ain n to th them emse selv lves es or so some meon onee wi with th wh whom om th they ey are familiar. familiar. These recent research findings suggest, at the very least, that peop pe ople le do no nott ne nece cess ssar arily ily fo form rm ab abst stra ract ct rep eprres esen enta tatio tions ns fo forr th thee mo most st fa fami milliar event sequences they experience in daily life. People may alternatively crea cr eate te pr prot otot otyp ypic ical al sc scen enar ario ioss on th thee fly as pa part rt of th thee em embo bodi died ed si simu mula lati tion onss they engage in during discourse processing. A Case Study: The Indexical Hypothesis

Thee cl Th clai aim m th that at em embo bodi dime ment nt un unde derl rlie iess pe peop ople le’s ’s un unde ders rsta tand ndin ing g of la lang ngua uage ge  1997,  1999 is nicely developed in the “indexical hypothesis” Glenberg & Robertson,  2000 ). This view assumes (Glenberg, that three major steps; occur when language is understood in context. First, words and phrases aree in ar inde dexe xed d to ob obje ject ctss in th thee en envi virron onme ment nt or to pe perrce cept ptua uall sy symb mbol olss in lo long ng-term memory. Second, the affordance structures (i.e., the possible actions that can be done to an object by a person) are derived for each object in the situation. Third, the listener must combine or “mesh” the affordances according to the constraints on embodied possibilities in the real world. For instance, the affordances of a chair include those of sitting on it, or using it to hold off a snarling lion, but they cannot ordinarily be meshed with the goal of propelling oneself across a room. This constraint on the meshing of embodied affordances predicts that people will have an easier time understanding the sentence “Art used the chair to defend himself  against the snarling lion” than interpreting the sentence “Art used the chair to propel himself across the room.” One empirical test of the indexical hypothesis asked people to judge

the me the mean anin ingf gful ulne ness ss of af affo forrde ded d (i (i.e .e., ., wi with th co cohe herren entt me mesh sh)) an and d no nona naff ffor orde ded d (i.e., those without coherent mesh) sentences, such as those shown above. The results, not surprisingly, surprisingly, showed that people judged the afforde afforded d sentenc te nces es to be si sign gnifi ifica cant ntly ly mo morre se sens nsib ible le th than an th thee no nona nafffo ford rded ed on ones es.. Fu Furt rthe herrmore, a second study measured the speed with which people read these two types of sentences. This study showed that nonafforded sentences took significantly longer to comprehend than the afforded ones. These psycholinguistic findings highlight the importance perceptual, embodied information when people combine conceptual representations during ordinary language understanding. Consider the sentence “John scratched his back with a floppy disk.” One might object that we understand this sentence, but not the sentence “John scratched his back with thread,” because of background

 

206

 

Embodiment and Cognitive Science

knowledge. Glenberg and Robertson (2000) demonstrated that this backgroun gr ound d kno knowle wledge dge can cannot not be pr pre-s e-stor tored ed pr propo oposit sition ionss fr from om whi which ch re relat lation ionss (e.g., a disk can be used for back-scratching) are inferred inferred by a formal process. First, understanding the experimental sentences did not seem to depend pe nd on ha havi ving ng ha had d ex expe peri rien ence cess si simi mila larr to th thos osee de desc scri ribe bed d by th thee se sent nten ence cess (e.g., experience using a disk to scratch one’s back). The stimuli used by Glenb Gle nber erg g an and d Ro Robe bert rtso son n (2000) de desc scri ribe bed d no nove vell si situ tuat atio ions ns (e (e.g .g., ., in an anot othe herr scen sc enar ario io a pe pers rson on us uses es an up upri righ ghtt va vacu cuum um cl clea eane nerr as a co coat at tr tree ee)) wi with th wh whic ich h thee re th read ader erss we werre un unlik likel ely y to ha have ve ha had d ex expe peri rien ence ce.. Se Seco cond nd,, as asso soci ciat ativ ivee re rela la-tion ti onss be betw twee een n co conc ncep epts ts th that at me mesh sh su such ch as flo flopp ppy y di disk sk an and d ba back ck-s -scr crat atch chin ing g are no stronger from the associative relations between concepts that did nott me no mesh sh,, su such ch as th thre read ad an and d ba back ck-s -scr crat atch chin ing. g. No None neth thel eles ess, s, re read ader erss wo woul uld d readily reject the sensibleness of sentences formed from nonmeshing concepts. Third, people required about the same time to read and understand sent se nten ence cess th that at su supp ppos osed edly ly re requ quir ired ed fo form rmal al in infe fere renc nces es as se sent nten ence cess th that at di did d not (e.g., She used the back scratcher to scratch her back). Fourth, understanding innovative denominalknowledge verbs (e.g.,because to crutch) which  be no pre-stor pre-stored ed propositional the for verbs werethere usedcan for the first time in the experiment. Thus, although background knowledge must be used in understanding, the background knowledge appears to be very flexible rather than pre-stor pre-stored. ed. This flexibility is provided by perceptual symbols from which new affordances can be derived to potentially mesh, under the guidance of syntax. Language at many levels consists of instructions for constructing an embo em bodi died ed me ment ntal al si simu mula lati tion on of wh what at th thee la lang ngua uage ge is ab abou outt (t (tha hatt is is,, an em em- bodied mental model). Thus, noun phrases are instructions for retrieving (indexing) representations from which affordances can be derived (Glen berg & Robertson, 2000), verbs of manner are instructions for retrieving motor programs or plans that can potentially select and act on those affordances dan ces,, ver verb-a b-arg rgume ument nt con constr struct uction ionss (e. (e.g., g., the dou double ble obj object ect con constr struct uction ion)) provide a general framework (e.g., transfer) that must be accomplished when simulating the effect of the verb (Kaschak & Glenberg,  2000), and temp te mpor oral al ad adve verb rbss pr prov ovid idee in inst stru ruct ctio ions ns fo forr co cont ntro rolli lling ng th thee ma mann nner er in wh whic ich h multiple models are combined. Thus, “while” is an instruction to mesh or simulate how two actions can be performed simultaneously, “after” is an

instruction to simulate the current clause and then the next, and before is an instruction to simulate the first clause, simulate the second clause, and then to check that the simulation of the first clause (temporally second event) will mesh with the end-state of the simulation of the second clause (temporally first event). This general approach considers language as instructions for constructing simulations. There appear to be text limits, however, on One the extent to which embodied experience constrains comprehension. set of studies investigated the type of inferences people construct when reading (Graesser, (Graesser, Singer, &

 

Language and Communication

 

207

Althou hough gh re reade aders rs eas easily ily dra draw w inf infer erenc ences es abo about ut why Trabasso,   1994). Alt acto ac tors rs/ /ev even ents ts ta take ke pl plac ace, e, an and d wh why y th thee wr writ iter er in incl clud uded ed so some meth thin ing g in a te text xt,, people do not necessarily draw inferences about how some action/event occurred. For example, when readers saw “The cook tripped the butler,” they th ey im imme medi diat atel ely y in infe ferr rred ed so some meth thin ing g ab abou outt wh why y th thee co cook ok di did d wh what at he di did d (e.g., for “revenge”), but readers did not draw specific embodied inference en cess abo bout ut ho how w an ev eveent oc occcur urrred (e (e.g .g., ., th thee co cook ok us useed his “fo foot ot”” to do th thee tripping). Readers construct embodied representations representations for text only to the exte ex tent nt th thos osee in infe ferren ence cess en enab able le th them em to un unde ders rsta tand nd th thee pl plot ot an and d th thee wr writ iter er’s ’s rationaleforincludingsomethinginthetext.Narrativetextsmightbeeasier to un unde ders rsta tand nd th than an ex expo posi sito tory ry on ones es,, pr prec ecis isel ely y be beca caus usee th thee ev even ents ts in na narr rraatives tiv es ar aree mo more re ac actio tionn-dr driv iven en an and d em embo bodi died ed al allo lowi wing ng th them em to tr trac ack k pr prot otag ag-onists’ goals, than are seen in expository writing. These psycholinguistic studies provide additional evidence that people’s construals of meaning are constrained by their recognition of embodied possibilities alluded to  by texts. Conclusion

A common theme in cognitive science is that language understanding is a modular activity with little interaction with conceptual and experiential knowledge, knowledge, espe especially cially during the early stages of pro processi cessing. ng. Even if  contextual information is employed to assist appropriate understanding of what speakers/writers mean, this knowledge is assumed to be represented in an abstract, disembodied format (e.g., list of propositions). This chapter presents a variety of empirical research that points to an opposite conclusion. Embodied activity plays a role in a least some aspects of languag gu agee ev evol olut utio ion, n, th thee pr proc oces essi sing ng of sp spee eech ch an and d wo word rd me mean anin ing, g, ho how w pe peop ople le understand why various words and phrases have the meanings they do, and an d pe peop ople le’s ’s im imme medi diat atee co comp mpre rehe hens nsio ion n of ve verb rbal al ex expr pres essi sion onss an and d wr writ itte ten n discourse. As was seen in Chapter  5 for aspects of higher-order cognition, embodied activity shapes parts of on-line communication and the off-line knowledge that is accessed during linguistic processing. A significant part of embodied activity subserves simulation processes thatt ope tha operat ratee dur during ing lan langua guage ge und unders erstan tandin ding. g. In fac fact, t, lan langua guage ge und unders erstan tandding within real-world communicative contexts may be best described as a kind of embodied simulation, rather than the activation of pre-existing, disembodie disem bodied, d, symb symbolic olic knowledge. knowledge. None of this implies that all aspects of language and communication, including some body movements used

to express meaning, are rooted in embodiment. But there is sufficient evi dence to suggest that many aspects of language and communication arise from, and continue to be guided by, bodily experience.

 

7 Cognitive Development

Developmental psychologists have long debated the role of early embodied ie d ac actio tion n in co cogn gnit itiv ivee de deve velo lopm pmen ent. t. Si Sinc ncee Pi Piag aget et’s ’s (1952, 1954) wri writin tings gs on how ho w se sens nsor orim imot otor or ac activ tivity ity un unde derl rlie iess di diff ffer eren entt as aspe pect ctss of co cogn gnit itiv ivee gr grow owth th,, psychologists have considered ways of linking patterns that emerge from young children’s bodily and perceptual experience with later intellectual development. Many psychologists now argue that children’s acquisition of significant conceptual abilities is rooted in innately given knowledge, whereas others emphasize the child’s active looking and listening skills. Yet ne neit ithe herr of th thes esee po posi sitio tions ns gi give vess ap appr prop opri riat atee at atte tent ntio ion n to th thee ch child ild’s ’s se self lf-generated movements and felt sensations in learning to perceive, think, and engage in intelligent behavior. Scholars such as Dewey ( 1934)   and Montessori (1914) earlier emphasized the importance of “learning by doing,”but there has not been enough concern with how kinesthetic action serves as the potential building block for conceptual development. This chapter describes ideas and empirical evidence in support of the embodied grounding for cognitive development. Piaget’ss Contribution Piaget’

 Jean Piaget’s seminal investigations of child development assumed that growth is a form of individual adaptation to the environment. Even small infants exhibit intelligent behavior, not by thinking, but by acting physically in the world (Piaget, 1952). Two principles of biological adaptation, assimilation and accommodation, provide the mechanism for development of intelligent action. Assimilation refers to the process by which infa in fant ntss us usee th thei eirr ex exis isti ting ng ab abili ilitie tiess wh when en re resp spon ondi ding ng to en envir viron onme ment ntal al ch chal al-lenges. Accommodation refers to the process of changing one’s existing abilities to adaptively deal with some task or situation. Most actions involve vol ve som somee com combin binati ation on of ass assimil imilati ation on and acc accomm ommoda odatio tion. n. For ins instan tance, ce,

 between 6 and 12 mo mont nths hs,, in infa fant ntss le lear arn n to ea eatt so soli lid d fo food ods. s. Th Thee ch chil ild d be begi gins ns 208

 

Cognitive Development

 

209

to assimilate the food by employing the tongue and lip action it uses for  breast-feeding.. But these  breast-feeding these movements are inadequate inadequate for dealing with soft, solid foods, and infants must then coordinate their tongue movements in new ways to accommodate their own actions to the shape of the food, incl in clud udin ing, g, at ti time mes, s, th thee sh shap apes es of sp spoo oons ns.. Bo Both th as assi simi mila latio tion n an and d ac acco comm mmoodation are necessary parts of how the child learns to act in the real world. Piaget’s development theory of intellectual development posited first years stage of sorimotor that occurs in the child’s firsta two of senlife. The sensorimotor stage is rooted in the infant’s embodied exploration of  the environment, or specifically, infants’ growing understanding of their  bodies and how their bodies relate to objects and other people in the world (Piaget & Inhelder, 1969). The sensorimotor stage is divided into six substages: (1)   Re Refle flex x sc sche heme mess (0 –1 mo mont nths hs): ): in inbo born rn refl eflex exes es,, su such ch as su suck ckin ing, g, lo look ok-ing, and crying, establish the infant’s first connections with the world. (2)  Primary circular reactions (1–4 months): repeated actions (circular) indu in duce ce co coor ordi dina nati tion on wi with thin in th thee in infa fant nt’s ’s ow own n bo body dy,, su such ch as co coor ordi dina na-tion of muscles in the mouth to suck on the thumb, where the initial occurrence happen by chance. (3)   Secondary circular reactions reactions (4–8 months): repeated actions involve coordination between the infant’s action and the environment, such as ki kick ckin ing g to ma make ke a ha hang ngin ing g mo mobi bile le mo move ve,, a di disc scov over ery y th that at ha happ ppen enss  by chance. (4)  Coordination of secondary circular reactions (8–12  months): goaldirected actions not motivated by chance occur, such as using one hand to hold an object while the other hand explores it. (5)  Tertiary circular reactions (12–18 months): familiar secondary reactions are used to make new things happen, such as when the child explores how different different objects fall from his or her high-chair high-chair,, which involves trial-and-error problem solving. (6)   In Inve vent ntio ion n of ne new w me mean anss th thrrou ough gh me ment ntal al co comb mbin inat atio ion n (18–24 months): the ability to think before acting by representing actions as mental images or symbols emerges, a process of problem solving without trial and error. Piaget’s theory was developed from his observations of his own young children. For example, evidence for substage   3, secondary circular reactions, is seen in the following observations of  4  4-month-old Laurent: “At 4 months 15 da days ys,, wi with th an anot othe herr do doll ll ha hang ngin ing g in fr fron ontt of hi him, m, La Laur uren entt tr trie iess to grasp it, then shakes himself to make it swing, knocks it accidentally accidentally,, and theen tr th trie iess si simp mply ly to hi hitt it it.. At 4 months 18 da days ys,, La Laur uren entt hi hitt my ha hand ndss wi with th-outt tr ou tryi ying ng to gr gras asp p th them em,, bu butt he st star arte ted d by si simp mply ly wa wavi ving ng hi hiss ar arms ms ar arou ound nd,,

 

210

 

Embodiment and Cognitive Science

and only ly af afte terw rwar ards ds we went nt on to hi hitt my ha hand nds. s. Th Thee ne next xt da day y, fin final ally ly,, La Laur uren entt and on immediately hits a doll hanging in front of him” (Piaget, 1952: 167–8). Piaget interpreted this sequence of events as evidence of infants using their own actions to create interesting effects in the world. This example also al so il illus lustr trat ates es ho how w di dise sequi quilib libri rium um se serv rves es as th thee pr prim imar ary y ca cata taly lyst st of de deve vellopment. Whenever assimilation and accommodation processes fall short of ad adap apta tati tion on,, th thee ex expe peri rien ence ce of di dise sequ quili ilibr briu ium m fo forc rces es th thee ch child ild to di disc scov over er new of knowing, or theschemes, world. But most an infant’s early skillsways and knowledge, knowledge that startimportantly, out as sens sensorimot orimotor or actions serv se rvee as th thee fo foun unda dati tion on fo forr lat later er co conc ncep eptu tual al sc sche heme mes, s, in invo volv lvin ing g id idea eas, s, co conncepts, and thoughts. The cognitive operations used in i n manipulating sym bolic repr representations esentations share a formal structure with sensorimotor activity that is governed by the laws of physics. The logic inherent in coordination of action is presumably reconstructed at the level of internal thought, ultimately enabling objective logico-mathematical knowledge. Thus, as the infant develops, overt actions give way to internal actions on “images of  abstra abs tract ct obj object ectss and the their ir dis displa placem cement ents” s” (Pi (Piage aget, t, 1954: 4). Pi Piag aget et as assu sume med, d, ther th eref efor ore, e, th that at th thee ch child ild’s ’s em embo bodi died ed ac activ tivit itie iess ar aree a ne nece cess ssar ary y co comp mpon onen entt in coming to know the world, but higher-order forms of thought become divorced vor ced fr from om ear earlie lierr sen sensor sorimo imotor tor beh behavi aviors ors as the they y bec become ome “in “inter terior iorize ized.” d.” Recent Studies on Physical Reasoning

Many of Pia Many Piaget get’s ’s ori origin ginal al obs observ ervati ations ons on sen sensor sorimo imotor tor dev develo elopme pment nt hav havee  been more rigorously studied by developmental psychologists over the past few decades. This research suggests that young infants are capable of sophisticated physical reasoning about objects and their properties, far  beyond that observed by Piaget. A main complaint motivating this work is that Piaget confused an infant’s motor competence with its conceptual abilities. Of course, this criticism assumes that what is conceptual may have ha ve li litt ttle le to do wi with th wh what at is mo moto tori ric, c, or mo morre fu full lly y em embo bodi died ed,, ab abou outt yo youn ung g children’s concepts. The question here is whether this new evidence from developmental psychology demonstrates that sensorimotor activity is not a prerequisite for cognitive development. Consider the case of object permanence. Object permanence refers to the child’s understanding that objects continue to exist even when they are hidden from view. Piaget assessed this ability by examining young children’s searching behavior. For example, when a toy is shown to an infant (ages   0   to   4  months) and then hidden under a cushion, the baby fails to seek out the object. Piaget assumed that objects for these babies were not differentiated from their own bodily actions. In the next stage of object permanence, infants begin to search for partially hidden, but not fully hidden objects. Later on, infants can search for fully hidden objects,  but only if objects are hidden in the same spatial location on repeated trials.

 

Cognitive Development

 

211

Thus,, if th Thus thee to toy y is hid idde den n un unde derr a di difffe ferren entt cu cusshi hion on in fu full ll vi view ew of th thee ch chil ild, d, theinfantwillsearchforthetoyunderthecushionwhereithadbeenhidden on previous trials. This “A-not-B” error occurs between  8  and 12  months of age. Eventually, children move into a stage in which the A-not-B error disappears and then can correctly search new locations for objects (when they th ey ch chan ange ge fr from om A to B) B).. No None neth thel eles ess, s, in infa fant ntss ar aree st stil illl un unab able le to se sear arch ch fo forr an object under different cushions without first observing the object being hidden. Finally, Finally, in stage 6 of object permanence, around 15 to 18 months, infants hidden objects untilisthey One systematically difficulty with search Piaget’sfor observational studies thatare thelocated. object concept is measured through analysis of infants’ searching behavior. But infants may possess some object concept, and recognize that objects are different from their own actions, before they can successfully reach for obje ob ject cts. s. On Onee re reas ason on fo forr ba babi bies es ma maki king ng AA-no nott-B B er errror orss in ob obje ject ct pe perm rman anen ence ce tasks is that they fail to inhibit a predominant action tendency to search for the object under the cushion where it was first, but is not now, now, hidden (Diamond, 1991; Reiser, Doxey, McCarrell, & Brooks,  1982). This suggests that babies may have some early object concept, but give misleading evidence of a lack of this concept because of their tendency to reach into the place where the object was first hidden. Significant research using methods that do not require babies to reach for objects suggests that babies may understand a good deal about the existence of objects. For example,   5-month-old infants in one study sat in front of a screen that rotated through   180  degrees toward and away from the baby (Baillargeon, Spelke, & Wasserman,  1985). After the baby habituated to this event, a box was placed in the path of the screen at the far end of the apparatus. When the screen began its 180-degree rotation, it gradually occluded the block, and when it reached  90  degrees, the entire  box was hidden. At this point, babies were shown one of two critical events. “Pos “P ossi sibl blee ev even ents ts”” we were re th thos osee in wh whic ich h th thee sc scrree een n ro rota tate ted d un unti till it ha had d pa pass ssed ed through  180  degrees, when it stopped, having apparently made contact with the box. “Impossible events” were those in which the screen rotated unti un till it ha had d pa pass ssed ed th thrrou ough gh th thee fu full ll 180-de -degr gree ee ro rotat tation ion,, app appar arent ently ly hav having ing  180 moved throug through h mili where wher the was located. Althoug Although hme thelo full degr de gree ee rright otat ot atio ion n wa wass fa fami liar areto theebox th babi ba bies es,, th they ey sp spen ent t mo morre ti time look okin ing g atthee im th impo poss ssib ible le ev even entt th than an th thee po poss ssib ible le on one. e. Th This is res esul ultt su sugg gges ests ts th that at ev even en 5monthmon th-old oldss und unders erstan tand d som someth ething ing abo about ut obj object ect per perman manenc ence, e, bec becaus ausee the they y seem se em su surp rpri rise sed d th that at th thee sc scre reen en co could uld pa pass ss th thrrou ough gh th thee bo box x in th thee imp impos ossi sibl blee event. One difficulty with Baillargeon et al.’s (1985)  study is that the experimental task may lead babies to form a strong perceptual expectation that the rotating screen will stop. This expectation may not require that the infant form a representation of the occluded object. Some scholars claim that infants may have some knowledge about the continued existence of 

 

212

 

Embodiment and Cognitive Science

occluded objects, but may not set up repr representations esentations of individual objects until around 1 year of age. To test this idea, Xu and Carey (1996) compared 10-month-old babies’ reactions to two occluding conditions, a property-kind condition and a

spatio-temporal condition. In the first case, infants sat in front of a screen, and a truck was brought out from the right side and then returned behind it. Similarly, a toy kitten was brought out from the left side of the screen, and then moved behind it. Babies habituated to these repeated events. The screen was then removed and either one or two of the toys appeared. For this the spatio-temporal condition, the same sequence of events occurred,  but time the truck and kitten were brought out simultaneously from  behind the screen. Finally, Finally, a baseline condition simply measured infants’ looking times for one or two objects that were not occluded. Analysis of babies’ looking times showed that they stared longer at the two-object outcome in the baseline and property-kind conditions, but looked longer at the single-object outcome in the spatio-temporal condition. This pattern of data indicated that infants generally prefer looking at two objects, but could overcome this preference in the spatio-temporal condition. Xu and Carey argued that the   10-month-olds were unable to perc pe rcep eptu tual ally ly di dist stin ingu guis ish h be betw twee een n th thee to toy y an and d ki kitt tten en to re reco cogn gnize ize th that at th ther eree were two different objects hidden behind the screen. Thus, young infants may only possess a generalized understanding of objects. Only later will infants be able to represent the specific identities of objects. This work is representative of studies examining infants’ physical reasoni so ning ng in th that at ch child ildrren en’s ’s se sens nsor orim imot otor or be beha havi vior or an and d ex expe peri rien ence ce is as assu sume med d to pl play ay a mi mini nima mall rol olee in ho how w th theey co come me to acq cqu uir iree con once cept ptua uall kn know owle led dge about objects and their properties. In fact, eliminating the child’s reaching in standard object permanence tasks is seen as a good way of removing motor behavior from the assessment of conceptual knowledge. Not surprisingly, perhaps, there are criticisms of studies employing the preferential looking method, particularly in regard to whether these measures adequately assess infants’ cognitive, as opposed to purely perceptual, capacities (Bogartz & Shinsky, 1998; Haith, 1997). But a large number of habituation-dishabituation studies demonstrate that youngtheir infants recognize many other object properties and the rules governing behavior. Thus,  3 -to4-month-olds are sensitive to object substance and different physical limits on object motion, such as that one solid object cannot move through another, another, and that an object much larger than an opening cannot pass through it (Spelke et al.,  1992). Babies at this age also recognize something about the effect of gravity and will look surpri sur prised sed whe when n a mov moving ing obj object ect sto stops ps in mid midair air wit withou houtt sup suppor portt (Si (Sitsk tskoor oorn n & Smitsman, 1995). Even 2 1/2-month-old babies expect a stationary object to move when a moving object collides with it (ref). Babies  3   1/2 months old recognize when an object is compressible or not (e.g., a sponge vs. a

 

Cognitive Development

 

213

wooden block) or when it is taller or shorter than the height of a screen (Baillargeon, 1987a, 1987 b  b)). By 5–6 mo mont nths hs of ag age, e, in infa fant ntss kn know ow th that at a la larg rger er moving object can cause a stationary object to travel further (Baillargeon, 1994. Beginning around 6 months of age, infants recognize that one object placed on top of another will fall unless a larger portion of its bottom surfac sur facee con contac tacts ts the low lower er obj object ect (Ba (Bailla illarg rgeon eon,, 1994 Baillar Baillargeon, geon, Need Needham, ham, & DeVos, 1992). Habituation studies also show that young infants not only may under-

stand stan d so some meth thin ing g ab abou outt ob obje ject ctss an and d th thei eirr pr prop oper erti ties es,, bu butt al also so ma may y re repr pres esen entt spatial relations between objects. For instance, in once study,  5 .5-montholds were habituated to displays of a tall rabbit and a short rabbit passing around a screen and appearing on the other side (Baillargeon & Graber, 1997). In Infa fant ntss we werre th then en sh show own n th thee sa same me di disp spla lay y ex exce cept pt th that at a wi wind ndow ow ha had d  been cut off the top part of the screen. If a tall rabbit has traveled the area  behind the screen, then it should appear in the window. window. In fact, when the short rabbit passed behind the modified screen, infants continued to ha bituate. However, However, when the tall rabbit passed behind the modified modified screen with wi thou outt ap appe pear arin ing g in th thee wi wind ndow ow (a (an n im impr prob obab able le ev even ent) t),, in infa fant ntss di dish shab abitituated (i.e., showed longer looking times) Infants presumably developed expectations that an object moving in a visible trajectory would appear when it was briefly occluded. A co cont ntrrol gr grou oup p of ba babi bies es we werre tr trea eate ted d id iden enti tica call lly y ex exce cept pt th that at be befo forre th thee experimentbegantheybrieflysawtworabbitsstandingoneithersideofthe screen. These infants did not dishabituate to the impossible event. Infants seemingly understood the display in terms of a rabbit passing behind the scrree sc een n an and d st stop oppi ping ng,, wh while ile an anot othe herr ra rabb bbit it ap appe pear ared ed at th thee ot othe herr si side de.. La Late terr work showed that  3 1/2-month-old infants behave in the same manner (Baill (Ba illar argeo geon n & DeV DeVos, os, 1991). Th Thus us,, ve verry yo youn ung g in infa fan nts ap app pea earr to be ab able le to represent repr esent spatial relations. Finally, Finally, differ di fferent ent studies reveal that 10-monthold babies can remember the location of objects when they are out of sight for as long as 70 sec second ondss (Ba (Bailla illarg rgeon eon,, DeV DeVos, os, & Gra Graham ham,, 1989). An And d whe hen n watching a blue ball disappear behind a screen with a red ball appearing on the other side,   10-month-olds infer that the first object launched the second object behind the partition (Cohen & Oakes,  1993). Thee st Th stud udie iess me ment ntio ione ned d he herre, ag agai ain, n, rep eprres esen entt on only ly a fe few w of th thee do doze zens ns of  experiments showing young have significantly more sophisticate ca ted d kn know owle ledg dgee ab abou outthat t ob obje ject ctss an and dinfants thei th eirr pr prop oper erti ties es th than an cla claime imed d by Piage Pia get. t. Ther Th eree is an em emer ergi ging ng bo body dy of wo work rk,, ho howe weve verr, sh show owin ing g th that at ol olde derr ch chil ildr dren en,,  between 2 and 3 years of age, lack aspects of this knowledge. For instance, in a task where  2- and 3-year-olds had to find a ball after watching it roll  behind a screen and stop, toddlers under under 3 years old performed no better than would have been expected if they were simply guessing at the ball’s location (Berthier et al.,  2000). Follow-up studies provided the children with more visual information about the ball’s trajectory by replacing the

 

214

 

Embodiment and Cognitive Science

opaque window screen with a transparent one (Butler, (Butler, Berthier, Berthier, & Clifton, 2002). This additional visual information did not help the 2-year-olds, but did assist the  2   1/2-year-old toddlers to some degree. Finally, even when the children were given a full view of the ball’s trajectory until it came to rest against a wall, the majority of  2  2-year-olds still could not find the ball. Analysis of the children’s gaze showed that if they looked at the ball as the screen was lowered and then fixated on it until the door opened, they were correct almost 90% of the time. Thes Th esee fin findi ding ngss su sugg gges estt th that at to todd ddle lers rs do no nott ha have ve kn know owle ledg dgee of co cont ntin inuuity and solidity in the way that younger infants appear to have it, at least as me meas asur ured ed by ha habi bitu tuat atio ion n ta task skss wh when en ch chil ildr dren en lo look ok lo long nger er at im impo poss ssib ible le even ev ents ts.. On Onee po poss ssib ibil ility ity is th that at to todd ddle lers rs’’ pr prob oble lems ms in th thee se sear arch ch ta task sk req equi uire re

them to predict where the ball should be, something that is not measured in typical habituation studies. Moreover, Moreover, the toddlers in i n the above studies had to coo coord rdina inate te the their ir pr predi edicti ctions ons wit with h the app appro ropri priate ate act action ionss of re reach aching ing out for the ball at the right spots. spots. It appea appears rs then, that 3  to  4-month-olds can reason about an object’s motion being constrained by continuity and solidarity, as demonstrated by habituation studies, but they cannot yet reason about after-the-fact incongruous events (Keen,  2003). Infants’ percept ce ptua uall re reco cogn gniti ition on of im impo poss ssib ible le ev even ents ts in ha habi bitu tuat atio ion n st stud udie iess ma may y re refle flect ct  just one small part of young children’s eventual ability to make predictions about objects and events in the real world.

Three Theories of Cognitive Development: Does Experience Matter? The question, again, remains whether success on any of the experimental tasks described above depends on sensorimotor experience. Piaget may have drastically underestimated infants’ concepts and physical reasoning skills,, but this does not neces skills necessarily sarily imply that embod embodied ied experience experience has littl lit tlee rol olee in ve very ry ea earl rly y co conc ncep eptu tual al de deve velop lopme ment nt.. So Some me ps psyc ycho holo logi gist stss ar argu guee that very young babies possess considerable knowledge about objects beforre th fo they ey ha have ve at atta tain ined ed th thee ca capa paci city ty to ph phys ysica ically lly ma mani nipu pula late te th them em ar arou ound nd the age of  6  6 months. This suggests to these scholars that infants’ developing in g ph phys ysic ical al re reas ason onin ing g sk skill illss mu must st be ba base sed d on in inna nate te,, mo modu dula larr kn know owle ledg dge, e, which wh ich be beco come mess mo more re el elab abor orat atee as ch chil ildr dren en co come me in into to co cont ntac actt wi with th di diff ffer eren entt aspects of the physical world (Leslie, 1994; Spelke, 1988, 1990, 1991). Un-

der this view, infants are born with substantial beliefs about how objects move mo ve in co cont ntin inuo uous us pa path thss an and d do no nott ch chan ange ge sh shap apee or pa pass ss th thrrou ough gh on onee an an-other,, based on unchanging, possibly innate, principles such as cohesion, other  boundedness, and rigidity (Spelke, 1994; Spelke & Newport, 1998). These initial perceptually based descriptions of concepts are gradually refined as the child learns to attend to relevant features of the environment. Some of th thes esee ea earl rly y co conc ncep epts ts ma may y be mo modu dula larr in be bein ing g in info form rmat atio iona nall lly y en enca caps psuulate la ted d fr from om ot othe herr ki kind ndss of ph phys ysic ical al an and d sp spat atia iall kn know owle ledg dgee (e (e.g .g., ., a ge geom omet etri ricc

 

Cognitive Development

 

215

spatial module; see Herner & Spelke, 1994, but also see Newcombe,  2002 for evidence refuting this proposal). A slightly different perspective on conceptual development claims that theinfantisendowedwithcertaindomain-specificbiases(ratherthanmodules ul es), ), wh whic ich h in in inte tera ract ctio ion n wi with th th thee ex exte tern rnal al en envi viro ronm nmen entt be beco come me pr prog ogrres es-sively modularized, or specialized to acquire knowledge as development proceeds (Karmiloff-Smith,  1992). Although infants (like older children) use domain-general processes, such as representational redescription, to recode sensorimotor input into accessible formats across a variety of domains, it is unlikely that there are domain-general stages of change such as proposed by Piaget. Thus, infants possess at birth the primitive forms of knowledge that are present in adults, with development occurring in a gradual, continuous manner (Case, 1992; Karmiloff-Smith, 1992). A third perspective on the development of infants’ physical reasoning claims that children do not have innate beliefs about objects. Instead, infants come into the world with highly constrained mechanisms that guide

their reasoning about objects (Baillargeon,  1994,  1995,  2000). Infants first learn preliminary aspects of a concept in an all-or-none manner that capture its essence, and later on begin to identify discrete and continuous variables relevant to the concept to form a more elaborate conceptual representation. Some So me of Ba Baill illar arge geon on’s ’s st stud udie iess de desc scri ribe bed d ab abov ovee pr prov ovid idee ev evid iden ence ce in su suppportt of thi por thiss cla claim. im. Inf Infant ants’ s’ re reaso asonin ning g abo about ut sup suppor port, t, col collis lision ion,, and unv unveili eiling ng rel elat atio ions ns sh show owss ho how w in init itia iall co conc ncep epts ts ar aree rev evis ised ed ov over er ti time me to pr prov ovid idee mo morre elaborated concepts. Thus, the concept of support is first understood in an all-or-none manner as contact or no contact. With greater perceptual experience, primarily through looking, infants incorporate discrete (e.g., locus of support) and continuous (e.g., amount of support) information. Bail Ba illa larg rgeo eon n an and d co coll llea eagu gues es cl clai aim m th that at th this is ac acqu quis isit itio ion n se sequ quen ence ce is no nott du duee to thee gr th grad adua uall un unfo fold ldin ing g of so some me in inna nate te be beli lief ef ab abou outt su supp ppor ort, t, bu butt ar aris ises es fr from om constr con strain ained ed lea learni rning ng mec mechan hanism isms, s, suc such h tha thatt inf infant antss mak makee app appro ropri priate ate gen gen-eralizations about objects and their behaviors in an ordered manner. For example, infants recognize that a small object cannot pass through a gapless opening before understanding that a large object cannot pass through a sm smal alll ga gap. p. An in inna nate te-b -bel elie ieff vie view w of co conc ncep eptu tual al de deve velo lopm pmen entt wo woul uld d ha have ve predicted that infants should understand both possibilities at the same time, because of some core principle of penetrability. penetrability. Under Und er Bai Baillar llargeo geon’s n’s vie view w, chi childr ldren en acq acquir uiree imp import ortant ant con concep ceptua tuall scheme sche mes, s, su such ch as ob obje ject ct pe perm rman anen ence ce,, th thrrou ough gh pe perc rcep eptu tual al me mean ans, s, su such ch as by looking and listening, rather than through acting on the world. Consider the case of a 3-month-old infant who realizes that an object will fall when released midair and stop fully when it hits the ground. The child may initially tia lly un unde ders rsta tand nd th this is ph phys ysic ical al ac actt be beca caus usee sh shee rep epea eate tedl dly y se sees es ad adult ultss dr drop op things off tables and throw clothes in hampers. Only when the child can

 

216

 

Embodiment and Cognitive Science

independently put objectssupported on surfaceswill andthe seechild how acquire they may fall off if the surface is not sufficiently a full mastery of the concept. Most generally, infants infer the rules of how objects behave once they have experienced many opportunities of watching objects move, collide, fall, and so forth. These visual experiences are initially constrained  by a learning mechanism that is dedicated to acquiring event-general expectations (e.g., a general principle covering all occlusion, containment, and covering events), which later on develops into a mechanism enabling event-specific expectations (i.e., different principles about occlusion, containment, and covering events, respectively) (Baillargeon,  2004). In fact, exposing infants to different contrastive information about physical events, such as those pertaining to height, appear to benefit   9 1/2month-old infants’ detection of violations of specific principles (Wang, Baillargeon, & Brueckner, 2004). Overall, this position places primary emphasis on the child’s looking skills skill s as the key ingredient in successful differentiated physical reasoning. The Importance of Embodied Action in Physical Reasoning

Each of the above theories of cognitive development gives little attention to the role of full-bodied sensorimotor activity in how the child learns to reason about the physical world. Spelke (1998)  correctly argues that the simple fact that behavior on some task has changed over time does not necessarily imply that experience is the cause of that development. Various maturational triggered by the environment, may drive behavioral processes, change andperhaps development overall, apart from any particular experiences the child has looking at or manipulating objects. Yet appeals to nativist accounts of cognitive development still ignore fundamental knowledge (including deeply nonrepresentational nonrepresentational information) that children acquire from their bodily experiences in interaction with the physical and cultural world. I agree wholeheartedly with Spelke (1998)  when she suggests four guidel gui deline iness for inv invest estiga igator torss stu studyi dying ng cog cognit nitive ive dev develo elopme pment, nt, one of whi which ch states that “All accounts of the findings of infant studies require evidence. In particular, particular, those who would explain infant’s performance by appealing to sensory or motor processes must provide evidence for these processes processes,, on a par with those who would explain infant’s performance in terms of  perceptual or cognitive processes” (Spelke,  1998:   41). But scholars who embrace the idea that concepts are acquired through nonsensory or disembodied processes must also share in the burden of having to explicitly seek out an embodied alternative as part of their experimental work. The fact that infants are not adept at manipulating objects until around 6 mo mont nths hs ol old d mi miss sses es th thee ob obvi viou ouss po poin intt th that at in infa fant ntss st stil illl ha have ve ma many ny co comp mple lex x

 

Cognitive Development

 

217

 bodily interactions with objects (e.g., touching them, being placed on and inside of them, having their mouths them). Newborns often bring objects placed in their hands up toon their mouths for oralwill contact (Lew & Butterworth, 1997). This kind of oral exploration enables infants to learn important qualities of objects, such as their solidity, boundedness ne ss,, an and d ri rigi gidi dity ty,, pr prop oper erti ties es th that at ar aree wi wide dely ly as assu sume med d to be in inna nate tely ly gi give ven. n. Othe Ot herr st stud udie iess sh show ow th that at ne newb wbor orns ns’’ ha hand nd an and d ar arm m mo move veme ment ntss ar aree fa farr fr from om random. rando m. In one study of hand and arm movement, movement, newb newborns orns were presented with three conditions: a person facing them, a ball moving slowly in front of them, and a control condition with neither a person or ball (Roseblad & von Hofsten, 1994). These newborns more often flexed their fingers and moved their hands in the social condition (i.e., a person facing them) than in the other two conditions. When newborns saw the rolling  ball, they were more likely to extend their fingers (as if to anticipate a grasp), to move their thumbs and index fingers (as if to grasp), and to extend their arms forward (as if to reach). A different study indicated that newborns resisted when adults attempted to manipulate their arms and could visually follow the movements of their arms if they could see them (van der Meer, van der Weel, & Lee,  1995). Finally, newborns presented with an object in different locations directed more arm and hand movements in the direction of the object (Bloch,  1990). These different results together suggest that even if very young infants cannot actually reach or grasp, they can engage in motor activities that are adaptive to the social and physical environment.

Although   2 and   3 month old infants initial exploration of an object tends to be oral, 4- and 5-month-old infants tend to examine objects visu 1989)mittens  2 -week ally (Rochat, . Yet infants who had experiences of wearing sticky (i.e., mittens with palmsenrichment that stick to the edges of objects and allow infants to pick them up) later on showed more object engagement than did inexperienced peers (Needham, Barrett, & Peterson, 2002). Th Thus us,, ex expe peri rien ence ce wi with th ac acti ting ng on ob obje ject ctss ma may y be cr crit itic ical al to in incr crea easi sing ng infa in fant nts’ s’ en enga gage geme ment ntss wi with th ob obje ject ctss an and d de deve velo lopi ping ng th thei eirr ob obje ject ct ex expl plor orat atio ion n skills. As in infa fant ntss gr grow ow an and d ex expe peri rien ence ce di dirrec ectt co cont ntac actt wi with th ob obje ject cts, s, no nott ju just st wi with th their hands, they learn a great deal about notions such as support, continuity, and boundedness directly from their kinesthetic experiences – for exam ex ampl ple, e, in infa fant nts’ s’ se self lf-p -prrod oduc uced ed lo loco como moti tion on an and d th thei eirr rea eact ction ionss to th thee de deep ep side of a visual cliff (Bertenthal, Campos, & Barrett, 1984; Bertenthal et al., 1994). Infants who had early experiences with self-produced locomotion

(eithe (eit herr th thei eirr ow own n na natu tura rally lly ac acqui quirred ex expe peri rien ence cess or ar artifi tifici cial ally ly ac acqu quir ired ed ex ex-peri pe rien ence cess us usin ing g an in infa fant nt wa walk lker er)) ex exhi hibi bite ted d wa wari rine ness ss to towa warrd th thee de deep ep si side de of the cliff (as described by an increase in heart rate or sudden avoidance of the deep side), compared to infants who did not have locomotion experience. Self-produced locomotion does not necessarily lead infants to fear

 

218

 

Embodiment and Cognitive Science

heights when they first learn to walk, because infants may have to relearn thee co th cons nseq eque uenc ncee of de deep ep sl slop opes es fo forr ea each ch do doma main in of mo move veme ment nt (e (e.g .g., ., cr craw awll 1997, 2000; Clearfield, 2000). Being able to walk ing walking) (Adolph, alsovs. increases social interactions. Thus, an upright infant is more likely to  be able to look, vocalize, and smile at adults (Gufstafson, 1984). Finally, infants with more locomotion experience, through either crawling or assisted walking, are more likely to persist in searches for hidden objects (as examined in object permanence tasks) (Bai & Bertenthal,  1992; Bertenthal et al., 1984; Kermoian & Campos, 1988). The development of walking also assists blind infants’ search for hidden objects (Bigelow, (Bigelow, 1992). Other studies observing a link between infants’ leg movements and the movements of a hanging mobile provide additional support for the claim that producing actions with observable effects on objects is highly reinforcing reinfor cing for young infants (Rovee-Collier & Hayne, 2000). Infants who exemplify more active exploration strategies (exploring more, switching more between oral and visual modalities of exploration) are also better able to segregate a visual display into its component parts, compared to infants with much less active exploration strategies (Needham, 2001). The abo above ve stu studie diess dem demons onstra trate te imp import ortant ant link linkss bet betwee ween n inf infant ants’ s’ per percep cep-tual, cognitive, and action-based abilities. None of these results imply that sensorimotor processing alone may be responsible for children’s understanding of physical events. But, at the very least, the evidence suggests that sensorimotor experience contributes to children’s understanding of  obje ob ject ctss an and d th thei eirr be beha havi vior ors. s. In Infa fant nts’ s’ ea earl rlie iest st co conc ncep epts ts of ob obje ject ctss ar aree ti tied ed no nott  just to their visual experiences, but to noting how objects change in differdifferentt ci en cirrcu cums msta tanc nces es,, bo both th wh when en th that at ch chan ange ge oc occu curs rs th thro roug ugh h th thee in infa fant nt’s ’s ow own n movements and when infants move around objects when carried around  1993  by adults For instance, a blanket when the feeding baby is  being put (Bloom, down to rest).and disappears when itappears is picked up for

or playing. Thus, infants’ theories of objects must arise partly from their embodied actions in relation to objects. It is unsurprising, then, that children’s first words express something about objects that move (e.g., “ball”) (Bloom,  1993). “Both conceptual categories and eventual linguistic categories build on an infant’s nascent theories about objects, motion, space, and causality, and these theories originate in the early experiences that come about with movement and change in location” (Bloom, 1993: 86). One possibility is that the movement of objects resembles certain aspects of an infant’s own tactile-kinesthetic actions. In this regard, there are two broad types of onset of motion, self-instigated motion and caused motion. Adults think of biological motion as having certain rhythmic, but unpre unp redic dictab table, le, cha charac racter terist istics ics,, whe where reas as mec mechan hanica icall mot motion ion is tho though ughtt of as undeviating, unless the moving object is deflected in some way. Infants’ concentrated attention on moving objects may easily lead them to analyze the animate trajectories of objects in motion. moti on.

 

Cognitive Development

 

219

ery y yo youn ung g in infa fant nts, s, in fa fact ct,, ar aree se sens nsit itiv ivee to th thee di diff ffer eren ence ce be betw twee een n so some me-Ver thingstartingtomoveonitsownandsomethingbeingpushedorotherwise made ma de to mo move ve (L (Les esli lie, e, 1988). Se Self lf-m -mot otio ion n is th thee st star artt of an in inde depe pend nden entt tr traa jectory where no other object or trajectory is involved. Noticing that dogs  bob up and down as well as follow irregular paths when they move is one example of this. When 1 - to  2 -year-olds played with little models of  a variety of animals and vehicles, they often responded to the animals by maki ma king ng th theem ho hop p al alon ong g th thee ta tab ble le,, but ma made de th thee ve veh hic icle less sc sco oot in a st strrai aigh ghtt linee (Ma lin (Mandl ndler er,, Bau Bauer er,, & McD McDono onough ugh,, 1991). Th Thus us,, ve very ry yo youn ung g ch chil ildr dren en ap ap-pear pe ar to un unde ders rsta tand nd di diff ffer eren ence cess in th thee mo move veme ment nt of an anima imate te an and d in inan anim imat atee objects. Perceived differences differences in motion patterns can be used to ground infants’ earl ea rly y un unde ders rsta tand ndin ing g of th thee di dist stin inct ctio ion n be betw twee een n an anima imate te an and d in inan anim imat atee ob ob- jects (Premack, 1990;  Mandler,  1992). Self-propelled, irregularly moving thin th ings gs te tend nd to be an anim imat ate, e, an and d co cont ntac actt-pr prop opel elle led, d, sm smoo ooth thly ly mo movi ving ng th thin ings gs tend to be inanimate. However, the animate/inanimate distinction may arise from a belief, not bodily information, about whether an object has the right source of energy (internal vs. external) and is made of the right kind ki nd of st stuf ufff to in init itia iate te it itss ow own n ac acti tion onss (G (Gel elma man, n, Du Durrgi gin, n, & Ka Kauf ufma man, n, 1995). Thus, preschoolers can make judgments related to animacy based on still pictures of animate-like versus nonanimate-like objects. But this finding may ma y be due to in infa fant ntss’ un unde derrst stan andi ding ngss of how va vari riou ouss sh shaape pess rela late te to animate action (i.e., perception of curvilinear contours, faces). People often infer inf er dyn dynami amic, c, som someti etimes mes emb embodi odied, ed, inf inform ormati ation on fr from om sta static tic per percep cepts ts (se (seee Chapter 3 ). Features are important in the perception of objects and their  behaviors because these are correlated with dynamic information that children experience from their own bodies and from watching others. Perceptual Meaning Analysis: Mandler’s Theory

My earlier brief review of three broad theories of cognitive development sugges sug gested ted tha thatt few app appro roach aches es app appro ropri priate ately ly ack acknow nowled ledge ge inf infant ants’ s’ sen sensosorimoto rim otorr abi abiliti lities es in the their ir phy physic sical al re reaso asonin ning g ski skills lls.. One Onethe theory ory tha thatt emb embrac races es

a sl slig ight htly ly mo morre em embo bodi died ed vi view ew of ch child ildrren en’s ’s co cogn gnit itiv ivee de deve velo lopm pmen entt ma main in-tains that infants’ capacity to abstract certain kinds of information from percep per ceptua tuall dis displa plays ys dev develo elops ps con concur curre rentl ntly y wit with h sen sensor sorimo imotor tor ski skills lls,, rat rather her than tha n bei being ng a sub subseq sequen uentt dev develo elopme pment nt (Ma (Mandl ndler er,, 1992, 2004). Ea Earl rly y co conc ncep eptt formation does not depend on physical interactions with objects, but instead arises from independent analyses of certain perceptual experiences. Innate perceptual mechanisms begin very early on to generate abstract, nonpr non propo oposit sition ional al ima images ges tha thatt ar aree “si “simpl mplifie ified d and con conden densed sed re relat lation ionshi ships ps of spatial structure” (Mandler,  1992: 591 –2). Thus, the process of perceptual tu al an anal alys ysis is ex extr trac acts ts va vari riou ouss as aspe pect ctss of th thee sp spat atia iall st stru ruct ctur uree of ob obje ject ctss an and d thei th eirr mo move veme ment ntss in sp spac ace, e, us usua ually lly fr from om vi visu sual al ex expe peri rien ence ce,, al alth thou ough gh to touc uch, h,

 

220

 

Embodiment and Cognitive Science

audition, and one’s own movements are also included. Perceptual meaning analysis redescribes the spatial and movement structure of perceptual disp di splay layss to to,, on once ce mo morre, to an anal alog ogic ic re repr pres esen enta tati tion ons, s, or ima image ge sc sche hema mas, s, th that at comp co mpos osee th thee pr prim imiti itive vess of an ac acce cess ssib ible le co conc ncep eptu tual al sy syst stem em (s (see ee Ch Chap apte terr 4). One possibility is that image schemas underlie many concepts acquired early in a child’s life, such as animacy, inanimacy, agency, containment, and support relations (Mandler, 2004). As described in Chapter  4, image schemas are not images in the sense of neces necessaril sarily y havin having g detail detailed ed infor information mation about objec objectt movem movement ent such as speed and direction. Unlike visual images, for example, image schemas are not usually conscious, and are best thought of as topographical representations that can be complex, despite their primitive nature (e.g., a CONTAINMENT schema consists of a boundary plus an inside and outside). Although image schemas may be temporary constructions, created on the fly as part of people’s embodied simulations, Mandler (1992, 2004) maintains the more traditional view that image schemas are permanent representations in memory. Nonetheless, let me now explore more fully thee id th idea ea th that at im imag agee sc sche hema mass un unde derl rlie ie as aspe pect ctss of yo youn ung g ch child ildrren en’s ’s co conc ncep epts ts.. Cons Co nsid ider er fir first st th thee id idea ea th that at va vari riou ouss im imag agee sc sche hema mass un unde derl rlie ie yo youn ung g ch chil il-dren’s understanding of animacy. animacy. The contingency of animate movement not only involves such factors as one animate object following another, another, as described by the image schema LINKED PATHS, but also involves avoiding in g ba barr rrie iers rs an and d ma maki king ng su sudd dden en sh shif ifts ts in ac acce cele lera rati tion on.. Ad Adul ults ts ar aree se sens nsiti itive ve to all of these aspects of animate movement (Stewart,  1983), but it is not yett kn ye know own n wh whet ethe herr in infa fant ntss ar aree res espo pons nsiv ivee to su such ch mo move veme ment nt,, ev even en th thou ough gh they appear to be perceptually salient. Nor has anyone considered how factors such as barrier avoidance might be represented in image-schema form (Mandler, 1992). Several FORCE schemas, such as BLOCKAGE and DIVERSION, may be useful in describing barrier avoidance, but these schema sche mass ne need ed to be fu furt rthe herr di difffe ferren enti tiat ated ed to ac acco coun untt fo forr an anim imat atee an and d in inan an-imat im atee tr traj ajec ecto tori ries es.. On Onee mi migh ghtt re repr pres esen entt an anim imat atee an and d in inan anima imate te di diff ffer eren ence cess in response to blockage as a trajectory that shifts direction before contacting a barrier versus one that runs into a barrier and then either stops or  bounces off from from it (Mandler, (Mandler, 1992). Studies of babies’ understanding of causality may reveal how FORCE and CAU CAUSED SED-MO -MOTIO TION N sch schema emass sha shape pe und unders erstan tandin ding g of cau causal sality ity.. Ima Imaggine a film clip of one ball rolling out from behind a screen from the right

and st and strrik ikin ing g a se seco cond nd ba ball ll sit itua uate ted d in th thee mi midd ddle le of th thee sc scrree een n, fo foll llow owed ed by the second ball rolling off the screen to the left. When watching this film, observers experience the first ball as having caused the movement of the second. Now imagine the same event with a brief delay between the time thee fir th first st ba ball ll st stri rike kess an and d th thee ti time me th thee se seco cond nd ba ball ll mo move ves. s. Th Thee de dela lay y rem emov oves es the appearance of a causal relation. But now imagine seeing a third film clip, which is just a copy of the first film clip played in reverse. Now a ball

 

Cognitive Development

 

221

moves in from the left, strikes the ball in the middle, and causes it to roll off to the right. Aside from the direction of movement, the third clip has exactly the same contiguity and successive relations as the first. Just like the first film clip, we see this clip as causal. The change in direction makes an important difference, however. however. In the first clip, people perceive the ball on the left as th thee agent with causal powers that are transfixed to the ball in the center of the screen. In the third clip, the ball on the right is i s the causal agent. Even 6-month-old infants are sensitive to causal relationships (Leslie & Keeble, 1987). Wh When en in infa fant ntss wa watc tch h th thee se seco cond nd cl clip ip,, pl play ayed ed fa fast st fo forw rwar ard d an and d theen in reve th verrse se,, th theey do no nott ge gett ve verry exc xcit ited ed by th thee reve verrsa sal. l. Wh Wheen th they ey se seee the first clip played forward and then in reverse (like the third clip), they get very excited by the reversal. The first reversal is perceived as a change in the locus l ocus of causal power power,, but the second reversal is not thus perceived. These findings suggest to some that contiguity and succession are insufficient to explain how we think about causation (Leslie & Keeble, 1987). In addition additi on to conce concepts pts of spatia spatiall and temporal properties, properties, infants have an inna in nate te co conc ncep eptt of ca caus usal al po powe werr. Sp Spat atia iall an and d te temp mpor oral al pr prop oper erti ties es ar aree gu guid ides es to at attr trib ibut utin ing g ca caus usal al po powe werr, bu butt th they ey ar aree no nott co cons nsti titu tuti tive ve of ca caus usal al po powe wers rs.. Under this view, casual power cannot be identified with any perceptual representations. Thee ma Th main in pr prob oble lem m wi with th th this is co conc nclu lusi sion on ab abou outt ca caus usal alit ity y is th that at it ne negle glect ctss the infant’s understandin understanding g of its own body as both sour source ce and rec recipien ipientt of causal forces. Infants and young children, similarly to adults, manipulate objects and feel the pushing and pulling of objects that make contact with them. In these situations, characteristic kinesthetic and somatosensory so ry ex expe peri rien ence cess ar aree sp spec ecific ifical ally ly fe felt, lt, an and d in infa fant ntss ma may y us usee su such ch ex expe peri rien ence cess to ground their early notions of causal powers. They may project characteristic embodied experiences onto perceived objects. For example, the pattern of contiguity and succession that children see when billiard balls collide are most significantly similar to the patterns produced by seeing their own interactions with objects. Thus, part of infants’ ability to recognize causal relations may be due to image schemas like FORCE that is derived through somatosensory and kinesthetic experiences, and projected onto inanimate objects when they act as causes. One suggestion is that infants’ understanding of contingent interactions amon am ong g ob obje ject ctss is al also so pr prec eced eded ed by th thei eirr an analy alysi siss of th thee re reci cipr proc ocal al gi give ve-a -and nd-take occurring during adult-infant turn-taking (Mandler, (Mandler, 2004; Murray & Trevarthen, 1985), which is certainly an embodied activity. activity. Severa Sev erall LIN LINK K sch schema emass may als also o str struct uctur uree you young ng chi childr ldren’ en’ss und unders erstan tandding of causal relations between animate and inanimate objects. A LINK is

established as the children regularly encounter one event followed by another, such as seeing that a spoon always falls to the floor when pushed off the side of the highchair. Of course, the ability to know that one has

 

222

 

Embodiment and Cognitive Science

moved move d on ones esel elf, f, as op oppo pose sed d to be bein ing g mo move ved d by so some meon onee or so some meth thin ing g el else se,, gives rise to the SELF-MOTION schema, which also surely is critical to the perception of many causal events involving animate and inanimate objects. Different studies with  7 -month-olds show that they look longer at object movements that begin without being compelled by contact with anot an othe herr ob obje ject ct (S (Spe pelk lkee et al al., ., 1995). In th this is wa way y, in infa fant ntss ma may y ac acqu quir iree so some me of  their understanding of objects and the circumstances in which they move from their own felt understandings of their bodily experiences. These image schemas provide enough information to understand what a concept of “starting to move” is like without the more detailed perceptual analysis that characterizes each perceptual event (Mandler (Mandler,, 2004). Furthermore, studies reveal that infants as young as 3 months of age can distinguish between correct and incorrect human actions (Bertenthal,  1993), suggesting that th at co conc ncep epts ts of an anima imacy cy ar aree ro root oted ed in im imag agee sc sche hema mass lik likee SE SELF LF-M -MOT OTIO ION. N. CONTAINMENT is another image schema that is critical to cognitive development. Some concept of containment seems to be responsible for the better performance  9 -month-old infants show on object-hiding tasks when wh en th thee oc occl clud uder er co cons nsis ists ts of an up upri righ ghtt co cont ntain ainer er,, ra rath ther er th than an an in inve vert rted ed  1980

container or a).screen (Freeman, Lloyd, &arSinha, ;  Lloyd, & Freeman, 1981 Thes Th esee in infa fant ntss al alrrea eady dy ap appe pear to ha have ve a co conc ncep eptt ofSinha, cont co ntai ainners as places where things disappear and reappear. Image schemas may expl ex plai ain n so some me of th thes esee da data ta.. Fo Forr ex exam ampl ple, e, th thee CO CONT NTAI AINM NMEN ENT T sc sche hema ma ha hass three structural elements (interior (interior,, boundary, boundary, and exterior) that primarily arise from two sources: (1) perceptual analysis of the differe differentiation ntiation of figure from ground, that is, seeing objects as bounded and having an inside that th at is se sepa para rate te fr from om th thee ou outs tsid idee (S (Spe pelk lke, e, 1988), a n d (2) per percep ceptua tuall ana analys lysis is of ob obje ject ctss go goin ing g in into to an and d ou outt of co cont ntain ainer ers. s. Th Thee lis listt of co cont ntai ainm nmen entt re rela lati tion onss that babies experience is long. Babies eat and drink, spit things out, watch theeir bod th odie iess be bein ing g cl clot oth hed an and d un uncl clot oth hed ed,, ar aree ta take ken n in and out of cr crib ibss an and d rooms, and so on. An infant’s understanding of opening and closing is also al so related to the development of containment. Piaget (1952)  documented in detail the actions   9- to   12-month-old infants performed while they were learning to imitate acts that they could not see themselves perform, such as blinking. Before infants accomplished the correct action, they sometimes opened and closed their mouths, opened and closed their hands, or covered and unco un cove vere red d th thei eirr ey eyes es wi with th a pi pillo llow w. Pi Piag aget et’s ’s ob obse serv rvat atio ions ns te test stif ify y to th thee pe perrceptual analysis in which the infants were engaging and their analogical understanding of the structure of the behavior they were trying to reproduce. Such understanding seems a clear case of an image schema of the spatial movement involved when anything opens or closes, regardless of  the particulars of the thing itself. Alth Al thou ough gh bo bodi dily ly ex expe peri rien ence ce ma may y be th thee ba basi siss fo forr un unde ders rsta tand ndin ing g of co conntainment, it is not obvious that bodily experience per se is required for

 

Cognitive Development

 

223

perceptual analysis to take place (Mandler,  1992). Infants have many opportun por tuniti ities es to ana analyz lyzee sim simple ple,, eas easily ily vis visibl iblee con contai tainer nerss suc such h as bot bottle tles, s, cup cups, s, and dishes and the acts of containment that make things disappear into and an d reap app pea earr ou outt of th theem. In Ind dee eed d, it mi migh ghtt be eas asie ierr to an anal alyz yzee th thee si sigh ghtt of  milk going in and out of a cup than milk going into or out of one’s mouth. Neve Ne vert rthe hele less ss,, wh whic iche heve verr wa way y th thee an anal alys ysis is of co cont ntai ainm nmen entt ge gets ts st star arte ted, d, on onee would expect the notion of food as something that is taken into i nto the mouth to be an early conceptualization. Another aspect that seems to be involved in an early concept of a container is that of support. True True containers not only envelop things but supportthemaswell.Infantsasyoungas 3 mon months ths ar aree sur surpri prised sed whe when n sup suppor portt relations between objects are violated (Needham & Baillargeon, 1993). Infants   5 1/2  months old are surprised when containers without bottoms appe ap pear ar to ho hold ld th thin ings gs (K (Kol olst stad ad,, 1991). Simila Similarly rly,, 9-mo -month nth-ol -old d inf infant antss cou could ld  judge whether a block could be supported by a box open at the top only whentheywereabletocomparethewidthsoftheblockandtheboxinasinglee gl gl glan ance ce as on onee wa wass lo lowe werred in into to th thee ot othe herr (S (Sit itsk skoo oorn rn & Sm Smit itsm sman an,, 1991). Finally, Baillargeon (1993)   demonstrated that   12 1/2-month-old infants could determine whether a cloth cover with a small protuberance could hide a small tigerdirectly toy only when were tofindings comparesuggest the sizethat of  the protuberance with thatthey of the t he toy.able toy. These the notions of containment and support may be closely related from an early age. A primitive image-schema of SUPPORT might require only a representation of contact between two objects in the vertical dimension (Mandler, 1992). One could argue that development of the notion of object permanence can be thought of as the development of several different image schemas and the workings of transformations between them. Following Mandler (1992), the transformations LANDMARK, to BLOCKAGE, to REMOVAL OF BLOCKAGE, and finally back to LANDMARK underlie the demonstration of object permanence in the   4 1/2-month-olds. The reason the 3 1/2-m -mon onth th-o -old ldss do no nott ex exhi hibi bitt ob obje ject ct pe perm rman anen ence ce is th that at th they ey ei eith ther er ha have ve not de not deve velo lope ped d on onee or mo morre of th thes esee im imag agee sc sche hema mass or ar aree no nott ye yett ca capa pabl blee of  tran tr ansf sfor ormi ming ng th them em.. Th Thee sp spec ecifi ificc ex expl plan anat atio ion n req equi uire ress mo more re sp spec ecific ific te test stss to determine which is true, but we suspect it has to do with blockage and remova mo vall of bl bloc ocka kage ge.. Th This is fo foll llow owss fr from om th thee fa fact ct th that at 3 1/2-mont -month-old h-old infan infants ts can ca n al alrrea eady dy fo focu cuss on in indi divid vidua uall ob obje ject ctss an and d th thus us ap appe pear ar to ha have ve de deve velo lope ped d the image schema for LANDMARK. Finally, consider the concept of agency. A traditional belief is that infants learn agency from observing the effects of their actions on the world (Gibson, 1988). An infant discovers his or her own agency by observing that th at hi hiss or he herr so solit litar ary y mo move veme ment ntss br brin ing g ab abou outt de desi sire red d co cons nseq eque uenc nces es,, su such ch as keeping a picture in focus rather than out of focus (Kalnins & Bruner, 1973).

 

224

 

Embodiment and Cognitive Science

But this view of how agency develops does not properly acknowledge a child’s kinesthetic activity (Sheets-Johnstone,  1999). For example, very early in a young infant’s life, the simple act of moving its lips, tongue, and mouth on its mother’s breast directly exposes the child to its own causal powers. Numerous other body actions, such as chewing, swallowing, and  bending and extending fingers, provide infants with a tactile-kinesthetic sens se nsee of ag agen ency cy.. Th This is un unde ders rsta tand ndin ing g of ag agen ency cy is no nott an aw awar aren enes esss de deri rive ved d from just seeing our efforts in the world, but rather is rooted in our own tactile-kinesthetic experiences. Of course, young children, like adults, do not engage in activity for no purpose, but aim to achieve specific goals (e.g., touching objects, obtaining food). Developmental studies show that 5-month-old infants begin to distinguish between goal-directed actions and those that occur accidentally (Woodward, (Woodward, 1999). By  9  months of age,  babies appear to be able to understand something about the paths along which wh ich ob obje ject ctss wi will ll tr trav avel el (G (Ger erge gely ly et al al., ., 1995). Th Thes esee fin findi ding ngss ar aree co cons nsis iste tent nt with the idea that an image schema of SOURCE-PA SOURCE-PATH-GOAL, along with ANIM AN IMAC ACY Y an and d SE SELF LF-M -MOT OTIO ION, N, he help lp in infa fant ntss re reas ason on ab abou outt th thee be beha havi vior or of  movi mo ving ng ob obje ject cts. s. On Once ce mo more re,, rec ecur urri ring ng as aspe pect ctss of in infa fant nts’ s’ us usin ing g th thei eirr bo bodi dies es and an d pa part rtss of th thei eirr bo bodi dies es,, ra rang ngin ing g fr from om mo movi ving ng th thei eirr ey eyes es to ga gaze ze at sp spec ecifi ificc objects to moving their whole bodies to reach people and objects, provide part of the foundation for more complex physical reasoning skills. Object Permanence Again

I st star arte ted d my di disc scus ussi sion on of rec ecen entt co cogn gnit itiv ivee de deve velo lopm pmen entt wo work rk by fo focu cusi sing ng on experimental studies of object permanence. These studies aimed to explore when and how infants acquire the object concept and reason about the behavior of objects. I was critical of these studies for not exploring the importance of the child’s own actions, such as its reaching behavior, in theoretical accounts of concept acquisition. But there are several newer studies that have attempted to model the child’ chi ld’ss per perfor forman mance ce on obj object ect per perman manenc encee tas tasks, ks, spe specifi cifical cally ly inv invest estiga igatin ting g infants’ A-not-B errors. For example, Munakata et al. (1997) developed a connectionist model to address developmental decalage between infants’ success on the object concept search task when assessed by reaching and succ su cces esss wh when en as asse sess ssed ed by lo look okin ing g pr pref efer eren ence ces, s, an ad adva vanc ncee th that at oc occu curs rs se sevveral months earlier. The traditional explanation of successes and failure in search tasks are principle-based – that is, early successes imply an allor-nothing knowledge of principles (e.g., object permanence) and failures are attributed to ancillary deficits (e.g., means-end abilities). Munataka et al. argued that the principle-based approach leads, for example, to the premature inference that  3   1/2-month-old infants who look longer at the “impossible” disappearing event (see Baillargeon,  1993) have knowledge of object permanence.

 

Cognitive Development

 

225

They propose an alternative “adaptive processes” account of the ac-

quisition of the object concept in which knowledge is graded in nature rather than all-or-none, evolves with experience, and is embedded in specific processes underlying overt bodily behavior. The adaptive processes approach attributes success to the ability to represent occluded objects, whic wh ich h in te term rmss de depe pend ndss on th thee co conn nnec ecti tion on am amon ong g ma many ny rel elev evan antt ne neur uron ons, s, an ab abil ility ity th that at is ac acqui quirred th thro roug ugh h a pr proc oces esss of st strren engt gthe heni ning ng th thes esee co conn nnec ec-tions. Failures occur because different behaviors require different different degrees of dev develo elopme pment nt in the re relev levant ant und underl erlyin ying g pr proce ocessi ssing ng sys system tem and re resul sultin ting g internal repr representations. esentations. This connectionist model shows how an infant could gradually learn to represent occluded objects over time, thus accounting for successes and failure without assuming principles and ancillary deficits on the ob ject search tasks. Thelen et al. (2001)  propose a dynamical alternative to Munakata et al.’s (1997)  Hebbian network model of the A-not-B task. Dyna Dy nami micc sy syst stem emss ps psyc ycho holo logi gist stss vi view ew de deve velo lopm pmen entt in te term rmss of th thee in infa fant nt’s ’s active engagement with the environment through movement, rather than depending on theoretical constraints or neurological programs. As Esther Thelen and Linda Smith argue, “Development does not happen because internal maturational processes tell the system how to develop. Rather, development happens through and because of the activity of the system  1994   305). Under this view, cognition is strucitself” (Thelen & Smith, tured in movement activity. A: young child’s ability to integrate information from across different sensory modalities is not a result of development, but the very basis from which development emerges. Development is seen as an emergent property of the whole system and can only be understood in terms of the complex interaction of psychological, biological, and physical components. A key feature of dynamic systems is that they are self-organizing – they arrive at new states simply through their own functio fun ctionin ning, g, wit withou houtt spe specifi cificat cation ion fr from om the env envir ironm onment ent or det determ ermina inatio tion n from within. With a continuous change in one or more control parameters (akin to but not equivalent to independent variables), new states may emerge spontaneously as a function of nonlinear interactions between the systems’ components. Quite importantly, importantly, the development of behavior that appears to be discontinuous or disorderly at the performance level may arise from underlying processes that themselves are continuous and orderly (e.g., an infant’s vocabulary acquisition or first steps), which is a key feature of all self-organizing systems. Dynamic systems theory is good at interpreting multiple levels of performance (e.g., infants are competent at the perceptual level when assessed by visual preference, but not when assessed by reaching). A dynamical systems view of the A-not-B error claims that this  behavior is not specific to any particular point in development, but may arise ari se in dif differ ferent ent sit situat uation ionss in whi which ch chi childr ldren en pr produ oduce ce goa goal-d l-dir irect ected ed act action ionss

 

226

 

Embodiment and Cognitive Science

to re reme memb mber ered ed loc locat atio ions ns.. Mo Morre sp spec ecifi ifica call lly y, th thee er erro rorr is at attr trib ibut uted ed to th thee in in-teraction of many factors. These include visual and attentional processes that function in the perception perception of objec objects, ts, motor pro processe cessess used in planning and executing hand and arm movements toward a target location, short-term memory processes active in maintaining task-related informa-

tion when relevant perceptual cues are absent, and long-term memory processes that maintain results from past actions. In a typical A-not-B experiment, infants start by locating the hidden object under location A over repeated trials, which produces a relatively stable long-term memory of the A location. At the start of the first B trial, infa in fant ntss pl plan an a mo move veme ment nt to towa warrd th thee B lo loca cati tion on to ret etri riev evee th thee ob obje ject ct.. Ho Howwever, in the absence of appropriate perceptual cues that specify that the object is in the B location when the object is hidden, the motor plan to reach toward B decays, especially over longer time intervals. After several seconds, a plan to move to A, which has been strongly established from previous trials, begins to dominate and the infant consequently makes the A-not-B error by reaching for the object in the A location. Several Sever al expe experimen riments ts teste tested d this theory by having 2 -year -year-olds -olds first see and then find a toy at the A location over several trials. When the toy was then hidden at the B location, the toddler’s toddler ’s search was biased toward toward the direction of A. The magnitude of the pull toward A depended on the number of times the toy had been previously hidden at that location. Furthermore, the longer the children had to wait before reaching for the toy at the B location, the more likely they were to make a response toward A. The bias toward location A was also present even when the A trials, both during training and in the main experimental task, varied over an  8-inch region. Finally, shifts in the babies’ position and visual perception influenced the probability of making the A-not-B error. When  8- to  10-montholds stood up on the B trials, they significantly reduced their A-not-B errors (Smith et al., 1999). These results are consistent with the dynamical claim that the error arises from an interaction of four factors: the graded natu na turre of sp spat atia iall me memo mory ry,, th thee se sequ quen ence ce of ev even ents ts in th thee ta task sk,, th thee li limi mite ted d xx location in the task space, and the delay in the B task. The dynamical approach to development is significant because it em braces the idea that cognition cognition is connected to bodily action. action. A child’s new abilities emerge through the dynamic indeterminacy of self-organization. Unlike most theories, the dynamical perspective explains development in terms of multiple causes and connections and acknowledges that even small, unexpected factors may critically shape the course of development. Morreo Mo eove verr, dy dyna nami mica call sy syst stem emss th theo eori ries es re reco cogn gniz izee th thee im impo port rtan ance ce of st stud udyying the whole system (i.e., the child) in understanding development, and not assuming that cognitive growth is based on the acquisition of isolated competences. Although dynamical systems theory has been most successful in describing motor and perceptual development (Bertenthal &

 

Cognitive Development

 

227

Pinto,  1993;  Butterworth,  1993;  Goldfield,  1993;  Thelen,  2000;  Thelen & Smith, 1994; va van n Ge Geer ert, t, 1991), th theere is an in incr crea eassin ing g bod ody y of wo work rk sh sho owi wing ng how ho w em emot otio iona nall an and d pe pers rson onal alit ity y de deve velo lopm pmen entt ma may y al also so be ch char arac acte teri rize zed d in dynamical terms (Granott & Paziale, 2002). This work is clearly consistent with wi th my “e “emb mbod odim imen entt pr prem emis ise” e” by ex expl plic icitl itly y lo look okin ing g fo forr th thee po poss ssib ible le ro role less that embodied activity has in human development. Children with Physical Handicaps

One challenge to the idea that sensorimotor experience is critical to cogni tive growth comes from the study of physically handicapped individuals. Although sensorimotor activity may provide part of the input for perceptual systems, high-level symbolic thought may still arise in its absence. As one group of scholars argued, “Since motor movements are not assumed to play an important role in their conceptual abilities, there is no reason to believe that motorically impaired children need be differentially disadvant va ntag aged ed in th thee de deve velo lopm pmen entt of th thes esee ea earl rly y fo foun unda datio tions ns of th thou ough ght, t, un unle less ss additional damage to non-motor areas of the brain have occurred” (Berko et al., 1992: 229). In fact, some studies report research consistent with the idea that motor experience is not crucial for intellectual development. This includes studies looking at academic achievement in children who are congenital amputees (Clarke & French, 1978), object permanence in children with severe quadriplegic cerebral palsy (Eagle, 1985), stage 6  object permanence and intersubjectivity in one   3-year-old child with congenital upper and lower limb deformities (Kopp & Shaperman, 1973), and cognitive gains in psychometric measures in thalidomide children with at least one partial limb (Gouin-Decarie, 1969). Butt ma Bu many ny st stud udie iess ha have ve be been en in inte terp rpre rete ted d as su supp ppor orti ting ng th thee vi view ew th that at mo mo-torr ex to expe peri rien ence ce is cr cruc ucia iall fo forr co cogn gnit itive ive de deve velo lopm pmen ent. t. Co Cons nsid ider er th thee de deve velo loppment of motion. First, crawling allows different kinds of interaction with thee en th envi viro ronm nmen entt th than an ar aree po poss ssib ible le wi with th th thee no nonm nmob obile ile in infa fant nt,, an and d cr craw awli ling ng experience is well established as the cause of transition in spatial location coding (Campos et al., 2000). Studies with children whose vision was obscured by congenital cataracts, and who underwent operations to remove them th em at di difffe ferren entt ag ages es,, sh show ow ho how w cr cruc ucia iall no norm rmal al ex expe pect ctab able le in inpu putt ca can n be to thedevelopmentofnormalvision(Maueretal.,1999). Adven Adventitiou titiously sly blind individuals perform better at distance judgments than congenitally blind children,suggestingtheimportanceofexpectableearlyexperienceforbasic spatial functioning (Reiser, Lockman, & Pick,  1980). Severely motorically handic han dicapp apped ed gr group oupss ar aree del delaye ayed d in the their ir com comple plex x spa spatia tiall re relat lation ion and pla plannning skills (McDonnell,  1988; Rothman, 1987). Moreover Moreover,, nonambulatory chil ch ildr dren en wi with th ce cere rebr bral al pa pals lsy y sh show owed ed mo more re si sign gnific ifican antt de defic ficie ienc ncie iess th than an di did d children with three unaffected limbs (McDonnell,  1988;  Rothman,  1987).

 

228

 

Embodiment and Cognitive Science

Children with cerebral palsy may, of course, have other neurological impairments. Determining which of these alternative views on sensorimotor experience and cognitive development is correct is a tricky problem. There are several methodological issues with most studies of cognitive development in the physically handicapped, including the diversity of procedures used, yielding data that are incommensurable across studies, and the fact that performance on some conceptual tasks is facilitated by bimanual manipulations of the objects and so forth. Thus, differences in the results between limb-deficient and nonhandicapped children may reflect motor limitations rather than conceptual deficiencies. The degree of impai pairme rment ntatisleast also als o cri critic tical. al. Tha Thalid lidomi omide deor individ ind ividual ualss that and con congen genita itall amp ampute utees es possess one functional limb segment could potentially be employed in the formation of motor schemes. These children will circum-

vent the expected impairments by compensating creatively for the handicaps (Eagle, 1985). Even in extreme conditions of physical impairment, a child may still have control of at least one means of interaction with the environment (Sinclair (Sinclair,, 1971). Fo Forr ex exam ampl ple, e, ey eyee mo move veme ment ntss or pe perrha haps ps mo mo-tor activities involved in chewing could be sufficient for motor schemes to develop. Severely quadriplegic children are more impaired than congenital amputees and thalidomide children. Thus, in addition to impairment of both arms and legs, the trunk, eye movements, and mouths are often spastic, rigid, or exceptionally hypotonic. But congenitally physically impaired infants often have more diffuse brain damage, extending to nonmotor areas. In each case, it is difficult to identify the unique contribution of organic as opposed to experiential factors in any observed cognitive delays (Eagle,  1985). Even when organic contributions can be ruled out, other sorts of deprivation, such as social deprivations, are often associated with physical handicaps. Butt mo Bu most st im impo port rtan antly tly,, ph phys ysic ical ally ly ha hand ndic icap appe ped d in indi divi vidu dual alss ma may y st still ill ex ex-perien per ience ce lar large ge amo amount untss of tac tactile tile-ki -kines nesthe thetic tic inf inform ormatio ation n fr from om eye eye,, mou mouth, th, and an d he head ad mo move veme ment ntss an and d th thee fe feel elin ings gs ar aris isin ing g fr from om ot othe herr bo bodi dily ly fu func ncti tion ons. s. These bodily experiences may be sufficient for the coordination of many action schemes that underlie cognitive growth. Moreover, Moreover, recent research shows that normally occurring sensory experience has a major role in species-specific instinctive behaviors that have traditionally been thought to be an outcome of heredity (Gottlieb,  2002). In this view, the nervous syst sy stem em do does es no nott de deve velo lop p fu full lly y or no norm rmal ally ly wi with thou outt th thee be bene nefit fit of no norm rmal ally ly occurring sensory experience. Young Young infants and children may not necessarril sa ily y ne need ed to ha hav ve all th theeir li limb mbss, or be ab able le to mo move ve th thei eirr bo bod die iess in no norrma mall ways, for them still to benefit enormously from simple and complex action patterns as they develop fundamental concepts. Thus, even disabled children will have sufficient bodily experiences to form a wide variety of 

 

Cognitive Development

 

229

schemas, mas, inclu including ding CONT CONTAINME AINMENT NT,, BALAN BALANCE, CE, SOUR SOURCE-P CE-PA ATHimage sche GOAL, LINKAGE, and so on. Multimodal Perception

Many st Many stud udie iess in de deve velo lopm pmen enta tall ps psyc ycho holo logy gy sh show ow th that at yo youn ung g ch chil ildr dren en fin find d abstract similarities between different sensory experiences. Infants seem able to connect visual information with tactile information. For example, 6-month-old infants were given one of two pacifiers to suck that had different textures (Meltzoff & Borton, 1979). The surface of one pacifier was smooth, whereas the other one had a ribbed surface. At first, the infants sucked on the their pacifiers without seeing the objects. In a second part of experiment, the infants were shown large pictures of both Asthe expected, the majority of the infants preferred looking at thepacifiers. pacifier that they had just been sucking. This preference for congruence suggests that even  1 -month-old infants have some understanding of cross-modal equivalence. Babies can also make cross-modal connections between auditory and

visual information. For instance,  4 -month-old infants viewed simultaneous films of two rhythmic events; a woman playing “peek-a-boo,” and a  baton hitting a wooden block (Spelke, (Spelke, 1976). While they were viewing the films, an audiotape, centered between the two screens, was played that was appropriate to one of the films. The infants preferred to look at the visual events that matched the auditory source. This too is evidence for some understanding of cross-modal equivalence. Infant Infa ntss as yo youn ung g as 2 to 5 mo mont nths hs ol old d ar aree ca capa pabl blee of pe perrce ceiv ivin ing g co cohe here rent nt unitary multimodal events, such as the relationship between a person’s face and voice, on the basis of temporal synchrony and shared rhythm between the movements of the mouth and the timing of the speech (Dodd, 1979; Lewkowicz, 1996), as well as between the shapes of the lips and the corresponding vocal sounds (Kuhl & Meltzoff, 1982). By age 5–7 months, infants can match their own body motions, experienced proprioceptively, proprioceptively, with a visual display of the motion, on the basis of shared temporal and spatial information (Bahrick & Watson,  1985;  Bahrick,  1995;  Rochet, 1995; Schumaker, 1986). Thus, when a 5 -month-old observes a live video display of her own legs moving, alongside one of an infant’s legs, she can distinguish the two and prefers to watch the novel display of the other infant. Detection of intermodal relations is not just a case of association of two experiences that happen to occur simultaneously. For example,  3-monthold infants were familiarized with different visible and audible filmed events (Bahrick,  1988). One film depicted a hand shaking a clear plastic  bottle containing one very large marble. The other film depicted a hand shaking a similar bottle containing a number of very small marbles. Four

 

230

 

Embodiment and Cognitive Science

conditions varied in their pairings of film fil m and sound tracks as to whether the appropriate track (one or many marbles) was paired with a film or whether a track was synchronous with the film or not. Only one group of  infants was acquainted with films paired with the appropriate, synchronized sound tracks. After familiarization, an internal preference test was given to each group of infants with two films presented side by side while a si sing ngle le ce cent ntra rall tr trac ack k pl play ayed ed.. Th Thee da data ta sh show owed ed th that at le lear arni ning ng di did d oc occu curr wi with th great gr eater er fam familia iliariz rizati ation, on, re resul sultin ting g in a pr prefe efere rence nce for mat matchi ching ng the film spe speccified by its appropriate sound track. But, most importantly, learning was confined to just one group of infants, namely, namely, those most familiar with the appropriate synchronized pairing of sight and sound. Equal opportunity to associate with an inappropriate sound track did not lead to a preference for that combination on the t he preference preference test. These findings show that ve very ry yo youn ung g ch child ildrren exhi ex hibi bittmodalities. an ab abili ility ty to ac acqu quir iree ab abst stra ract ct rel elat atio ions ns be betw twee een n events in different sensory A different line of research on how children find abstract similarities  between differe different nt sensory experiences comes from work on synesthesia. In one early study, infants were challenged to construct a similarity relationship between two events that shared no physical features or history of  co-occurrence (e.g., a pulsing tone and paired slides of a dotted line and a solid line). Nine- to   12-month-old infants looked longer at the dotted line than at the solid line in the presence of a pulsing tone, suggesting

that a metaphorical match was construed (Wagner et al., 1981). Similarly, they looked more at an arrow pointing upward when listening to an ascending tone and at a downward arrow when listening to a descending tone. The infants were thus able to recognize an abstract dimension that underlay two physically and temporally dissimilar events (e.g., discontinuity in the pulsing tone and discontinuity in the dotted line). Another study demonstrated that four-year-olds already perceive and conceive of  similarities between pitch and brightness (e.g., low pitch equals dim; high pitc pi tch h eq equa uals ls br brig ight ht)) an and d be betw twee een n lo loud udne ness ss an and d br brig ight htne ness ss (e (e.g .g., ., so soft ft eq equa uals ls dim; loud equals bright). These findings are especially important because they parallel the idea that adults project image-schemas from one domain onto another, another, for example, conceptualizing quantity in terms of verticality (e.g., MORE IS UP and LESS IS DOWN). Finally, more recent research examined whether infants can construe an abstract unity between a facial expression of emotion (e.g., joy) and an audi au dito tory ry ev even entt (e (e.g .g., ., an as asce cend ndin ing g to tone ne), ), ev even ents ts th that at al also so sh shar aree no ph phys ysic ical al 1990)expressions features history of did co-occurrence (Phillips et al.,facial . The 7-month-old infants inorthis study not categorize different of joy and anger. But the infants did look significantly longer at joy, surprise, and sadness when these facial expressions were were matched with ascending, pulsing, and descending and continuous tones, respectively. Because the auditory and visual events in this experimental task were substantially

 

Cognitive Development

 

231

difffe ferren ent, t, in infa fant ntss ha had d to ac actt up upon on th thee ev even ents ts wi with thin in a sh shor ortt pe peri riod od of ti time me to di  bring meaning (i.e., determine equivalences) to the disparity disparity.. Thus, infants had ha d to de dete term rmin inee th thee eq equi uiva vale lenc ncee be betw twee een n fa faci cial al ex expr pres essi sion onss an and d au audi dito tory ry events. This is a striking demonstration of how infants metaphorically match disparate events to construe some meaning in facial expression expressionss of  emotion. There The re ar aree thr three ee bas basic ic pri princi nciple pless of int interm ermoda odall per percep ception tion (Ba (Bahri hrick, ck, 2000). First, global, abstract intermodal relations are detected earlier than are moree spe mor specifi cificc nes nested ted re relat lation ions. s. For exa exampl mple, e, glo global bal re relat lation ionss inv involv olvee sha share red d sync sy nchr hron ony y, as wh when en th thee si sigh ghtt an and d so soun und d of a ha hamm mmer er hi hitt ttin ing g th thee gr grou ound nd ar aree synchronized. Nested relations are more specific and reveal details about an ev even ents ts,, su such ch as th that at a sp spec ecifi ificc ob obje ject ct ma make kess a sp spec ecifi ificc so soun und d wh when en hi hitt ttin ing g the ground, whereas a compound object (e.g., a tray of cutlery) makes a complex set of sounds. The second principle claims that amodal relations are perceived earlier than arb than arbitr itrary ary re relat lation ions. s. Thu Thus, s, a per person son’s ’s voi voice ce is alw always ays syn synchr chroni onized zed with his or her mouth movements in such a way that the synchrony provides amodal information. But the precise sound of the voice is arbitrary and cannot be specified in advance. Moreover, it is not possible to predict the specific sound that a red object will make when dropped onto the floor floor.. The thi third rd pri princi nciple ple hol holds ds tha thatt the det detect ection ion of amo amodal dal re relat lation ionss fac facilit ilitate atess perc pe rcep eptu tual al le lear arni ning ng ab abou outt ar arbi bitr trar ary y rel elat atio ions ns.. Fo Forr in inst stan ance ce,, wh when en an in infa fant nt perceives the synchrony between his or her mother’s face and voice, he or shelearnstoassociatetheuniquesoundsofthevoicewiththatperson.Ifthe two tw o mo moda dali litie tiess ar aree no nott sy sync nchr hron oniz ized ed,, th then en th thee as asso soci ciat atio ion n of fa face ce an and d vo voice ice will wi ll no nott be le lear arne ned. d. Th This is pr prin inci cipl plee gu guid ides es th thee yo youn ung g in infa fant nt th thrrou ough gh a ma maze ze

of sights, sounds, and other natural mental combinations and provides a way wa y of or orga gani nizi zing ng pe perc rcep eptu tual al ex expe peri rien ence ce th that at le lead adss in infa fant ntss to mo more re ma matu turre knowledge of adults. The experimental work on multimodal perception is quite relevant to an embodied perspective on cognitive development. This work clearly demonstrates that young children are capable of making cross-modal, or cross-sensory, connections, which enable them to understand important aspects of objects and events in the world. Yet these cross-modal connections also form the bases of image schemas that underlie many concrete and abstract concepts. Imitation

Imitation is cognitive Piaget lack la ck th thee sk skil illlan to embodied imit im itat atee ad adul ult t ac acti tion ons, s,activity. such su ch as to tong ngue ue suggested prot pr otrrus usio ion nthat and an d infants mout mo uth h open op enin ing, g, be befo fore re th thee ag agee of 8 or 9 mo mont nths hs.. Bu Butt in infa fant ntss as yo youn ung g as 42 minutes after birth have crude imitative abilities, such as when they initiate moving tongue protrusions, mouth openings, and lip movements of an adult

 

232

 

Embodiment and Cognitive Science

(Meltzoff, 1990; Gallagher & Meltzoff, 1996). This ability to match visual with proprioceptive actions may be assumed to represent a “supramodal representational repr esentational system” or an embryonic body schema (Meltzoff, 1990). Piaget argued that deferred imitation is not exhibited under 18 months when whe n inf infant antss ar aree una unable ble to for form m men mental tal re repr prese esenta ntatio tions ns apa apart rt fr from om imm immedi edi-ate environmental stimuli. Yet 6-month-olds exhibit deferred imitation, or thee ab th abil ilit ity y to im imit itat atee an ac acti tion on se seen en 24 ho hour urss ea earl rlie ierr, su such ch as an ad adul ult’ t’ss fa faci cial al expression (Collie & Hayne, 1999; Meltzoff & Moore,  1994). At 9  months of age, infants can copy adults’ actions on objects (Meltzoff, 1990) and can reproduce repr oduce event sequences after a delay (Carver & Bauer, 1999). Later on,  by 14 mo mont nths hs,, in infa fant ntss ca can n mo mode dell a ra rang ngee of ad adul ult, t, as we well ll as pe peer er,, be beha havi vior orss towar tow ard d dif differ ferent ent obj object ectss (Ha (Hanna nna & Mel Meltzo tzoff ff,, 1993; Meltzoff, 1990). Mel Meltzo tzoff  ff  (1995) demonstrated that  18-month-olds will go beyond imitation per se and will perform the intended (i.e., initiated but not completed) actions of a model following a brief retention interval. These studies clearly show that deferred imitation is not a final achievement of the sensorimotor period but emerges early on and becomes more sophisticated by the end of  toddlerhood. Three hypotheses have been proposed to explain infants’ surprising imitative abilities (Meltzoff & Moore, 2000). First, imitation may be based on reinforcement from the parents. But parents are unaware of having some of the behaviors imitated in the first four weeks of life. Parental reinforcement, therefore, seems implausible as the source of early infant imitation. Second, imitation may be based on an innate releasing mechanism (Anifeld, 1996). Ac Acco corrdi ding ng to th this is id idea ea,, lip pr prot otru rusi sion on an and d se sequ quen entia tiall fin finge gerr move mo veme ment nt ar aree fix fixed ed-a -act ctio ion n pa patt tter erns ns th that at ge gett rel elea ease sed d wh when en in infa fant ntss se seee co corrresponding adult gestures. However, the variety of infants’ gestures suggests that their imitative abilities are not due to a small set of fixed-action patterns.

A final possi possibility bility is that imitati imitation on is based on the infan infantt s capacity to represent visually and proprioceptively perceived information on a form common to both modalities. Under this view, view, the infant compares sensory information from her own bodily movements to a “supramodal” representation of the visually perceived gesture gesture and construct a match between them th em.. Me Melt ltzo zofff an and d Mo Moor oree (1992, 1997) co cont nten end d th that at tr true ue im imit itat atio ion n is ma made de possible by active intermodal mapping (AIM) mechanisms that enable infants fan ts to mak makee cr cross oss-mo -modal dal equ equiva ivalen lence ce bet betwee ween n bod body y tra transf nsform ormati ations ons tha thatt are seen or heard and those that are felt on the basis of their own tactilekinesthetic action to generate the matching response. This idea is consiste tent nt wi with the th e po poss ssib ibili ility ty th that at ch child ildrren en’s ’s ea earl rly y se sens nsor orim imot otor or ac acti tion s pr prov ovid idee part of th the basis for abstract representational systems used inons successful imitation. Yet the psy psycho cholog logica icall pri primiti mitive ve und underl erlyin ying g inf infant ants’ s’ imit imitati ation on may not necessarily be due to either a body schema, or some developing

 

Cognitive Development

 

233

representational system. Instead, imitation may reflect primal animation of the tactile-kinesthetic body (Sheets-Johnstone,  1999). Whether infants are copying pure body movements of others, or how others interact with objects, their imitation reflects a dynamically attuned body and an organizing kinesthetic liveliness. What the infant sees is a replication of the dynamics of its own felt movement. Under this view view,, imitation is i s just one facet of learning to move oneself. In some cases, perhaps, reproducing reproducing an even ev entt ma may y be ev evok oked ed no nott th thrrou ough gh me memo mory ry,, bu butt vi viaa th thee me merre pr pres esen ence ce of an object and the actions that it may typically elicit or afford. This possibility emphasizes how sensorimotor activity is integral to infants’ understanding of objects, even when an infant has not previously performed these actions on a specific object. Of course, this idea is limited to conventional motor patterns and cannot explain babies’ developing abilities abili ties to recreate novel sequences or actions (see Mandler Mandler,, 2004). Anot An othe herr in inst stan ance ce of im imit itat atio ion n is ev evid iden entt in ol olde derr in infa fant nts’ s’ an analo alogi gica call re reen en-actm ac tmen entt of pr prev evio ious usly ly se seen en ev even ents ts.. Pi Piag aget et ar argu gued ed th that at se sens nsor orim imot otor or be beha havvior becomes representational via a process of interiorization. Overt behavior serves as the model for understanding real-world events via a process of analogy, specifically “motor analogies.” Piaget observed that his own children imitated certain spatial relations that they had seen in the physical world using their own bodies. For instance, Piaget’s children imitated thee op th open enin ing g an and d cl clos osin ing g of a ma matc tchb hbox ox by op open enin ing g an and d cl clos osin ing g th thei eirr ha hand ndss and mouths. Piaget suggested that this behavior showed that the infants are trying to understand the mechanism of the matchbox through a motor analogy, reproducing a kinesthetic image of opening and closing. Mental imag im ager ery y wa wass th then en ar argu gued ed to de deve velo lop p ou outt of kin kines esth thet etic ic im imita itatio tions ns as a res esult ult of progressive interiorization. In fact, various studies demonstrate that infants as young as 10 months can make spatial relational mappings, which can be readily transferred to new objects by infants of   13  months (Bauer,  1996;  Chen, Sanchez, & Campbell, 1997). In Chen et al. (1997), infants were presented with a doll that was out of their reach. The doll was also behin ind d a box and had a string attached to it that was lying on a visible cloth. Infants could bring the doll within reach by performing a series of actions, such as removing the

 box and pulling the cloth so that they could pull the string attached to the toy. Once infants could successfully perform this action and pull the doll toward themselves, they were presented with two different scenarios, each using identical i dentical tools (cloths, boxes, and strings). However, However, these new problems differed in that the cloths, boxes, and strings were all differren fe entt fr from om th thos osee en enco coun unte tere red d be befo forre. Mo More reov over er,, in th this is ne new w pr prob oble lem, m, tw two o stri st ring ngss an and d tw two o cl clot oths hs we werre pr pres esen ente ted, d, al alth thou ough gh on only ly on onee pa pair ir co coul uld d rea each ch the toy. Children 10 and 13 months old were tested in this experiment. Some of  thee ol th olde derr in infa fant ntss wo work rkeed ou outt th thee so solu luti tion on to rea eacchi hing ng th thee to toy y on th theeir ow own, n, where whe reas as oth others ers mod modele eled d the their ir par parent ents’ s’ sol soluti ution on bef befor oree suc succes cessfu sfully lly sol solvin ving g

 

234

 

Embodiment and Cognitive Science

the problem. Once the first problem was solved, the 13-month-old infants transferred an analogous solution to the second and third problem. But the 10-m -mon onth th-o -old ldss fo focu cuse sed d on on onee sa sali lien entt pe perrce cept ptua uall cu cuee be befo forre th they ey co coul uld d analogously move the first solution onto the second and third problems. These data generally support Piaget’s contention analogical  based on sensorimotor schemes, is a critical part that of how infantstransfer, learn to represent repr esent and reenact real-world events (see Gibbs, Gi bbs, 1994 for a further discussion of the relevant developmental evidence here). Developing a Theory of Others’ Minds

One of the hallmarks of child development is the acquisition of a “theory of mind.” Evidence of the child’s developing theory of mind comes from several sources, most notably false-belief experiments. In these studies, a participant is asked about the thoughts and actions of another person or character who lacks certain information that the participant knows. A partic par ticipa ipant nt kno knows, ws, for ins instan tance, ce, tha thatt a can candy dy box con contai tains ns pen pencil cils. s. Som Someon eonee else enters the room and the participant is asked, “What will the other pers pe rson on sa say y is in th thee ca cand ndy y bo box? x?”” Fo Four ur-y -yea earr-ol olds ds ty typi pica cally lly re resp spon ond d co corr rrec ectl tly y that th at th thee ot othe herr pe pers rson on wi will ll th thin ink k th ther eree ar aree ca cand ndie iess in th thee bo box. x. Bu Butt 3-year-olds are unable to see that the other person may falsely believe that there are can ca ndi dies es in th thee bo box x an and d res espo pon nd th that at th thee ot othe herr pe pers rson on wil illl say th thaat th ther eree ar aree actually pencils in the box. Several theories have been proposed to account for this developmental milestone. The “theory theory” claims that children’s theory of mind is generated as an innately specified domain-specific mechanism, or a theory of mind module, particularly designed for reading others’ minds (Baron-Cohen, 1995;  Leslie,  1991; Tooby & Cosmides,  1995). At first, the child develops a first-order belief that allows him or her to distinguish her own beliefs from someone else’s. Later, around the age of  3  or  4, the child acquires an ability to notice another person’s thoughts about a third person’s thoughts (a second-order belief attribution). In contrast to theory theory, simulation theory argues that a person understands something of another individual’s mind by pretending to be in that person’s shoes. More specifically, an observer tries to make his or her own mind emulate the thought processes that the other person is experiencing given the present situation. By recr recreating eating the other person’s presumed thought processes, the observer will come to understand that person’s point of view.

There is a large literature devoted to the debate between the “theory theory” and “simulation” views. Both positions maintain that the child’s deve de velo lopi ping ng th theo eory ry of mi mind nd co come mess ab abou outt as th thee ch chil ild d de deve velo lops ps a th theo eorret etic ical al stance involving the possible existence of mental states in others. A common mo n as assu sump mpti tion on un unde derl rlyi ying ng bo both th th theo eori ries es is th thee “m “men enta talis listic tic as assu sump mpti tion on””

 

Cognitive Development

 

235

(Gallagher,  2001), which states that “to know another person is to know that person’s mind, and their means to know their beliefs, desires, or intentional tentio nal state states” s” (Galla (Gallagher gher,, 2001: 91). Pe Peop ople le us usee th thei eirr “t “the heor ory” y” to ex expl plai ain n and predict the behavior of others. Thee fa Th fals lsee-be beli lief ef pa para radi digm gm,, ho howe weve verr, fa fails ils to ac acco coun untt fo forr ma many ny as aspe pect ctss of  primar pri mary y int inters ubject ectivit ivity y ava availa ilable ble childr chi ldren en ar aroun ound d the age of  that th at ma may y  be critical inersubj children’s abilities to to read others’ minds (Bloom &4 German, 2001). Something clearly happens around age  4  and the false-belief task may ma y ta tap p in into to so some me pa part rtss of th this is sp spec ecia iali lize zed d co cogn gniti itive ve ab abil ility ity.. Bu Butt ex expl plai aini ning ng and predicting, as required by the false-belief task, are indeed specialized acti ac tivi viti ties es,, an and d do no nott refl eflec ectt mu much ch ab abou outt ho how w ch chil ildr dren en,, or ot othe hers rs,, no norm rmal ally ly inte in tera ract ct wi with th on onee an anot othe herr, an and d to so some me ex exte tent nt,, rea ead d ea each ch ot othe her’ r’ss mi mind nds. s. Ex Ex-perime per imenta ntall par partic ticipa ipants nts mus mustt ado adopt pt thi third rd-pe -perso rson n per perspe specti ctive ve to sol solve ve the these se probl pr oblems ems,, whe where reas as sec second ond-pe -perso rson n int intent ention ionss ar aree the typ typica icall way wayss we int inter er-act with one another. Moreover, false-belief tasks require metarepresentationa tio nall pr proc oces esse sess th that at ar aree co cons nsci ciou ous, s, as op oppo pose sed d to th thee pr prima imary ry un unco cons nsci ciou ouss way people engage with and read others. In general, the empirical demonstrations that a child acquires a theory of mind at some age so he or she can consciously explain or predict what someone else, whom her or she is not interacting with, knows is not proof  that th at th thee ch child ild’s ’s pr prim imar ary y un unde ders rsta tand ndin ing g of ot othe hers rs is roo oote ted d so sole lely ly in a th theo eory ry of mind capabilities. Ther Th eree is a th thir ird d po poss ssib ible le th theo eory ry to co cons nsid ider er.. Th This is al alte tern rnat ativ ivee su sugg gges ests ts th that at children’s theory of mind is not primarily based on developing a theory of the other, or constructing an internal simulation. Instead, understanding another person is a form of embodied practice (Gallagher, (Gallagher, 2001). Even prior to devel developing oping an ability to rea read d other other’s ’s minds, young children children engage in embodied practices that are sensorimotor, perceptual, emotional, and an d no nonc ncon once cept ptua ual. l. Th Thes esee pr prac acti tice cess “c “con onst stit itut utee ou ourr pr prim imar ary y ac acce cess ss fo forr un un-derstanding others, and continue to do so even after we attain a theory of  mind abilities” (p. 85).children may acquire a “theory of mind” without Infants and young necessarily developing a mentalistic theory of the other. Even before the acqu ac quis isit itio ion n of a pu puta tati tive ve th theo eory ry-o -off-mi mind nd mo modu dule le,, in infa fant ntss ex exam amin inee th thee bo boddies and expressive movements of others to recognize people’s intentions or to find the meaning of some object. As early as  9 –14  months, infants look to the eyes of others to help interpret the meaning of an ambiguous event eve nt (Ph (Philli illips, ps, Bar Baronon-Coh Cohen, en, & Rut Rutter ter,, 1992). Th Thus us,, a ch chil ild d ca can n un unde ders rsta tand nd that another person is looking at a door and may have some intention vis-`a-vis a-vis that door from the person’s expressive body movements. Infants 5–7 mo mont nths hs ol old d ca can n de dete tect ct re relat latio ions nshi hips ps be betw twee een n vi visu sual al an and d au audi dito tory ry in info forrmation specifying emotional expressions (Walker,  1982). This perception of emotion is especially noted from the movement of others, rather than

from an infant having a theory or simulation of an emotional event. As

 

236

 

Embodiment and Cognitive Science

mentioned earli mentioned earlier er,, 5-m -mon onth th-o -old ldss re reco cogn gnize ize th thee em emot otio iona nall na natu turre of hu huma man n movement, as demonstrated in their preferential attentiveness to human shapes in point-light displays of actions (Bertenthal et al.,  1994). The emotional states of others are not something that infants must infer, but something that is directly perceived in the movement and experience of other’s  behaviors (Allison & Puce, & McCarthy McCarthy,,  2000). This also explains why 11-month-olds can recognize the intentional boundaries in some continuous scenes (Baldwin & Baird,   2001), and why   18-month-olds can

understand what other persons intend to do, and may even, under some situations, complete actions that another person had not completed (Meltzoff, 1995). None of these abilities demand that the child must infer the mental states of others. Gall Ga llag aghe herr ar argu guees th that at eve ven n bef efor oree a ch chil ild d de dev vel elop opss a th theo eory ry of mi mind nd,, he or she already has understandings of others, and this experience includes “(a) “( a) an un unde ders rsta tand ndin ing g of wh what at it me mean anss to be an ex expe peri rien enci cing ng su subj bjec ect, t, (b (b)) an understanding of what it means that certain kinds of entities (but not others) er s) in th thee en envi viro ronm nmen entt ar aree in inde deed ed su such ch su subj bjec ects ts,, an and d (c (c)) an un unde ders rsta tand ndin ing g that in some way these entities are similar to and in other ways different from oneself” oneself ” (Gallagher, 2001: 86). Meltzoff and Brooks (2001) speculate, along these lines, that infants’ construal of human acts as intentional and goal go al-d -dir irec ecte ted d ar aris ises es fr from om a “l “lik ikee me me”” an anal alog ogy y th that at ac acts ts as th thee st star arti ting ng po poin intt for social cognition. Most generally, this alternative claims that primary intersubjectivity is not just a precursor to children’s developing theory of mind. Instead, primary intersubjectivity comprises a set of embodied practices that are primary, not just in the developmental sense, but in the sense that anyone can explain or predict what other people believe, desire, or intend in the practice of their own minds (Gallagher, 2001). Fina Fi nall lly y, th thee ch child ild’s ’s ac acqu quis isit itio ion n of a th theo eory ry of mi mind nd al also so in invo volv lves es en ente teri ring ng a community of minds, where people with different minds communicate for common purposes and understandings (Nelson et al., 2003). Thus, the  burden constructing a model minds restthe onlarger the child’s individualofcognitive processes, butof rather “isdoes a giftnot from community that incorporates these constructs into its language and its talk about concerns of people within the community” (Nelson et al.,  2003: 43 ). This view is consistent with an experientialist view of cognitive development in which the sources of what children know are seen as rooted both in the cond co ndit itio ions ns of ex expe peri rien ence ce in th thee sp spec ecia iall so soci cial al an and d cu cult ltur ural al wo worl rld, d, an and d in th thee phen ph enom omen enol olog ogy y of ex expe peri rien enci cing ng on th thee pa part rt of th thee ch chil ild. d. Th Thee res esul ults ts of tw two o studies of  3  3- and 4-year-old children engaging in differe different nt theory-of-mind tasks and discussing their responses suggests that changes in the understanding of their own and other’s mental states reflect their participation in communities in which social interactions are directed toward specific pragmatic purposes.

 

Cognitive Development

 

237

Mental Imagery

Embodied action may be critic Embodied critical al to childr children’s en’s developing developing imagery abilities. Studies show that the ability of children (ages  5–8) to form dynamic mental images of objects is greatly facilitated if they first engage in overt manipulation of the objects (Wolff & Levin,  1972). Interestingly, this was true tr ue eve ven n wh when en th thee ch chil ildr dren en had no vi vissua uall ac acce cesss to th thei eirr mo move veme ment ntss or to the objects being manipulated. Other studies show that handling objects is cr crit itic ical al to ch chil ildr dren en’s ’s su succ cces esss in me ment ntal al rot otat atio ion n ta task sks, s, at le leas astt up un unti till th thee ages ag es of 9 and 10 (Zabalia, 2002). Bo Both th se sets ts of fin findi ding ngss su supp ppor ortt Pi Piag aget et’s ’s id idea ea on the role of overt and covert activity in the creation of mental imagery. imagery. The importance of embodied movement in imagery is also shown in a different study in which children either watched someone build four simple block buildings, looked at the completed buildings, pretended to  build the building along with the experimenter, experimenter, or constructed the buildings themselves while looking at pictures of them (Corriss & Kose,  1998). Afterwards, the buildings were removed and the children were asked to reconstruct the configurations. Children were more accurate when they either imagined or actually constructed the buildings, rather than looking at the finished products or watching the construction. These results show that if children in some way actthe physically physically, , eitherthey by mimicking mimi theclearer building process or by performing construction, havecking a much mental image of the building. Imagining actions of oneself is more effective than imagining the actions of another person. The results support the theory that imagination is an action-based process.

Conclusion

Developme Develo pmenta ntall psy psycho cholog logist istss con contin tinue ue to deb debate ate the ro role le tha thatt sen sensor sorimo imotor tor activity plays in cognitive development. Although there is a large body of  eviden evi dence ce sug sugges gestin ting g spe specifi cificc way wayss tha thatt dif differ ferent ent emb embodi odied ed exp experi erienc ences es underl de rlie ie va vari riou ouss as aspe pect ctss of co cogn gniti itive ve gr grow owth th,, to too o mu much ch of th thee de deve velo lopm pmen enta tall work ignores the child’s own tactile-kinesthetic in bootstrapping cognition. Let me present one final exampleexperiences of this problem. Childr Chi ldren’ en’ss und unders erstan tandin ding g of the con concep ceptt of bal balanc ancee has bee been n ext extens ensive ively ly studied by developmental psychologists, especially in the context of children’s physical reasoning skills. Most generally, this work demonstrates that as children get older, they are to take account of more variables in deal de alin ing g wi with th di difffe ferren entt ba bala lanc ncin ing g ta task sks. s. Fo Forr ex exam ampl ple, e, th thee st stud udy y of ho how w ch chil il-dren learn to balance objects is primarily seen as a matter of learning how to co comb mbin inee in info form rmat atio ion n ab abou outt di diff ffer eren entt ph phys ysic ical al va vari riab able les. s. Of co cour urse se,, bo both th chil ch ildr dren en an and d ad adul ults ts ca can n en enga gage ge in di diff ffer eren entt ba balan lanci cing ng ac acti tivi vitie ties, s, be bett tter er th than an theey can ex th expl plai ain n how th theey do it it.. Th Thee ta tassks us useed to as asse sess ss ch chil ildr dren en’s ’s un unde derrstandings of the balance concept involve showing a child some pictures of 

 

238

 

Embodiment and Cognitive Science

a seal trying to balance a ball on its nose and asking the child to select the onee in wh on whic ich h th thee se seal al wa wass ba bala lanc ncin ing g th thee ba ball ll co corr rrec ectl tly y (P (Pin inee & Me Mess sser er,, 2003). A different task required children to actually place a set of six blocks on each side of a fulcrum so that all the blocks would balance. In one study, 5- and 6-y -yea earr-ol old d ch child ildre ren n en enga gage ged d in th thes esee an and d ot othe herr ba bala lanc ncin ing g ta task skss ov over er the course of   5  days. Not surprisingly, the children improved over this time, but they still experienced difficulties in explaining correct balancing  behavior,, suggesting that they did not have adequate conscious access to  behavior representations about balancing. representations I have no doubt about the accuracy of these findings. But it is remarkable that the study of children’s developing ideas about balance did not acknowledge the child’s own tactile-kinesthetic experiences of balancing across sensory modalities and full-bodied sensations and movements. Balance Balan ce is not an abstract idea or principle that we learn as a rule, but is something that is a constant part of our embodied experiences as living organisms. The image schema for BALANCE arises from these varying embodied activities, and, as argued in earlier chapters, is a critical part of many abstract concepts. My plea for developmental psychologists is to find clever ways of assessing how concepts such as balance, or balancing, may ma y be gr grou ound nded ed in em embo bodi dime ment nt,, an and d no nott to as assu sume me th that at so solv lvin ing g ph phys ysic ical al reasoning problems somehow develops as a pure cognitive competence apart from sensory, embodied experience. My argument in favor of an embodied perspective on cognitive development is not entirely consistent with Piaget’s claim, because he underestimated young children’s cognitive abilities, and assumed, incorrectly, that higher-level forms of cognition are abstracted away from sensorimotor action. As I suggested in previous chapters, many of our embodied experiences are incorporated into our conceptual and linguistic repre represensentations. Moreover, our phenomenological bodily experiences continue to reaffirm and ground cognitive symbols.

 

8

Emotion and Conscio Consciousness usness

The stream of consciousness reveals many insights into who we are, how we think, and what we feel. Consider the case of Charlie Kaufman, a characte ac terr in th thee fil film m Adaptation  Adaptation,, pl play ayed ed by Ni Nich chol olas as Ca Cage ge.. Ka Kauf ufma man n is a sc scrree eennwriter who has been hired to adapt a book published by Susan Orleans on people peo ple who ar aree fan fanati atical cal abo about ut or orchi chids. ds. Unf Unfort ortuna unatel tely y, Kau Kaufma fman n str strugg uggles les with this assignment and worries about an upcoming meeting with his agent. As he paces in his living room one morning, trying to come up with ideas for the adaptation, Kaufman thinks to himself (in voice over): I’m old. I’m fat. I’m bald. (reaches for notebook, catches sight of bare feet) My toenails have turned strange. I am old. I am – (flipping through notebook, paces) I have nothing. She’ll think I’m an idiot. Why couldn’t I stay on that diet? She’ll pretend not to be disappointed, but I’ll see that look, that look – (passes mirror, glances quickly at reflection, looks away) God, I’m repulsive. (another glance) But as re repu pulsi lsive ve as I th think ink?? My Bo Body dy Dy Dysm smor orphi phicc Dis Disor order derco confu nfuses sesev every eryth thing ing.. I me mean, an, I know people call me Fatty behind my back. Or Fatso. Or facetiously Slim. But I also reali realize ze this is my pervert perverted ed form of self-ag self-aggrandize grandizement, ment, that no one talks about me at all. What possible interest is an old, bald, fat man to anyone?

Many people, like despondent Charles Kaufman, struggle with the ebb and flow of their emotional states, and emotions often dominate the stage of consciousness. Each of us may not have the same negative self-image that Kaufman has, but all people typically experience running narratives in th thee mi mind nd’s ’s th thea eate terr ab abou outt th them emse selv lves es,, ot othe herr pe peop ople le,, an and d th thee wo worl rld d ar arou ound nd them th em.. Ou Ourr in inne nerr vo voic ices es ty typi pica cally lly pr prod oduc ucee bi bits ts an and d pi piec eces es of th thou ough ghts ts ra rath ther er than fully formed sentences. This narrative is tightly linked to the self,  because we feel that the voice is the “I” talking. The language-like quality of th thee st strrea eam m of co cons nsci ciou ousn snes esss su sugg gges ests ts to ma many ny th that at co cons nsci ciou ousn snes esss is qu quite ite divorced from the body, again reaffirming Descartes’ famous observation, “I think, therefore I am.” But the highly private, idiosyncratic nature of  239

 

240

 

Embodiment and Cognitive Science

consci ousnes nesss has mos mostly tly dis discou courag raged ed sci scient entific ific psy psycho cholog logist ists, s, and oth others ers,, conscious from studying these aspects of the mind. Fortunately, the situation in cognitive science has now changed. Both emotions and consciousness are now recognized as ideal phenomena through which to study the relationship between mind and body. One important conclusion that is slowly emerging from this work is that both emotions and consciousness partly arise from, and are expressed in, em-

 bodied action. This chapter describes the importance of embodiment in emotion and consciousness.

Emotion

The Language of Emotions Thee wa Th way y we sp spea eak k ab abou outt em emot otio ion n pr prov ovid ides es a si sign gnifi ifica cant nt cl clue ue to un unde ders rsta tand nd-ing emotional experience. A poignant example of how people experience thei th eirr em emot otio ions ns in te term rmss of em embo bodi died ed mo move veme ment nts, s, wh whic ich h ha have ve bo both th te text xtur uree and depth, is seen in the following conversation between a man and his psycho psy chothe therap rapist ist whe where re the two par partic ticipa ipants nts neg negoti otiate ate the mea meanin ning g and per per-sonal son al imp implic licati ations ons of sev severa erall met metaph aphors ors (Fe (Ferra rrara, ra, 1994: 139–41). Ho Howa warrd is a client in his thirties who is talking with Judy, Judy, his therapist, in their third session. One month earlier, Howard had been fired from his job as an orderl de rly y in a ho hosp spit ital al be beca caus usee of su susp spic icio ion n ov over er so some me mi miss ssin ing g dr drug ugs. s. Ho Howa warrd

maintained that he did not steal the drugs and was eventually reinstated. judy:   “When you have a problem, what do you do with it?” howard:   “I usually let it be a problem. I don’t usually do anything much or I . . . I was thinking about that the other day.” judy:   “Does the problem go away if you don’t do anything about it?” howard:  “No, it gets worse . . . or it just complicates things as you go further down the road.” judy:   “Can you look at your own life, kind of on a continuum? Look down the road of that line and see what that’s gonna do . . . in your life?” howard:   “Look on down the road.” judy:   “Y “Yeah, eah, kinda visual visualize ize what un . . . your own life will be like if  you don’t deal with some of it . . . your problems . . . Can you see how it might complicate . . . your life?” howard:  “It will wil l just continue the way it is.” judy:   “Kind of like a snowball . . . effect.” howard:   “No no not a snowball. Just kinda floating, floating down the river.” judy:   “Floating down the river river.” .” howard:   “That’s what I’m doing now. now. That’s what I was afraid I was gonna go back into all this. I said something the first time I talked to you about.”

 

Emotion and Consciousness

 

241

judy:   “Yeah.” howard:   “Floating and being afraid of going back into floating. That’s

 just you know, know, floating, drifting . . . ” judy:   “So you’re adrift right now?” howard:   “Y “Yeah. eah. And feel dead and I feel like I’m – I drink to feel a little  bit deader. deader. No, that’s not true.” judy:   “Feel depressed . . . or numb?” howard:  “Yeah.” judy:   “Numb, you feel?” howard:  “Yeah. Yeah.” judy:   “What’s it like to be floating down the river? Tell me more.”

howard:  “It’s comfortable. It’s safe . . . Everything just keeps on an even

keel, you know.” judy:   “Mmmhmm.” howard:  “You’re just kinda floating . . . ” judy:   “Kind of in a canoe? . . . going down the river, or” howard:   “No, more like a great ole big barge . . . on a great old big river.” judy:   “Barge, very stable, kinda.” howard:  “Yeah, plenty of room to spread out and . . . sit in the sun. Yeah, and you don’t have to worry about falling off the edge.” judy:   “Mmhmm.” howard:  “And sun, you know, it’s kinda hazy. It’s not really clear sun. It’s kinda hazy.”  Judy;    “Kinda “Mmhmm.” howard: half asleep, that what’s it’s like . . . ” judy:   “What happens when you kind of come to the . . . falls, the falls that are down there, about two miles down the river?” howard:  “Get the hell off the river.” judy:   “That’s certainly one way to handle it. Get out.” howard:   “I feel a lot of discomfort. That’s what happened just last month. I hit those falls last month.” (noise) judy:   “I don’t know why it did that. So that’s what happened . . . um this . . . last time there was kind of um . . . an external situation that sort of forced you out of your boat.” howard:   “It was uncomfortable but I was, I was pretty, pretty, I was enjoying it too. And I didn’t want to go back to just floating. It was uncomfortable and I was out, I don’t know, know, I been floating a long time.” judy:   “Mmhmm . . . Well, you’ve found that it works for you . . . in a sense.” howard:  “What works for me?” judy:   “Floating.” howard:  “Because I’m . . . stay . . . comfortable and” judy:   “In a sense, but it may . . . now be inappropriate. It may not be working as well . . . as it did in the past.”

 

242

 

Embodiment and Cognitive Science

Yeah, I’d like to have a little excitement now and then” howard:   “Mnnn. Yeah, judy:   “Some rapids” howard:   “Y “Yeah eah (laughs) Something I can keep in control of maybe and not drown. But . . . yeah, I think I am bored.” This conversation is notable because of the way embodied metaphors structure the discourse. For instance, the client, Howard, introduces the metaphorical idea of his future life as being “down the road.” The therapist pi st pi pick ckss up on th this is id idea ea an and d as asks ks Ho Howa warrd to el elab abor orat atee on th thee me meta taph pho or in her questions abut visualizing the road ahead and whether he feels adrift at present. Howard provides more detail when he rejects the therapist’s question about life being like a “snowball effect” (i.e., a forceful felt movement of bei ment being ng swe swept pt awa away y unc uncont ontro rolla llably bly). ). Ins Instea tead, d, his exp experi erienc encee fee feels ls like “Jus “J ustt ki kind ndaa flo float atin ing, g, flo float atin ing, g, do down wn th thee ri rive verr.” Cli Clien entt an and d th ther erap apis istt fu furt rthe herr extend their metaphorical characterization of how Howard feels about his

life by ta life talk lk of ke keep epin ing g on an ev even en ke keel el,, ex expe peri rien enci cing ng som omee ra rapi pids ds,, an and d not having “to worry about falling falli ng off the edge.” The metaphorical meanings here seem quite appropriate, and perhaps highly desirable, in this therapeutic context. Howard and Judy’s conversation illustrates the value of felt movement in talking talki ng about one’s emotional experiences. When Wh en un unde derrgoi oin ng a st strron ong g em emo oti tion onal al ex expe peri rien ence ce,, we fe feel el as if we ar aree in thee gr th gras asp p of an em emot otio ion, n, as if we ar aree be bein ing g sw swep eptt aw away ay by it itss ho hold ld an and d fo forrce ce.. Many cognitive linguistic studies on the metaphorical nature of emotion talk illustrate the importance of movement in people’s emotional experiences (Harker & Wierzbicka, 2001; Kovecses, 2000a, 2000 b  b;; Y  Yu, u, 1999). For instance, Kovecses (2000a,  2000 b  b))  provides numerous examples of how emotions are understood as forces that appear to change people’s embodied positionings. Consider some of these conceptual metaphors, and relevantt lin van lingui guisti sticc exa exampl mples, es, tha thatt ar aree spe specific cific ins instan tantia tiatio tions ns of the gen generi eric-l c-leve evell EMOTION IS FORCE metaphor: EMOTION IS AN OPPONENT “He was seized by emotion.” “He was struggling with his emotions.” “I was gripped by emotion.” “She was overcome by emotion.” EMOTION IS A WILD ANIMAL “His emotions ran away with him.” “She kept her emotions in check.” “He couldn’t hold back his feelings.” EMOTION IS A SOCIAL FORCE “He was driven by fear.” fear.” “His whole life was governed by passion.” “He was ruled by anger.”

 

Emotion and Consciousness

 

NATURAL TURAL FORCE EMOTION IS A NA “I was swept off my feet.” “I was overwhelmed by her love.” EMOTION IS A MENT MENTAL AL FORCE “Our emotions often fool us.” “His emotions deceived him.” “She was misled by her emotions.” EMOTION IS INSANITY “She was beside herself with emotion.” EMOTION IS PHYSICAL AGITATION “The speech stirred everyone’s feelings.” “I am all shook up.” “He was slightly ruffled by what he heard.” “The children were disturbed by what they saw.” saw.” EMOTION IS A PHYSICAL FORCE “When I found out, it hit me hard.” “That was a terrible blow blow.” .” “She knocked me off my feet.”

243

They gravitated toward each other immediately immediately.. “I was magnetically drawn dr awn to her.” “I am attracted to her.” “That repels me.”

These different conceptual metaphors together illustrate the most pervasive folk theory of the emotion process in English (Kovecses,  2000 b  b)): (1) cause of emotion – force tendency of the cause of emotion – (2) self has emo em oti tion on – for orcce te ten nde den ncy of em emo oti tion on – (3) se self lf’s ’s fo forrce te tend nden ency cy – em emot otio ion’ n’ss force tendency – (4) resultant effect. This schema reflects our basic understanding of emotions as different physical/embodied forces interacting with one other. Emotions as Felt Movements The lin lingui guisti sticc evi eviden dence ce sug sugges gests ts tha thatt emb embodi odimen mentt is cen centra trall to und unders erstan tandding in g em emot otio iona nall ex expe peri rien ence ce.. Mo Most st co cogn gniti itive ve th theo eori ries es ad admi mitt th that at an im impo port rtan antt  body component in the emotion process is the readiness to take action (Lazarus, 1991; Oatley, 1992). This readiness to act is a corporeal felt urge to do so some meth thin ing g – ap appr proa oach ch so some meon one, e, st stri rike ke so some meth thin ing g or so some meon one, e, to touc uch h something, run away from something or someone, and so on. Emotion is not identical to simple action such kicking, embracing, running, and so on, but reflects a change in postural attitude, or an affective sense of such action (Sheets-Johnstone, 1999). Thee fu Th fund ndam amen enta tall rel elat atio ion n be betw twee een n em embo bodi died ed ac actio tion n an and d em emot otio ion n is ca cappturedbytheideathatto“bemoved”referstofeelingasifoneisinadifferent

 

244

 

Embodiment and Cognitive Science

position in regard to one’s situation. The word “emotion” is derived from the La the Lati tin n “e “e”” (o (out ut)) an and d “m “mov over ere” e” (t (to o mo move ve). ). Th Thee em emph phas asis is on mo move veme ment nt in emotion is a recurrent theme in the psychological literature. For example, Adle Ad lerr (1931: 42) de defin fined ed em emot otio ions ns as “p “psy sych chol olog ogic ical al mo move veme ment nt fo form rms, s, lim lim-ited in time,” whereas Arnold and Gasson (1954: 294) suggested “that an emotion or an affect can be considered as the felt tendency toward an ob ject judged suitable, or away away from an object object judged unsuitable, reinforced reinforced  by specific bodily changes according to the type of emotion.” Having an emotion may clearly involve some sense of bodily movement. One early study surveyed people about their sense of movement when they were thinking of different emotion terms (Manaster, Cleland, & Brooks,  1978). Parti Participan cipants ts rated their felt movem movement ent (i.e., eithe eitherr towar toward d others or away from others) for  140  emotion terms. The results showed that there were a group of  20  20 emotion words that tended to move people toward others (e.g., love, jolly, affectionate, sexy, confident, sentimental), and an d an anot othe herr gr grou oup p of 20 em emot otio ion n wo worrds th that at mo move ved d pe peop ople le aw away ay fr from om ot othhers (e.g., hate, humiliated, sulky sulky,, bitter bitter,, guilty guil ty,, aggravated). Other studies show sh ow a st stro rong ng as asso socia ciatio tion n be betw twee een n af affe fect ct an and d sp spat atia iall po posi sitio tioni ning ng,, su such ch th that at people are faster to classify positive words as “good” when they are presented toward the top of a computer screen, and negative words as “bad” when wh en th they ey ar aree in th thee lo lowe werr po porrti tio on of th thee sc scrree een n (M (Mei eier er & Ro Robi bins nson on,, 2004). Thes Th esee res esul ults ts ar aree co cons nsis iste tent nt wi with th ou ourr me meta taph phor oric ical al id idea eass th that at GO GOOD OD IS UP and an d BAD IS DO DOWN WN (L (Lak akof offf & Joh ohn nso son n, 1980, 1999). Fi Find ndin ings gs su such ch as th thes esee support the idea that the primary feeling of having an emotional experi-

ence is that of being moved. Each emotion reflects different, sometimes subtle, bodily movements. We We may at times experience some emotion as a state of passively being moved rather than as moving ourselves. Emotional experience involves our perceptible source of meaningful chan ch ange ge in a si situ tuat atio ion n an and d in ou ours rsel elve vess un unde derr so some me ci circ rcum umst stan ance ce.. Em Emot otion ionss aris ar isee as we be beco come me di disp spla lace ced d an and d di disl sloc ocat ated ed to an anot othe herr po posi sitio tion n in ad adap aptiv tivee response to some situation. One way of characterizing the felt dimension of emotional experience is in terms of “affective space,” or the space we move mo ve th thro roug ugh h as we ex expe peri rien ence ce di dist stin inct ct em emot otio ions ns (C (Cat atal aldi di,, 1996). Th This is id ideea of affective space is nicely illustrated by considering how we hesitate in advancing when worried, gently blossom when in love, distinctly loiter about when sad or depressed, or suddenly burst forward when feeling outraged. People experience their emotions as movements toward something in themselves. When people feel joy, they have repositioned themselves as  being “on top of of the world,” or when they feel feel emotionally troubled, troubled, then they experience a burden on “one’s shoulders,” where there is a downward turn of the body as the head drops and the person slouches. Feeling superior to another makes us feel as if we are “looking down” on that

 

Emotion and Consciousness

 

245

person, or that he or she is “beneath us.” Feeling admiration for another makes us “look up” to that person. Different emotions imply varying levels of removedness from others. When I fe When feel el ov over erwh whel elme med, d, th thee wo worl rld d se seem emss to too o cl clos ose, e, su sufffo foca cati ting ng me me.. Be Bein ing g in lo love ve su sugg gges ests ts a cl clos osen enes esss or pr prox oxim imit ity y to ou ourr lo love ved d on one, e, wh wher erea eass ha hatr tred ed repositions us away from others. When I feel lonely lonely,, I experience my body as separated from others. Feeling fear drives us away from others, and is accompanied, like all emotions, by a complex set of action tendencies. Consider the following phenomenological account of the kinesthetics involved in fear (Sheets-Johnstone, 1999: 269): “An intense and unceasing whole-body tension drives the body forward. It is quite unlike the tension one feels in a jogging run, for instance, or in i n a run to greet someone. There is a hardness to the whole body that congeals it into a singular tight mass; thee dr th driv ivin ing g sp spee eed d of th thee mo move veme ment nt co cond nden ense sess ai airb rbor orne ne an and d im impa pact ct mo moun unts ts into a singular continuum of motion. The head-on movement is at times erratic; there are sudden changes of direction. With these changes, the legs le gs mo move ve su sudd dden enly ly ap apar art, t, mo mome ment ntar aril ily y wi wide deni ning ng th thee ba base se of su supp ppor ortt an and d  bending at the knee, so that the whole body is lowered. The movement is each time abrupt. It breaks the otherwise unrelenting and propulsive speed of movement. The body may suddenly swerve, dodge, twist, duck, or crouch, and the head may swivel about before the forward plunging run with its acutely concentrated and unbroken energies continues.” As th this is pa pass ssag agee rev evea eals ls,, af affe fect ctiv ivee sp spac acee ha hass a se sens nsuo uous us fe feel el to it it,, a te text xtur uree that th at ma make kess it ne neit ithe herr pu purrel ely y me ment ntal al,, or re redu duci cibl blee to th thee ph phys ysiol iolog ogic ical al bo body dy.. In fact, the cognitive component of emotion may be based on these felt, tactile dimensions of emotional feeling (Cataldi,  1996). This is precisely why we speak of having been “touched” when we have been emotionally

affected by something, as when a situation is felt as touching. It would  be surprising if emotion and touch did not overlap given that we talk of  “feeling” in connection with the body. For instance, we sense “butterflies in our stomach” when feeling anxious (such as when first falling in love). This feeling cannot be objectively defined as a spastic stomach apart form some situation that elicits this response. After all, we may feel a spastic stomach without experiencing any particular emotion, such as when something we ate upsets our stomachs. This is why when feeling apprehensive we experience butterflies fluttering in our stomach, rather than, more simply, having a spastic stomach. The embodied feeling here is kinesthetically similar to butterflies fluttering because we sense our apprehension as an intermingling of things we can ca n to touc uch h on th thee “o “out utsi side de”” (e (e.g .g., ., bu butt tter erfli flies es flu flutt tter erin ing) g) wi with th th thee fe feel elin ing g “i “innside” our stomach. Thus, emotions are not simply or completely “mental sensations” but rely on tactile, felt feelings from the outside that become part of our inner emotional experiences.

 

246

 

Embodiment and Cognitive Science

Moving through affective space has a textured, palpitably felt dimension, just in the way that we can feel different textures of substances we touch with our skin. Physical substances we touch have a depth to them, and this is precisely why our emotions are also experienced at different leve le vels ls of de dep pth th.. Th Thee la lan ngu guag agee pe peop ople le us usee to ta talk lk ab abo out th thee nu nuaanc ncees of th theeir emotions, once again, reveals important aspects of the textured, in-depth feel of different emotions. Consider some of the felt textures associated with different emotions (Cataldi,  1996). For example, when feeling very frig fr ight hten ened ed,, we fe feel el ou ourr bo bodi dies es to be fr froz ozen en so soli lid, d, al almo most st “p “pet etri rifie fied” d”;; we ar aree radiant with love or bask in pride, or drown in sorrow, or effervescently  bubble over in happiness, or blissfully walk on air in joy joy,, or wallow in selfpity, or cautiously trend on pins and needles when feeling apprehensive. We feel steamy when lustful; we feel dry, stifled, and stale when bored. Yet being serene feels smooth, whereas gratitude has a plush, or lavish feel. When we are simply worn out, we may feel affectively stuck in some situations, as when we are in a pinch or a jam. Most generally, each emotion is distinguishable by skin-deep textures that are felt when we move through affective space. Furthermore, the greater the emotional extreme, the more depth we feel in our textured experiences. An emotional state need not be revealed in immediate overt acti ac tion ons, s, bu butt it ce cert rtai ainl nly y imp implie liess th thee hi high gh pr prob obab abili ility ty of ac actio tions ns th that at wi will ll so soon on  be directed outward from an individual into the world (Freeman, 1999). Such states are easily recognized and explained as intentional in many situations, but in others they seem to boil up spontaneously and illogically within an individual in defiance of intent. In cases when we do not move our bodies, we feel our emotions as if  something within us has moved (De Rivera, 1977). An emotion may have distinctive kinetic forms that are dynamically congruent with it, but these forms are not identical with the emotion. We may distinguish between an emot em otio ion, n, in te term rmss of it itss af affe fect ctiv ivee fe feel el an and d an any y po post stur ural al at atti titu tude dess it ex exhi hibi bits ts,, and the actual kinesthetic form that manifests the emotion. People may corporeallyexperienceanemotion,eventhoughtheactualbodymovement

does not occur. Thus, people may inhibit the movement associated with an emotion if necessary. We may learn to mentally simulate our actions – movingquickly,flailingourarmsaround,gettingredintheface,andsoon– without engaging in these actions. In this way, way, emotions are kinesthetic or potentially kinesthetic, and what is kinetic may be affective or potentially affective (Sheets-Johnstone, 1999). Basic Emotions and Facial Expressions Mostt cog Mos cognit nitive ive sci scient entist istss ign ignor oree the imp import ortanc ancee of ful full-b l-bodi odied ed mov moveme ement nt in empi em piri rica call st stud udie iess of em emot otio ions ns.. Ho Howe weve verr, ov over er th thee pa past st 30 ye year ars, s, on onee ki kind nd of   bodily action, facial expressions, has been extensively studied. One widely held he ld be beli lief ef is th that at sp spec ecifi ificc fa faci cial al ex expr pres essi sion onss in indi dica cate te,, an and d de defin fine, e, un unive ivers rsal al

 

Emotion and Consciousness

 

247

“basic emotions” (Ekman,  1992, 1994). This view holds the following assumptions (adopted from Fernandez-Dols, Carrera, & Casado, 2002): (a)   There are a small number (seven plus or minus two) of basic (a) emotions. (b) Each (b)  Each basic emotion is genetically determined, universal, and discrete. (c)   Each basic emotion is a coherent pattern of facial behavior, (c) behavior, physiology, and instrumental action. (d) Any (d)  Any state lacking its own distinct facial expression is not a basic emotio emo tion. n. The There reis is con consen sensus sus abo about ut the exi existe stence nce of six bas basic ic emo emotio tions: ns: happiness, surprise, fear, anger, disgust, and sadness. (e)   Al (e) Alll em emot otio ions ns ot othe herr th than an th thee ba basi sicc on ones es ar aree su subc bcat ateg egor orie iess or mi mixt xtur ures es of the basic emotions. (f) Expressions (f)  Expressions of emotion are spontaneous. Voluntary facial expressions can simulate spontaneous ones. (g)   Differen (g) Differentt cultures establish different display rules. Display rules inhibit, exaggerate, or mask spontaneous expression. (h)   Th (h) Thee ex expr pres essi sion onss of ba basi sicc em emot otio ions ns ar aree ea easi sily ly rec ecog ogni nize zed d by al alll hu huma man n  beings. (i) The (i)  The ability to recognize an expression of basic emotion is innate rather than culturally determined. (j)  (j)   The true criteria for the existence of a basic emotion are to be found in people’s facial movements. Verbal reports of emotion can be bypassed. (k)   The meaning of a facial expression of basic emotion is invariant (k) across changes in the context in which it is produced. Thesee sp Thes spec ecifi ificc as assu sump mptio tions ns ar aree su supp ppor orte ted d by em empi piri rica call st stud udie iess in wh whic ich h part pa rtic icip ipan ants ts ar aree sh show own n a sm smal alll se sett of pi pict ctur ures es of pr prot otot otyp ypic ical al fa faci cial al ex expr pres es-sions and asked to assign each one to a category of basic emotion (e.g., happiness, sadness, anger, fear, disgust, and surprise). The results show that people in many different cultures regularly associate specific facial expressions with particular emotions (Ekman, 1985). Howeve How everr, the there re ar aree sev severa erall met method hodolo ologic gical al pr probl oblems ems wit with h the these se stu studie diess that th at ra rais isee qu ques estio tions ns ab abou outt th thei eirr in inte terp rprret etat atio ion n (F (Fer erna nand ndez ez-D -Dol olss et al al., ., 2002; Russell,  1995). First, the experimenters in these studied did not usually

speak the participants’ languages in the cross-cultural studies. Many of  the terms used to specify basic emotions may have had a different meaning in g fo forr th thee na nati tive ve sp spea eake kers rs th than an fo forr th thee ex expe peri rime ment nter ers. s. Se Seco cond nd,, mo most st of th thee stud st udie iess us used ed fo forc rced ed-c -cho hoic icee re resp spon onse se fo form rmat atss th that at ma may y ha have ve co comp mpel elle led d pa parrticipants to pair specific faces with particular emotion terms in a way they would not do in ordinary life. Thus, the “recognition” of emotion in these facial experiments is not necessarily representative of universal entities known by an individual, but is rather an attribution in which people link

 

248

 

Embodiment and Cognitive Science

some emotions to some facial patterns using lay explanations that lack any an y ne nece cess ssar ary y or su suffi ffici cien entt re rela lati tion onsh ship ip to th thee ac actu tual al ex expe peri rien ence ce of em emot otio ion n and its behav behavioral ioral consequences consequences.. For example, people may frown when they threaten someone, and most people are angry before they threaten some so meon one. e. An id idea eall de desc scri ript ptio ion n wo woul uld d sa say y th that at pe peop ople le fr frow own n wh when en th they ey ar aree angry.. A realistic description of the situation would say that the person is angry angry and afterward frowns when threatening another person. Empirical studies show, in fact, that people will associate facial expressions with a person’s actual statement to someone about his or her emotion (e.g., “Y “You ou harmed my son!”), but not to statements reflecting a person’s thoughts (e.g., “So, he harmed my son!”) (Fernandez-Dols et al.,  2002). This suggests that facial expressions may reflect something that happens after the emotional experience, but not necessarily simultaneous with it. The idea that facial expressions are natural, automatic expressions of  inner emotional experiences is also at odds with empirical work showing that adults’ spontaneous facial expressions during intense emotional episodes are surprisingly rare (Ferndanez-Dols & Ruiz-Belda, Rui z-Belda, 1997). Lack of faci fa cial al ex expr pres essi sion on ca can, n, inevillain part pa rtic icul ular cont co ntex exts ts,, -Dols, be as,in info form rmat ive asalys inte in tens e facial displays displ ays (Carrera-L (Carr era-Levilla in &ar Fernandez Fern andez-Dols 1994 ). ativ One On eean anal ysis isnse of  pain pa inti ting ngss in th thee Pr Prad ado o su sugg gges ests ts th that at sm smile iless ar aree di disp spla laye yed d by vu vulg lgar ar,, dr drun unk, k, or crazy models or by children. Smiling was not linked, as it is today today,, with  beauty.. A fixed, open smile was a sign of simpleness, not of happiness  beauty (Fernandez-Dols et al., 2002). Furthermore, a significant body of work suggests that many emotional facial expressions are strategic, or intentionally produced (Gibbs, 1999a). For instance, one study observed the facial displays of bowlers (Kraut (Kr aut & Joh Johnst nston, on, 1979). Ob Obse serv rver erss we were re po posi siti tion oned ed bo both th be behi hind nd th thee wa wait it-ing pit and in the back of the pin-setting machine at the end of the lane. This allowed observers observers to chart the bowlers’ behavior behavior as they rolled rolled the  ball, watched the ball roll, and as they pivoted to face the members of  the bowling party. Bowlers rarely smiled when facing the pins, but did smile frequently when they pivoted to face their friends in the waiting pit. The outcome of the roll, which one might expect to affect the bowler’s emotion, bore little relationship to the production of smiles. A different study analyzed the facial displays of Olympic gold medal winners during the award ceremonies (Fernandez-Dols & Ruiz-Belda, 1995). The athletes only smiled when interacting with others and rarely did so alone. Even though the athletes were judged by observers to be very happy, the fact that they only smiled when looking at or talking to others suggests the importance of psychological beliefs and desires

(e.g., intentions to communicate information about oneself to others) in regulating emotional behavior. Oneisbehavior that mimicry. is widelyWe considered a readout of of inner emotional states that of motor cringe when we hear another’s fear,

 

Emotion and Consciousness

 

249

and grit our teeth when confronted with someone’s anger. But we now know that motor mimicry is communicative. For instance, one empirical demonstration of the intrinsically intentional function of facial displays showed that the timing of a wince of empathetic pain depended on the availability of the display to its intended audience (Bavelas et al., 1986). In one study, an experimenter staged an event where he dropped a color TV moni mo nito torr on onto to hi hiss ap appa parren ently tly al alrrea eady dy in inju jure red d fin finge gerr in fu full ll vi view ew of th thee ex expe perrimen im enta tall pa part rtic icip ipan ant. t. Wh When en th thee ex expe peri rime ment nter er di dire rect ctly ly fa face ced d th thee pa part rtic icip ipan ant, t, the participant participant fre frequentl quently y displa displayed yed a sympa sympathetic thetic wince, but when the expe ex peri rime ment nter er tu turn rned ed aw away ay ri righ ghtt af afte terr dr drop oppi ping ng th thee TV TV,, an any y in init itia iall wi winc ncin ing g  by the participant quickly ceased. Again, many aspects of our nonverbal disp di splay layss ar aree sp spec ecifi ifica call lly y di dirrec ecte ted d to an au audi dien ence ce an and d ti time med d to be re reco cogn gnize ized d  by the intended recipient. recipient. Finally, other research showed that observers could easily identify the type of odor (good, bad, or neutral) from the emotional facial expressions of th thee ra rate ters rs on only ly wh wheen th thee ra rate ters rs kn kneew th that at th theey wer eree bei eing ng ob obse serrve ved, d, not when whe n the they y tho though ughtt the they y wer weree mak making ing the their ir rat rating ingss alo alone ne (Gi (Gilbe lbert, rt, Fri Fridlu dlund, nd, & Sabini, 1987). Similarly, Similarly, people emit few spontaneous facial expressions wh when eati ea ting ngons swee sw eet t an and salt sa lty y the sand sa ndwi ches es by them th selv es,er , bu but t em emit many ma nyton fa-fa ciallen cia expre exp ressi ssions when whe ndeat eating ing these sewich sandwi san dwiche ches semse with wit h lves other oth people peo pleit(Br (Brigh ighton et al., 1977). These findings contradict the traditional view that facial expressions are by nature nonverbal readouts of pure, or genuine, emotions rather than social, communicative displays. A person’s overt emotional behaviorr is no io nott a ge genu nuin inee “s “spi pill llov over er”” fr from om in inne nerr vi vica cari riou ouss ex expe peri rien ence ces, s, bu butt ha hass a dist di stin inct ctly ly co comm mmun unic icat ativ ivee fu func nctio tion. n. Ev Even en wh when en we ta talk lk to ou ours rsel elve vess an and d de de-ploy pl oy fa faci cial al di disp splay layss in th thee co cour urse se of th thes esee ac acts ts,, we ar aree be bein ing g co comm mmun unic icat ativ ive. e. Thus Th us wh when en we ar aree al alon onee we of ofte ten n tr trea eatt ou ours rsel elve vess as in inte tera ract ctan ants ts (F (Fri ridl dlun und, d, 1994). An emotional expression need not be fully conscious for it to be understoo st ood d as an in inte tent ntio iona nall ac act. t. Re Rese sear arch ch sh show owss th that at pe peop ople le ha have ve th thee te tend nden ency cy to automa aut omatica tically lly mim mimic ic and syn synchr chroni onize ze mov moveme ements nts,, fac facial ial exp expre ressi ssions ons,, pos pos-turres tu es,, an and d em emot otio iona nall vo voca cali liza zati tion onss wi with th th thos osee di disp spla laye yed d by ot othe hers rs – a ph pheenomenon called “emotional contagion” (Hatfield, Cacioppo, & Rapsom, 1992). Emotional contagious behaviors still express actions, as people intend to emotionally converge with those around them. We We can, of course, consciously display, for example, pretend smiles in situations where we do not feel especially happy. These emotional displays often have a different bodily appearance than do behaviors that are less self-conscious (Ekman, 1994). Yet many of the socially determined, intentionally based emotional expressions observed in the above-mentioned studies (e.g., the smiles of Olympic athletes when receiving their medals) are quite genuine and have none of the characteristics usually associated with pretend emotional displays. It is a mistake, then, to assume that only involuntary, involuntary,

 

250

 

Embodiment and Cognitive Science

nonaction emotional displays are “genuine.” Expressing oneself emotionally can be intentional and genuine at the same time. Emotions and Bodily Changes Psychologists have long debated the relationship between emotional experience and bodily changes, beyond those conveyed by the face. William  James, one hundred years ago, suggested that the feelings of emotion are associated with the bodily changes that accompany or follow the perception of some exciting fact (James,  1882). He noted, “The various permutations and combinations of which these organic activities as susceptible makes it it abstractly possible that no shades of emotion, however slight, should be without a bodily reverberances as unique, when taken in its totality, as is to emotional mood itself” (James, 1898: 1066). Neurophysiologists have focused their attention on the anatomical structures that underlie emotion. Papez (1937), for example, proposed that there exists an entire circuit comprising the interconnections between the hypothalamus, anterior thalamus, cingulate gyrus, and hippocampus. Limbic structures, such as the amygdala, have been shown to play a crucial role in emotions (Aggleton & Mishkin,   1986;  Damasio,  1994;

LeDoux, 1998). But psych psychophys ophysiologi iologists sts have follow followed ed James in explo exploring ring the exten extentt to which emotions are differentiated by autonomic nervous system activity. Much of this research has centered on the “specificity-debate,” or whether various emotions can be distinguished by unique patterns of  somatovisceral arousal (Cacioppo et al.,  1993;  Ekman & Davidson,  1994; Panksepp, 1998). Som Somee psy psycho chophy physio siolog logist istss att attemp emptt to dem demons onstra trate te spe specicificit fic ity y by co comp mpar arin ing g tw two o or mo morre em emot otio ions ns on th thee ba basi siss of su such ch me meas asur ures es as skin temperature, heart rate, respiration, finger temperature, skin conductanc ta nce, e, fa faci cial al te temp mper erat atur ure, e, an and d bl bloo ood d pr pres essu sure re (A (Ax, x, 1953; Ekman Ekman,, Leven Levenson, son, & Friesen, 1983; Levenson et al.,  1992). For example, heart rate and temperature go up when people feel angry, but go down when they feel sad. Yet other psychophysiologists argue that bodily awareness is insufficient to distinguish between different emotional experiences (Mandler, 1984;  Schachter & Singer,  1962). Bodily awareness in emotion experience is merely awareness of general arousal, and emotion experience is based crucially on cognitive attributions of the cause of bodily arousal. An emotion is arousal plus a cognitive label for it in terms of anger, sadness, joy, and so on. In some cases, people misattribute arousal to some source that is not responsible for what they are consciously feeling. Some empirical research has explored people’s felt bodily sensations when wh en ex expe peri rien enci cing ng di diff ffer eren entt em emot otio ions ns.. On Onee st stud udy y as aske ked d a gr grou oup p of  psychology students to imagine a scene in which they might experience a specific emotion and to write down the characteristics that define that specific emotional experience (Parkinson,  1995). Two-thirds of 

 

Emotion and Consciousness

 

251

participants’ emotional definitions referred to some bodily symptoms. Many emotions corresponded to very specific changes in the body. Anger, for instance, was closely associated with tension, rising temperature, and feeling hurt, whereas fear was related to feeling nausea, cold sweat, and increased heart rates. Specific body symptoms were less related to happiness and sadness. Consequently, bodily changes may not accompany all emotional experiences to the same degree. A different study asked participants to remember different emotional experiences and mark a schematic diagram of the areas of the front and  back of the human body that were involved in what they felt (Nieuwenhuyse, Offenberg, & Frijda, 1987). People generally responded that certain emotions were deeply associated with localized internal body symptoms. For example, fear is localized in the abdomen and anal areas. Some emotions seem to spread over the entire body, such as when we feel a certain glow gl ow fr from om be bein ing g in lo love ve.. It is di difffic ficul ultt to un unde ders rsta tand nd th thee ca caus usal al rel elat atio ions nshi hip p  between body symptoms and emotion in this study study,, however however,, because it did not distinguish between bodily symptoms, consequences of emotions for future human actions, and experiences compatible with our responses to emotion. Another project attempted to demonstrate that body movements and postures are specific for certain emotions to some degree (i.e., that body movements and postures reflect not only the quantity of an emotion, but also als o its qua quality lity)) (W (Walb albott ott,, 1998).Asampleof 224 vid videot eotape apes, s, in whi which ch act actors ors and actresses portrayed the emotions of joy, happiness, sadness, despair, fear, terror, cold anger, hot anger, disgust, contempt, shame, pride, guilt, and boredom was analyzed for their body movements and postures. The results showed that 66 % of the movement and posture categories distinguished significantly between emotions or subclasses of emotion studied. For instance, an erect body posture is very rare when one experiences the emot em otio ion n of sh sham ame, e, sa sadn dnes ess, s, or bo borred edom om.. Du Duri ring ng th thes esee em emot otio iona nall st stat ates es ac ac-torss muc tor much h mor moree oft often en cho chose se a col collap lapsed sed bod body y pos postur ture. e. Lif Liftin ting g the sho should ulders ers,, on the other hand, is typical for elated joy and hot anger anger,, but rather infrequen qu entt fo forr al alll ot othe herr em emot otio ions ns.. Mo Movi ving ng th thee sh shou ould lder erss fo forw rwar ard d is fr freq eque uent nt fo forr disg di sgus ust, t, as we well ll as fo forr de desp spai airr an and d fe fear ar,, co comp mpar ared ed to ot othe herr em emot otio ions ns.. Fo Forr di diffferent types of head movements and head postures significant differences also arose between emotions. Orientation of the head directly toward the camera is least frequent during boredom experiences. On the other hand, moving the head downward is most typical of disgust. Moving the head  backward, that is, raising the chin, can be observed rather rather frequently during boredom, but also during pride and elated joy, joy, compared to the other emotions. The most significant variation among emotions is seen in different types of hand and arm postures and movements. Lateralized hand/arm movements are most frequent during hot anger, cold anger, and interest,

 

252

 

Embodiment and Cognitive Science

that is, rather active emotions. Arms stretched out to the front indicate the same sa me th thrree em emot otio ions ns an and d in ad addi diti tion on,, el elat ated ed jo joy y in co comp mpar aris ison on to th thee ot othe herr emotions. Arms stretched sideways are especially typical for terror, but less le ss of ofte ten n us useed fo forr al alll ot othe herr emo moti tio ons ns.. Cr Cros osssin ing g ar arms ms in fr fron ontt of th thee bod ody y is rath ra ther er fr freq eque uent nt du duri ring ng pr prid idee ex expe peri rien ence ces, s, as we well ll as du duri ring ng di disg sgus ust. t. Op Open en-ing and closing of the hands is again typical of some active emotions, such as hot anger and elated joy, but also of despair and fear. Overall, a discriminant analysis showed that it is possible to correctly classify different emotions, above chance level, based onpatterns the analysis of movement patterns alone.farThis implies that distinctive of movement and postural behavior may be associated with at least some emotions. The results of all these studies are consistent with the idea that people  believe differe different nt emotions to be associated with differen differentt localized bodily symptoms. Whether these beliefs about links between the body and emotional experiences reflect actual physiological responses to emotion is less clear. Visceroception research shows, in fact, that people are very poor perceivers of their physiological changes (Pennybaker,  1982;  Rime, Philippot, & Cisamolo,  1990). Many scholars argue that the process of  symp sy mpto tom m pe perc rcep epti tion on is at le leas astt pa part rtly ly de dete term rmin ined ed by in indiv divid idua uals ls’’ pe pers rson onal al and cultural expectations about their physiological states. If a culture has a fo folk lk ps psyc ycho holo logi gica call st ster ereo eoty type pe th that at pe peop ople le ar aree “h “hot ot”” wi with th an ange gerr, or “c “col old” d” with fear, they will then retrospectively report feeling hot with anger regarrdl ga dles esss of wh whet ethe herr th they ey ac actu tual ally ly fe felt lt ho hott wh when en th they ey we werre an angr gry y (P (Phi hill llip ippo pott & Ri Rime me,, 1997). Th Thus us,, th thee rep epor orti ting ng of th thee bo bodi dily ly se sens nsat atio ions ns as asso soci ciat ated ed wi with th emot em otio ions ns ma may y be a th theo eory ry-d -dri rive ven n pr proc oces ess, s, re refle flect ctin ing g va vari riou ouss ki kind ndss of so soci cial al schemata. One way to understand the relation of physiology to emotional experience is to examine individuals with physical handicaps. Does decreased  bodily sensation reduce the intensity of emotions? One study interviewed 25 adult men, with no psychiatric problems, who had suffered spinal in juries and lost all sensation below the site of the injury (Hohmann, 1966). The patients were questioned about their sexual feelings, fear, grief, sentimentality,, and overall emotionality. timentality emotionality. Most of the men reported decreases in sexual feeling since their injury. Those with injuries at the neck level reported large decreases. One man said that before injury his sexual feelings were “a hot, tense feeling all over my body,” but since the accident “it doesn’t do anything for me” (p.  148 ). One man, whose injury was at the high chest level, talked about fear. One day he was fishing on a lake when a storm came up and a log punctured his boat. He said “I knew I was sinking, and I was afraid all right, but somehow I didn’t have that feeling of trapped panic that I know I would have had before” (p. 150). Along with decreases in sexual feelings, fear, and anger, most participants reported an increase in feelings that might be called sentimentality, sentimentality, feeling tearful and choked up on occasions such as partings. But some of 

 

Emotion and Consciousness

 

253

these results may be due to participants getting older and changing their cognit cog nitive ive app apprai raisal salss of eve events nts tha thatt lea lead d to dif differ ferent ent emo emotion tionss and emo emotio tional nal intensity.. For example, one participant said about anger “Now I don’t get intensity a feeling of physical animation . . . Sometimes I get angry when I see some

injust inju stic icee. I ye yell ll an and d cu cusss an and d rai aisse hel elll be beca caus usee if you do don n t do it so some meti time mess I’ve I’ ve le lear arne ned d th that at pe peop ople le wi will ll ta take ke ad adva vant ntag agee of yo you, u, bu butt it ju just st do does esn’ n’tt ha have ve the heat in it. It’s a mental kind of anger” (p. 151). This Th is st stud udy y se seem emss to su supp ppor ortt Ja Jame mes’ s’ss cl claim aim th that at em emot otio ion n be begi gins ns in bo bodi dily ly sens se nsat atio ion. n. Bu Butt ot othe herr re rese sear arch ch pr pres esen ents ts a di diff ffer eren entt co conc nclu lusi sion on.. Fo Forr ex exam ampl ple, e, one study stu dyous interv int ed subj su bjec ts 4wh who o ha had d su suff ffer ered ednd spin sp inal al inju juri ries es duri du ring ng the pr previ evious 1erview toiewed 8 ye year ars37 s (m (mea ean nects of  .5 ye year ars) s) (Ber (B ermo mond et al., al .,in 1991 ). Pa Part rtic iciipants pan ts wer weree ask asked ed abo about ut the int intens ensitie itiess of the their ir phy physio siolog logica icall dis distur turban bances cesin in relation to the subjective intensities of emotional experiences. Specifically, Specifically, part pa rtic icip ipan ants ts we were re as aske ked d to re reme memb mber er tw two o ex expe peri rien ence cess of fe fear ar,, on onee fr from om be be-forre th fo thei eirr in inju juri ries es an and d on onee fr from om af afte terw rwar ard, d, an and d de desc scri ribe be wh what at ca caus used ed th them em to fee feell thi thiss way way.. Ove Overal rall, l, par partic ticipa ipants nts re repor ported ted mor moree sig signifi nifican cantt exp experi erienc ences es of fear following the injury than before, even though purely physiological disturbances in the postinjury emotion had diminished. Partic Par ticipa ipants nts als also o rat rated ed the their ir fea fearr, ang anger er,, gri grief, ef, sen sentim timent entalit ality y, and joy joyful ful-ness ne ss on sc scal ales es in indi dica cati ting ng in incr crea ease sess an and d de decr crea ease sess si sinc ncee th thei eirr in inju jury ry.. Ne Neit ithe herr in th thee wh who ole gr grou oup, p, nor in th thee 14 in indi divi vidu dual alss wi with th th thee gr grea eate test st se sens nsor ory y lo loss ss (i.e (i .e., ., th thos osee wi with th ne neck ck in inju juri ries es), ), wa wass th ther eree an any y ge gene nera rall de decr crea ease se in ra rate ted d em emootional intensity. Most participants reported little change on most scales, though some reported some increases in intensity since their injury injury.. These findings are difficult to reconcile with James’s predictions. One problem with the research on physical handicaps and emotions is that people in Western cultures are often aware of bodily sensations that are part of an emotion without being aware of them as an emotion. One study found that 18% of all consultations with general medical practitioners were found to be cases of anxiety or depression in which the persons complained of bodily symptoms but were not aware of specific affective or cognitive symptoms and had poor recognition of their emotional states (Bridges & Goldberg, 1992). A different way to think about the role of the body in emotional experience is opos to osal seealifsu particular body movements may induce emotions. Onee pr On prop sugg gges ests ts th that at some so me faci fa cial al ex expr pres essi sion ons s ha have vespecific emot em otio iona nal l ef effe fect ctss  by constricting flow through through blood vessels in the face (Zajonc, Murphy Murphy,, & Inglehart, 1989). These facial constrictions affect blood flow through parts of the brain, which then produce temperature changes that are affectively expe ex peri rien ence ced d as po posi siti tive ve or ne nega gati tive ve.. To te test st th this is id idea ea,, na nativ tivee Ge Germ rman an sp spea eakkerss we er werre as aske ked d to rea ead d a nu numb mber er of st stor orie iess al alou oud. d. So Some me of th thes esee st stor orie iess we werre filled with words requiring them to make movements with their mouths and lips that were just like the facial expressions of disgust, whereas the other stories had few such words. Participants liked the stories inducing

 

254

 

Embodiment and Cognitive Science

the disgust expression less than they did other stories, even though the two kinds of stories were virtually identical in content. Participants in another study were asked simply to hold a pen in the mouth, thus making the muscle movements characteristic of a smile without the participants realizing it (Strack, Martin, & Stepper,  1988). As they did this, participants judged whether different cartoons were humorous. These participants thought the cartoons were funnier than did a control group who made the same judgments without holding a pencil in their mouths.

Participant Partic ipantss in a rela related ted study were asked to judge the personality personality of  aTroccoli, fictitious person who described in very termswhile (Berkowitz &  1990 ). Half of was the subjects heard the neutral descriptions holding a pen between their teeth without using their lips. Holding a pen like this forces the face into an expression similar to that of smiling. The remaining mainin g parti participan cipants ts hear heard d the same descr description iptionss while biting down on a towel, which provokes an expression s similar to frowning. Participants who were smiling rated the fictitious person in far more positive terms than th an di did d pa part rtic icip ipan ants ts wh who o we were re fr frow owni ning ng.. Th This is fin findi ding ng su sugg gges ests ts th that at wh when en thee bo th body dy is pl plac aced ed in into to a si situ tuat atio ion n th that at is hi high ghly ly co corr rrel elat ated ed wi with th an em emot otio ion n (e.g (e .g., ., sm smil ilin ing g or fr frow owni ning ng fa face ce), ), th this is co cons nstr trai ains ns ot othe herr co cogn gnit itiv ivee (t (tha hatt is is,, em em- bodied) processing. A different experiment induced people to draw their eyeb ey ebrrow owss to toge geth ther er in a wa way y th that at mi mimi mick cked ed a sa sad d fa face ce (L (Lar arse sen, n, Ka Kasi sima mati tis, s, & Frey, 1992). These peoples’ judgments of pictures were sadder than a control tr ol gr grou oup, p, al alth thou ough gh th they ey di did d no nott kn know ow th that at th thei eirr ey eyeb ebro row w po pose se ha had d imp implie lied d sadness. All of the above studies suggest that making certain facial expressions prompts people to experience slightly different affective states. But facial feedback may not be a necessary component for emotional experience. Studies show that stroke patients who had lost the ability to make facial expressions had no loss in emotional experience (Ross & Mesulam, 1979), and patients with Moebius syndrome with congenital loss of facial movement have no apparent deficit in emotional experience (Cole,  1997). This clinica clin icall dat dataa imp implie liess tha thatt fee feedba dback ck fr from om fac facial ial exp expre ressi ssions ons is not an ess essent ential ial component of emotional experience. Do particular body movements, other than facial expressions, initiate specific emotions? in one study adopted three different tures without beingParticipants told what these body poses represented (Duclos etposal., 1989). In on onee po post stur ure, e, ch char arac acte teri rist stic ic of fe fear ar,, th thee pa part rtic icip ipan ants ts ke kept pt th thei eirr he head adss faci fa cing ng fo forw rwar ard d as th they ey le lean aned ed th thei eirr up uppe perr bo bodi dies es ba back ckwa warrd wh whil ilee tw twis isti ting ng them slightly and dipping one shoulder, similarly to how they would react when a sudden danger had unexpectedly appeared. Another posture, posture, generally associated with sadness, required people to fold their hands in their laps, drop their heads forward, and let their bodies sag. The third posture was an anger pose. Participants placed their feet flat on the floor,

 

Emotion and Consciousness

 

255

clenched their hands tightly, and leaned their bodies forward. After each pose was held for a brief period, the participants rated their feelings at that time. Participants reported the strongest feelings for the mood that was cha charac racter terist istica ically lly ass associ ociate ated d wit with h eac each h pos postur ture. e. Thu Thus, s, emo emotio tion-r n-rela elated ted  body movements, as well as facial expression expressions, s, appear to initiate at least some so me ki kind ndss of em emot otio iona nall ex expe peri rien ence ces, s, ev even en if on only ly at a low le leve vell of in inte tens nsity ity.. Of cou course rse,, emo emotio tion-r n-rela elated ted mus muscul cular ar mov moveme ements nts can af affec fectt our tho though ughts ts as well as our feelings. People make different attributions when sad than when angry. angry. Sadness heightens the chance that people will wil l attribute their life circumstances to situational forces, whereas when feeling angry, angry, peoplee ar pl aree mo morre ap aptt to reg egar ard d so some me pe pers rson on as sp spec ecifi ifica call lly y res espo pons nsib ible le fo forr wh what at happens them. Moreover Moreover, differences canin betheir produced bytogetting people to, adopt eitherina causal sad orattributions an angry pose

faces and bodies (Keltner, (Keltner, Ellsworth, & Edwards, 1993). One case study provides some support for the idea that merely participating in the action tendency associated with some emotion may easily caus ca usee on onee to fe feel el th that at em emot otio ion n (D (Dam amas asio io,, 2003). A 65-ye -yearar-old old wom woman an wit with h Parkinson’s disease had recently undergone a treatment of having tiny electrodes implanted bilaterally in the brain stem. These electrodes emit a low-intensity,, high-frequency electrical current that alters the functioning low-intensity of the motor nucleus. Most patients experience remarkable recovery recovery after thee su th surrge gery ry an and d ar aree qu quic ickl kly y ab able le to wa walk lk no norm rmal ally ly an and d pr prec ecis isel ely y mo move ve th thei eirr hand ha nds. s. Bu Butt de dete term rmin inin ing g wh wher eree th thee el elec ectr trod odes es sh shou ould ld be imp impla lant nted ed in or orde derr nott to cr no crea eate te un unwa want nted ed si side de ef effe fect ctss ca can n be tr tric icky ky.. Th This is pa pati tien ent, t, fo forr ex exam ampl ple, e, experi exp erienc enced ed an une unexpe xpecte cted d eve event nt whe when n the ele electr ctric ic cur curre rent nt pas passed sed thr throug ough h onee of th on thee co con nta tact ct sit itees on he herr le left ft sid idee. She im imme medi diat atel ely y slu lump mped ed,, ca casst he herr eyes down and to the right, leaned to the right, and appeared very sad. Soon So on th thee wo woma man n st star arte ted d cr cryi ying ng,, an and d th then en so sobb bbin ing, g, an and d sh shee st star arte ted d de desc scri ribbing in g how sad sh shee fe felt lt,, an and d th that at she ha had d no en ener erg gy le left ft to co con nti tin nue li livi ving ng.. Th Thee surrge su geon on in ch char arge ge of th thee tr trea eatme tment nt re reali alize zed d th thee pr prob oble lem m an and d ab aban ando done ned d th thee procedure. procedur e. Just a few seconds later, later, the patient’s behavior returned to normall an ma and d sh shee se seem emed ed co comp mple lete tely ly pe perp rple lexe xed d ab abou outt wh what at ha had d ju just st ha happ ppen ened ed.. This case illustrates how activation of the brain stem nucleus that controls specific motor actions, realized as an ensemble of movements of the faci fa cial al mu musc scle les, s, mo move veme ment ntss of th thee mo mout uth, h, ph phar aryn ynx, x, la lary rynx nx,, an and d di diap aphr hrag agm, m, alll ne al nece cess ssar ary y fo forr cr cryi ying ng,, wi will ll fa faci cili lita tate te on one’ e’ss fe feel elin ing g qu quit itee sa sad. d. Ev Even en th thou ough gh no event had such occurred to induce sadness, nor was the patient to experiencing sadness beforehand, emotion-related thoughtprone (feeling sad) could affect the emotion-action sequence that had begun. A di difffe ferren entt ki kind nd of res esea earrch fo focu cuse sess on bo bodi dily ly cu cues es to rea eadi ding ng ot othe herr pe peoople’ pl e’ss em emot otio ions ns.. So Some me st stud udie iess su sugg gges estt th that at ad adul ults ts us usee si six x sp spec ecifi ificc bo body dy cu cues es to infer people’s different emotional states (Boone & Cunningham,  1998): (1) frequency of upward arm movements, (2) the duration of time the actor’s arms are kept close to the body, (3) the amount of muscle tension,

 

256

 

Embodiment and Cognitive Science

(4) the duration of time the actor leans forward, ( 5) the number of directional changes in an actor’s face and torso, and (6) the number of tempo changes an actor makes in a particular sequence of movements. Experiment me ntss in indi dica cate te th that at th thee si six x cu cues es he help lp ob obse serv rver erss di disc scri rimin minat atee di diff ffer eren entt em emootions when watching actors move without speaking. For instance, anger was distinguished by a greater number of directional changes in the face and tor torso so and a gr great eater er num number ber of tem tempo po cha change nges. s. Obs Observ ervers ers dis distin tingui guishe shed d happinessfromsadnessandfearbyagreaternumberofupwardarmmovements and by the increase in time in which the arms are positioned away fro fr om th thee to torrso so,, su such ch as wh when en an ac acto torr th thrrow owss hi hiss ha hand ndss abo bov ve hi hiss hea ead d an and d kept his arms outstretched. Actors were seen as sad when they kept their heads down with a slumped body posture (i.e., less muscle tension). Fear was detected when an actor’s body was rigid and his head was kept up and alert. Other empirical research investigated people’s recognition of emotion from gait (Montpare, Goldstein, & Clausen,  1987). Participants viewed videotaped displays of walkers and judged which of four emotions (hap-

piness, sadness, anger, and pride) the walkers were expressing. Walkers read brief scenarios describing emotional situations and were instructed to imagine themselves in the situation and to walk accordingly. Participants performed better than chance levels in identifying sadness, anger, happiness and pride from the walkers’ gait. People were less proficient at identifying pride compared to sadness or anger. A further analysis of  the walkers who conveyed the most consistent emotions to observers reveal ve aled ed th that at an angr gry y ga gait itss we werre mo most st he heav avyy-fo foot oted ed,a ,and nd sa sad d ga gait itss ha had d th thee lo lowwestt am es amou ount nt of ar arm m sw swin ingi ging ng.. An Angr gry y an and d pr prou oud d ga gait itss ha had d th thee lo long nges estt st stri ride de lengths. Happy gaits were faster-paced than the other gaits. Consistent with these findings, other work demonstrates that people can recognize happ ha ppy y da danc nces es wh when en al alll th they ey ca can n se seee ar aree po poin ints ts of li ligh ghtt pl plac aced ed at th thee da danc ncer ers’ s’ main joints (Brownlow et al.,  1997). Not surprisingly, depressed people tend to gesture less and hold their heads down more than nondepressed peoplee (Segr peopl (Segrin, in, 1998). Th Thes esee da data ta su supp ppor ortt th thee id idea ea th that at em emot otio iona nall in info form rmaation is revealed in gait. Thee di Th diff ffer eren entt em empi piri rica call st stud udie iess lo look okin ing g at th thee re rela latio tion n of pa part rtic icula ularr bo body dy movements and emotional experiences, including reading of other people’s emotions, tend to individually analyze the influence of one factor (facial movements, body postures, perception of gait) in defining specific em emot otio ions ns.ly . Itwis is so some time mes harrt d, ha this th is case ca se,, to dete term rmin ine e wh whet ethe r sc scho hola lars rs necess nec essari arily wish h meti to cla claim imstha that theinind indepe epende ndent nt de variab var iable le studie stu died dher necess nec essari arily ly serves as the sole causal reason for different different emotional experiences. Howeverr, som eve somee psy psycho cholog logist istss ope openly nly adv advoca ocate te a com comple plex x vie view w of the bod body’s y’sro role le in emotional experience. For example, Berkowitz (2000) claims that emotional tio nal exp experi erienc encee ari arises ses fr from om the int integr egrati ation on of bod bodily ily sen sensat sation ionss with oth other er associated mental representations, including previously acquired conceptions of how one customarily feels in a certain class of situations. Suppose

 

Emotion and Consciousness

 

257

that Joe is faced with a bully who has just insulted him. Joe’s body reacts quickly. His heart beats faster, his face becomes hot, his mouth clamps shut, his brows draw together, and his fists clench. Joe also may recall other times when he has been insulted and the feelings that he had experienced on these occasions, as well as the stories he had read, seen, and heard about anger-provoking occurrences. Joe’s mind actively integrates alll of th al thes esee in inpu puts ts,, gu guid ided ed to so some me de degr greee by hi hiss con once cept ptio ion n of wh what at an ange gerr is like, with the result that Joe feels/thinks “I am angry.” angry.” This integrated view of emotions suggests that certain beliefs about  both the body and emotional expressions expressions are are partly constitutive of subjectivel tiv ely y fe felt lt em emot otio ions ns.. Un Unde derr th this is pe pers rspe pect ctive ive,, na naiv ivee th theo eori ries es ab abou outt em emot otio ions ns should play an important role in people’s reports about their affective states. In fact, a variety of research indicates that there is at least some cultural var tural variat iation ion in bod bodily ily and men mental tal emo emotio tion n exp experi erienc ence. e. Lin Lingui guisti sticc stu studie diess show that speakers from different cultures differ in regard to how much attention is paid to the body in emotion talk. For example, Russian participants discursively construct emotions as an action process expressed in a nu numb mber er of ex exte tern rnal al be beha havi vior ors, s, wh wher erea eass Am Amer eric ican an pa part rtic icip ipan ants ts pr pres esen entt emotions as internal states (Pavlenko, 2002; Wierzbicka, 1999). Empirical studies reveal other cultural differences in emotional experience en ce.. Fo Forr ex exam ampl ple, e, em emot otion ionss ar aree lo loca cate ted d an and d ex expe peri rien ence ced d in th thee bo body dy mu much ch

more frequently by Chinese and Taiwanese than by North Americans (Kleinman, 1982). Anger in Chinese individuals is frequently located and expe ex peri rien ence ced d in th thee ch ches estt an and d he hear art, t, de depr pres essi sion on is of ofte ten n ex expe peri rien ence ced d in te term rmss of something pressing into the chest, or down on the head, and grief may  be experienced in terms of a kind of back pain. Chinese, in contrast to North Americans, very rarely describe emotion experiences in terms of  “intrapsychic feelings,” such as personal thoughts. Rather, Rather, they comment on emotion experience only to a caused situation and to its somatic and intrapersonal effect and not personal cognition. Thus, an individual’s or a culture’s tacit folk model of self and of emotion will influence the form of  emotional experiences. But despite this cultural variation, there remains a large degree degree of similarity across cultures in their association of body experience and emotion. One large study of people in  37  countries in five continents examined in ed pe peop ople le’s ’s em embo bodi died ed st stat ates es (b (bas ased ed on 10 que questi stions ons)) for sev seven en emo emotio tional nal states (Scherer and Wallbott,  1994). Although some statistical differences appeared across cultures, the overall amount of variance explained was much larger for the effects of emotion than for the effects of culture. A re rela late ted d st stud udy y as aske ked d Am Amer eric ican an,, Ge Germ rman an,, Po Polis lish, h, an and d Ru Russ ssia ian n st stud uden ents ts to rep epor ortt in wh whic ich h pa part rtss of th thee bo body dy th they ey fe felt lt an ange gerr, en envy vy,, fe fear ar,, an and d je jeal alou ousy sy (Hupka et al.,  1996). Extending the number of body questions to  31 , the researchers resear chers actually showed significant effects of culture for 8 out of these 31 variables in the case of anger anger,, for  6 of them in the case of envy, envy, for  9 of  themforthecaseoffear,andfor 6 ofthemforthecaseofjealousy.Thus,there

 

 

258

Embodiment and Cognitive Science

table 8.1. Emotion

Evaluation

Action

World-Focused

 Joy

Enhanced

Sadness

Diminished

Buoyant, light, easy to move, able Heavy, unable, weak

Anger

Impeded, compressed, pushed back

Ready to push out

Fear

About to be overwhelmed,

Self Se lf-p -pro rote tect ctin ing g

Open, inviting, nonresistant, supportive Empty, closed,  burdening, lacking in attractiveness Impeding, Impedi ng, compressing, compressing, requiring force to remove blocking agent Over Overwh whel elmi ming ng,, piercing,

Shame

pierced, destroyed Stained

Pride

Augmented

Shrinking, self-occluding Increasing the exposed self 

disintegrative The impinging gaze of  others The welcoming gaze of  others

appear to be some differences of people’s bodily expressions of emotions across cultures. However, once again, cross-cultural similarity seemed to pred pr edom omin inat ate. e. In fa fact ct,, ot othe herr st stud udie iess sh show owed ed th that at be betw twee een n 6% and 8% of the  bodily sensation variance is accounted for by cultural differe differences nces (Philippot & Rime, 1997). These analyses showed that some emotions, including the

social emotions (joy, (joy, guilt, shame, and disgust), yield yi eld more cultural variations tio ns tha than n oth others ers (an (anger ger,, fea fearr, sad sadnes ness, s, and sur surpri prise) se).. Sim Similar ilarly ly,, cer certai tain n bod bod-ily sensations, such as temperature and respiratory changes or muscular sensations, are marked by more cultural variations than others. The relation between culture and body expressions of emotion is clearly complex. Fina Fi nall lly y, on onee re rece cent nt th theo eory ry pr prop opos osed ed th that at an any y em emot otio ion n st stat atee is a co comb mbin inaation tio n of ev eval alua uativ tivee de desc scri ript ptio ions ns (E (ED) D),, ba base sed d on co cogn gniti itive ve ap appr prai aisa sals ls,, an and d an action act ion att attrib ribute ute (AA (AA)) (La (Lambi mbiee & Mar Marcel cel,, 2002). A si sing ngle le ED is no nott in inev evit itab ably ly associated with any one AA. A single specific emotion episode (e.g., sadness) may vary over time or in the degree to which the experience is of the ED or AA. This is seen especially in coping with emotions where the differe fe rent nt co copi ping ng st stra rate tegi gies es in invo volv lvee fo focu cusi sing ng on ei eith ther er ev eval alua uati ting ng as aspe pect ctss (e (e.g .g., ., negati neg ative ve tho though ughts) ts),, or bod bodily ily re respo sponse nsess (e. (e.g., g., one one’s ’s br breat eathin hing). g). The There re is als also o no one-to-one determinate relation between a particular ED and a particular AA; each emotional state is defined by a combination of the two. This theory also acknowledges how emotions can be both self-focused (bodily physic phy sicali ality) ty) and wor worldld-foc focuse used d (ho (hodol dologi ogical cal spa space) ce).. Pr Prese esente nted d in Tabl ablee 8.1 aree ex ar exam ampl ples es of se seve vera rall ty typi pica call em emot otio ions ns,, th thei eirr ch char arac acte teri rist stic ic ED EDss an and d AA AAs, s, and an d th thei eirr wo worl rldd-fo focu cuse sed d sp spac acee (a (ada dapt pted ed fr from om La Lamb mbie ie & Ma Marc rcel el,, 2002: 238).

 

Emotion and Consciousness

 

259

conc nclu lusi sion on,, th thee es esse senc ncee of an any y em emot otio ion n is no nott co comp mple lete tely ly ca capt ptur ured ed by In co the way it is articulated in the body. body. For instance, if I am in i n some situation that me feel frustrated theloudly point of extreme feel as if makes I am about to blow up, I to may voice someagitation, sounds orI may sounds words words, ,I may begin to clench my fists and shake them deliberately, deliberately, as if fighting an imag im agin inar ary y op oppo pone nent nt,, or I ma may y co cons nsci ciou ousl sly y sw swea earr so some meth thin ing g mu must st be do done ne and even take action to reposition myself given the situation. All these felt expressions of gestures, thoughts, and behavior combined may make me feel as if I am right there “in touch” with the essence of anger. But the esse es senc ncee of an ange gerr is ne neve verr co comp mple lete tely ly ca capt ptur ured ed by th thee pa part rtic icul ular ar bo bodi dily ly di dissplays I produce. In some instances, I may express anger through complete silence and a blue face. I may, on other occasions, when angry, display a red edde dene ned d fa face ce wi with th te tens nsed ed ey eyeb ebrrow owss an and d pu purs rsed ed li lips ps bu butt sp spea eak k wi with th a ca calm lm voic vo ice. e. Th Thes esee ob obse serv rvat atio ions ns le lend nd cr cred eden ence ce to th thee id idea ea th that at af affe fect ctive ive pr proc oces esse sess are “emotional gestalts” that emerge from a complex of “interacting environmental, bodily, and cognitive variables” (Thagard & Nerb,  2002: 275). Let me explore this idea in more detail below.  A Dynamical View of Emotional Expression Chapter 3 presented a dynamical model of intentional actions that is relevant va nt to ex expl plor orin ing g em emot otio iona nall ex expr pres essi sion onss (G (Gib ibbs bs & Van Or Orde den, n, 2003). Im Imag ag-ine that you walk down the street, come across someone you know, and smile. Why did you do this? Was your smile intentional or an automatic response to seeing someone you knew? As shown above, there is much research to demonstrate that people may strategically express emotions in the sense of intending to communicate specific messages. But are other emotional expressions, such as having sweaty palms when nervous, also

int intent ention ional? al?psychological Thiss fol Thi folk-l k-leve evelldynamics analys ana lysis is doe does not ade adequa quatel tely y cap captur tureeAint intent ention ion-ality or the of semotional expression. dynamic

systems perspective on emotional expressions, as self-organized critical states, may yield a unified view of emotional expressions as a natural consequence. Dynamic systems have a capacity for self-control whereby they reduce a set of potential actions (e.g., the large set of potential ways to greet a friend) to that which is actually expressed, such as a particular smiling demeanor. Self-organization reduces the degrees of freedom for acti ac tion on un unti till a hu huma man n fa face ce be beco come mess a co cont ntex extt-ap appr prop opri riat atee “s “spe peci cial al de devic vice, e,”” a sm smil ilin ing g de devi vice ce,, fr frow owni ning ng de devi vice ce,, or wh what atev ever er wi will ll su suit it th thee si sing ngul ular ar se sett of  circumstances in which the action is situated. This capacity is creative and exqu ex quis isite itely ly co cont ntex extt-se sens nsit itiv ive, e, in th thee se sens nsee th that at it pr prod oduc uces es a si sing ngul ular ar ac acti tion on tailored to a particular context. Under the dynamical view, emotional expressions are on a par with intentional contents. The intention one feels to purposefully smile, or raise one’s hand to wave hello, or enact some other greeting, all result from a person’s capacity for self-organization. Intentions attendant on

 

260

 

Embodiment and Cognitive Science

self-o rgani anizat zation ion ent entail ail a pot potent ential ial to pur purpos posefu efully lly smi smile, le, for exa exampl mple, e, eve even n self-org  before the desire to smile reaches awareness (Ellis, 2002; Shaw, 2001). Intent te ntio iona nall ac actio tions ns,, su such ch as pu purp rpos osef eful ully ly sm smil ilin ing, g, st star artt wi with th th thee id idea ea th that at se self lf-organized dynamical areThus globally stable even though they may compose local sourcesstructures of disorder. a complex system can be driven toward local instabilities in the interaction of external circumstances and the system’s own internal i nternal dynamic processes. For example, feeling happy when seeing a friend can precipitate local instab ins tabilit ility y, not onl only y neu neuro rolog logica ically lly,, but als also o in abs abstra tract ct re relat lation ionss at cog cognit nitive ive and an d em emot otio iona nall “l “lev evel els. s.”” By fo form rmin ing g an in inte tent ntio ion, n, sa say y to sm smil ile, e, wh when en se seei eing ng a fr frie iend nd,, a co cogn gnit itiv ivee ph phas asee-ch chan ange ge ma may y fin find d a lo loca cally lly mo more re st stab able le tr traj ajec ecto tory ry (i.e., a better match between the “friend bearing” situation and the possi bilities for friendly discourse). The new intention restructures (“prunes”) the vast set of behavioral possibilities, excluding all but a potential set of  frie fr iend ndly ly ac acti tion ons. s. Th Thes esee in inte tent ntion ional al li limit mitss on th thee po pote tent ntia iall se sett av avoi oid d th thee ne need ed to consider and evaluate every logical and physical possibility for action (Shaw & Turvey, 1999). Thus, the emergent intention to let a friend know of your happiness to see him or her prunes the set of possibilities down to the act of smiling, excluding other possibilities such as writing the person a note, shaking his or her hand, whispering to him or her, and so forth. A dynamical view also accommodates other aspects of emotional expression. For instance, people sometimes experience an emotion without displaying any outward bodily reaction. People may also minimize, or inhibit, an emotional behavior in some situations, or even substitute one expression for another (e.g., smiling when feeling angry). These noncorrespondencesmaybeexpectedtooccurifemotionalexpressionsarerecruited for display in the interaction of intentions and circumstances (rather than causal chains). This approach may also explain how people can produce certain facial expressions are codeable asThis being emotional whenby it is unlikely that emotion is that being experienced. may be explained incorporating nonemotional facial expressions within the framework of  self-organization. For example, a face can produce a continuum of muscle actions that

may potentially combine in an infinite number of configurations. Never theless, only a limited subset of these configurations actually occur. The face fa ce as assu sume mess on only ly a ci circ rcum umsc scri ribe bed d se sett of pa patt tter erne ned d st stat ates es du duee to co cons nstr trai aint ntss imposed by lower-level synergistic relationships among muscle actions (i.e., coordinative structures of motor control). There are limits (embodied constraints) on the ways muscles can come together in combination. Such coor co ordi dina nativ tivee st stru ruct ctur ures es mi migh ghtt em emer erge ge in se seve vera rall di diff ffer eren entt wa ways ys be besi side dess ap ap-parent evolutionary sources. Another source might be the experience of  an em emot otio ion n it itse self lf (e (e.g .g., ., su surp rpri rise se). ). Ho Howe weve verr, th thee sa same me fa faci cial al ac acti tion on en ense semb mble le might also appear if only some of its components were produced in an instrumental action.

 

Emotion and Consciousness

 

261

For example, brow raising is a facial action that may occur alone or in comb co mbin inat atio ion n wi with th ot othe herr fa faci cial al ex expr pres essi sion ons. s. On Onee de deve velo lopm pmen enta tall st stud udy y wi with th 5- and  7 -month-old infants showed that raised brow movements significantly co-occurred with heads-up and/or eyes-up movements (Michel, Camras, & Sullivan,  1992). Raised eyebrows occurred more often when a raised head and/or gaze was required to look at things. This suggests that raised eyebrows are part of a coordinative motor structure involving actions of the head, eyes, and brows. The operation of this coordinative structure may determine whether infants produce a variant of expressing inte in terres estt in invo volv lvin ing g ra rais ised ed ey eyeb ebro rows ws wh when en th they ey ar aree di disp spla layin ying g th this is em emot otio ion. n. Again, raised eyebrows may sometimes be produced when head and/or gaze ga ze is li liffte ted d bu butt th thee emo moti tion on of in inte terres estt is not pr preese sent nt.. Ot Othe herr st stud udie iess sh sho ow that infants produce surprised expressions in situations in which their  brows are raised, as when an infant opens its mouth to orally explore an object (Camras, Lambrecht, & Michel, 1996). Mostt gen Mos genera erally lly,, the these se re resul sults ts sug sugges gestt tha thatt coo coord rdina inativ tivee str struct uctur ures, es, eme emerrgent in one context, can be recruited in other contexts for a variety of  purposes. This recruitment would take place over lower-level synergistic rela re latio tions nshi hips ps am amon ong g fa faci cial al mu musc scle le ac acti tion ons. s. As a co cons nseq eque uenc nce, e, th thee em emot otio ionnrele re leva vant nt fa faci cial al co confi nfigu gura rati tion on mi migh ghtt so some meti time mess be pr prod oduc uced ed wh when en no em emootion was being experienced. Unique and exclusive ties may not be found  between emotions and their correspond corresponding ing facial expression expressions, s, or other bodily actions; no singular causal chain can be traced through the body. body. A key consequence of these ideas is the elimination of the conventional distinction between intentional and automatic emotional expression. It simply disappears once we view emotional expressions as emergent dynamical structures.  Knowing How We Feel: The Interaction of Emotion and Consciousness

Most of us have some sense of how we usually feel at any given moment. Yet as de desc scri ribe bed d ea earl rlie ierr, pe peop ople le of ofte ten n mi misu sund nder erst stan and d wh what at th they ey ar aree fe feel elin ing, g, and on some occasions we struggle to define how we feel. Philosopher Naitka Newton (2000) asks “how can I know how I feel?” and attempts to answ an swer er th this is qu ques estio tion n in th thee co cont ntex extt of a dy dyna nami mica call ap appr proa oach ch to em emot otio ion n an and d cons co nsci ciou ousn snes ess. s. A ty typi pica call res espo pons nsee to th this is qu ques esti tion on fo focu cuse sess on ou ourr ab abil ilit ity y to obse ob serv rvee ou ourr ow own n me ment ntal al st stat ates es,, in incl clud udin ing g th thos osee rel elat ated ed to va vari riou ouss th thou ough ghts ts and desires, which are critical to our having some understanding of our

own min own minds ds.. Bu Butt Ne Newt wton on su sugg gges ests ts,, al alte tern rnat ativ ivel ely y, th that at kn know owin ing g ho how w we fe feel el is best characterized as a self-organized process. Consider the following situation (Newton, 2000: 102): I want to keep writing this paper. I also want to stop and take a nap. Both images are attractive, in that there are no vivid obstacles accompanying the imagery. But I cannot image doing both at the same time. If I imagine taking a nap now, then

 

262

 

Embodiment and Cognitive Science

that imagery activates pleasurable images of drifting off to sleep, but also other imagery of resuming the writing later with great difficulty. difficulty. If I imagine continuing to write, that imagery activates anticipatory imagery of increasing discomfort, but also other imagery of being able to sleep later without worrying about the paper. The latter imagery wins the competition, at least on this occasion.

Newton Newt on su sugg gges ests ts th that at he herr de deci cisi sion on ab abou outt ho how w sh shee fe feel elss ca can n be de desc scri ribe bed d in dynamical terms about self-organization processes that operate at two leve le vels ls:: th thee se self lf-o -org rgan aniz izin ing g sy syst stem em as a wh whole ole (s (ser ervi ving ng in inte tern rnal al ho home meos osta tasi siss while wh ile se seek ekin ing g in inte tera ract ctio ion n wi with th th thee en envi virron onme ment nt to sa sati tisf sfy y or orga gani nism smic ic go goal alss such as nouri nourishmen shmentt and rep reprod roductio uction), n), and consc consciousn iousness. ess. Consc Consciousn iousness ess is also a self-organizing activity (see below) and is directed toward intentional action. But knowing what one consciously feels is not a matter of  observing the contents of mind, or introspecting about our present experience. Instead, we know our minds by enacting mental states. For instance, Newton knows what she feels in the case of her paper writing by noting, “I feel a reluctance to go on working, but it makes me anx an xio ious us to th thin ink k of ta taki king ng a na nap p be befo forre I am fin finis ish hed ed,, so I bas asic ical ally ly fe feeel as if  I wan antt to go on un unti till th thee pap apeer is do don ne” (N (New ewto ton n 2000: 103). Sh Shee is th ther eref efor oree spec sp ecifi ifica call lly y co conc ncep eptu tual aliz izin ing g he herr fe feel elin ings gs,, no nott by ob obse serv rvin ing g th them em,, bu butt ra rath ther er  by “imaginatively perform[ing] alternative actions in an effort to identify the emotion they best satisfy” (Newton 2000: 103). In this way, the feeling she had arose from enacting relevant activities in imagination. Knowing how ho w on onee em emot otio iona nally lly fe feel elss ab abou outt so some me si situ tuat atio ion n de depe pend ndss up upon on th thes esee ki kind ndss of ima imagin ginat ativ ive, e, an and d in so some me ca case sess re real al-w -wor orld ld,, em embo bodie died d ac acti tion on.. En Enac actm tmen entt facilitates competition between various subsystems of the self-organizing whole, with the result that they can become better organized and directed towa to warrd th thee pu purrsu suit it of a sin ingl glee go goal al.. As we wi will ll no now w se see, e, co cons nsci ciou ouss st stat atees of  mind mi nd ar aree em emot otio iona nally lly dr driv iven en re resp spon onse sess of a go goal al-o -ori rien ente ted d bo body dy in dy dyna namic mic interaction with the environment. Consciousness

The topic of consciousness is at the heart of the mind-body problem. Do conscious thoughts exist in a realm separate from the body and bodily experience? Or might consciousness be intimately tied to embodied activity, not only within the brain, but also in the full body in action? Of  course, the traditional dualist position asserts that consciousness must have some nonmaterial existence. Many contemporary scholars believe, however, that consciousness must have some neural correlates (Crick & Koch, 1996; Metzinger, 2000; and see Jackendoff, 1987 and Prinz, 2001 for a related claim that there is an intermediate level of internal representa-

tion between sensory processing and higher-level thought). In fact, there

 

Emotion and Consciousness

 

263

is a huge effort in recent research to characterize the neural correlates of  consciousness employing neuroimaging techniques. Butt as Ch Bu Chal alme mers rs (1996: 384) cor corre rectl ctly y ar argue gues, s, “Th “Thee fac facts ts abo about ut con consci scious ous-ness do not just fall out of facts about the structure and firing of neural processes.” There remains an explanatory gap between these processes and an d th thee ex expe peri rien enti tial al le leve vell – th thee so so-c -call alled ed “h “har ard d pr prob oble lem” m” of co cons nsci ciou ousn snes ess. s. Even if there are important correlations between neural systems and conscio sc ious us ex expe peri rien ence ce,, it is no nott cl clea earr th that at th this is ev evid iden ence ce pr prov oves es th thee ex exis iste tenc ncee of a content match between the two (Noe & Thompson, 2004). For instance, as Noe and Thompson (2004: 1  11 1–12) note in regard to the evidence of certain cell firings during perception of a vertical line, the perceptual experience “of a vertical line is never just a matter of the registration of the presence of a vertical line in this way. The perceptual experience as of a vertical line will represent the line as against a background, and as occupying a certain position in egocentric space, that is, as occupying a certain spatial relation to you, the embodied perceiver.” perceiver.” Thus, the receptive field content and the content of the perceptual experience are two different kinds of  content. In line with my discussion in Chapter   3  of the problems of reducing bodily experience to neural firings in the somatosensory cortex, there is no way to determine the receptive-field content of neurons apart from consideration of the sensorimotor content of the entire animal in action (Varela et al., 1991). My argument, following this perspective, is that understanding how consciousness arises from animate motion, and more spec sp ecifi ifica call lly y th thee dy dyna namic mical al in inte tera ract ctio ions ns of br brai ains ns,, bo bodi dies es in mo motio tion, n, an and d th thee environment, is the best way to close this explanatory gap and ultimately solve the “hard problem.” Let me discuss this idea more fully. Defining Consciousness and Its Functions The essence of consciousness is experience: those things that are before your mind right now, such as different perceptions, bodily sensations, thoughts, feelings, images, and so on. As described at the beginning of 

this chapter, chapter, the elements of mental experience appear to float in a continuous narrative stream that endures as long as we are awake or alive. Most consciousness scholars today do not believe that this stream represents represents all there is to consciousness, or even that there is a single, definite stream of  consci con scious ousnes ness. s. “(i “(i)ns )nstea tead, d, the there re ar aree mul multip tiple le cha channe nnels ls in whi which ch spe specia cializ lized ed circuits try, try, as parallel pandemoniums, to do their various things, creating mult mu ltip iple le dr draf afts ts as th they ey go . . . so some me ge gett pr prom omot oted ed to fu furt rthe herr fu func ncti tion onal al ro role less  by the activity of a Virtual Machine in i n the brain” (Dennett, 1992: 253 –4). Thus Th us,, de defin finin ing g co cons nsci ciou ousn snes esss in te term rmss of th thee st stre ream am of co cons nsci ciou ousn snes esss al alon onee underestimates the complexity of mental experience, even if this stream reveals much about the process and content of consciousness. Nevertheless, conscious ideas and images are always owned in a highly physical

 

264

 

Embodiment and Cognitive Science

and body-based way (Donald, 2001; James, 1892). We own our conscious thou th ough ghts ts in th thee sa same me wa way y th that at we fe feel el ow owne ners rshi hip p of ou ourr bo bodi dies es.. Co Cons nsci ciou ouss experiences are not ethereal, ethereal, but quite often have a raw feel to them that is testimony to their embodied nature. Cognitive scientists do not yet completely understand the physical nature of this embodied ownership, yet the sense of consciousness as something that is part of us, and our bodies, is quite real. The psychologist George Miller once wrote that “Consciousness is a word worn smooth by many tongues” to note that consciousness is used to ref efer er to ma many ny as aspe pect ctss of me ment ntal al ex expe peri rien ence ce,, ra rang ngin ing g fr from om si simp mple le aw awar areeness to complex reflexive experiences of self-consciousness. At the very least, though, consciousness has several unique features, compared to other oth er eve events nts suc such h as sle sleep/ ep/com coma, a, hab habitua ituated ted eve events nts,, unc uncons onscio cious us pr probl oblememsolving, involuntary actions (Baars, 1988): 1.   Consc Conscious ious expe experienc riencee involv involves es gener generally ally bro broadcas adcastt infor information mation..

Thus,, th Thus this is in info form rmat atio ion n is av avai ailab lable le to al alll ef effe fect ctor orss an and d ac actio tion n sc sche hema mas. s. 2.   Conscious events are are internally consistent. This distinguishes consciousness from dreams, even when the content of dreams are generally broadcast. 3.   Co Cons nsci ciou ouss ev even ents ts ar aree in info form rmat ativ ivee (i (i.e .e., ., th they ey pl plac acee a de dema mand nd fo forr ad adap ap-tation on other parts of the system). 4.   Conscious events require access by the self-system. 5.   Conscious events may require perceptual/imaginal events events of some duration. One popular position holds that consciousness primarily functions as a spot sp otli ligh ght, t, dir irec ecte ted d to a po poin intt in th thee th thea eate terr of th thee mi mind nd by att tteent ntio ion n (B (Baa aarrs, 1988). Consciousness integrates multiple sensory inputs and disseminates them th em to a wi wide de au audi dien ence ce of di diff ffer eren entt mo modu dule less wi with thin in th thee ne nerv rvou ouss sy syst stem em.. In thiss way thi way,, con consci scious ous ope operat ration ionss con confer fer a num number ber of evo evolut lution ionary ary adv advant antage agess (Mandler, 1975). For example: 1.   Con Consci scious ousnes nesss ena enable bless the cov covert ert tes testin ting g of pos possib sible le way wayss of int intera eractct-

ing wi ing with th th thee im imme medi diat atee en envi viro ronm nmen ent. t. Th This is co cons nsid ider erat atio ion n of co comp mple lex x input-output contingencies eliminates the need for overt testing of  actions that may have harmful consequences. 2.   Consc Consciousn iousness ess makes it possi possible ble to ref reformul ormulate ate long-range long-range plans, involving the retrieval of information for long-term l ong-term memory, memory, modification of that information, remembering new plans, and so forth. 3.   Consciousness provides provides a troubleshooting troubleshooting function for systems that normally operate unconsciously, but only become conscious when they fail. For example, if one is driving a car and the brakes suddenly fail, awareness is immediately redirected to the task in hand, enabling repair work to get under way.

 

Emotion and Consciousness

 

265

These functions enable people to react reflectively rather than automatically and provide for more adaptive transactions between the organism and the environment. Most generally generally,, consciousness is directly tied to action ti on,, bo both th wh when en th thes esee ac acti tion onss ar aree ph phys ysic ical ally ly pe perf rfor orme med d an and d wh when en th they ey ar aree  just mentally entertained. Yet Yet consciousness does not exist along a single dimension. At the very least, three types of consciousness may be distinguished guish ed (Shan (Shannon, non, 1997). Th Thee mo most st ba basi sicc fo form rm is se sens nsua uall be bein ing, g, or co corp rpor orea eall consciousness (Sheets-Johnstone, 1998). Sensual experience differentiates the living or animate from the inanimate or dead. Inanimate objects do not respond to the environment in the dynamic moment-by-moment way that living organisms do. The second form of consciousness is mental awareness, which ordinarily provides something about the contents, but not processes, of cognition. Finally, reflection is perhaps the highest form of consciousness, where creatures not only are aware of their present em bodied situations, but can reflect on themselves in terms of both past and prospective actions. Consciousness and Enactment  Conscious experience is fundamentally grounded in perceptually guided activity in the environment. As we move about the world, various infor-

mation to our perceptual systems (i.e.,spatial affordance) that specifiesbecomes points ofavailable view sequentially occupying different locations. The things that we are most conscious of are those that offer opportunities ti es fo forr ac acti tion on.. Th Thee af affo forrda danc nces es th that at ar aris isee fr from om ou ourr bo bodi dily ly in inte tera ract ctio ions ns wi with th objects produce in us a fleeting, and usually inhibited, inner movement that brings them into consciousness. This perceptual grounding, together with our subjective experience of our bodies supports the experience of  consciousness. The serial and unitary nature of conscious experience is a fund fu ndam amen enta tall co cons nseq eque uenc ncee of th thee em embo bodie died d an and d si situ tuat atio iona nall ch char arac acte terr of th thee mind (Carlson,  1997;  Damasio,  1994). This view of consc conscious ious experience experience differs from Dennett’s (1992) claim that the serial character of consciousness is due to culturally transmitted memes (or ideas). Thee id Th idea ea th that at co cons nsci ciou ousn snes esss de depe pend ndss up upon on mo move veme ment nt ha hass a lo long ng,, al albe beit it sporadic, history in psychology psychology.. Washburn ( 1916) argued that consciousness is tied to “tentative movements” that are diminished versions of real action. He speculated that the cortex, not the muscles, was the locus of  these tentative movements for consciousness. Bodily action and attitude were crucial to the ways of meaning in Tichener’s ( 1909) view. Thus, the meaning a person gives to a situation depends on the bodily sensations experienced in that context. Several contemporary cognitive scientists argue that consciousness is clearly associated with basic sensory and motor processes (Ellis,  2002; Newton,  1996,  2000). Several neuroscientists have argued that thought aris ar ises es fr from om th thee ac acti tiva vati tion on of se sens nsor orim imot otor or im imag ages es (D (Dam amas asio io,,   1994;

 

266

 

Embodiment and Cognitive Science

Edelman,  1992)  that capture memories of how someone has in the past moved or felt his or her body in certain ways. The philosopher Newton

(1996, 2000) proposed that our conscious experience of embodied images grounds our concept of intentionality. This directedness upon an object, thee ha th hall llma mark rk of in intu tuit itiv ivee me ment ntal al st stat ates es,, co conc ncer erns ns th thee ex expe peri rien ence ced d di dirrec ecte teddness of a physical action toward a goal. As Newton states “Our apparent introspective awareness of the intentionality of mental states is our conscious sci ous exp experi erienc encee of the sen sensor sorimo imotor tor ima imager gery y tha thatt con consti stitut tutes es the these se sta states tes”” (Newton, 2000: 105). People experience consciousness when sensorimotor imag im ager ery y in wo work rkin ing g me memo mory ry,, co cons nsis istin ting g of se sens nsor ory y an and d mo moto torr as asso soci ciat atio ions ns distributed across the cortex, is combined with ongoing sensory and somatosensory input. This image provides something of what it would be like to prefer a certain action (Newton, 2000). Movement is related to consciousness, because consciousness is rooted in animate motion. Some psychologists emphasize that the major function tio n of co cons nsci ciou ousn snes esss is en enac actm tmen entt (S (Sha hann nnon on,, 1997). Ena nact ctme ment nt is a me ment ntal al activity in which people simulate concrete, embodied action in the real world without overtly performing this action (e.g., imagining a conversation, mentally rotating an object to imagine it from a different viewpoint). Shannon (1997) provides an example of enactment where he was thinking about a forthcoming trip to reside at the Cite Universitate in Paris. The conscious contents of Shannon’s thoughts were conversation with S, a friend who had previously stayed at an theimagined same university. The conversation emerged as a sequence of ideas in Shannon’s mind: 1.   They gave me a room at the Cite Universitate. Universitate. 2.   Do you know whether whether they give you sheets there? there? 3.   Oh, I can ask ask S if they give you sheets at the Universitate.

This mundane sequence is nonetheless remarkable. Shannon specifically enacted in his mind a conversation in which he posed to S a question, yet only afterward did he decide to perform the desired action in the real world. Enacting the conversation in the mind was necessary for the decision to actually ask S the question. In this manner, enactment is not simply a thought sequence we entertain but action actually performed. Enactment is not pure mental computation, but is fundamentally constituted by action. Consciousness enables people to find natural, efficient ways of performing actions in the world. Different thought experiments (Gedan (Ge danken kenexp experi erimen mentum tum)) ar aree won wonder derful ful exa exampl mples es of ena enactm ctment ent in whi which ch peoplee menta peopl mentally lly cre create ate entiti entities es and explore hypothetical hypothetical state statess of action through active manipulation of these entities. Recognizing the tight relationship between kinesthetic action and cognition (i.e., the decisions we make) demonstrates that there is not a mysterious gap between cognitive abilities and consciousness (Shannon, 1997). In fact, the ability to control our con consci scious ously ly hel held d tho though ughts ts may mak makee use of the sam samee cer cerebr ebro-c o-cer erebe ebella llarr

 

Emotion and Consciousness

 

267

dynami amics cs emp employ loyed ed in the con contr trol ol of or organ ganism ism-en -envir vironm onment ental al int intera eracti ctions ons dyn (Ito, 1993). Neu Neuro roima imagin ging g stu studie diess sho show w, con consis sisten tentt wit with h thi thiss ide idea, a, tha thatt del deluusions of control often seen in schizophrenia may be related to problems in the functioning of the cerebro-cerebellar loops that enable people to recognize that our thoughts originate with us (Frith, Blakemore, & Wolport, 2000).

The varied detailed bodily feelings that arise from movements provide the fundamental grounding for consciousness (Sheets-Johnstone,  1999). Consciousness is not a neural state that acts in the preparation of action,  but fundamentally emerges from action. The basic kinesthetic abilities by which any creature distinguishes parts of itself as an animate form constitute a “corporeal consciousness.” A creature’s corporeal consciousness is primarily focused on the movement of its own body. As creatures move, they th ey br brea eak k fr from om th thei eirr re rest stin ing g po posi siti tion onss an and d in initi itiat atee mo move veme ment nt by res espo pond nd-ing appropriately to the surrounding environmental context. There is an inherent kinesthetic specificity of animate form that provides for a wide range of movement possibilities – a series of “I cans,” which constitute a creator’s sense of agency. agency. Creatures differ sensually in their own proprioceptive actions in the present movement as they begin crawling, undulating, flying, elongating, contracting, and so on. Unde Un derr th this is vi view ew,, co cons nsci ciou ousn snes esss is no nott so sole lely ly li limi mite ted d to hu huma man n experience; it is not solely a natural product, but a visceral, complex biological “Consciousness is thus notof in living matter,form it is that a dimension of livingfaculty. form, in particular, a dimension moves” (Sheets-Johnstone, 1999:   60). A primitive form of intentionality exists in any animal that exists in worlds of sensorimotor representation used in the pursuit of goals (Newton, 1996). Conscious thoughts differ depending on whether the focused content is the self or the world. Consider the situation of pressing your forefinger on the horizontal edge of a table. Simply through a shift in attention, you can experience either (a) the sensation on the inner end of your finger,, of indentation and pressure, which has a shape and orientation, or (b) ger the perception of the edge of an external object, which has shape, texture, orientation, mass, and location. The single informational state due to reception in your finger in mechanical contact with another object can lend itself to awareness of either of the above, or both. Attention to the world yields yie lds hap haptic tic per percep ceptua tuall exp experi erienc encee (th (thee tab table le edg edge); e); att attent ention ion to one one’s ’s bod body y yields tactile sensation (felt pressure on the finger). In attention, either the external world or body becomes figure. Some interesting differences exist between these two attentional patterns. ter ns. Fir First, st, bod bodily ily sen sensat sation ionss mov movee with withone one’s ’s own mov moveme ement, nt, whe where reas as extern te rnal ally ly pe perrce ceiv ived ed ob obje ject ct fe feat atur ures es do no not. t. Se Seco cond nd,, th they ey ar aree kn know own n un uniqu iquel ely y through proprioception awareness. Third, they have hedonic qualities of a kind different from externally perceived objects or features. These

 

268

 

Embodiment and Cognitive Science

differences differences in spatial reference frames, spatial properties, epistemic properties, and hedonic properties emphasize that even though one’s body is expe ex peri rien ence ced d as lo loca cate ted d in th thee wo worl rld, d, th thee tw two o ar aree no nott ho homo moge gene neou ouss pa part rtss of  thee sa th same me ex expe peri rien enti tial al wo worl rld. d. Th This is is wh why y th thee fo foll llow owin ing g sy syllo llogis gism m is in inva valid lid:: There is a pain in my hand, my hand is on the table, therefore the pain is on the table. These The se dis distin tincti ctions ons als also o sug sugges gestt int inter erest esting ing dif differ ferenc ences es bet betwee ween n the own own-ership of the experiential content. The contents of bodily sensation are experienced as owned in a way in which the contents of haptic perception (a surface, edge, object) are not. Tactile pressure, pain, itch, is “mine” or

“yours” and “I am hot or in pain.” The haptic perception of the table edge is mine or yours, but the perceived table edge is not. Such a distinction in experiences could only obtain if one’s body and its states are experienced as one’s (physical) self. In general, any sensation and experienced state of  action readiness are part of the experienced states of the body that is me (Gallagher & Marcel, 1999).  Altered States of Consciousness Altered, or nonordinary, states of consciousness have not been widely studied or discussed within the cognitive sciences. But one remarkable phenomenological investigation of the various effects of Ayahuasca on conscious experience reveals several insights into the embodied nature of  some aspects of conscious experience. Shannon (2002) examined in great

detail the reports of dozens of Ayahuasca users, including his own, particularly in the context of indigenous Amazonian cultures. Ayahuasca Ayahuasca is a  brew made from several plants that is used in religious rites and tribal ceremoni mo nies es,, an and d mo most st rec ecen entl tly y by va vari riou ouss sy sync ncrret etic ic rel elig igio ious us gr grou oups ps th that at br brin ing g together Christian and Amerindian traditions. Ayahuasca inebriation includess feelin clude feelings gs of over overall all heigh heightened tenedsensi sensitivity tivity,, enha enhanced nced meani meaningfuln ngfulness, ess, faster mentation, and more energy, along with more specific effects such as vis visual ual ima imager gery y, enh enhanc anced ed met metaph aphori oricit city y (i. (i.e., e., see seeing ing-as -as), ), mor moree syn synest esthehesia, and greater fluidity (i.e., openness to new ways of seeing). Ayahuasca Ayahuasca drin dr inke kers rs of ofte ten n rep epor ortt th that at th thee di divi vide de be betw twee een n th thee se self lf an and d no nons nsel elff is si sign gnif if-ican ic antl tly y di dimi mini nish shed ed an and d th that at th they ey so some meti time mess ex expe peri rien ence ce a ma marv rvel elou ouss se sens nsee of transcendence beyond the normal human condition. Shannon’s (2002) analysis of Ayahuasca Ayahuasca experience revealed that there are two types of nonordinary conscious experience (Consciousness 4 and 5), which are extensions beyond three ordinary types of consciousness (Consciousness 1, 2,and 3) (Sh (Shann annon, on, 1997). Consc Consciousn iousness ess 1 co cons nsis ists ts of th thee undifferentiated quality of sensual being, relating to the fact that sentient agen ag ents ts ar aree in to touc uch h wi with th th thee ex exte tern rnal al,, re real al wo worl rld. d. Co Cons nsci ciou ousn snes esss 2 is a di diff ffer er-entiated, well-defined state that encompasses all thought sequences, mental ima images ges,, dr dream eams, s, and day daydr dream eams. s. Con Consci scious ousnes nesss 3 is selfself-cons consciousn ciousness ess or a second-order ability to reflect upon the mind’s own productions. All

 

Emotion and Consciousness

 

269

of these types of consciousness are interrelated and compose a coherent, unified system. In normal, waking life, we continually float between these three types of conscious experiences. There The re ar aree two add additi itiona onall typ types es of con consci scious ousnes nesss tha thatt cha charac racter terize ize non nonor or-dinary states of consciousness under the influence of Ayahuasca. Consciousness   4, like Consciousness  1 , consists of direct sensual experience from contact with the real world. However, in Consciousness   4, undifferentiated mentation appears to be independent, externally given, rather than th an a pr prod oduc uctt of on one’ e’ss ow own n mi mind nd.. Co Cons nsci ciou ousn snes esss 5 refe ferrs to wh what at is som omeetimes tim es ca call lled ed “s “sup uper” er” or “c “cos osmi mic” c” co cons nsci ciou ousn snes ess, s, in wh whic ich h on one’ e’ss ex expe peri rien ence ce transcends human agency. The two types of nonordinary consciousness are described as being similar to skilled bodily performance, specifically in the paradox of being immersed in bodily activity, while still being able to focus on aspects of 

the pr the pres esen entt en envi viro ronm nmen ent. t. Aya yahu huas asca ca ex expe peri rien ence ce is lik liken ened ed,, fo forr ex exam ampl ple, e, to one’s playing one’s own mind in the way that a master musician plays a musical instrument. A master pianist can immerse himself or herself into the act of playing the piano and seem to be at one with the instrument. Yet he or she can also maintain some distance from this activity to critically reflect on his or her performance. Thus, there is a constant, dynamic flow between the feeling of total immersion and of critical reflection. In a similar way, way, the master “ayahuasquero” exhibits the contradictory skill of   being grounded in the world and being able to soar high above without constraint. Grounding is achieved through straight body posture while sitting, stable breathing, having a relaxed psychological attitude, and contempla tem platio tion n wit with h dis distra tractio ction. n. Thi Thiss ena enable bless dri drinke nkers rs to “im “immer merse se the themse mselve lvess in th thee ot othe herl rly y re real alms ms of Aya yahu huas asca ca”” (S (Sha hann nnon on,, 2002: 352). Bu Butt dr drin inke kers rs ca can n imme im medi diat atel ely y sh shif iftt th thei eirr at atte tent ntio ion n to re real al-w -wor orld ld ev even ents ts,, ra rang ngin ing g fr from om as assi sist st-ing others undergoing a difficult Ayahuasca experience to chasing a dog from the place where the Ayahuasca session is occurring. A different analogy for characterizing nonordinary consciousness is that of dancing. Dancers often find themselves in states of consciousness that are quite similar to those labeled as “altered,” such as “flow” states (Csikszentmihalyi, 1990). As the dancer is fully engaged in movement, he or sh shee fe feel elss im imme mers rsed ed in a se sepa para rate te rea eali lity ty of so sort rts, s, th thus us se sepa para rati ting ng hi hims msel elf  f  or herself from the domain of life existing outside the dancer at that moment. When dancing with others, as in pas de deux, dancers sometimes feel their personal identities being transformed so they become one with their partners. But skilled dancers still can be aware of what is going on,  both with themselves and in the immediate immediate environment. environment. Shannon Shannon claims that this example perfectly illustrates important aspects of Ayahuasca Ayahuasca ine briation. Hallucination experience is not simply perceptual, because it also pertains to action. When Ayahuasca drinkers experience perceptual visions,

 

270

 

Embodiment and Cognitive Science

they they of ofte ten n st stop op be bein ing g me merre sp spec ecta tato tors rs an and d as assu sume me rol oles es as ac acto tors rs.. A pe pers rson on may step into the scene of his vision, but remain still and simply observe. Yet he ca can n al also so mo move ve ab abou outt th thee sc scen ene, e, so some meti time mess in inte tera ract ctin ing g wi with th ot othe herr be be-ings and objects in the scene as if in reality, reality, and sometimes moving about with no interaction with others. On occasion, this embodied immersion is acco ac comp mpan anie ied d by fe feel elin ings gs of me meta tamo morp rpho hosi sis, s, as if on onee ha had d tr tran ansf sfor orme med d in into to another person or creature. Some enactments are especially meaningful to people because they involve personal performances that they would not ordinarily accomplish in real life. Consider the following brief example (Shannon, 2002: 158): “I was climbing a very high mountain. I have never done any mountaineering in my life and the feat was quite difficult for me. I almost got to the top but could not carry on any further. Then a fairy came, gave me a push and I reached the summit. It was a most gratifying experience.” The importance of enactment in aspects of nonordinary consciousness is no nott li limit mited ed to in inte tern rnal al th thou ough ghts ts,, be beca caus usee Aya yahu huas asca ca dr drin inke kers rs of ofte ten n en enac actt aspects of their visions, or engage in activities such as singing, dancing, or playing instruments, often in remarkable ways that give overt evidence of 

this special state of mind. Consciousness and Self-Organization Many cognitive scientists believe that the first step toward constructing a scientific theory of consciousness is to discover the neural correlates of  consciousness. As noted above, this strategy ignores the “hard problem” of explaining the gap between neural structures and experiential content.

But even understanding the neural basis of consciousness is best done at the level of dynamical brain signatures (large-scale dynamical patterns of  activity over multiple frequency bands) rather than the structural level of specific circuits or classes of neurons. This dynamical approach to understanding consciousness gives little reason to search for matches in the content conte nt betwe between en inter internal nal menta mentall rep repres resentat entations ions and consc conscious ious expe experienc rience. e. The processes crucial for consciousness cut across brain-body-world divisions, as part of an individual’s embodied capabilities, rather than being limited to neural events in the t he head (Thompson & Varela, 2001). The re relat lation ionshi ship p bet betwee ween n neu neural ral dyn dynami amics cs and con consci scious ous sit situat uated ed agents can be described in terms of the participation of neural processes in the “cycle of operation” that constitutes the agent’s life (Thompson & Varela,  2001). Three kinds of cycles can be distinguished (Thompson & Varela, 2001). Cycles of organismic regulation of the entire body. The main basis for this regulation is the autonomic nervous system, in which sensors and effect fe ctor orss to an and d fr from om th thee bo body dy li link nk ne neur ural al pr proc oces esse sess to bu buil ild d ho home meod odyn ynam amic ic processes of the internal organs and viscera. Emotional states – reflecting the links between the autonomic nervous systems and the limbic system

 

Emotion and Consciousness

 

271

via the hypothalamus – are part and parcel of homeodynamic regulation. In this way way,, organism regulation has a pervasive affective dimension that manifests in the range of affective behavior and feeling that makes up sentience – the feeling of being alive (often referred to as primal or core consciousness). Cycles of sensorimotor coupling with the environment provide the organi ga nism sm wi with th a se sens nsee of ho how w it mo move vess ba base sed d on wh what at it se sens nses es.. Th Thee su subs bstr trat atee of these cycles is the sensorimotor pathways of the body body,, which are mediated at ed in th thee br brai ain n by mu multi ltipl plee ne neoc ocor ortic tical al reg egio ions ns an and d su subc bcor orti tica call st stru ruct ctur ures es.. Transient neural assemblies mediate the coordination of sensory and motor surfaces and sensorimotor coupling with the environment constrains and modulates this neural dynamic. Finally,, cycles of intersubjective interaction provide for the recognition Finally of inte in tent iona nal l me mean anin ing g the in ve verb rbal al an and d no nonv nver erba ball ac acti tion ons. s. Ne Neur ural al stru st ruct ctur ures es,, such asntio the amygdala, ventromedial frontal cortices, and the right somato ma tose sens nsor ory y rel elat ated ed co cort rtic ices es,, ar aree kn know own n to be im impo port rtan antt in so soci cial al co cogn gnit itio ion n  based on our reading of other people’s bodies. Intersubjectivity involves distinct forms of sensorimotor coupling, such as seen in mirror neurons. These neurons display the same pattern of activity both when the animal accomplishes certain goal-directed hand movements and when the agent observ obs erves es ano anothe therr per person son per perfor formin ming g the sam samee act action ion.. Thu Thus, s, the re recog cognit nition ion of in inte tent ntio iona nall me mean anin ing g of ac acti tion onss in ot othe hers rs ap appa parren ently tly de depe pend ndss on pa patt tter erns ns of ne neur ural al ac acti tivi vity ty in pr prem emot otor or ar area eass th that at ar aree si simi mila larr to th thos osee in inte tern rnal ally ly ge genn-

erated to produce the same type of action. Most generally, these three levels of cycles suggest how consciousness depends on the manner in which brain dynamics are embedded in the somaticc an mati and d en envi virron onme ment ntal al co cont ntex extt of th thee an anim imal al’s ’s li life fe.. Th Thee co coup uple led d dy dyna nammics of bra brain, in, bod body y, and env envir ironm onment ent exh exhibi ibitt sel self-o f-org rgani anizat zation ion and eme emerg rgent ent proc pr oces esse sess at mu mult ltip iple le le leve vels ls an and d th that at em emer erge genc ncee in invo volv lves es bo both th up upwa warrd ca cauusation and downward causation. Upward causation occurs when neural activity influence cognitive operations and phenomenological experience. Down Do wnwa ward rd ca caus usat atio ion n oc occu curs rs at mu multi ltipl plee le leve vels ls in th thes esee sy syst stem ems, s, in incl clud udin ing g that of conscious cognitive acts in relation to local neural activity activity.. Although conscious cognitive acts may be emergent phenomena, they can still have causal effects on local neuronal activity. This suggests that one can observe a moment of consciousness and its substrate large-scale neural assemblies at the level of local properties of neuronal activity. activity. One case study supports this claim (Varela,  2002). When an epileptic patient was engaged in different particular cognitive tasks (i.e., visual and auditory discrimination tasks), this activity influenced specific effects in the local activity given by an epileptic discharge. Thus, deterministic temporal patterns within the apparently random fluctuation of human epileptic activity can be modulated during cognitive tasks. An analysis of the periodic orbit in this person’s brain activity showed that the act of perception

 

272

 

Embodiment and Cognitive Science

contributes in a specific manner to “pushing” the epileptic activities toward unstable periodic orbits, a clear case of downward causation. In this way wa y, co cons nsci ciou ouss ex expe peri rien ence ce,, as a un unifie ified d gl glob obal al pa patte ttern rn of br brai ain n ac activ tivit ity y, ma may y have a downward causative effect on local neural activity activity.. At first glance, the idea of downward causation seems at odds with the classic, but still controversial finding that conscious will (i.e., a person deciding to act) may not be the ultimate cause of simple hand movements (Libet,  1985). But Libet’s studies only refer to simple one-way causation from conscious will to simple body action. The point of Varela’s work is to offer a dynamical model of consciousness that does not see consciousness as merely localized brain activity. Instead, consciousness is best understood as a whole-organism activity in which the person is situated or coupled pl ed wi with th th thee wo worl rld d in te term rmss of dy dyna nami mica call in inte tera ract ctio ions ns of br brai ain, n, bo body dy,, and and world. A different study, in fact, showed synchronic patterns of brain activ tivity ity co corr rrel elat atin ing g wi with ongo on goin ing g ex expe peri ence cess (i (i.e .e., ., a pe pers rson on’s ’sperformance sens se nsee of pr prep epaaration and quality ofthperception) in rien trained participants’ on a depth-perception task (Lutz, Lachaux, Martinerie, & et al.,  2002). But the relation between presenting the visual object and the ongoing brain activity is dependent upon what happens before and after an individual experimental trial. Thus, understanding the dynamics of a single moment of co cons nsci ciou ousn snes esss is no nott me mere rely ly a ma matt tter er of mo mome ment ntar ary y br brai ain– n–su subj bjec ectiv tivee ex ex-perience correlations, because any conscious experience is characterized in terms of both the ongoing activity preceding stimulation and the activity following it. This work, more generally, generally, aims to show how first-person data can, and must, be used to interpret neural data when constructing theories of mind and body. This perspective is clearly in an early stage of theoretical development (see Varela & Shear,  1999). Yet the dynamical

theory offers the prospect of closing the explanatory gap between minds and bodies. One implication of the dynamical perspective on consciousness is that consciousness may extend beyond the confines of an individual’s head, and may be temporally extended beyond any single moment in time (see Donald, 2001; Wilson, 2004 for similar claims). Each of us may be aware of  ourr imm ou immed edia iate te ex expe peri rien ence ce wi with thin in a fe few w se seco cond ndss of tim time. e. Bu Butt mo most st mo mome ment ntss of awareness are temporally extended, and include longer time-scales of  minutes, hours, and days, when we consciously think about learning a complex skill, creating a narrative, following the directions on a map, or some other complex set of instructions (Wilson,  2004). These different instances of awareness make use of environmental and cultural tools that help off-load(see cognition into real be world. As istothe case in many cases of reasoning Chapter  5 ),the it may difficult distinguish purely internal aspects of consciousness from those that extend through the body and out into the physical/cultural world. Similarly to seemingly singular acts of perception, such as observing a mug in front of us (see Chapter 3),

 

Emotion and Consciousness

 

273

many ny mo mome ment ntss of co cons nsci ciou ouss aw awar aren enes esss ar aree st stru ruct ctur ured ed ar arou ound nd th thee se sens nsor oriima motor contingencies of what we may may,, can, or will wil l do with the objects that are the immediate focus of attention. Even feelings of bodily pain also are not just perceived as passive sensations, but are directed outward outward toward how a bodily part, such as an injured knee, would feel engaging in various physical actions. These different experiences reflect the “as-if” quality of consciousness, which encompasses both what we experienced in the past and what we could experience in the future. Consciousness, then, is not simply tied to a very short moment in time, such as a few seconds in our phenomenal “stream of consciousness,” and instead expands across longer past, present, and future time-scales that are fundamentally tied to complex interactions of brains, bodies, and world. Conclusion

Emotions andemotions consciousness are boththrough directlyaffective tied to human action. We feel different as movement space that defines who we are at any moment in time. Different emotions have varying effects on the body, body, even if it may be impossible to strictly define individual i ndividual emot em otio ions ns in te term rmss of sp spec ecifi ificc bo bodi dily ly se sens nsat atio ions ns.. Bu Butt em emot otion ional al ex expr pres essi sion onss can ca n be ch char arac acte teri rize zed, d, an and d ev even en pr pred edic icte ted, d, by sp spec ecifi ificc dy dyna nami mica call pa patt tter erns ns of  intera int eracti ction on bet betwee ween n bra brain, in, bod body y, and wor world. ld. Thi Thiss dyn dynami amical cal per perspe specti ctive ve on emotional expression dissolves the traditional divide between behaviors thatt ar tha aree aut automa omatic ticall ally y gen genera erated ted and tho those se tha thatt ar aree int intent ention ionall ally y pr produ oduced ced.. Both kinds of expressions arise as emergent products of self-organizing proc pr oces esse sess th that at co cons nstr trai ain n th thee de degr gree eess to wh whic ich h va vari riou ouss em emot otio ions ns ar aree fe felt, lt, ex ex-pres pr esse sed, d, an and d ac acte ted d on on.. As su such ch,, em emot otio ions ns cu cutt ac acro ross ss br brai ain, n, bo body dy,, an and d wo worl rld, d, and are thus neither purely mental nor purely physiological phenomena. Conscious experiences are also not purely mental, but exist as kinds of  actions, even when these actions are not manifested by full-bodied behavior. These actions are not abstract and colorless, because people usually

experience distinct bodily sensations when they engage in conscious reflection. Of course, consciousness occurs at different levels of experience, yet each level is constituted by its own sensations of felt movement, either in terms of the body’s direct interaction with the world or when we imag im agin inee ou ours rsel elve vess en enga gagi ging ng in pa past st or fu futu ture re ac actio tions ns.. A dy dyna namic mical al sy syst stem emss fram fr amew ewor ork k is mo most st ca capa pabl blee of de desc scri ribi bing ng ho how w co cons nsci ciou ousn snes esss em emer erge gess fr from om interactions of brain, body, and world rather than being a specific result of brain-state activations, or just some immaterial substance with no tie to human bodies. Emotions and consciousness are tightly linked in enabling us togoals. consider appropriate courses of action for immediate and long-term Cognit Cog nitive ive sci scienc encee has tra tradit dition ionall ally y voi voiced ced ske skepti pticis cism m of firs first-p t-pers erson on inv invesestigationsofhumanexperience.Butitisclearthatthestudyofbothemotions

 

274

 

Embodiment and Cognitive Science

and con consci scious ousnes nesss dem demand andss fur furthe therr und unders erstan tandin ding g of how firs first-p t-pers erson on experiences of how we feel and are conscious correspond to third-person properties of brains and bodies that are beyond our phenomenological awareness. This is the task of phenomenological cognitive science, which aims ai ms to cl clos osee th thee tr trad adit itio iona nall ex expl plan anat ator ory y ga gaps ps in th thee st stud udy y of em emot otio ions ns an and d consciousness, and which aims to give embodied action its rightful place in the scientific study of the mind.

 

9 Conclusion

Bodily experiences matter greatly in mental life. No longer is cognition divorced from considerations of the body and our phenomenological experiences of our bodies, because mind and body are deeply intertwined. The previous chapters describe the mass of empirical evidence in favor of an embodied view of thought and language. This work is representative of a second wave in the history of cognitive science that dramatically differs from the traditional view of mind as purely symbolic, computational, and disembodied. Although there are alternative ways of explaining aspects of some of this evidence, the collective weight of  this work is very impressive in suggesting a unified view of mind and  body..  body Most discussions of embodiment in cognitive science focus exclusively on particular topics, such as the two visual systems (Chapter   3)  or em bodied grounding grounding for metaphor metaphor (Chapters 4 and 6) debates. Scholars then draw dra w gen genera erall con conclu clusio sions ns abo about ut the pos possib sibilit ility y of emb embodi odied ed cog cognit nition ion fr from om cons co nsid ider erat atio ion n of th thes esee sp spec ecific ific re rese sear arch ch ar area eas. s. On Onee mo motiv tivat ation ion fo forr th this is bo book ok was to provide a fuller picture of embodiment that cuts across the many areas ar eas of cog cognit nitive ive sci scienc encee re resea searc rch, h, inc includ luding ing per percep ceptio tion/a n/acti ction, on, con concep cepts, ts, mental imagery, memory, language, development, consciousness, and so on. I have aimed to more completely represent various disciplinary and subdisciplinary approaches to embodied cognition than is typically done in discussions of the mind-body problem. This broad sweep of the empirical literature makes it difficult to offer a single, explicit model that best characterizes the precise ways in which bodily experience shapes each aspect pe ct of pe perc rcep epti tion on,, co cogn gniti ition on,, an and d la lang ngua uage ge.. Bu Butt th thee wo work rk pr pres esen ente ted d in th this is

 book surely demonstrates in myriad ways that embodied activity is central to mental life. A key part of my search for the embodied mind is the embodiment premise (repeated here from Chapter 1). This states: 275

 

276

 

Embodiment and Cognitive Science

People’s subjective, felt experiences of their bodies in action provide part of the fundam fun dament ental al gr grou ound nding ing fo forr lan langua guage ge and th thoug ought ht.. Co Cogni gniti tion on is wh what at oc occur curss wh when en the body engages the physical, cultural world and must be studied in terms of  the dynamical interactions between people and the environment. Human language and thought emerge from recurring patterns of embodied activity that constrain ongoing intelligent behavior. We must not assume cognition to be purely internal, symbolic, computational, and disembodied, but seek out the gross and detailed ways that language and thought are inextricably shaped by embodied action.

This pr This prem emis isee refl eflec ects ts a me meth thod odol olog ogica icall im impe pera rati tive ve fo forr co cogn gnit itiv ivee sc scie ienc nce. e. Accordingly, cognitive scientists must not assume that any aspect of perception, cognition, or language arises from disembodied processes unless an explicit search has failed to find mind-body connections. Too much of  the debate about embodiment in cognitive science is done in the abstract, where scholars take principled positions about the autonomy of perception, cognition, action, or language and then only pursue research fitting with these ideals. Embodied cognition is often dismissed, or seen as irrelevant to the true goals of cognitive science, without appropriate efforts to seek out the “gross and detailed ways” in which mind and body are linked. Experimental psychologists, in particular particular,, have historically aimed to reduce bodily effects on human performance in laboratory studies, precisely because of the prevailing belief that embodied action has little to do with the essence of cognition or language. Yet th thee wo work rk de desc scri ribe bed d in th this is bo book ok cl clea earl rly y of offe fers rs a dr dram amat atic ical ally ly di diff ffer eren entt view vi ew of th thee ma many ny po poss ssib ibili iliti ties es fo forr ho how w em embo bodi died ed ac actio tion n an and d ex expe peri rien ence cess of  thee bo th body dy ar aree rel elat ated ed to a wi wide de va vari riet ety y of hu huma man n co cogn gnit itiv ivee pe perf rfor orma manc nce. e. Of  course, this does not imply that all searches for the embodied foundations of perception, cognition, or language will necessarily find mind-body corresp re spon onde denc nces es.. No Noth thin ing g in wh what at I ha have ve ar argu gued ed in th this is bo book ok ne nece cess ssar arily ily in indi di-cate ca tess th that at th thee min mind d is co comp mple lete tely ly,, ir irre redu duci cibl bly y em embo bodie died. d. Bu Butt th ther eree is su sure rely ly enough empirical evidence to suggest that a careful search for mind-body connections will often find such links. For this reason, cognitive scientists embracing disembodied views of mind and language should take on the challenge of doing the right scientific thing by looking for embodiment in cognition before espousing any theoretical position that denies the body its rightful place in understanding the human mind. Cognitive science includes ideas and research from scholars in several rel related ated discip disciplines, lines, includ including ing psych psychology ology,, lingui linguistics, stics, philos philosophy ophy,, computer science (artificial intelligence), neuroscience, and anthropology. anthropology. Schola Sch olars rs fr from om man many y oth other er aca academ demic ic dis discip ciplin lines, es, suc such h as bio biolog logy y, edu educat cation ion,, literature, and the arts, also contribute to the ongoing discourse about the origins and functions of the human mind that is, in my view, part of the wider web of cognitive science. Despite the continued acknowledgment

 

Conclusion

 

277

from almost all cognitive scientists of the need for interdisciplinary research and perspectives in formulating comprehensive theories of the human ma n mi mind nd,, th ther eree rem emai ains ns a st strron ong g te tend nden ency cy fo forr sc scho hola lars rs to pr priv ivil ileg egee me meth th-ods and data from their own respective academic fields. For instance, cognitive psychologists often dismiss ideas from linguistics and philosophy, precisely because these disciplines do not engage in scientific work using hyp hypoth otheti eticoco-ded uctive ive method hods, such h as those se tion employ emp loyed edllin the natura nat ural litiv scisci ence en ces. s. At the th e deduct same sa me time ti me,met , th ther ere e iss,asuc strron st ong g rtho educ ed ucti onis ist t pu pull with wi thin in cogn co gnit iveescience to explain matters about cognition in terms of specific brain states and patterns of neural activation. Cond Co nduc uctin ting g ri rigo gorrou ouss sc scie ient ntifi ificc st stud udie iess on hu huma man n co cogn gnit itio ion n is su surrel ely y im im-portant, and the data from brain-imaging studies, for example, provide important constraints on the embodied mind, as evident in my discussion of this work in the previous chapters. But I reject claims that cognitive linguistics and phenomenological evidence have little bearing on cognitive science theories because they are not based on behavioral or neuropsychological studies. The systematic study of language, including examining possible language-mind and language-mind-body correspondences, is vital to understanding abstract thought and the grounding of  symbols in embodied experience. Cognitive scientists who dismiss cognitive linguistic work are missing critically important empirical, even if  not experimental, evidence on the embodied mind and language. There are various questions about the reliability of the methods employed by cognitive linguists (Gibbs,  1996), but systematic exploration of linguistic structure and behavior should clearly be part of the methodological tools in cognitive science’s bag of tricks for understanding the embodied mind. At the very least, cognitive scientists must, once more, be able to explain why wh y it is th that at pe peo opl plee ta talk lk in th thee em emb bod odie ied d wa ways ys th theey do do,, wi with thou outt ap app pea eall to embodied experience, before they can reject cognitive linguistic and cognitive psychological claims about the fundamentally embodied character of language and thought. Thee te Th tend nden ency cy to pr priv ivil ileg egee on one’ e’ss ow own n wo work rkin ing g me meth thod odss in de deba bate tess ab abou outt the mind-body problem is surely quite natural. But there is a major downside to this normal scientific bias. By insisting that data from phenomenological reports, behavioral measures, or functional neuroimaging, to take three examples, are the best ways to discover principles of cognition, or even the embodied mind, cognitive scientists of whatever stripe too narrowly define the causal locus of what is “cognitive.” Part of the problem here is that cognitive scientists too often assume that there are single causes underlying complex human behaviors. In most cases, these singular causes are highly localized functional or anatomical mechanisms that are far removed from the whole organism acting in complex environments. For example, in cognitive psychology and neuropsychology,

 

278

 

Embodiment and Cognitive Science

observed performance in laboratory tasks (e.g., the overall variability in response times, error rates, recognition rates) is divided into component effects using linear statistical models (e.g., analysis of variance), and these comp co mpon onen entt ef effe fect ctss ar aree as assu sume med d to or orig igin inat atee in ca caus usal al co comp mpon onen ents ts of mi mind nd.. Thus, behavior is understood to be the sum of strictly separable pieces, plus pl us so some me no nois ise. e. Fu Furt rthe herm rmor ore, e, th thee pr pres esen ence ce of an ef effe fect ct is eq equa uate ted d wi with th th thee presence of a mental structure/r structure/representation epresentation and the absence of an effect is equated with = thestructures” absence of logic a mental structure/re structure/representation. This “effects is deeply flawed,presentation. especially in its implicit assumption that behavioral tasks can be unpacked to reveal individual components components of mind that are the single causes for behav behavior ior.. One place where this effects   =  structure fallacy has had unfortunate consequences in cognitive science is in the use of double-dissociation logic in behavioral and neuropsychological studies (Plaut,  1995; Shallice, 1988; Van Orden, Pennington, & Stone, 2001). Under the standard logic, a dou ble dissociation between performances on materials from two differe different nt experimenter-defined categories is assumed to rely on separate modules specialized for processing the different categories of materials. Consider some typical findings from this paradigm. A double dissociation between living things and artifacts in picture naming and property verification has been used to argue for separate modules for different semantic categories (Warrington & McCarthy, 1987). A double dissociation in reading abstract versus concrete nouns has been used to argue for separate modules for abstract and concrete nouns (Warrington, 1981). Moreover, a dou ble dissociation in production of past tenses of exception words versus regular words has been used to argue for separate brain mechanisms for words and rules (Pinker, 1991; Pinker & Ulmann,  2002). Various paralleldistributed/connectionist processing models have offered alternative accoun co unts ts fo forr th thes esee da data ta in te term rmss of a si sing ngle le,, in inte tegr grat ated ed in inte tera ract ctiv ivee me mech chan anis ism m instead of separate modules (Farah & McClelland,  1991; Plaut, 1995). But more generally, the observed patterns of dissociation demonstrating autonomous or independent representations representations (single causes) simply reaffir reaffirm m the inevitableinconsequent of assuming thatJansen there op were representations the first place (Van Orden, de autonomous Haar, & Bosman, 1997; Van Orden et al., 2001). The effort to explain human performance in terms of singular causes is also evident in contemporary brain-imaging studies, which also adhere to the “effects = structure” fallacy. fallacy. Thus, dissociations in performance on some experimental task are correlated with varying patterns of neural activity. Researchers then conclude from such studies that the neural basis forr so fo some me co cogn gnit itiv ivee be beha havi vior or is roo oote ted d in pa part rtic icul ular ar br brai ain n si site tess or pa patt tter erns ns of  neural activation. Some scholars even contend that this type of neuropsychological evidence is the real location and causal basis for the embodied mind.

 

Conclusion

 

279

As noted earlier, these types of arguments about the neural basis for thought and language are far from my own vision of the embodied mind. I cl clea earl rly y do no nott ig igno norre res esea earrch fin findi ding ngss fr from om ne neur urop opsy sych chol olog ogy y, as ev evid iden entt in my frequent mention of this work in the previous chapters. But there is something deeply problematic in how cognitive scientists typically interpret neuropsychological findings, especially in their aim to identify specific ci fic br brai ain n si site tess as th thee si sing ngul ular ar ca caus usal al lo loca cati tion onss fo forr di diff ffer eren entt ty type pess of hu huma man n performance. First, showing that a particular brain area is “lit up” under certain conditions says g nothing about what the rest of the is doing, and indeed contributin contr ibuting to human perfor performance mance. . Secon Second, d, brain the brain does nott wo no worrk in is isol olat atio ion, n, bu butt is par artt of an or org gan anic ic wh whol olee th thaat in incl clud udees th thee ne nerrvous vo us sy syst stem em an and d ki kine nest sthe hetic tic se sens nsat atio ions ns of th thee bo body dy in ac acti tion on.. As Asse sert rtin ing g th that at spec sp ecifi ificc br brai ain n si site tess ar aree th thee ca caus usal al loc locii of pa part rtic icul ular ar ki kind ndss of co cogn gniti itive ve pe perf rfor or-mance man ce com comple pletel tely y mis misses ses the ful full-b l-bodi odied ed nat natur uree of cog cognit nition ion.. Emb Embodi odimen mentt shapes cognitive performance not only as a distal, or ultimate cause, but also as a proximate cause, in the sense that bodily experience continually influences ongoing cognition over the course of an individual’s lifetime. Finally,, painting a picture of human cognition as performance-brain state Finally correspondences corresponde nces creates a distorted image of mind as completely defined  by what is on the inside of the skull/body skull/body.. Yet the traditional separation of mind and environment misses the important ways in which minds are shap sh aped ed by by,, an and d ex exte tend nded ed in into to,, th thee ph phys ysic ical al/c /cul ultu tura rall wo worl rld d th thrrou ough gh bo bodi dily ly action. The strong tendency in cognitive science to posit autonomous, disem bodied repre representations sentations is also clearly seen in theories on the modularity of mind. For example, Fodor (1983)  argued that the mind is made up of  geneti gen etical cally ly spe specifi cified, ed, ind indepe epende ndentl ntly y fun functio ctionin ning g mod module ules, s, suc such h as tho those se re re-spon sp onsi sibl blee fo forr vi visi sion on,, mo moto torr ac actio tions ns,, an and d la lang ngua uage ge.. In Info form rmat atio ion n fr from om th thee en en-vironment passes first through a system of sensory transducers that transform fo rm th thee da data ta in into to fo form rmat atss ea each ch sp spec ecia iall-pu purp rpos osee mo modu dule le ca can n pr proc oces ess. s. Ea Each ch modu mo dule le th then en ou outp tput utss da data ta in a co comm mmon on fo form rmat at fo forr ce cent ntra ral, l, do doma main in-g -gen ener eral al processing. These modules are assumed to be hardwired (not assembled from more primitive processes), of fixed neural architectur architecturee (specified genet netica ically) lly),,todom domain -speci ecific, fic, fast, t, and inform inf ormati ationa lly enc encaps apsula ulated ted (i.e., (i. e., insensitive theain-sp operation offas other modules oronally central cognitive goals). In recent years, various cognitive scientists have revised Fodor’s original idea to suggest that many aspects of higher-order cognition may also  be modular modular.. For example, evolutionary psychologists argue that instead of a uniform learning procedure, single long-term memory, and a small set of inference engines, the mind is really a grab bag of quite specialized knowle kno wledge dge-an -and-a d-acti ction on sto storie ries, s, dev develo eloped ped in a pie piecem cemeal eal fas fashio hion n (ov (over er evo evo-lutio lu tiona nary ry ti time me)) to se serv rvee sp spec ecifi ific, c, ad adap aptiv tivel ely y im impo port rtan antt en ends ds (S (Spe perb rber er,, 2001; Cosmides & Tooby,  1997). Thus, there are specific modules for thinking about abo ut spa spatia tiall re relati lations ons,, too tooll use use,, com compr prehe ehensi nsion, on, soc social ial und unders erstan tandin ding, g, and

 

280

 

Embodiment and Cognitive Science

so on. Unlike earlier proposals on modularity, these newer theories even assume that there are modules working within other modules, such as the compr com prehe ehensi nsion on mod module ule bei being ng emb embedd edded ed wit within hin the spe specia cializ lized ed the theory ory-of -of--

mind module (Sperber, 2000). Modularity theory embraces the traditional view of the mind as being composed of autonomous components, even if it holds the additional belief that these t hese components are domain-specific and evolutionarily determined. Although there may be specific modules for different kinds of bodily action, modularity theory downplays the role of embodied experience in the development, and continued operation, of  cognitive processes. I am enthusiastic about any effort to think about human cognition in terms of devices that solve adaptive problems for full-bodied organisms in complex environments. Indeed, there may be a variety of specialized, geneti gen etical cally ly det determ ermine ined d dev device icess tha thatt und underl erlie ie dif differ ferent ent kin kinds ds of hum human an per per-forman for mance. ce. But mod modula ularit rity y the theori orists sts app appear ear bli blind nd to the imp import ortanc ancee of liv lived, ed, embodied experience in cognition in their desire to reduce cognitive process ce sses es to sp spec ecia iali lize zed d mo modu dule les. s. Ma Many ny be beha havi vior orss ma may y ap appe pear ar to be mo modu dula larr and domain-specific. But these effects may be better understood as functional outcomes of an individual’s self-organizing abilities rather than as evidence of underlying, causal mechanisms. As noted earlier, part of the problem here rests rests with the guiding assumption that complex human behavi ha vior orss ma may y be ca caus used ed by au auto tono nomo mous us co comp mpon onen ents ts of min mind. d. Am Amon ong g mo moddularity scholars, there is no effort given to seeing what is common across modules, or even to describing how these modules interact to produce adaptive behavior. Forr ex Fo exam ampl ple, e, co cons nsid ider er tw two o di diff ffer eren entt hu huma man n be beha havi vior ors, s, tw two o pe peop ople le ta talk lk-ing with one another, another, and two people walking across a room and through a narrow doorway. What do these different events have in common? Most cognitive scientists would answer that these are very different different human activities, and indeed, these behaviors would probably be studied by very differentt types of scholars. But both events require a kind of coordination differen that is rooted in physical action. Even the two people talking are engaged in a kind of embodied coordination that is established through their body posi po sitio tions ns an and d sp spee eech ch ac acti tion ons. s. Ju Just st as th thee tw two o pe peop ople le wa walk lkin ing g th thrrou ough gh a na narrrow doorway tacitly negotiate enters doorway first, the spea sp eake kers rs mu must stmust also al so co coop oper erat ate e in su such chwho a wa way y as tothe meet me et thei th eirr in indi divi vidu dual al two and an d  joint goals. Both events then rely on embodied coordination for their successful completion, and as such have something in common that is i s rooted in em embo bodi died ed ex expe peri rien ence ce.. In th this is wa way y, mo modu dula lari rity ty th theo eori rist stss wh who o po posi sitt a va vast st numb nu mber er of in inde depe pend nden entt mo modu dule less of mi mind nd co comp mple lete tely ly mis misss wh what at is co comm mmon on acrros ac osss di diff ffer eren entt mo modu dule les, s, so some me of wh which ich ma may y in invo volv lvee em embo bodi died ed ac acti tion on.. Of  course, it is not surprising that modularity scholars miss embodied regularities in cognitive performance, because they never look for mind-body corre cor respo sponde ndence ncess in the their ir emp empiri irical cal and the theor oreti etical cal end endeav eavors ors.. Thi Thiss neg neglec lectt is a major problem for contemporary cognitive science.

 

Conclusion

 

281

hast sten en to ad add d he herre th that at em embr brac acin ing g an em embo bodi died ed vi view ew of co cogn gnit itio ion n do does es I ha not at all imply that there must be domain-general mechanisms underlying in g hu huma man n co cogn gniti itive ve be beha havi vior or.. Af Afte terr al all, l, th ther eree ma may y be di diff ffer eren entt sp spec ecial ializ ized ed,, domain dom ain-sp -speci ecific fic mec mechan hanism ismss tha thatt dri drive ve ada adapti ptive ve beh behavi avior or.. But the these se mec mechhanisms may be variously embodied to some degree (i.e., shaped by em bodied action both ontogenetically and phylogenetically) and may also be part pa rt of sp spec ecifi ificc pa patte ttern rnss of in inte tera ract ctio ion n of br brai ain, n, bo body dy,, an and d wo worl rld d ra rath ther er th than an functional, independent components of the mind/brain.

Dynami Dyna micc sy syst stem emss th theo eori ries es of offe ferr al alte tern rnat ativ ivee wa ways ys of th thin inkin king g ab abou outt co coggnitive performance that properly acknowledge embodied experience and do not assume that cognition should be reduced to single, autonomous components of mind. As described earlier, this alternative view invokes a reciprocal form of causality in which every part of a system is always present in each behavior of that system. Each of the parts continuously affects, along a different time scale, the overall behavior of the system to the point that its independent contribution cannot be sorted out from the behavior of the whole. Most important, the performance of cognitive systems emerges from coordinated activity among interdependent sensorimotor ensembles. These strongly nonlinear, qualitative transformations show the impossibility of reducing cognitive performance to singularly causal cau sal neu neural ral ass assemb embles les,, or eve even n sin singul gularl arly y cau causal sal com compon ponent ent osc oscilla illatio tions. ns. Thiss sel Thi self-o f-org rgani anizin zing g ges gestal taltt allo allows ws flui fluid d con contin tinuit uity y bet betwee ween n act action ion and per per-ceptio cep tion, n, and or organ ganism ism and env envir ironm onment ent.. Not sur surpri prisin singly gly,, thi thiss app appro roach ach to cogn co gnit ition ion no nott on only ly ex expl plai ains ns ma many ny em empi piri rica call fin findi ding ngss wi with thou outt ap appe peal alin ing g to specific underlying representations, but more importantly, places proper attention to the fully embodied whole organism in the scientific study of  mental life. It is somewhat unclear whether this theoretical approach may be able to explain all aspects of adaptive human performance. Nonetheless, the previous chapters make it evident that the dynamical systems approach is amenable to describing the diversity of human experiences ranging from low-level aspects of perception/action up to emotion and consciousness. Dynamical systems theory most certainly best captures my argument that embodied cognition arises from, and is sustained through, ongoing interactions between brains, bodies, and world. My argument in favor of a dynamical systems approach to cognition may see may eem m at od odd ds wi with th my cla laim imss, in va varrio ious us pla lace cess, th that at emb mbod odie ied d ac acti tion on underlies part of people’s conceptual representations. Dynamical systems theo th eori rist stss ge gene nera rally lly ai aim m to de desc scri ribe be hu huma man n pe perf rfor orma manc ncee as a se self lf-o -org rgan aniz ized ed process without any need for explicit, internal mental representations as thee ca th caus usal al ba basi siss fo forr ad adap apti tive ve be beha havi vior or.. Ad Advo voca cate tess of em embo bodi died ed rep eprres esen enta ta-tions suggest that many conceptual symbols include significant information about the motoric actions involved when people perceive and think abou ab outt co conc ncrret etee ob obje ject ctss an and d ab abst stra ract ct id idea eas. s. Th Ther eree is mu much ch de deba bate te in co cogn gniti itive ve science over whether explicit “representations” are needed in cognitive

 

282

 

Embodiment and Cognitive Science

theories of the human mind. Many of these discussions are quite interesting (e.g., Clark,  1997,  1998; Doffner,  1999;  Markman & Deitrich,  2000; Prin Pr inzz & Ba Bars rsalo alou, u, 2001; va van n Ge Geld lder er,, 1998) an and d ech cho o som omee of th thee sam amee ar argu gu-ments that occurred 50  years ago at the birth of the cognitive revolution, when behaviorism was cast aside in the study of human thought processes. Defenders of “representations” conclude that dynamical systems may characterize reactive cognitive agents, but that some form of internal repr re pres esen enta tati tion onss is ne need eded ed to ac acco coun untt fo forr hi high gher er-o -orrde derr as aspe pect ctss of co cogn gniti ition on.. An embodied approach to the study of cognition does not, in my view view,, demand that researchers either embrace or abandon representationalism in theories of mind. I am clearly impressed with the power of dynamical

systems theory to account for a wide variety of perceptual and cognitive phenom phe nomena ena,, inc includ luding ing pur purely ely men mental tal act actss inv involv olving ing int intent ention ions, s, bel belief iefs, s, and desires. Yet Yet I remain open to the possibility that some aspects of cognition may ma y re requ quir iree in inte tern rnal al me ment ntal al re repr pres esen enta tati tion ons, s, at le leas astt so some me of wh whic ich h sh shou ould ld  be deeply shaped by embodied experience. experience. But I also reject the automatic refle re flex x to po posi sitt re repr pres esen enta tati tion onss as th thee dr drivi iving ng ca caus usal al fo forc rcee fo forr hu huma man n pe perf rfor or-mance, as is done far too often in cognitive science. Adopting a dynamical perspe per specti ctive ve pr prope operly rly ack acknow nowled ledges ges the bod body’s y’s ro role le in cog cognit nitive ive beh behavi avior or as  brain, body body,, world interactions. Under this perspective, there is less temptation tio n to si simp mply ly re redu duce ce co cogn gniti itive ve pe perf rfor orma manc ncee to si sing ngul ular ar,, au auto tono nomo mous us,, an and d disembodied components of mind. My hope is that cognitive science will cont co ntin inue ue th thee res esea earrch tr tren ends ds de desc scri ribe bed d in th this is bo book ok an and d fu furt rthe herr ex expl plor oree th thee lived body by properly acknowledging how embodied experiences shape and guide cognitive performance in real-world contexts.

 

References

Abbott, V., Black, J., & Smith, E. ( 1985). The representation of scripts in memory.  Journal of Memory and Language Language,, 24  24,, 179–99.

Adler, A. (1931). ). What  What life should mean to you. you. Oxford: Little, Brown. Adolph, K. (1997). Learning in the development of infant locomotion. Monographs locomotion.  Monographs of the Society for Research in Child Development, Development , 62  62.. Adolph, K. (2000). Specificity of learning: Why infants fall over a vertiable cliff. Psychological Science, Science, 11  11,, 290–5. Adolphs, R., Damasio, H., Tranel, D., Cooper, G., & Damasio, A. ( 2000). A role for somatosensory somatosensory cortices in the visual recognition recognition of emotio emotion n as revealed by three-dimensional lesion mapping. Journal mapping.  Journal of Neuroscience, 20, 20, 2683–90. Aggleton,J.,&Mishkin,M.( 1986). The Theam amyg ygda dala: la:Se Sens nsor ory y ga gate tewa way y to th thee em emot otio ions ns.. In R. 281 Plutchik & H. Kellerman (Eds.), (Eds.), Emotion:  Emotion: Theory, research, and experience (pp. –9). Orlando, FL: Academic Press. Aglioti, S., Goodale, M., & DeSouza, J. ( 1995). Size-co Size-contrast ntrast illusions deceive the eye but not the hand. Current hand.  Current Biology, Biology, 5  5,, 679–85. Agre, P., & Chapman, D. (1987). Pengi: An implementation of a theory of activity.  AAAI-87. Menlo Park: AAAI. Proceedings of  AAAI-87. Ahsen, A. (1995). Self-report questionnaires: New directions for imagery research.  Journal of Mental Imagery, Imagery , 19  19,, 107–22. Akshoomoff, N., Courchesne, E., & Townsend, J. (1997). Attent Attention ion coor coordinatio dination n and anticipatory control. International control. International Review of Neurobiology, Neurobiology, 411  411,, 575–98. Alib Al ibal ali, i, M. M.,, & Di DiRu Russ sso, o, A. (1999). Th Thee fu func ncti tion on of ge gest stur uree in le lear arni ning ng to co coun unt: t: Mo More re than keeping track. Cognitive track. Cognitive Development, Development, 14  14,, 37–56. Alibali, M., Kita, S., & Young, A. ( 2000). Gesture and the process of speech Processes, production: We think, therefore we gesture.  Language and Cognitive Processes, 15,, 593–613. 15 Alliso Alli son, n, T., Pu Puce ce,, R. R.,, & Mc McCa Cart rthy hy,, G. (2000). So Soci cial al pe perc rcep epti tion on fo forr vi visu sual al ca case sess of th thee 4, 267–78. STS region. Trends region. Trends in Cognitive Science, 4, Andrews, E. (1995). Seeing is believing: Visual categories in the Russian lexicon. explanation (pp. 363–77). In E. Contini-Morava & B. Goldberg (Eds.), Meaning (Eds.),  Meaning as explanation (pp. Berlin: Mouton de Gruyter.

283

 

284

 

References

Anifeld, M. (1996 ). Only modeling is matched by neonates. Developmental Review Review,  16,tongue , 16 , 149–61protrusion . Arditi, Ardit i, A., Holtzm Holtzman, an, J., & Kosslyn, S. (1988). Mental imagery and seeing experi 26,, 1–12. ences in congenital blindness. Neuropsychologia blindness.  Neuropsychologia,, 26 Arnold, M. (1946). On the mechanism of suggestion and hypnosis.   Journal of   Abnormal and Social Psychology Psych ology,, 41  41,, 107–28. Arnold, M., & Gasson, S. (1954). Feelings and emotions as dynamic factors in personality integration. In M. Arnold & S. Gasson (Eds.),   The human person (pp. 294–313). New York: Ronald. Asci, F. (2003). The effect of physical fitness training on trait anxiety and physical self-concept of female university students. Psychology students.  Psychology of Sports and Exercise, 4, 4, 255–64. Attn At tnea eave ve,, F., & Ol Olso son, n, R. (1967). Dis Discr crim imin inab abil ilit ity y of st stim imuli uli va vary rying ing in ph phys ysic ical al and retinal orientation. Journal orientation. Journal of Experimental Psychology, Psychology, 74  74,, 149–57. Ax,A.(1953). The ph phys ysio iolo logi gical cal di diff ffer erent entiat iatio ion n be betw tween een fe fear ar and an anger ger in hum humans ans.. Psychosomatic Medicine, Medicine, 15  15,, 433–42. Ayer, A. (1936). ). Language,  Language, truth, and logic. logic. London: Gollancz. Ayers, T., & Jonides, J. ( 1979). Differing suffix effects for the same physical suffix.  Journal of Experimental Psychology: Psycholog y: Human Learning & Memory, 5, 5, 315–21. Baar Ba ars, s, B. (1988). ). A  A cognitive theory of consciousness consciousness.. New York ork:: Cam Cambri bridge dge Univ Universi ersity ty Press. Babcock, M., & Freyd, J. (1988). Perception of dynamic information in static hand-

written forms. American forms. American Journal of Psychology, Psychology, 101  101,, 11  111 1–30. Bach-y-Rita (1996). Sustitucion sensorielle et qualia. In J. Proust (Ed.), Perception (Ed.),  Perception et intermodalite (pp. intermodalite  (pp. 81–100). Paris: Presses Universitaires de France. Baddeley, A. (1986). ). Working  Working memory. memory. Oxford, England: Clarendon Press. psychol hology ogy Badde Bad deley ley,, A., & Hi Hitc tch, h, G. (1974). Wor orki king ng me memo mory ry.. In G. Bo Bowe werr (E (Ed. d.), ), The psyc of learning and memory: Vol. 8 (pp. 8  (pp. 47–89). New York: Academic Press. Badd Ba ddel eley ey,, A. A.,, & Li Lieb eber erm man an,, K. (1980). Sp Spat atia iall wo work rkin ing g mem emor ory y. In R. Attention on and per perfor forman mance ce VII VIII  I    (pp.   521–39). Hil Nickerson Nickers on (Ed.),   Attenti Hillsd lsdale, ale, NJ: Erlbaum. Bahrick, L. (1988). Intermodal learning in infancy: Learning on the basis of two  Child Development, Development, 59  59,, kinds of invariant relational in audible and visible events. even ts. Child 197–209. Bahrick, L. (1995). Intermodal origins of self-perception. In P. P. Rochat (Ed.), The (Ed.),  The self  in infancy (pp. infancy (pp. 349–73). Amsterdam: North-Holland. Bahrick, L. (2000). Increasing specificity in development of intermodal perception. p erception. Infant ant deve develop lopmen ment: t: The ess essent ential ial re readi adings ngs (pp. 119–37). In D. Mu Muir ir & A. Sl Slat ater er (E (Eds ds.) .),, Inf Malden, MA: Blackwell. Bahrick, L., & Watson, S. (1985). Detection of intermodal proprioceptive-visual contingency as a potential basis for self-perception in infancy.   Developmental  963 73(.1992). Lo Psychology, , 21  21, ,thal BaPsychology Bai, i, D. D.,, & Be Bert rten enth al,,–B. Loco como moto torr st stru ruct ctur uree an and d th thee de deve velo lopm pment ent of sp spat atial ial Development, 63  63,, 215–26. search skills. Child skills. Child Development, Bailey, D. (1998). Getting a grip: A computational model of the acquisition of verb semantics for hand actions. Unpublished Ph.D. dissertation, International Computer Science Institute, University U niversity of California, Berkeley Berkeley.. Baillargeon, Baillar geon, R. (1986). Representing Representing the existe existence nce and locati location on of hidden objects: objects: Object permanence in 6- and 8-month-old infants. Cognition infants. Cognition,, 23  23,, 21–41.

 

References

 

285

Baill arge geon, on, R. (1987a). Ob Obje ject ct per perma manen nence ce in 3.5 and 4.5 month month-old -old infants infants.. DevelBaillar opmental Psychology, Psychology , 23  23, , 655–64 . Baillargeon, R. (1987 b). Young infants’ responding about the physical and a nd spatial Development, 2  2,, 179–200. properties of a hidden object. Cognitive object.  Cognitive Development, Baillargeon, Baillar geon, R. (1993). The object concept revisited: New direction in the investigation of infant’s physical knowledge. In C. Granud (Ed.), Visual (Ed.),  Visual perception and cognition in infancy (pp. infancy (pp. 265–313). Hillsdale, NJ: Erlbaum. Baill Bai llar arge geon, on, R. (1994). Ob Obje ject ct per perma manen nence ce in yo young ung inf infant ants: s: Fu Furt rther her ev evid idenc ence. e. Child Development,, 62 Development  62,, 1227–46. Baill Bai llar arge geon, on, R. (1995). Phy Physi sica call re reas ason oning ing in in infan fancy cy.. In M. Ga Gazza zzani niga ga (E (Ed.) d.),, Physical reasoning in infancy (pp. infancy (pp. 187–204). Cambridge, MA: MIT Press. Baillargeon, R. (2000). How do infants learn about the physical world. In D. Muir Infant ant dev develo elopmen pment: t: The ess essent ential ial re readin adings gs (pp. 195–212). Ma & A. Sl Slat ater er (E (Eds ds.) .),, Inf Mald lden, en, MA: Blackwell. Baillar Bail largeo geon, n, R. (2004). Inf Infants ants’’ phys physica icall rea reason soning ing.. Curr Current ent Dire Directions ctions in Psychol Psychological ogical Science,, 13 Science  13,, 89–94. Baillargeon, R., & DeVos, J. (1991). Object permanence in young infants: Further  Child Development, Development, 114  114,, 1227–41. evidence. Child evidence. Baillargeon, R., DeVos, J., & Graber, M. ( 1989). Location memory in  8 -month-old infants in a non-search AB task: Further evidence. eviden ce. Cognitive  Cognitive Development, 4, 4, 345– 67. Baillargeon, R., & Graber Graber,, M. (1988). Evidence of location memory in  8-month-old infants in a non-search AB task. Developmental task.  Developmental Psychology, Psychology, 24  24,, 502–11. Baillargeon, R., Needham, Needha m, A., DeVos, DeVos, J. (1992). The development of young infants’ intuitions about support. Early support.  Early Development and Parenting, Parenting, 1  1,, 69–78. Baillargeon, R., Spelke, E., & Wasserman, S. (1985). Object permanence in five-

 20,, 191 208. month old infants. Cognition infants. Cognition,, 20 Baker, L. (2000). ). Bodies  Bodies and persons. persons. New York: Cambridge University Press. Baldwin, D., & Baird, J. (2001). Discerning intentions in dynamic human action. Trends in Cognitive Sciences Sc iences,, 5  5,, 171–8. Ballard, D., Hayhoe, M., Pook, P., & Rao, R. ( 1997). Deictic codes and the embodiment of cognition. Behavorial cognition. Behavorial and Brain Sciences, Sciences, 20  20,, 723–67. Bargh,, J., Chen, M., & Burro Bargh Burrows, ws, L. ( 1996). Automaticity of social behavior: Direct effe ef fect ctss of tr trait ait co cons nstr truct uct and st ster ereo eoty type pe ac acti tiva vati tion on on ac acti tion on.. Journal of Personality and Social Psychology, Psychology, 71  71,, 230–44.  Mindblindness.. Cambridge, MA: MIT Press. Baron-Cohen, S. (1995). ). Mindblindness Barsalou, L. (1983). Ad hoc categories. Memory categories. Memory & Cognition, Cognition, 11  11,, 211–27. Barsalou, Barsalo u, L. (1985). Ideals, central tendency, and frequency of instantiation as determinants of graded structure in categories. Journal categories.  Journal of Experimental Psychology: Learning, Memory, & Cognition, Cognition, 11  11,, 629–54. Barsalou, L. (1987). The instability of graded structure in concepts. In U. Neisser (Ed.), Concepts (Ed.),  Concepts and conceptual development: Ecological and intellectual factors in categorization (pp. egorization –40). New York: Cambridge Universi University ty Press. Barsalo Barsalou, u, L.  (pp. (1989 101 ). Intra-concept similarity and its implications for inter-concept similarity.. In S. Vosniadou & A. Ortony (Eds.), similarity (Eds.), Similarity  Similarity and analogical reasoning (pp. 76–121). New York: Cambridge University Press. Barsalou, L. (1991). Deriving categories to achieve goals. In G. H. Bower (Ed.), The psychology of learning and motivation: Advances in research and theory, Vol. 27 (pp. 1–64). New York: Academic Press.

 

286

 

References

Barsalou, L. (1995). Flexibility, structure, and linguistic vagary in concepts: Manifestations of a compositional system of perceptual symbols. In A. Collins, S. Gathercole, memory. Hillsdale, NJ: Erlbaum. M. Conway, & P. Morris (Eds.),  Theories of memory. Sciences,  22  22,, Barsalou, Barsalo u, L. (1999a). Perceptual symbol systems. Behavioral systems.  Behavioral and Brain Sciences, 577–660. Barsalou, Barsalo u, L. (1999 b). Language La nguage comprehension: Archival memory or preparation for situated action. Discourse action. Discourse Processes, Processes, 28  28,, 61–80. Barsalou, L. (2002). Being there conceptually: Simulating categories in preparation for situated action. In N. Stein & P. Bauer (Eds.),  Representation, memory, and development: Essays in honor of Jean Mandler (pp. Mandler (pp. 1–15). Mahwah, NJ: Erlbaum. Barsalou, L. (2003). Situated simulation in the human conceptual system.   Lan guage & Cognitive Processes, Processes, 18  18,, 513–562. Barsalou, L., & Medin, D. (1986). Concepts: Fixed definitions or dynamic contextdependent representations? Cahiers representations? Cahiers de Psychologie Cognitive, Cognitive , 6  6,, 187–202. Bassili, J. (1978). Facial motion in the perception of faces and emotional expression.   Journal of Experimental Psychology: Human Perception and Performance, Performance ,   4, 373–9. Weste stern rn Apac Apache he lan languag guagee and cul cultur ture: e: Ess Essays ays in lin lingui guisti sticc anth anthro ropol pology ogy.. Basso,K.(1990). We Tucson: University of Arizona Ar izona Press. mind. Ch Bate Ba teso son, n, G. (1972). Steps to an ecology of mind. Chic icago ago:: Uni Unive vers rsit ity y of Ch Chic icag ago o Pr Press ess.. Bauer, P. (1996). What do infants recall recall of their lives? Memory for specific events  by one- to two-year-olds. American two-year-olds.  American Psychologist, 51, 51, 29–41. Bauer, P. (1997). Development of memory in early childhood. In N. Cowan (Ed.), The development development of memory in childho childhood od (pp.  (pp. 83–11 111 1). Hove, England: Psychology Press. Bavac-Cikoja, D., & Turvey, M. (1995). Does perceived size depend on perceived distance? An argument for extended haptic perception.   Percept Perception ion and Psy Psy-chophysics,, 57 chophysics  57,, 216–24. Bavelas, J., Black, A., Lemery, Lemery, C., & Mullet, J. (1986). “I show how you feel”: Motor mimicry as a communicative act. Journal act.  Journal of Personality and Social Psychology, Psychology,  50  50,,

322–9.

Beach, Beac h, K. (1988). The ro role le of ex exte terna rnall mn mnem emon onic ic sy syst stem em in ac acqui quiri ring ng an oc occu cupat patio ion. n. In M. Gr Grune unebe berg rg & P. Mo Morri rriss (E (Eds ds.), .),Pra Practic ctical al asp aspects ects of memo memory ry (pp. 342–6). Oxf Oxford ord:: Wiley. Beard Bea rdsw swor orth th,, T., & Bu Buck ckner ner,, T. (1981). Th Thee ab abil ilit ity y to rec ecog ogni nize ze on ones esel elff fr from om a vi vide deo o recording of one’s movement without seeing one’s body.  Bulletin of the Psychonomic Society, Society, 18  18,, 19–22. Becker, A., & Ward, T. (1991). Children’s use of shape in extending novel labels to animate objects: Identity versus postural change. Cognitive change. Cognitive Development, Development, 6  6,, 3–16.  The meanings of war & peace. peace. College Station: Texas A & M Press. Beer, F. (2001). ). The Beer, R. (1997). The dynamics of adaptive behavior. Robotics behavior.  Robotics and Autonomous Systems,, 20 tems  20,, 257–89. 2003). TheBehavior, Beer, R. ( Adaptive dynamics active agent. Adaptive agent. Behavior , 11  11,,of  209 –43. categorical perception in an evolved model Beitel,D.,Gibbs,R.,&Sanders,P.(2001).Theembodiedapproachtothepolysemyof  thee sp th spat atial ial pr prepo eposi siti tion on “o “on. n.”” In H. Cu Cuyc ycken kenss (E (Ed.) d.),, Polsemy in cognitive linguistics (pp. 241–60). Amsterdam: Benjamins.

 

References

 

287

Bergan, Chang, g, N. (in pre press) ss).. Sim Simulat ulation ion-bas -based ed lang language uage unde underst rstandandBergan, B., & Chan ing in embodied construction grammar. In J.-O. Ostman & M. Fried (Eds.), Construction grammar(s): Cognitive and cross-linguistic dimensions. dimensions . Amsterdam: Benjamins. Berko,J.,Burke,L.,Craven,J.,&Sarlo,N.( 1992).Theimportanceofmotoractivityin senso se nsori rimo moto torr de deve velo lopm pment ent:: A pe persp rspec ecti tive ve fo forr chi child ldre ren n wi with th phy physi sica call han handi dicap caps. s.  Human Development, Development , 35  35,, 226–40. Berk Be rkow owit itz, z, L. (2000). Causes and consequences of feelings. feelings. Ne New w Yor ork: k: Ca Camb mbri ridg dgee Un Uniiversity Press. Berkowitz, L., & Troccoli, B. (1990). Feelings, direction of attention, and expressed evaluations of others. Cognition others. Cognition and Emotion, Emotion, 4  4,, 305–25. Bermond, B., Nieuwenhuyse, B., Fasolti, S., & Schuerman, J. ( 1991). Spinal cord lesions, peripheral feedback, and intensities of emotional feelings. Cognition feelings.  Cognition and Emotion,, 5 Emotion  5,, 201–20. Berm Be rmud udez ez,, J. J.,, Ma Marrce cel, l, A. A.,, & Ei Eila lan, n, N. (E (Eds ds.) .) (1995). The body and the self . Cam Cambri bridge dge,, MA: MIT Press. Bertenthal, B. (1993). Infants Infants’’ percep perception tion of biomec biomechanical hanical motions: motions: Intrins Intrinsic ic and knowledgeknowle dge-bas based ed con constr straint aints. s. In C. Gra Granru nrud d (Ed. (Ed.), ), Visual perception and cognition in infancy (pp. infancy (pp. 175–214). Hillsdale, NJ: Erlbaum. Bert Be rten enth thal al,, B. B.,, & Ca Camp mpos os,, J. (1987). Ne New w di dire rect ctio ions ns in th thee st stud udy y of ea earl rly y ex expe peri rien ence ce.. Child Development, Development, 58  58,, 560–7. Bertenthal, Bertent hal, B., Campos, J., & Barret Barrett, t, K. (1984). Self-produced locomotion: An organizer of emotional, cognitive, and social development in infancy. In R. Ende development  (pp. 174–210). & R. Harmon (Eds.), Continuties (Eds.),  Continuties and discontinuities in development (pp. New York: Plenum. Bert Be rten enth thal al,, B. B.,, Ca Camp mpos os,, J. J.,, & Ke Kerm rmoi oian an,, R. (1994). An ep epig igen enet etic ic pe pers rspe pect ctiv ivee on th thee development of self-produced locomotion and its consequences. Current consequences.  Current Directions in Psychological Science, Science, 3  3,, 140–5. Bertenthal, Bertent hal, B., & Pinto, J. (1993). Complementary processes in the perception and production produ ction of human movem movements. ents. In L. Smith and E. Thelen (Eds.), A (Eds.), A dynamic systems approach to development: Applications (pp. Applications  (pp. 209–39). Cambridge, MA: MIT Press. Berth Ber thie ierr, N. N.,, DeB DeBlo lois is,, S. S.,, Poi Poire rerr, C. C.,, No Nova vack ck,, M. M.,, & Cl Clift ifton on,, R. (2000).Where’stheball? Two wo-- and th thre reee-yea yearr ol olds ds re reaso asons ns ab abou outt uns unseen een eve event nts. s. Develo Developmental pmental Psychol Psychology ogy,, 36,, 394–401. 36

movement. Ca Bertho Bert hoz, z, A. (2000). The brain s sense of movement. Camb mbri ridg dge, e, MA MA:: Ha Harv rvar ard d Uni Unive verr sity Press. Behaviorr Bigel Bi gelow ow,, A. (1992). Lo Loco como moti tion on and se sear arch ch be behav havior ior in bl blin ind d inf infant ants. s. Infant Behavio and Development, Development, 15  15,, 179–89. Blakemore, S-J., Wolpert, D., & Firth, C. ( 2000). Why can’t you tickle yourself? NeuroReport, 11, 11, R11–R16. Bloc Bl och, h, H. (1990). St Stru ruct ctur uree and fu funct nctio ion n of ear early ly sen senso sory rymo moto torr co coor ordi dinat natio ions. ns. In H. Bloch & B. Bertenthal (Eds.), Sensorymotor (Eds.), Sensorymotor organization and development in infancy and early childhood (pp. childhood (pp. 163–78). New York: Kluwer Academic. Bloom, L. (1993 ).   Language development from two to three. ). three. New York: Cambridge University Press. Bloom, P., & German, T. ( 2001). Two reasons to abandon the false belief task as a test of theory of mind. Cognition mind.  Cognition,, 77  77,, B25–B31.

 

288

 

References

Shinsky,, J. ( 1998). On the perception of partially-occluded objects in Bogartz, R., & Shinsky  Cognitive Development, Development, 13  13,, 141–63. 6-month-olds. -month-olds. Cognitive Boone, T., & Cunningham, J. ( 1998). Children’s decoding of emotion in expressive Developmental pmentalPsychol Psychology ogy,,  body movements: The development of cue attunement. Develo 34, 34 , 1007 Bor Borodi oditsk tsky y–,14 L..(2000). Met Metapho aphoric ric str struct ucturi uring: ng: Unde Underst rstandi anding ng tim timee thr throug ough h spat spatial ial  Cognition,, 75  75,, 1–28. metaphors. Cognition metaphors. Borodi Bor oditsk tsky y, L. (2001). Do Does es la langu nguag agee sh shape ape th thou ought ght?? Eng Englis lish h and Ma Manda ndarin rin sp speak eak-ers’ conception of time. Cognitive time.  Cognitive Psychology, Psychology, 43  43,, 1–22. Boroditsky, L., & Ramscar, M. ( 2002). The roles of body and mind in abstract thought. Psychological thought.  Psychological Science, Science, 13  13,, 185–9. Bosc Bo schk hker er,, M. M.,, Ba Bakk kker er,, F., & Mi Mich chael aels, s, C. (2002). Ef Effe fect ct of me ment ntal al im imag agery ery in re real alizi izing ng  Quarterly Journal of Experimental Psychology, Psychology, 55A  55A,, 775–92. affordances. Quarterly affordances. Botvinick, J., & Cohen, J. ( 1998). Rubber hands ‘feel’ touch that eyes see. Nature, see.  Nature, 391,, 756. 391  Outline of a theory of practice. practice . New York: Cambridge University Bourdieu, P. (1977). ). Outline Press. Bower, G., Black, J., & Turner, T. ( 1979). Scripts in memory for texts.   Cognitive Psychology, 11 Psychology,  11,, 177–220. Brandt, S., & Stark, L. (1997). Spontaneous eye movements during visual im Journal of Cognitive Neuroscience, Neuroscience, 9, ager ag ery y re refle flect ctss th thee co cont nten ents ts of th thee vi visu sual al sc scen ene. e. Journal 27–38. Brass, M., Bekkering, H., Wohlschlager ohlschlager,, A., & Prinz, W. (2000). Compatibility 6 between observed and executed finger movements: Comparing symbolic, spatial, and imitative cues. Brain cues. Brain & Cognition, Cognition, 44  44,, 124–43. Brecht, M., Singer, W., & Engel, A. ( 1998). Correlation analysis of corticotectal in79, 2394–2407. teractions in the cat visual system. Journal system.  Journal of Neurophysiology, 79, Bridgeman, B. (1983). Mechanisms of space constancy. In A. Hein & M. Jeannerod (Eds.), Spatially-oriented (Eds.),  Spatially-oriented behavior (pp. behavior (pp. 263–79). New York: Springer. Bridgeman, B. (2000). Interaction between vision for perception and vision for behavior.. In Y. havior Y. Rossetti & A. Revonsuo (Eds.), (Eds. ), Interaction  Interaction between dissociated implicit and explicit processing (pp. processing (pp. 17–40). Amsterdam: Benjamins. Brid Br idge gema man, n, B. B.,, Ki Kirc rch, h, M. M.,, & Sp Sper erli ling ng,, A. (1981). Se Segr greg egat atio ion n of co cogn gnit itiv ivee an and d mo moto torroriented systems of visual position perception. Perception perception.  Perception & Psychophysics, Psychophysics,   29 29,, 336–42. Brid Br idge gema man, n, B. B.,, Le Lew wis is,, S. S.,, He Heit it,, G. G.,, & Na Nagl gle, e, M. (1979). Re Relat latio ions ns be betw twee een n co cogn gnit itiv ivee and mot motoror-ori oriente ented d sys system temss of vis visual ual pos positi ition on per percept ception. ion. Journal  Journal of Experimental Psychology: Human Perception and Performance, Performance, 5  5,, 692–700. Bridgeman, B., Peery, S., & Anand, S. (1997). Interaction of cognitive and sensorimotor maps of visual space. Perception space.  Perception & Psychophysics, Psychophysics, 59  59,, 456–69.

Bridge Brid ges, s, K. K.,, & Go Gold ldbe berrg, D. (1992). So Soma mati tizat zatio ion n in pr prim imar ary y he healt alth h car care: e: Pr Prev evale alence nce and determinants. In B. Cooper & R. Eastwood (Eds.),  Primary health care and  psychiatric epidemiology epidemiology (pp.  (pp. 341–50). London: Routledge. Brighto Brig hton, n, V., Seg Segal, al, A., Werth erther er,, P., & Ste Steiner iner,, J. (1977). Fac Facial ial expr express ession ion and hedo hedonic nic response to taste stimuli. Journal stimuli. Journal of Dental Research, Research , 56  56,, B161. Brooks Broo ks,, L. (1968, 22 ). Sp Spat atial ial and verb rbal al co comp mpon onent entss of th thee ac actt of re reca call ll.. Canadi Canadian an Journ Journal al of Psychology, Psychology  22, , 349 –68 . ve Brooks, R. (1991). Intelligence without representations.  Artificial Intelligence, Intelligence,   47 47,, 139–59.

 

References

 

289

us . New Yor Broo ks,, R. (2002). Flesh and machine: How robots will change us. ork: k: Pan Panth theon eon.. Brooks Browder, J., & Gallagher, J. (1948). Dorsal cordotomy for painful phantom limb.  Annals of Surgery, 128, 128, 456–69. Browman, C., & Goldstein, L. (1992). Articul Articulatory atory phonol phonology: ogy: An overvi overview ew.. Phonetica,, 49 Phonetica  49,, 155–80. Browman, Goldstein, L. Mind (1995as). motion: Dynamics and articulatory phonology. In R. Explorations in the dynamics of cognition Port Po rt & T. C., van va n&Ge Gelde lderr (Ed (Eds. s.), ), Mind (pp. 175–94). Cambridge, MA: MIT Press. Brown, J., Collins, A., & Duguid, P. (1989). Situated cognition and the culture of   Educational Researcher, Researcher, 18  18,, 32–42. learning. Educational learning. Brow Br ownl nlow ow,, S. S.,, Di Dixo xon, n, A. A.,, Eg Egbe bert rt,, C. C.,, & Ra Radc dcli liff ffe, e, R. (1997). Pe Perc rcept eptio ion n of mo move veme ment nt and dancer characteristics for point-light displays of dance. Psychological dance.  Psychological Record, Record, 47,, 411–21. 47 Brugger, P., Regard, M., & Shiffrar, M. (2000). Hand movement observation in a person born without hands: Is body schema innate? Meeting of the Swiss Neurological Society, Society, London, England, Sept.  2000. Brugman, C., & Lakoff, G. (1988). Cognitive typology typology and lexical networks. networks. In S. resolution  (pp. 477 – Small, G. Gorrell, & M. Tanenhaus (Eds.), Lexical (Eds.),  Lexical ambiguity resolution (pp. 508). Palo Alto, CA: Morgan Kaufman. Brugma Brug man, n, C. C.,, & Mc McCa Caul ulay ay,, M. (1986). In Inte terac racti ting ng sem semant antic ic sy syst stem ems: s: Mi Mixt xtec ec exp expre ressSociety, 12  12,, 315–27. sions of location. Berkeley location.  Berkeley Linguistic Society,  A study of thinking. thinking . New York: Wiley. Bruner,, J., Goodnow, J., & Austin, G. (1956). Bruner ). A Bruno,N.(2001). Whe When n doe doess act action ionres resist ist vis visual ual illu illusio sions? ns? Tren rends ds in Cognit Cognitive ive Scienc Sciencee, 5, 379–82. Bryant, D., & Wright, G. (1999). How body asymmetries determine accessibility in spatial function.  Quarterly Journal of Experimental Psychology, Psychology,   52A 52A,,   487– 508. Bullitt-Jones, M. (1999). ). Hunger:  Hunger: A memoir of desire. desire. New York: Knopf. Burgess, C. (2000). Theory and operational definitions in computational memory  Journal of Memory and Language, Language , 43  43,, 482–8. models. Journal models. Burgess, C., & Lund, K. (2000). The dynamics of meaning in memory. In E. Dietrich & A. Markman (Eds.),   Cognitive Cognitive dynamics dynamics   (pp.   117–56). Mahwah, NJ: Erlbaum. Burton, G. (1992). Nonvisual judgments of the crossability of path gap. Journal gap.  Journal of  Experimental Psychology: Human Perception and Performance, Performance, 18  18,, 698–713. Butcher, C., & Goldin-Meadow, S. (2000). Gesture and the transition from one- to two-word speech: When hand and mouth come together. In D. McNeil (Ed.), Language and gesture: Window into thought and action (pp. action  (pp.   167–91). New York: Cambridge University Press. Butler, S., Berthier, N., & Clifton, R. ( 2002). Two-year-olds’ search strategies and Psychology, 38 38,, 581– visu vi sual al tr trac acki king ng in a hi hidde dden n dis displ plac acem ement ent ta task sk.. Developmental Psychology, 90. Butterworth, B., & Beattie, G. (1978). Gesture and silence as indicators of planning in speech. In R. Campbell & P. Smith (Eds.),  Recent advances in the psychology of 

language (pp. 347 60). New York: Plenum. language (pp. Butterworth, B., & Hadar Had ar,, U. ( 1989). Gesture, speech, and computational stages: A reply to McNeil. Psychological McNeil.  Psychological Review , 96  96,, 168 . Butterworth, G. (1993 ). DynamicReview, approaches to–74 infant perception and action: Old and an d ne new w th theo eori ries es ab abou outt th thee or orig igin inss of kn know owle ledg dge. e. In L. Sm Smit ith h & E. Th Thel elen en (E (Eds ds.) .),,

 

290

 

References

 A dynamic systems approach to development: developm ent: Applications (pp. Applications  (pp. 171–87). Cambridge, MA: MIT Press. Cacioppo, J., Klein, D., Bernston, G., & Hatsfield, E. (1993). The psychophysiology emotions (pp. 119–42). of emotion. In M. Lewis & J. Haviland (Eds.), Handbook (Eds.),  Handbook of emotions (pp. New York: York: Guilford Press. Camras, L., Lambrecht, L., & Michel, G. ( 1996). Infant “surprise” expressions as coordinative motor structures. structures. Journal  Journal. Mahwah, of Nonverbal Behavior , 20  20,, 183–95. Carlson, R. (1997 ). Experienced ).  Experienced cognition cognition. NJ:Behavior, Erlbaum. Carpenter, P., & Eisenberg, P. (1978). Mental rotation and the frame of reference in Psychophysics, 23  23,, 117–24.  blind and sighted individuals. Perception individuals.  Perception & Psychophysics, Carpent Car penter er,, W. (1874). Pr Prin incip ciple less of me ment ntal al phy physi siol olog ogyy, wi with th the their ir app appli lica catio tions ns to th thee tr trai ainning and discipline of the mind and the study of its morbid conditions . New York: Appelton. Carreiras, M., Carriedo, N., Alonso, M., & Fernandez, A. (1997). The role of verb tens te nsee an and d ve verb rb as aspec pectt in th thee fo fore regr groun oundi ding ng of in info form rmat atio ion n dur during ing re readi ading ng.. Memory & Cognition, Cognition, 25  25,, 438–46. Carrer Car rera-L a-Levil evillain lain,, P., & Fer Fernand nandez-D ez-Dols, ols, J-M J-M.. (1994). Ne Neut utra rall fa face cess in co cont ntex ext: t: Th Thei eirr 18, 281–289. emotional meaning and their function. Journal function. Journal of Nonverbal Behavior, 18, Carroll-Phelen, B., & Hampson, P. (1996). Multiple components of the perception of musical sequences: A cognitive neuroscience analysis and some implications for auditory imagery. Music imagery.  Music Perception, 13, 13, 517–61. Carver, L., & Bauer, P. ( 1999). When the event is more than the sum of its parts: 9-month-olds’ long-term ordered recall. Memory, recall.  Memory, 7, 7, 147–74. Cary,, M., & Carlson, R. ( 1999). External support and the development of problemCary solving routines. Journal routines. Journal of Experimental Psychology: Learning, Memory, and Cognition,, 25 tion  25,, 1053–70.  The mind’ mind’s staircase. staircase. Hillsdale, NJ: Erlbaum. Case, R. (1992). ). The Cass Ca ssell ell,, J., J.,Bic Bickm kmor ore, e, T., Ca Camp mpbe bell ll,, L., L.,V Vih ihja jamm mmss sson, on,H. H.,, & Yan, H. (2001).Morethan  just a pretty face: Conversational protocols and the affordances of embodiment. Knowledge-Based Systems, Systems, 14  14,, 55–64. Castiello, U. (1996). Grasping a fruit: Selection for action. Journal action.  Journal of Experimental Psychology: Human Perception & Performance, 23, 23, 582–603.  Emotion, depth, and flesh. flesh. Albany: State University of New York Cataldi, S. (1996). ). Emotion, Press. Chalmers, D. (1996). ). The  The conscious mind. mind. New York: Oxford University Press. Chambers, D., & Reisberg, D. (1992). What an image depicts depends on what an image means. Cognitive means. Cognitive Psychology, Psychology, 24  24,, 145–74. Chen, Z., Sanchez, R., & Campb Campbell, ell, T. T. ( 1997). From beyond to within their grasp: Psychology, Analo Ana logi gical cal pr prob oblem lem so solv lving ing in 10- and 13-month-olds. Developmental Psychology, 33,, 790–801. 33  Matter and consciousness. consciousness. Cambridge, MA: MIT Press. Churchland, P. (1984). ). Matter Churchland, Churc hland, P. (1985). Red Reduct uction ion,, qual qualia, ia, and the dir direct ect int intro rospec spectio tion n of brai brain n sta states tes..  Journal of Philosophy Philosophy,, 82  82,, 8–28. Churchla Chur chland, nd, P., Ram Ramach achandr andran, an, V., & Sej Sejino inowsk wski, i, T. (1994).Acritiqueofpurevision. In C. Koch & J. Davis (Eds.), (Eds.), Large-scale  Large-scale neuronal theories of the brain (pp. brain  (pp. 23–60). Cambridge, MA: MIT Press. Cien Ci enki ki,, A. (1998). St Strai raigh ght: t: An im imag agee sc schem hemaa and it itss me meta taph phor orica icall ex exte tensi nsion ons. s. Cognitive Linguistics, Linguistics, 9  9,, 107–49.

 

References

 

291

 Situated cognition: On human knowledge and computer representaClancey, W. (1997). ). Situated tions.. New York: Cambridge University Press. tions Clar Cl ark k, A. (1996). Being there: Putting brain, body, and world together. together. Ca Camb mbrid ridge ge,, MA MA:: MIT Press. Science, 21  21,, 461–81. Clark, A. (1997). The dynamic challenge. Cognitive challenge.  Cognitive Science, Clark, A. (2003). ). Natural-born  Natural-born cyborgs: Minds, technologies, and the future of human intelligence.. New York: Oxford University Press. intelligence Clark, ). Using language language. . New York: Cambridge University Clarke,H. E.((1996 2001). ). Using Meaning and the specification of motion in music. Musicae music.Press.  Musicae Scientiae,, 5 tiae  5,, 213–34. Clarke, S., & French, R. (1978). Can congenital amputees achieve academically?  American Corrective Therapy Journal, Journal , 32  32,, 7–11. Clearfield, M. (2000). The role of locomotor experience in the development of navigational memory. memory. Unpublished Ph.D. dissertation. Clement, C. (1987). Applying general principles to novel problems as a function of learning history: Abstraction from examples vs. studying general statements. Dissertation Abstracts International, 48, 48, 585. Coccia, M., Bartolini, M., Luzzi, S., Provinciali, L., & Ralph, M. ( 2004). Semant Semantic ic memo me mory ry is an am amod odal, al, dy dynam namic ic sy syst stem em:: Ev Evide idenc ncee fr from om th thee int intera eract ctio ion n of na nami ming ng and object use in semantic dementia. Cognitive dementia.  Cognitive Neuropsychology, Neuropsychology, 21  21,, 515–27. Cohen, L., & Oakes, L. (1993). How infants perceive a simple causal event. Develevent.  Developmental Psychology, 29, 29, 421–433. Colcom Colc ombe be,, S. S.,, & Wye yerr, R. (2001). The ro role le of pr prot otot otyp ypes es in th thee me ment ntal al re repr prese esent ntat atio ion n Psychology, 44  44,, 67–105. of temporally related events. Cognitive events.  Cognitive Psychology, Cole, J. (1995). ). Pride  Pride and the daily marathon. marathon. Cambridge, MA: MIT Press. Cole, J. (1997). On “being faceless”: Selfhood and facial embodiment.  Journal of  Consciousness,, 4 Consciousness  4,, 467–84. Cole, M., Hood, L., & McDermott, R. (1997). Concepts of ecological validity: Their differing implications for comparative cognition. In M. Cole & Y. Engestroem  Mind, culture, and activity (pp. activity  (pp. 48–58). New York: Cambridge University (Eds.), Mind, (Eds.), Press. Collie, R., & Hayne, H. (1999). Deferred imitation by 6- and 9-month-old infants: Developmen mental tal Psy Psychob chobiol iology ogy,, 35 35,, More Mo re ev evide idenc ncee fo forr de decl clara arati tive ve me memo mory ry..   Develop 83–90. Mental im image agery ry an andd the their ir tr tran ansf sfor orma matio tions ns.. Coop Co oper er,, L. L.,, & Sh Shep epar ard, d, R. (1982).   Mental Cambridge: MIT Press. Connell, J. (1989). A colony architecture for an artificial machine. AI Tech Report 1152, MIT AI Labs, August. Corballis, M. (1994). Neuropsychology of perceptual functions. In D. Zaidel (Ed.), Neuropsychology handbook of perception and cognition ( cognition  ( 2nd ed.) (pp.  83 –104). San Diego, CA: Academic Press. Corriss, D., & Kose, G. (1998). Action and imagination in the formation of images. Perceptual and Motor Skills, Skills, 87  87,, 979–83. Cosm Co smid ides es,, L. L.,, & Too ooby by,, J. (1997). Dis Dissec sectin ting g the com comput putati ational onal arc archit hitect ecture ure of soc social ial inference mechanisms. Characterizing human psychological adaptations. Ciba Foundation symposium, No. 208 (pp. 132–61). New York: Wiley.  Journal of Experimental Craighero, Craighe ro, L. (1996). Grasping a fruit: Selection Selection for action. action. Journal Psychology: Human Perception and Performance, Performance, 22  22,, 582–603.

 

292

 

References

Craighero, L., Fadiga, L., Rizzolatti, G., & Umilta, C. ( 1999). Action for perception: A motor-visual attentional effect. Journal effect. Journal of Experimental Psychology: Human Perception and Performance, Performance, 25  25,, 1673–92. Creem, S., & Proffitt, D. (1998). Two memories for geographical slant: Separation Review, and interdependence of action and awareness. Psychonomic awareness.  Psychonomic Bulletin and Review, 5, 22–36. Crick,F.,&Koch,C.(1996). Why Whyneu neuro rosc scien ience ce ma may y be ab able le to exp expla lain in co consc nscio ious usnes ness. s. Scientific American, 273, 273, 84–85. 2004). Cognit Crott,W.,&Cruse,A.( Cognitive ive Lingui Linguistics. stics.New New York: Cambr Cambridge idge Universi University ty Press. Csikszentmihalyi, M. (1990). ). Flow:  Flow: The psychology of optimal experience. experience. New York: Perennial.  Embodiment and experience. experience. New York: Cambridge UniverCsordas, T. (Ed.) (1994). ). Embodiment sity Press. Cutting, J., Proffitt, D., & Kozlowski, L. (1978). A biomechanical invariant for gait  Journal of Experimental Psychology: Human Perception and Performance, Performance , perception. Journal perception. 4, 357–72. Dama Da masi sio, o, A. (1989). Time ime-lo -locke cked d mul multir tiregi egional onal ret retro roact activat ivation ion:: A sys system tem-lev -level el pro pro-33, 25–62. posal for the neural substrate of recall and recognition. Cognition, recognition.  Cognition, 33, Dama Da masi sio, o, A. (1994). Descartes Descartes’’ error: Emotion, reason, and the human brain. brain . Ne New w Yor ork: k: G.P. Putnam & Sons.  The feeling of what happens: Body and emotion in the making of  Damasio, A. (1999). ). The

consciousness.. New York: Harcourt Brace & Co. consciousness  Looking for Spinoza: Joy, sorrow, and the feeling brain. brain . New York: Damasio, A. (2003). ). Looking Harcourt. Damasio, A., & Damasio, H. (1994). Cortical systems for retrieval of concrete knowledge: The convergence zone framework. In C. Koch & J. Davis (Eds.), LargeLar ge-sca scale le neu neuro ronal nal theo theorie riess of the bra brain in   (pp.   61–74). Ca Camb mbri ridg dge, e, MA MA:: MI MIT T Press. Decety, J., & Grezes, J. (1999). Neural mechanisms subserving the perception of  human action. Trends action. Trends in Cognitive Science, 3, 3, 172–78. Decety,, J., Grezess, J., Costes, N., Perani, D., Jeannerod, M., Procyk, E., Decety E ., Grassi, F., & Fazio, F. (1997). Brain activity during observation of actions: Influence of action content and subject’s strategy. strategy. Brain, 120, 120, 1763–77. Dece De cety ty,, J. J.,, Je Jean anne nerrod od,, M. M.,, & Pr Prob obla lanc nc,, C. (1989). The ti timi ming ng of me ment ntall ally y re repr pres esent ented ed  Behavioral Brain Research, Research, 34  34,, 35–42. actions. Behavioral actions. Decety Dec ety,, J., Pera Perani, ni, D., Jean Jeanner nerod, od, M., Bet Bettin tinard ardi, i, V., Tadar adary y, B., Wood oods, s, R., Maz Mazzio ziotta tta,, Nature,, 371 371,, 600–2.  J., & Fazio, F. (1994). Map Mapping ping mot motor or rep repres resenta entatio tions ns wit with h PET PET.. Nature DeLoache, J., Uttal, D., & Rosengren, K. (2004). Scale errors offer evidence for a perception-action dissociation early in life. Science life. Science,, 304  304,, 1027–9.  Consciousness explained. explained. Boston: Little Brown. Dennett, D. (1992). ). Consciousness De Rivera, J. (1977). ). A  A structural theory of the emotions. emotions. New York: International Universities Press. philoso osophi phical cal wri writin tings gs of Des Descar cartes tes:: vol. 2. New York: Descartes, R. (1984).   The phil Cambridge University Press. Dewey, J. (1896). The reflex arc concept in psychology.  Psychological Review, Review,   3, 357–70. Dewey, J. (1934). ). Art  Art as experience. experience. New York: Minton, Balch. Dewey, J. (1938). ). Logic:  Logic: The theory of inquiry. inquiry. New York: Henry Holt.

 

References

 

293

Diamond, A. (1991). Neuro-psychological insights into the meaning of object con-

Thee ep epige igene nesi siss of mi mind nd:: Es Essa says ys cept de cept deve velo lopm pmen ent. t. In S. Ca Care rey y an and d R. Ge Gelm lman an (E (Eds ds.) .),, Th on biology & cognition (pp. cognition (pp. 67–110). Hillsdale, NJ: Erlbaum. Dijksterhuis, A., & van Knippenberg, A. (1998). The relation between perception and behavior, or how to win a game of Trivial Pursuit.  Journal of Personality and Social Psychology, Psychology, 74  74,, 865–77. Dijksterhuis, A., Bargh, J., & Miedema, J. (2001). Of mice and mackerels: Attention and an d au auto toma mati ticc so soci cial al be beha havi vior or.. In H. Bl Bles esss & J. Fo Forrga gass (E (Eds ds.) .),, Subjective experience in social cognition and behavior (pp. behavior (pp. 37–51). Philadelphia: Psychology Press. DiPelligrinoi, G., Fadiga, Fadiga Fogassi, L.,Brain Ga llese, Gallese, V., &, 91 V., Rizzolatti, standing motor events., L., events. Experimental  Experimental Research Research,  91, , 176–80.G. (1992). UnderdiSessa, A. (1993). Toward Toward an epistemology of physics. Cognition physics.  Cognition & Instruction, Instruction, 10  10,, 105–225. Dodd Do dd,, B. (1979). Li Lipr prea eadin ding g in in infan fancy cy:: At Atte tent ntio ion n to spe speec ech h in and out of sy sync nchr hron ony y. Cognitive Psychology, Psychology, 11  11,, 478–84. Doffner, G. (1999). The connectionist route to embodiment and dynamicism. In A. Riegler, M. Peschl, & A. von Stein (Eds.),  Understanding representation in the cogniti cogn itive ve sci scienc ences: es: Doe Doess re repr prese esenta ntatio tional nal nee needd re reali ality ty?? (p (pp. p. 23–32). New York ork:: Kluw Kluwer er..  A mind so rare: The evolution of human consciousness. consciousness . New York: Donald, M. (2001). ). A Norton. Dorm Do rman an,, M. M.,, St Stud udde dert rt-K -Ken enne nedy dy,, M. M.,, & Ra Raph phae ael, l, L. (1977). Sto Stop-c p-conso onsonant nant recognition: Release bursts and formant transitions as functionally equivalent, context-dependent cues. Perception cues. Perception & Psychophysics, 22, 22, 109–22. Douglas, M. (1970). ). Natural  Natural symbols. symbols. New York: Pantheon. Downing, L. (2000). ).   Negation, text worlds, and discourse: The pragmatics of fiction. fiction . Mahwah, NJ: Erlbaum. Duclos, S., Laird, J., Schneider, E., Sexter, M., Stern, L., & Van Lighten, O. (1989). Emotion-specific effects of facial expressions and postures on emotional experience. Journal ence.  Journal of Personality and Social Psychology, Psychology , 57  57,, 100–8. Eagle, R. (1985). Deprivation of early sensorimotor experience and cognition in the severely involved cerebral palsy child. Journal child.  Journal of Autism, and Developmental Disorders,, 15 Disorders  15,, 269–83. Brig ight ht ai airr, br bril illi lian antt fir fire: e: On th thee ma matt tter er of th thee mi mind nd.. Ne Edel Ed elma man, n, G. (1992). Br New w Yor ork: k: Ba Basi sicc Books. Eime Ei merr, M. (1995). Sti Stimulu mulus-r s-resp esponse onse com compat patibi ibilit lity y and auto automat matic ic res respons ponsee activation: Evidence from psychophysiological studies. Journal studies.  Journal of Experimental Psychology: Human Perception and Performance, Performance, 21  21,, 335–59. Ekman, P. (1985). Telling lies: clues to deceit in the marketplace, politics, and marriage. New York: Norton. Ekman, P. (1992). Are there basic emotions? Psychological emotions? Psychological Review, Review, 99  99,, 550–3. Ekman, P. (1994). Strong evidence for univers universals als in facial expressions: expressions: A reply to Russell’s mistaken critique. Psychological critique. Psychological Bulletin, Bulletin, 115  115,, 268–87. Ekman, P., & Davidson, R. (Eds.) (1994). ). The  The nature of emotions. emotions. New York: Oxford University Press. Ekman, P., P., Levenson, R., & Friesen, Fri esen, W. W. (1983). Autonomic nervous system activity distinguishing among emotions. Science emotions. Science,, 221  221,, 1208–10. Ellis, N., & Hennelly Hennelly,, R. ( 1980). A bilingual word-length effect: Implications for intell te llig igenc encee te test stin ing g an and d th thee re relat lativ ivee eas easee of me ment ntal al ca calc lcula ulati tion on in Wel elsh sh and En Engl glish ish.. British Journal of Psychology, Psychology, 71  71,, 43–52.

 

294

 

References

 Questioning consciousness. consciousness. Amsterdam: Benjamins. ). Questioning Ellis, R. (1995). Ellis,R.(2002). Ef Effer ferent ent brai brain n pr proce ocesses sses and the enac enactiv tivee appr approach oachto to cons conscio ciousne usness. ss.  Journal of Consciousness Studies, Studies, 7  7,, 40–52. Ellis, R., & Tucker, M. (2000). Micro-affordances: The potentiation of components Psychology, 91  91,, 457–71. of action by seen objects. British objects. British Journal of Psychology,

Emmorey, K. (2002). ). Language,  Language, cognition, and the brain: Insights from sign language research.. Mahwah, NJ: Erlbaum. research Emmorey, K., Kosslyn, S., & Bellugi, U. (1993). Visual imagery and visual-spatial langu lan guag age: e: Enh Enhanc anced ed im imag agery ery ab abil ilit itie iess in de deaf af and he heari aring ng AS ASL L si sign gners ers.. Cognition Cognition,, 46,, 139–81. 46 1998 Engelkamp, J.,(& ). Memory ).  Memory for actions. actions .tor Hove, England: Psychology Enge En gelk lkam amp, p, J. J., Zimm Zi mmer er,, H. (1984 ). Mo Moto r pr prog ogra ramm mmee inf infor orma mati tion on asPress. a sep separa arabl blee  Psychological Research, 46, 46, 283–297. unit. Psychological unit. Epstein, W. (1973). The process of ‘taking-into-account’ in visual perception. Perception,, 2 Perception  2,, 267–85. Farah, M., Hammond, K., Levine, D., & Calvanio, R. (1988). Visual and spatial mental imagery: Dissociable systems of representation. Cognitive representation. Cognitive Psychology, Psychology, 20  20,, 439–62. Farah, M., & McClelland, J. (1991). A computational model of semantic memory impairment: Modality specificity and emergent category-specificity. Journal category-specificity.  Journal of  Experimental Psychology: General, General, 120  120,, 339–57.  Do you see what I mean? Plains Indian sign talk and the embodiment Farnell, B. (1995). ). Do of action. action. Austin: University Univers ity of Texas Texas Press. Feldman, J., & Narayanan, S. (2004). Embodiment in a neural theory of language.

Brain & Language, Language, 89  89,, 385–92. Fernandez-Do Fernande z-Dols, ls, J-M J-M., ., & Car Carrer rera, a, P. (1994). Neu Neutr tral al fac faces es in co cont ntext ext:: The Their ir em emot otio ional nal Behavior, 11  11,, 287–99. meaning and their function. Journal function.  Journal of Nonverbal Behavior, Fernandez-Dols, J-M., Carrera, P., Casado, C. ( 2002). The meaning of expression: Vie iews ws fr from om ar artt an and d ot othe herr so sour urce ces. s. In L. An Anol olli li,, R. Ci Cice ceri ri,, & G. Ri Riva va (E (Eds ds.) .),, Say not to say: New perspectives on miscommunication (pp. 117–34). Am Amst ster erda dam: m: IO IOS S Pr Pres ess. s. Ferna Fe rnande ndez-D z-Dol ols, s, J-M J-M., ., & Ru Ruiz iz-Be -Beld lda, a, M. (1995). Ar Aree sm smil iles es si sign gnss of ha happi ppines ness? s? Go Gold ld meda me dall wi winn nner erss at th thee Ol Olym ympi picc ga game mes. s. Journal  Journal of Personality and Social Psychology Psychology,, 69,, 1113–19. 69 Ferrara, K. (1994). ). Therapeutic  Therapeutic ways with words. words . New York: Oxford. Fery, J.-C. (2003). Differentiating visual and kinesthetic imagery in mental practice Canadian Journal of Experimental Psychology, Psychology, 57, xx–xx. Fillm Fi llmor ore, e, C. (1982). Fra Frame me se sema mant ntic ics. s. In Li Ling nguis uisti ticc So Socie ciety ty of Ko Kore reaa (Ed (Ed.) .),, Linguistics in the morning calm (pp. calm (pp. 11  111 1–38). Hansin, Seoul. Finke, R. (1989). ). Principles  Principles of mental imagery. imagery. Cambridge: MIT Press. Fink Fi nke, e, R. R.,, & Fr Frey eyd, d, J. (1985). Tra ransf nsfor orma mati tion onss of vi visua suall me memo mory ry ind induce uced d by im impl plied ied  Journal of Experimental Psychology Psychology:: Learn Learning, ing, Memory Memory,, motion mot ionss of pat patter tern n elem element ents. s. Journal and Cognition, Cognition, 11  11,, 780–94. Finke, R., Freyd, J., & Shyi, G. ( 1986). Implied velocity and acceleration induce General, transformations of visual memory.  Journal of Experimental Psychology: General, 115,, 175–88. 115 Fishe Fi sherr, S. (1990). Th Thee ev evol olut utio ion n of psy psych chol olog ogic ical al co conc ncept eptss ab abou outt th thee bo body dy.. In T. Ca Cash sh images:: Develo Development, pment, deviance, deviance, and change (pp. change (pp. 3–20). & T. Pruzinsky (Eds.), Body (Eds.),  Body images New York: Guilford.

 

References

 

295

Fitzpatrick, P., Carello, C., Schmidt, R., & Corey, D. (1994). Haptic and visual perPsychology, 6  6,, 265–88. ception of an affordance for upright posture. Ecological posture.  Ecological Psychology, Flanagan,, O. (2002). Flanagan ). The  The problem of the soul: Two visions of mind and how to reconcile them.. New York: Basic Books. them Flanagan, R., & Beltzer, M. (2000). Independence of perceptual and sensorimotor prediction in the size-weight illusion. Nature illusion. Nature Neuroscience, Neuroscience, 3  3,, 737–41. Flan Fl anag agan an,, R. R.,, Ki King ng,, S. S.,, Wol olpe pert rt,, D. D.,, & Jo Joha hans nsso son, n, R. (2001). Sen Sensori sorimot motor or pre predic dictio tion n and me memo mory ry in ob obje ject ct ma manip nipula ulati tion on.. Canadian Journal of Experimental Psychology, Psychology, 55,, 87–95. 55 Fletcher, C., van den Broek, P., & Arthur, E. (1996). A model of narrative compre-

hens he nsio ion n an and d rec ecal all. l. In B. Br Brit itto ton n & A. Gr Grae aess sser er (E (Eds ds.) .),, Models of understanding text (pp. 142R. –64(). Mahwah, NJ:ofErlbaum. 1994 Fletcher, ). Levels representation in memory for discourse. In M. Gernsbacher Gernsb acher (Ed.),   Handbook Handbook of psyc psychol holingu inguist istics ics   (pp.   589–608). Sa San n Di Dieg ego: o: Academic Press. Flores d’Arcais, G., & Schreuder Schreuder,, R. ( 1982). Semantic activation during object naming. Psychological ing.  Psychological Research, Research, 49  49,, 153–9.  The modularity of mind. mind . Cambridge, MA: MIT Press. Fodor, J. (1983). ). The Fogel, A., & Hannan, Han nan, T. (1985). Manual actions of nine- to fifteen-week-old human infants infa nts duri during ng fac face-t e-to-fa o-face ce int interac eractio tions ns wit with h the their ir mot mothers hers.. Child Develo Development pment,, 56 56,, 1271–9. Foster, D. (1983). Visual discrimination, categorical identification, and categorical rating in brief displays of curved lines: Implications for discrete encoding pro Journal of Experimental Psychology: Human Perception and Performance, Performance,   9, cesses. Journal cesses. 785–806. Foster, J., & Strack, F. (1996). Influence of overt head movement on memory for valenced words: A case of conceptual-motor compatibility. compatibility . Journal of Personality and Social Psychology, 71, 71, 421–30. Fowler, C. (1986). An event approach to the study of speech perception from a Phonetics, 14  14,, 3–28. direct-realist perspective. Journal perspective. Journal of Phonetics, Fowler, C. (1987). Perceivers as realists, talkers too: Commentary on papers by Strange, Diehl et al., and Rakerd and Verbrugge. Journal Verbrugge.  Journal of Memory & Language, 26,, 574–87. 26 Fowler,C.(1994).Auditory“objects”–TheroleofmotoractivityinauditorypercepOrigins: s: Brain and selfself-orga organizati nization on tion ti on and sp speec eech h pe perc rcep epti tion on.. In K. Pri Pribr bram am (Ed (Ed.) .),, Origin (pp. 593–603). Hillsdale, NJ: Erlbaum. Fowler, C., & Rosenblum, D. (1991). The perception of phonetic gestures. In I.  Modularity and the motor theory of speech Mattin Mat tingly gly & M. Stu Studder ddert-K t-Kenne ennedy dy (Eds (Eds.), .), Modularity  perception (pp.  perception  (pp. 33–59). Hillsdale, NJ: Erlbaum. Franz, V., Gegenfurtner, K., Buelthoff, H., & Fahle, M. (2000). Grasping visual illusions: No evidence for a dissociation between perception and action.  Psychological Science, Science, 11  11,, 20–5. Free Fr eema man, n, N. N.,, Ll Lloy oyd, d, S. S.,, & Si Sinh nha, a, C. (1980). In Infan fantt se sear arch ch ta task skss re reve veal al ea early rly co conc ncept eptss  8,, 243–62. of containment and canonical usage of objects. Cognition objects.  Cognition,, 8 Freem Fr eeman, an, W. (1991). Th Thee phy physi siol olog ogy y of per percep cepti tion on..   Scientific Scientific Americ American an,,   264 264,, 78–87.  How brains make up their minds minds.. New York Freema Fr eeman, n, W. (2001). ). How ork:: Col Columb umbia ia Univ Universi ersity ty Press.

 

296

 

References

Review, 94  94,, 429–38. representation. Psychological Review, Freyd, J. (1987). Dynamic mental representation. Psychological Freyd, J., & Finke, R. (1984). Representational momentum. Journal momentum.  Journal of Experimental Psychology: Learning, Memory, and Cognition, Cognition, 10  10,, 126–32. Freyd, J., & Finke, R. (1985). A velocity effect for representational momentum. Bulletin of the Psychonomic Society, Society, 23  23,, 443–6. Freyd, J., & Johnson, J. ( 1987). Probing the time course of representational mo Journal of Experimental Psychology: Learning, Memory, and Cognition, Cognition,  10  10,, mentum. Journal mentum. 126–32. Freyd, J., & Jones, K. (1994). Representational momentum for a spiral path. Journal path.  Journal of Experimental Psychology: Learning, Memory, and Cognition, Cognition, 20  20,, 968–76. Freyd, J., Kelly, M., & DeKay, M. ( 1990). Representational momentum in memory for pitch. Journal pitch. Journal of Experimental Psychology: Learning, Memory, and Cognition, Cognition , 16  16,, 1107–J., 17& . Pantzer, T. (1995). Static patterns moving in the mind. In S. Smith, T. Freyd, War ard, d, & R. Fi Fink nkee (E (Eds ds.) .),, The creative cognition approach (pp. 181–204). Cam Cambri bridge dge::

MIT Press. Freyd,J.,Pantzer,T.,&Cheng,J.(1988).Representingstaticsasforcesinequilibrium.  Journal of Experimental Psychology: Psycholog y: General, General , 117  117,, 395–407. Friberg, A., & Sundberg, J. (1994). Does music performance allude to locomotion? A model of final ritardandi derived from measurements of stopping runners.  Journal of the Acoustical Society of America America,, 105  105,, 1469–84. Friberg, A., & Sundberg, J. (1999). Does music performance allude to locomotion? A model of final ritardandi derived from measurements of stopping runners.  Journal of the Acoustical Society of America, 105, 105 , 1469–84. Friberg, A., Sundberg, J., & Fryden, L. (2000). Music for motion: Sound level envelopes of tones expressing human locomotion. Journal locomotion.  Journal of New Musice Rsearch, Rsearch, 24,, 199–210. 24 Fridlund, A. (1994). ). Human  Human facial expression. expression. San Diego: Academic Press. Friedman, R., & Foerster, J. ( 2000). The effect of approach and avoidance motor actionss on the elements of creat action creative ive insight. Journal insight. Journal of Personality and Social Psychology,, 79 chology  79,, 477–92. Frith, C., Blakemore, S., & Wolport, D. (2000). Explaining the symptoms of  Reviews, schizophrenia: Abnormalities in the awareness of action. Brain action.  Brain Research Reviews, 31,, 357–63. 31 Gain Ga inot otti ti,, G. G.,, Si Silv lver eri, i, M. M.,, Da Dani niel ele, e, A. A.,, & Gi Gius usto toli lisi si,, L. (1995). Neur Neuroim oimagi agining ning cor correre 3,, 247 – lates of category-specific semantic disorders: A critical survey. Memory survey.  Memory,,  3 64. Galla Ga llagh gher er,, S. (1995). Bo Body dy sc sche hema ma an and d in inte tent ntio iona nali lity ty.. In J. L. Be Berm rmud udez ez & A. Ma Marc rcel el  The body and self  (pp. (Eds.), The (Eds.),  (pp. 225–44). Cambridge, MA: MIT Press. Galla Ga llagh gher er,, S. (2001). Th Thee pra pract ctic icee of mi mind: nd: The Theor ory y, si simu mulat lation ion,, or pr prim imar ary y interaction? Journal interaction?  Journal of Consciousness Studies, Studies, 8  8,, 83–103. Gallagher, S., & Marcel, A. (1999). The self in contextualized action. Journal action.  Journal of Consciousness Studies, Studies, 6  6,, 212–28. Galla Ga llagh gher er,, S. S.,, & Me Melt ltzo zoff ff,, A. (1996). Th Thee em emer ergi ging ng se sens nsee of se self lf an and d ot othe hers rs:: Me Merl rlea eauuPonty and recent developmental studies. Philosophical studies.  Philosophical Psychology, Psychology, 9  9,, 211–33. Gallese, V. (2000). The inner sense of action: Agency and motor representations  Journal of Consciousness Studies, 7, 23–40. Gallese, V., V., Ferari, P., & Umilta, M. (2002). The mirror matching system: A shared manifold for intersubjectivity intersubjectivity.. Behaviorial and Brain Sciences, Sciences, 25  25,, 35–6.

 

References

 

297

Gallese, V., & Goldman, A. ( 1998). Mirron neurons and the simulation theory of  2, 439–50. mind reading. Trends reading. Trends in Cognitive Science, 2, Gardner, H. (1983). ). Frames  Frames of mind: The theories of multiple intelligences. intelligences . New York: Basic Books.  The mind’ mind’s new science: A history of the cognitive revolution. revolution . New Gardner,, H. (1985). Gardner ). The York: Basic Books. Gardner, R., Martinez, R., & Sandoval, Y. ( 1987). Obesity and body image: An evaluation of sensory and non-sensory components. Psychological components.  Psychological Medicine, 17, 17, 927–32. Garr Ga rrod od,, S. S.,, & Sa Sanf nfor ord, d, A. (1985). On th thee re realal-ti time me ch chara aract cter er of in inte terpr rpret etat atio ion n dur durin ing g reading. Language reading.  Language and Cognitive Processes, Processes, 1  1,, 43–61. Garry, M., & Polaschik, D. (2000). Imagination and memory. Current memory.  Current Directions in Psychological Science, Science, 9  9,, 6–10. Geeraerts, (1993 vagueness.  Cognitive Lin guistics,, 4  guistics  4,D. , 223 –72). . Vagueness’s puzzles, polysemy’s vagueness. Cognitive  Diachronic ic prot prototype otype semantics: A contri contribution bution to histor historical ical lexiGeeraerts, Geeraert s, D. (1997). ). Diachron cography.. Oxford: Clarendon. cography society. New York: Cambridge Geertz, C. (1979).  Meaning and order in Morocan society. University Press.

Gelman, R. (1991). Epigenetic foundations of knowledge structures: Initial and transcendent construction. In S. Carey and R. Gelman (Eds.), Epigenesis (Eds.),  Epigenesis of mind: Essays in biology and cognition (pp. cognition (pp. 293–322). Hillsdale, NJ: Erlbaum. Gelman, R., Durgin, F., & Kaufman, L. (1995). Disting Distinguishing uishing between animate animatess and inanimates: Not by motion alone. In D. Sperber & D. Premack (Eds.), Causal (Eds.),  Causal cognition (pp. cognition  (pp. 150–84). New York: Oxford University Press. Gent Ge ntner ner,, D. D.,, Im Imai, ai, M. M.,, & Bo Boro rodi dits tsky ky,, L. (2002). As ti time me go goes es by by:: Un Unde ders rsta tand ndin ing g ti time me Processes, 17  17,, 537–65. as spatial metaphor metaphor.. Language and Cognitive Processes, Georgo Geor gopo poul ulos os,, A. A.,, Lu Luri rito to,, J. J.,, Pe Petr trid ides es,, M. M.,, Sc Schw hwar artz tz,, A. A.,, & Ma Mass ssey ey,, J. (1989). Me Ment ntal al rotation of the neuronal n euronal population vector ve ctor.. Science  Science,, 243  243,, 234–36. Gergely, G., Nadasdy, Z. Csiba, G., & Biro, S. (1995). Taking the intentional stance at 12 months of age. Cognition age.  Cognition,, 56  56,, 165–93. Gergely, G., & Watson, J. (1999). Early socio-emotional development: Contingency perception and the social-biofeedback model. In P. Rochat (Ed.),   Early social cognition: Understanding others in the first months of life (pp. life  (pp.   101–36). Mahwah, NJ: Erlbaum.  The saturated self . New York: Basic Books. Gergen, K. (1991). ). The Gerl Ge rlac ach, h, C. C.,, La Law w, I. I.,, & Pa Paul ulso son, n, O. (2002). Wh When en act actio ion n tu turn rnss in into to wo word rds: s: Ac Acti tiva vati tion on  Journal of mot motoror-bas based ed kno knowle wledge dge duri during ng cat catego egoriz rizatio ation n of man manipul ipulabl ablee obj object ects. s. Journal of Cognitive Neuroscience, Neuroscience, 14  14,, 1230–9. Geurts, K. (2002). ). Culture  Culture and the senses: Bodily ways of knowing in an African community.. Berkeley: University of California Press. nity Gibbs, R. (1994). ). The  The poetics of mind: Figurative thought, language, and understanding. understanding. New York: York: Cambridge University Un iversity Press.  61,, 309–19. Gibbs, R. (1996). Why many concepts are metaphorical. Cognition metaphorical.  Cognition,, 61 Gibbs, R. (1999a).   Intentions in the experience of meaning. meaning. New York: Cambridge University Press. Gibbs,, R. (1999 b). Moving metaphor out Gibbs ou t of the head he ad and an d into the cultural cultura l world. In R. Gibbs & G. Steen (Eds.),   Metaphor in cognitive linguistics   (pp.   145–66). Amsterdam: Benjamins.

 

298

 

References

Gibbs, R., Beitel, D., Harrington, M., & Sanders, P. ( 1994). Taking a stand on the meanings of “stand”: Bodily experience as motivation for polysemy.  Journal of  Semantics,, 11 Semantics  11,, 231–51. Gibbs,R.,&Berg,E.(2002). Ment Mental al ima imager gery y and emb embodie odied d act activi ivity ty.. Journal of Mental Imagery,, 26 Imagery  26,, 1–30. Gibbs,, R., & Franks, H. (2002). Embodied metaphor in women’s narratives about Gibbs Communication, 14  14,, 139–65. their experiences with cancer. Health cancer. Health Communication, Gibbs, R., Lima, P., & Francuzo, E. ( 2004). Metaphor in thought and language is grounded in embodied experience. Journal experience. Journal of Pragmatics, Pragmatics, 36  36,, 1189–210. Gibbs, R., & Matlock, T. (2000). Psycholinguistics and mental representations. Cognitive Linguistics, Linguistics, 10  10,, 263–9. Gibbs, R., & O’Brien, J. ( 1990). Idioms and mental imagery: The metaphorical mo 36,, 35–68. tivation for idiomatic meaning. Cognition meaning.  Cognition,, 36 Gibb Gi bbs, s, R. R.,, St Stro rom, m, L. L.,, & Sp Spiv ivey ey-Kn -Know owlt lton on,, M. (1997). Con Concep ceptual tual met metapho aphorr in men mental tal imagery forTenney, proverbs. proverbs. Journal Mental Imagery Imagery, , 21  21,, 83in –110 . Gibbs, R., & Y.  Journal (1980). of The concept of scripts understanding stories.  Journal of Psycholinguistic Research, Research, 9  9,, 275–84. Gibbs, R., & Van Orden, G. (2003). Are emotional expressions intentional? A selfEmotion, 4  4,, 1–16. organizational approach. Consciousness approach. Consciousness & Emotion, Gibson, E. (1988). Exploratory behavior in the development of perceiving, acting, Psychology, 39  39,, 1–41. and the acquiring of knowledge. Annual knowledge. Annual Review of Psychology, Review, 69  69,, 477–90. Gibson, J. (1962). Observations on active touch. Psychological touch.  Psychological Review, Gibson, J. (1966). ). The  The senses considered as perceptual systems. systems . Boston.  Psychological Review Review,, Gibson, J. (1968). What gives rise to the perception of motion? motion? Psychological

57, 335–46. 57, Gibson, J. (1979).  The ecological approach to visual perception. perception . Boston: Houghton Mifflin. Gilbert, A., Fridlund, A., & Sabini, J. ( 1987). Hedonic and social determinants of  facial displays to odors. Chemical odors. Chemical Senses, Senses, 12  12,, 355–63. Sciences, 20  20,, 1–55. Glenberg, A. (1997). What is memory for? Behavioral for?  Behavioral and Brain Sciences, Glenberg, A. (1999). Why mental models need to be embodied. In G. Rickert & C. Habel (Eds.), Mental (Eds.),  Mental models in discourse processing (pp. processing  (pp.  7  77 7–90). Amsterdam: Elsevier. Glenberg, A., & Robertson, D. ( 2000). Symbol grounding and meaning: A comparison of high-dimensional and embodied theories of meaning. Journal meaning.  Journal of Memory and Language, Language, 43  43,, 379–401. Glenberg, A., Meyer, M., & Lindem, K. (1987). Mental models contribute to foreLanguage,, 26 26,, 69–83. ground gr ounding ing duri during ng tex textt com compre prehens hension ion.. Journal of Memory and Language Glenberg, A., Schroeder, J., & Robertson, D. ( 1998). Averting the gaze disengages Cognition, 26  26,, 651–8. the environment and facilitates remembering. Memory remembering.  Memory & Cognition, language. Ne Gluck Gl ucksb sber erg, g, S. (2001). Understanding figurative language. New w Yor ork: k: Ox Oxfo forrd Un Univ iver er-sity Press. Glucksberg, S. (2002). Emotion language: A new synthesis? Contemporary synthesis?  Contemporary Psychology,, 47 ogy  47,, 764–6. Glucksberg, S., Brown, M., & McGlone, M. (1993). Conceptual metaphors are not Cognition,   21 21,, automatically accessed during idiom comprehension. Memory comprehension.  Memory & Cognition, 711–19. Glucksberg, S., & Keysar, B. ( 1990). Understanding metaphorical comparisons: Review, 97  97,, 3–18. Beyond similarity. Psychological similarity. Psychological Review,

 

References

 

299

Goff, L., & Roediger, H. (1998). Imagination inflation for action events: Repeated Cognition, 26  26,, 20–33. imaginings leads to illusory recollection. Memory recollection.  Memory & Cognition, Goffman, E. (1959). ). The  The presentation of self in everyday life. life. New York: Doubleday.  54,, 787–815. Goffman, E. (1976). Response cries. Language cries. Language,, 54 Goldap, C. (1992). Morphology and semantics s emantics of Yucatec Yucatec space relators. Zeitschrift relators.  Zeitschrift  fur Phonetik, Sprachwissenschaft und Kommunikationsforschung, 45, 45 , 612–25.  Constructions.. Chicago: University of Chicago Press. Goldberg, A. (1995). ). Constructions Goldfield, E. (1993). Dynamic systems in development: Action systems. In L. Smith & E. Thelen (Eds.), A (Eds.), A dynamic systems approach to development: Applications (pp. 51–70). Cambridge, MA: MIT Press. Goldie, P. (2000). Explaining expressions of emotion. Mind emotion.  Mind,, 109  109,, 25–38. Gold Go lding inger er,, S. (1995). Ec Echo hoes es of ec echo hoes es?? An ep epis isod odic ic th theo eory ry of le lexi xica call ac acce cess ss.. Psychological Review, Review, 105  105,, 251–79. Goldman, A. (1970). ). A  A theory of action action.. Englewood Cliffs, NJ: Prentice-Hall. Goodale, M., & Humphrey Humphrey,, G. ( 1998). The objects of action and perception. Cogniperception.  Cogni 181 67 67, –207 . phy Gotion, Good odal ale, e, ,M. M., ,& Mur urph y, K. (2000). Sp Spac acee an and d th thee br brai ain: n: Di Difffe ferren entt ne neur ural al su subs bstr trat ates es for allocentric and egocentric forms of reference. In T. Metzinger (ed.), Neural (ed.),  Neural correlates of consciousness (pp. consciousness (pp. 189–202). Cambridge, MA: MIT Press. Good Go odwi win, n, C. (1981). Con Conver versat sation ional al or organ ganiza izatio tion: n: Int Intera eracti ction on bet betwee ween n spe speake akers rs and hea hearrers.. New York: Academic Press. ers Goossens, L., Pauwels, B., B. , Rudzka-Ostyn, M., Simon-Venderberger Simon-Venderberger,, J., & Varpays,  J. 1995.  By word of mouth: Metaphor, metonymy, and linguistic action in a cognitive  perspective.. Amsterdam: Benjamins.  perspective Gorf Go rfein ein,, D. (Ed (Ed.) .) (2001). On the con conseq sequen uences ces of mea meanin ningg sel selecti ection: on: Pri Princi nciple pless of re resol solvin vingg lexical ambiguity. ambiguity. Washington, DC: APA Books. Gottlieb, F. (2002). On the epigenetic evolution of species-specific perception: The dev develop elopmen mental tal mani manifol fold d conc concept epts. s.   Cognitiv Cognitivee Develo Development pment,,   17 17,,   1287–

310.

Gouin-Decarie, T. (1969). A study of the mental and emotional development of the thalidomide child. In B. Foss (Ed.). Determinants (Ed.).  Determinants of infant behavior (pp. behavior (pp. 167–187). London: Methuen. Linguistics, 8  8,, 267–90. Grady, J. (1997). Theories are buildings revisited. Cognitive revisited.  Cognitive Linguistics, Grady, J. (1999). A typolo typology gy of motiv motivation ation for concept conceptual ual metaph metaphor: or: Correlation Correlation vs. resemblance. In R. Gibbs & G. Steen (Eds.),  Metaphor in cognitive linguistics (pp. 79–100). Amsterdam: Benjamins. Graesser, A., Singer, M., & Trabasso, T. (1994). Constructing inferences during narReview, 101  101,, 371–95. rative text comprehension. Psychological comprehension. Psychological Review, Graesser, A., Woll, S., Kowalski, D., & Smith, D. ( 1980). Memory for typical and atypical actions in scripted activities. Journal activities.  Journal of Experi Experimental mental Psychology: Psychology: Human Learning and Memory, Memory, 6  6,, 503–15. Graf Gr afto ton, n, S. S.,, Fa Fadi diga ga,, L. L.,, Ar Arbi bib, b, M. M.,, & Ri Rizz zzol olat atti ti,, G. (1997). Pre Premot motor or cor cortex tex act activa ivatio tion n during observation and naming of familiar tools. Neuroimage tools.  Neuroimage,, 6  6,, 231–6. Granott, Granot t, N., & Paziale, J. (Eds.) (2002). ). Microdevelopment:  Microdevelopment: Transition processes in development and learning. learning. New York: Cambridge University Un iversity Press. Green, D. (2001). Understanding microworlds.   Quarterly Journal of Experimental Psychology, 54A, 54A, 879–911. Greenberg, J. (1978). ). Universals  Universals of human language: Vol. 3. 3 . Stanford, CA: Stanford University Press.

 

300

 

References

Gree eene ne,, R. R.,, & Sa Samu muel el,, A. (1986). Re Rece cenc ncy y an and d su suffi ffix x ef effe fect ctss in se seri rial al rec ecal alll of mu musi sica call Gr  Journal of Experimental Psychology: Learning, Memory and Cognition, Cognition ,   12 12,, stimuli. Journal stimuli. 517–24. Gregory,, R., & Wallace, J. (1963). Gregory ). Recovery  Recovery from early blindness: A case study. study. Monograph No. 2. Experimental Psychology Society. Society. Grezes, J., & Decety, J. (2001). Functional anatomy of execution, mental simulation, observation, and verb generation of actions: A meta-analysis. Human meta-analysis.  Human Brain  Mapping,, 12  Mapping  12,, 1–19. Grush, R. (2004). The emulation theory of representation: Motor control, imagery, Science , 27  27,, 377–396. and perception. Behavioral perception. Behavioral and Brain Science, Hadamard, J. (1945). ). The  The psychol psychology ogy of inven invention tion in the mathematical mathematical field. field. Princeton, NJ: Princeton University Press.  Journal Hadar, U. (1989). Two Two types of gesture and their role in speech production. production. Journal of Language and Social Psychology, Psychology, 8  8,, 221–8. Hadar Had ar,, U., Wenk enkertert-Olen Olenik, ik,D., D., Krau Krauss, ss,R., R., & Sor Soroke okerr, N. (1998).Gestureandthepro62 , 107–126. cessing of speech: Neuropsychological evidence. Brain evidence.  Brain & Language, 62, 1997). Haith, development of future&thinking essential the emer The for developmental genceM. of(skill inThe planning. planning . In S. Friedman Scholnick Scholn ickas(Eds.), The (Eds.),  psychology of planning: Why, how, and when do we plan plan?? (pp. 25–42). Mahwah, NJ Erlbaum. Halff, H., Ortony, Ortony, A., & Anderson, R. (1976). A context-sensitive representation of  Cognition, 4  4,, 378–84. word meaning. Menory meaning. Menory & Cognition, Hall, C., Bernoties, L., & Schmidt, D. (1995). Interference effects of mental imagery Psychology, 86  86,, 181–90. on a motor task. British task. British Journal of Psychology, Hamilton, E., & Cairns, H. (1961).   Plato: The collected dialogues. dialogues. Princeton, NJ: Princeton University Press. Hanna, E., & Meltzoff, A. (1993). Peer imitation by toddlers in laboratory and daycare contexts: Implications for social learning and memory. Developmental memory.  Developmental Psychology,, 29 Psychology  29,, 701–17. Hanrahan, C., Tetreau, B., & Sarrazin, C. (1995). Use of imagery while performing

dance movement. International movement. International Journal of Sport Psychology, Psychology , 26  26,, 413–30. Hardy, L., & Callow, N. ( 1999). Efficacy of external and internal visual imagery perspectives for the enhancement of performances on tasks in which form is

important. Journal of Sport and Exercise Psychology, important. Journal Psychology , 21  21,, 95–112. perspective. New Hark Ha rker er,, J. J.,, & Wie ierz rzbi bick cka, a, A. (E (Eds ds.) .) (2001). Emotion in crosslinguistic perspective. York: Mouton de Gruyter. Harman, K., Humphrey, Humphrey, G., & Goodale, M. ( 1999). Active manual control of object Biology, 9  9,, 1315–18. views facilitate object recognition. Current recognition. Current Biology, D, 42  42,, 335–46. Harnad, S. (1990). The symbol grounding problem. Physica problem.  Physica D, Hatano, G., & Osawa, K. (1983). Digit memory of grand experts in abacus-derived 5, 47–53. mental calculation. Cognition, calculation. Cognition, 5, Hatfield, E., Cacioppo, J., & Rapsom, R. (1992). Primitive emotional contagion. In M. Cl Clar ark k (Ed (Ed.) .),, Re Revi view ew of pe pers rson onal alit ityy an andd so soci cial al ps psych ychol ology ogy:: Vol ol.. 14. Emo Emoti tion on an andd so soci cial al behavior (pp. behavior  (pp. 151–77). Newbury Park, CA: Sage. Hatsopoulos, N., & Warren, W. ( 1991). Resonance tuning in rhyt rhythmic hmic arm movements. Journal ments.  Journal of Motor Behavior, Behavior, 28  28,, 3–14. Heale He aley y, A. (1982). Sho Short rt-t -term erm me memo mory ry fo forr or order der inf infor orma mati tion on.. In G. Bo Bowe werr (E (Ed. d.), ), The  psychology of learning and motivation: Vol. 16 16 (pp.  (pp. 191–238). New York: Academic Press.

 

References

 

301

Hecht, H., Vogt, S., & Prinz, W. (2001). Motor learning enhances perceptual Psychological Resea Research rch,,   65 65,,  judgment: A case for action-perception transfer.   Psychological 3–14.  Being and time. time. New York: Harper & Row. Heidegger,, M. (1962). Heidegger ). Being Hein He ine, e, B. (1989). Adpo Adposit sition ionss in Afri African can lang language uages. s. Lingui Linguistique stique Africa Africaine ine,, 2, 77–127. Heine, B. (1997). ). Cognitive  Cognitive foundations of grammar. grammar. New York: Oxford University Press. Heine, B., Ulrike, C., & Hunnemeyer, F. ( 1991). ). Grammaticalization:  Grammaticalization: A conceptual  framework . Chicago: University of Chicago Press.  Catch-22.. New York: Knopf. Heller, J. (1961). ). Catch-22 Hemingway, E. (1960).   The collected poems of Ernest Hemingway. Hemingway . San Francisco: Pirated Edition. Henderson, W., & Smyth, G. (1948). Phantom limbs. Journal limbs.  Journal of Neurology, Neurosurgery,, Psychiatry, 11, surgery 11, 88–117. Heptulla-Chatterjee, S., Freyd, J., & Shiffrar, M. (1996). Configural processing in the perception of apparent biological motion. Journal motion.  Journal of Experimental Psychology:  Human Perception and Performance, Performance , 22  22,, 916–29. 1994 Hermer, & Spelke, E. (370, ). A  Nature, 370 young L., children. children. Nature, , 57 –9.geometric process for spatial reorientation in Hertenstein, M. (2002). Touch: Its communicative function in infancy.   Human Development,, 45 Development  45,, 70–94. Hewes, G. (1983). The invention of phonemically-based language. In E. de Grolier (Ed.),  Glossogenetics: The origins and evolution of language   (pp.   143–62). Paris: Harwood Publishers. Heywood, C., & Kentridge, R. (2000). Affective blindsight?  Trends in Cognitive Sciences,, 4 Sciences  4,, 125–6. Hohmann, G. (1966). Some effects of spinal cord lesions on experienced emotional feelings. Psychophysiology feelings.  Psychophysiology,, 3  3,, 143–56. Hommel, B. (1995). Stimulus-response compatibility and the Simon effect: Toward Toward  Journal of Experimental Psychology: Human Perception an empiri empirical cal clarific clarification. ation. Journal and Performance, Performance, 21  21,, 764–75.

Hommel Homm el,, B. (1996). S-R com compat patibi ibilit lity y eff effect ectss wit withou houtt res respon ponse se unce uncerta rtaint inty y. Quarterly Journal of Experimental Psychology, 49A, 49A, 546–71. Hommel, B., Musseler, J., Aschersleben, G., & Prinz, Prinz , W. W. (2001). The theory of event coding (TEC): A framew framework ork for percep perception tion and action planning. Behavorial planning. Behavorial and Brain Sciences, Sciences, 24  24,, 849–937. Horn Ho rnst stei ein, n, S. S.,, & Mu Mull llig igan an,, N. (2001). Me Memo mory ry of ac acti tion on ev even ents ts:: Th Thee rol olee of ob obje ject ctss in

114, memory of self- and other-performed tasks. American tasks.  American Journal of Psychology, 114, 199–217. Howes, D. (2003). ). Sensual  Sensual relations: Engaging the senses in culture and social theory. theory. Ann Arbor: University of Michigan Press. Hubbard, T. (1990). Cognitive representation of linear motion: Possible direction and gravity effects in judged displacement. Memory displacement.  Memory & Cognition, Cognition, 18  18,, 299–309. Hubbard, T. (1995). Cognitive representations of motion: Evidence for friction fri ction and gravity analogues. Journal analogues. Journal of Experimental Psychology: Learning, Memory, and Cognition,, 21 nition  21,, 241–54. Hubb Hu bbar ard, d, T. (1996). Rep Repres resenta entatio tional nal mom momentu entum, m, cent centrip ripetal etal for force, ce, and curv curvilin ilinear ear impetus. Journal impetus.  Journal of Experimental Psychology: Learning, Memory, and Cognition, Cognition ,  22  22,, 1049–62.

 

302

 

References

Hubbard, T. (1999). How consequences of physical properties influence mental repres rep resent entati ation: on: The envi enviro ronmen nmental tal inv invaria ariants nts hyp hypoth othesis esis.. In P. Kill Killeen een & W. Utt Uttal al (Eds.), Fechner Day 99: The end of 20th-century psychophysics, Proceedings of the 15th annual ann ual mee meetin tingg of the Inte Interna rnatio tional nal Soc Societ ietyy for Psy Psychop chophys hysics ics (pp. 274–9). Tem empe, pe, AZ AZ:: International Society for Psychophysics. Hubbard, T., T., & Bharacha, Bhar acha, J. ( 1988). Judged displaced in apparent vertical and horPsychophysics, 44  44,, 211–21. izontal motion. Perception motion. Perception & Psychophysics, Humphrey, N. (1974). Vision in a monkey without striate cortex: A case study. Perception,, 3 Perception  3,, 241–55. Hupka, R., Zaleski, Z., Otto, J., Reidl, L., Tarabrina, N. ( 1996). Anger, envy, fear, and jealousy as felt in the body: A five-nation study. Cross-Cultural study.  Cross-Cultural Research, Research, 30  30,, 243–64.  Phenomenological psychology. psychology. The Hague: Martinus Nijhoff. Husserl, E. (1977). ). Phenomenological Husserl, E. (1980). ). Ideas  Ideas pertaining to a pure phenomenology and to a phenomenological  philosophy.. Boston: Kluwer Academic.  philosophy  Cognition in the wild. wild. Cambridge: MIT Press. Hutchins, E. (1995). ). Cognition Hutchinson, W., Davis, K., Lozano, A, Tasker, R., & Dostrovsky, J. ( 1999). Pain2, 403–5. related neurons in the human cingulated cortex. Nature cortex.  Nature Neuroscience, 2, Iacoboni, M., Woods, & Mazziotta, (1998). Brain-behavior relationships: Evidence from practiceR., effects or spatialJ.stimulus-response compatibility compatibility. . Journal of  Neurophysiology,, 76 Neurophysiology  76,, 321–31. Intos-Peterson, M., & Roskos-Ewoldsen, B. (1987). Sensory-perceptual qualities of  images.   Journal of Experimental Psychology: Learning, Memory, & Cognition, Cognition,   15 15,, 188–99. Ito, M. (1993). Movement and thought: Identical control mechanisms by the cere bellum. Trends  bellum.  Trends in Neuroscience, Neuroscience, 16  16,, 448–50. Ito, M. (1999). Imagined movement and response programming. Journal programming.  Journal of Mental Imagery,, 23 Imagery  23,, 71–84. Iverson, J., & Thelen, E. (1999). Hand, mouth, and brain: The dynamic emergence Studies , 11–12, 19–40. of speech and gesture. Journal gesture.  Journal of Consciousness Studies, Ivry, R., & Fiez, J. (2000). Cerebellum contributions to cognition and imagery. In M. Gazzaniga (Ed.), The (Ed.),  The new cognitive neuroscience (pp. neuroscience  (pp.   999–1011). New York: Plenum.  Jackendoff, R. (1987). ). Consciousness  Consciousness and the computational mind. mind . Cambridge, MA: MIT Press.  Jackendoff, R., & Aaron, D. (1991). Review of G. Lakoff & M. Turner, More than cool reason: A filed guide gui de to poetic metaphor. Language, metaphor. Language, 67, 67, 320–38. Quarterly, 32  32,, 127–36.  Jackson, F. F. (1982). Epiphenomenal qualia. Philosophical qualia. Philosophical Quarterly, Philosophy, 83  83,, 291–5.  Jackson, F. F. (1986). What Mary didn’t know. know. Journal of Philosophy,  Jackson, J. (1994). Chronic pain and the tension between the body as subject and experience  (pp.  2  201 01–28). New York: object. In T. Csordas (Ed.),  Embodiment and experience (pp. Cambridge University Press.

 Jacobson, E. (1932). Electrophysiology of mental activities.   American Journal of  Psychology,, 44 Psychology  44,, 677–94.  James, K., Humphrey Humphrey,, K., & Goodale, M. (2001). Manipulating and recognizing virtual objects: Where the action is. Canadian is.  Canadian Journal of Experim Experimental ental Psychology, Psychology,  111 1–20. 55,, 11 55  James, W. W. (1890). ). The  The principles of psychology. psychology. New York: MacMillan.

 

References

 

303

 Psychology: Briefer course. course. Cambr  James, W. (1892). ). Psychology: Cambridge, idge, MA: Harvar Harvard d Universi University ty Press.  James, W. (1895). The knowing of things together.  Psychological Review, 2, 2,   105– 24.  Jarvella, R., & Collas, J. (1974). Memory for the intentions of sentences. Memory sentences.  Memory & Cognition,, 2 Cognition  2,, 185–8.  Jeannerod, M. (1994). The representing brain: Neural correlates of motor intention and imagery. Behaviorial imagery. Behaviorial and Brain Sciences, Sciences, 17  17,, 187–245.  Jeannerod, M. (1995). Mental imagery in the motor cortex. Neuropsychologica cortex.  Neuropsychologica,,   33 33,, 1419–32.  Jeannerod, M. (1999).Toactornottoact:Perspectivesintherepresentationofaction. Quarterly Journal of Experimental Psychology: Human Experimental Psychology, Psychology ,  52  52,, 1–29.  Johansson, G. (1973). Visual perception of biological motion and a model for its analysis. Perception analysis.  Perception and Psychophysics, Psychophysics, 14  14,, 201–11.  The body in the mind. mind. Chicago: University of Chicago Press.  Johnson, M. (1987). ). The  Johnson, M. (1991). Knowing through the body body.. Philosophical Psychology, 4, 4, 3–20.  Johnson, M. (1993). Conceptual metaphor and embodied structures of meaning. Philosophical Psychology, Psychology, 6  6,, 413–22.  Johnson, S. (2000 ). Thinking Cognition, ahead: The case  74,, 33  judgments of prehension. Cognition prehension. , 74 –70for . motor imagery in prospective system .  Juarrero, A. (1999).   Dynamics in action: Intentional behavior as a complex system. Cambridge, MA: MIT Press.  Jusczyk, P. (1995). In Infan fants ts’’ det detect ectio ion n of th thee so sound und pat patte terns rns of wo word rdss in flu fluent ent sp speec eech. h. Cognitive Psychology, 29, 29, 1–23. Kaczmarek, K., & Bach-y-Rita, P. P. (1995). Tactile Tactile displays. In W. W. Barfield & Furness,  349 49–414). New T. (Eds.), Virtual (Eds.),  Virtual environments and advanced interface design  (pp.  3 York: Oxford University Un iversity Press. Kaiser, M., & Proffitt, D. ( 1987). Observers’ sensitivity to dynamic anomalies in  Perception and Psychophysics, Psychophysics, 42  42,, 275–80. collision. Perception collision. Kalnins, I., & Bruner, J. (1973). The coordination of visual observation and instrumental behavior in early infancy infancy.. Perception  Perception,, 2  2,, 307–14. Kandel Kan del,, S. S.,, Orl Orliag iaguet uet,, J. J.-P -P., ., & Viv ivian iani, i, P. (2000). Per Percept ceptual ual ant anticip icipati ation on in handwriting: The role of implicit motor competence. Perception competence. Perception & Psychophysics, 62,, 706–16. 62 Kant Ka nt,, I. I.,, (1787/1927). Im Imma manue nuell Kan Kant’ t’ss Cr Crit itiq ique ue of pur puree re reaso ason n : in co comm mmem emor orat atio ion n of the centenary of its first publication. New Ne w York: York: MacMillan.  Beyond modularity: A developmental perspective on cogniKarmiloff-Smith, A. (1992). ). Beyond tive science. science. Cambridge, MA: MIT Press. Kasc Ka scha hak, k, M. M.,, & Gl Glen enbe berrg, A. (2000). Co Const nstru ruct ctin ing g me meani aning ng:: The ro role le of af affo ford rdanc ances es and gramma grammatical tical constr construction uctionss in sentenc sentencee compr comprehension ehension..  Journal of Memory and Language, Language, 43  43,, 508–29. Kay ay,, P., & Fillmore, C. (1999). Gr Gram amma mati tical cal co cons nstr truct uctio ions ns and lin lingu guis isti ticc  75,, 1–33. generalizations: The “what’s X doing Y?” construction. Language construction. Language,, 75 Keefe, D., & McDaniel, M. (1993). The time course and durability of predictive inferences. Journal inferences.  Journal of Memory and Language, Language, 32  32,, 446–63.

Kee een, n, R. (2003). Re Repr pres esen enta tati tion on of ob obje ject ctss an and d ev even ents ts:: Wh Why y do in infa fant ntss lo look ok so sm smar artt and toddlers look so dumb? Current dumb? Current Directions in Psychological Science, Science , 12  12,, 79–83.

 

304

 

References

Ke Kell lly y, M. M.,, & Frey eyd, d, J. (1987). Exp Explora loratio tions ns of rep repres resenta entatio tional nal mom momentu entum. m. Cognitive Psychology,, 19 Psychology  19,, 369–401. Kels Ke lso, o, J. (1995).   Dynami Dynamicc pat patte tern rns: s: The se self lf-o -org rgan aniz izat atio ion n of br brai ain n an andd be beha havi vior or.. Cambridge, MA: MIT Press. Keltner Kelt ner,, D., D.,Ells Ellswor worth, th,P P., & Edw Edward ards, s, K. (1993). Beyo Beyond nd sim simple ple pes pessim simism ism:: Eff Effect ectss of  sadness sadn ess and ange angerr on soc social ial per percept ception. ion. Journal  Journal of Personality and Social Psychology Psychology,, 64,, 740–52. 64 Kennedy, J., Gabia, P., & Nicholls, A. (1991). Tactile pictures. In M. Heller & W. Schift (Eds.), The (Eds.), The psychology of touch (pp. touch (pp. 263–99). Hillsdale, NJ: Erlbaum. Kermoian, R., & Campos, J. (1988). Locomotor experience: A facilitator of spatial cognitive development. Child development. Child Development, Development, 59  59,, 908–17. Kerr, N. (1983). The role of vision in “visual imagery” image ry” experiments: Evidence from General , 112  112,, 265–77. the congenitally blind. Journal blind. Journal of Experimental Psychology: General, Kimura, D. (1973). The asymmetry of the human brain. Scientific brain.  Scientific American, 228, 228, 70–8. Kints Kin tsch ch,, W. (1988). The ro role le of kn know owled ledge ge in dis disco cour urse se co comp mpre rehen hensi sion: on: A construction-integration model. Psychological model. Psychological Review, Review, 95  95,, 163–82.  Comprehension: A paradigm for cognition. cognition . New York: Cambridge Kintsch, W. (1998). ). Comprehension: University Press. Kirsh, D. (1995). The intelligent use of space. Artificial space. Artificial Intelligence, Intelligence, 73  73,, 31–68. Kirsh, D., &Science, Maglio, P., 513 (1994 ). .On distinguishing epistemic from pragmatic action. Cognitive Science  18, , 18 –49 Klatt, D. (1989). Review of selecte selected d models of speech percep perception. tion. In W. Marseln Marseln- Lexical representations and processes (pp. processes (pp. 169–226). Cambridge, MA: Wilson (Ed.), (Ed .), Lexical MIT Press. Klatzky, R. (1994). On the relation between motor imagery and visual imagery. Behavioral and Brain Sciences, Sciences, 17  17,, 212–13. Klatzky, R., Lederman, S., & Metzger, V. (1985). Identifying objects by touch: An expert system. Perception system. Perception and Psychophysics, Psychophysics, 37  37,, 299–307. Klatzky, R., Loomis, J., Lederman, S., & Wake, H. (1993). Haptic identification identification of  objects and their depictions. Perception depictions.  Perception & Psychophysics, 54, 54, 170–78. Klatzky, R., Pellegrino, J., McCloskey, B., & Doherty, S. ( 1989). Can you squeeze a tomato?: The role of motor representations in semantic sensibility judgments.  Journal of Memory and Language Language,, 28  28,, 56–77. Kleinman, A. (1982). Neurasthenia and depression: A study of somatization and Psychiatry, 6  6,, 117–89. culture in China. Culture, China.  Culture, Medicine, and Psychiatry, Knoblich, G., & Flach, R. (2001). Predicting the effects of action: Interaction of  Science, 12  12,, 467–72. perception and action. Psychological action.  Psychological Science, Knoblich, G., Seigerschmidt, E., Flach, R., & Prinz, W. ( 2002). Authorship effects in th thee pr prod oduct uctio ion n of han handw dwrit riting ing st stro roke kes: s: Ev Evid idenc encee fo forr act actio ion n si simu mulat latio ion n dur durin ing g 55A, 1027–46. action perception. Quarterly perception. Quarterly Journal of Experimental Psychology, 55A, Knuf, L., Aschersleben, G., & Prinz, W. (2001). An analysis of ideomotor action.  Journal of Experimental Psychology: Psycholog y: General, General , 130  130,, 779–98. Kohler, E., Keysers, C., Umilta, M., Fogassi, L., Gallese, V., & Rizzolatti, G. ( 2002). Hearing sounds, understanding actions: Action representation in mirror neurons. Science rons.  Science,, 297  297,, 846–48. Koivisto-Alan Koivis to-Alanko, ko, P. (1998).Mechanismsofsemanticchangeinnounsofcognition:A generalmodel.InJ.Coleman&C.Kay(Eds.), Lexico Lexicology logy,, seman semantics tics and lexico lexicography graphy (pp. 35–54). Amsterdam: Benjamins.

 

References

 

305

Koleck, M., Bruchon-Schweitzer, M., Cousson-Gelie, F., Gillard, J., & Quintard, B. Skills, (2002). The body-image questionnaire: An extension. Perceptual extension. Perceptual and Motor Skills, 94,, 189–96. 94  Understanding of containment in 5.5 month-old infants. infants . Poster preKolstad, V. (1991). ). Understanding sented at the meeting of the Society for Research in Child Development, Seattle: Washington. Kopp, C., & Shaperman, J. (1973). Cognitive development in the absence of object manipulation during dur ing infancy. infancy. Developmental Psychology, 9, 9, 430. Kosl Ko slow owsk ski, i, L. L.,, & Cu Cutt ttin ing, g, J. (1977). Re Reco cogn gniz izin ing g the se sex x of a wa walk lker er fr from om a dy dyna nam mic Psychophysics, 21  21,, 575–80. point-light display display.. Perception & Psychophysics, Kosslyn, Kossly n, S. (1987). Seeing and imagining in the cerebral hemisphere: A computational approach. Psychological approach. Psychological Review, Review, 94  94,, 148–75. Kosslyn, S. (1994). Im Imag agee an and d br brain ain:: The re reso solu luti tion on of th thee im imag agery ery de deba bate te.. Cambridge, MA: MIT Press. Kosslyn, S., Cave, M., Provost, D., & von Gierke, S. (1988). Sequential processes in Psychology, 20  20,, 319–43. image generation. Cognitive generation. Cognitive Psychology, Kosslyn, S., DiGirolamo, G., & Thompson, W. ( 1998). Mental rotation of object versus hands: Neural mechanisms revealed by positron emission tomography. Psychophysiology,, 35 Psychophysiology  35,, 151–61. Kosslyn, S., Thompson, T hompson, W., W., Wraga, M., & Alpert, N. N . (2001). Imagining rotation by Neuroreport,, endogen endo genous ous ver versus sus exo exogenu genuous ous for forces ces:: Dis Distin tinct ct neur neural al mec mechani hanisms sms.. Neuroreport 12,, 2519–25B. 12 . (2001). About ham and eggs – perception and action, ecology, and Kotchoubey,  Ecological Psychology, Psychology, 13  13,, 123–33. neuroscience: A reply to Michaels (2000). ). Ecological Kourt Ko urtzi zi,, Z. Z.,, & Kan Kanwi wishe sherr, N. (2000). Ac Acti tiva vati tion on in hum human an MT MT/M /MST ST by st stat atic ic im imag ages es with implied motion. Journal motion. Journal of Cognitive Neuroscience, Neuroscience, 12  12,, 48–55. Kourtzi, Z., & Shiffrar, M. (1997). One-shot view invariance in a moving world. Psychological Science, Science, 8  8,, 461–6. Kourtzi, Z., & Shiffrar, M. (1999). Dynamic representations of human body move Perception,, 28  28,, 49–62. ment. Perception ment. Kovecses, Z. (2000a). a). Metaphor  Metaphor and emotion: Language, culture and body in human  feeling.. New York: Cambridge University  feeling Un iversity Press. Kovecses, Z. (2000 b). Force and emotion. In In L. Albertazzi (Ed.), Meaning (Ed.), Meaning and cognition (pp. tion  (pp. 145–68). Amsterdam: Benjamins. Krauss, R. (1998). Why do we gesture when we speak? Current speak?  Current Direction in Psychological Science, Science, 7  7,, 54–60. Krauss, R., & Hadar, U. (1999). The role of speech-related arm/hand gestures in word retrieval. In L. Messing & R. Campbell (Eds.),   Gesture, speech, and sign (pp. 93–116). New York: Oxford University Press. Kraut, R., Fussell, S., & Siegel, J. (2003). Visual information as a conversational Interaction, 18  18,, 13–49. resource in collaborative physical tasks. Human-Computer tasks.  Human-Computer Interaction, Kraut, R., & Johnston, R. ( 1979). Social and emotional messages of smiling: An eth etholo ologic gical al appr approac oach. h.   Journ Journal al of Pe Pers rson onal alit ityy an andd So Socia ciall Ps Psyc ychol holog ogyy,   37 37,, 1539–1553. Krist, H., Fieberg, E., & Wilkening, F. (1993). Intuitive physics in action and  judgment: The development of knowledge about projectile motion. motion. Journal  Journal of  Experimental Psychology: Learning, Memory, & Cognition, 19, 19, 952–966. Kugle Kug lerr, P., & Turv urvey ey,, M. (1987). Information, natural law, and the self-assembly of rhythmic movement. movement. Hillsdale, NJ: Erlbaum.

 

 

306

References

Kuhl, P., & Miller, J. ( 1975). Speech perception by the chinchilla: Voiced-voiceless 190, 69–72. distinction in alveolar plosive consonants. Science, consonants. Science, 190, Kuhl Ku hl,, P., & Me Melt ltzo zoff ff,, A. (1987). The bi bimo modal dal per perce cept ptio ion n of spe speech ech in in infan fancy cy.. Science Science,, 218,, 1138–41. 218 Kuzouk Kuz ouka, a, H. H.,, Oy Oyam ama, a, S. S.,, Yam amaza azaki ki,, K., Su Suzuk zuki, i, K. K.,, & Mi Mits tsuis uishi hi,, M. (2000). GestureMan: A mobile robot that embodies a remote instructor’s actions. Proactions.  Proceedings ceedin gs of the CSCW 2000 (pp. 2000 (pp. 155 – 62 62). New York: ACM. Lachs, K., & Pisoni, D. (2004). Specification of cross-modal source information in isolated kinematic displays of speech. Journal speech. Journal of the Acoustical Society Society of America, America, 116,, 507–18. 116 Laeng Lae ng,, B., & Teo eodo dore resc scu, u, DD-S. S. (2002). Ey Eyee sc scan anpat paths hs dur durin ing g vi visua suall im imag agery ery re reena enact ctss those of perception of the same visual scene. Cognitive scene.  Cognitive Science, Science, 26  26,, 207–31.  Women, fire, and dangerous things: What our categories reveal about Lakoff, G. (1987). ). Women, the mind. mind. Chicago: University of Chicago Press. Lako La koff ff,, G. (1990). The in inva varia riance nce hy hypo poth thesi esis: s: Is ab abst strac ractt re reaso asoni ning ng ba based sed on im image age- Cognitive Linguistics, 1, 1, 39–74. schemas? Cognitive schemas? Lakoff, G. (1993). The contemporary theory of metaphor. In A. Ortony (Ed.),  Metaphor and thought thought (pp.  (pp. 202–251). New York: Cambridge University Un iversity Press. by. Chicago: University of  Lakoff, G., & Johnson, M. (1980).   Metaphors we live by. Chicago Press.  Philosophy in the flesh. flesh. New York: Cambridge Lakoff, G., & Johnson, M. (1999). ). Philosophy University Press. 2000

Labrings Lako kofff, G. G., , & Nu Nune nez, z, into R. ( being. )..Whe Where reYork: mathe ma thema mati tics csBooks. comes com es fr from om:: Ho How w the em embo bodie diedd mi mind nd mathematics being New Basic Lambek,M.,&Strathern,A.(Eds.)(1998). Bodies Bodiesand and perso persons: ns: Compar Comparative ative perspe perspectives ctives  from Africa and Melanesia. Melanesia . New York: Cambridge University Press. Lambie, J., & Marcel, A. (2002). Consciousness and the varieties of emotion experience: A theoretical framework. Psychological framework. Psychological Review, Review, 109  109,, 219–59. Landau, J., Libkuman, T., & Wildman, J. (2002). Mental simulation inflates perforCognition, 30  30,, 372–9. mance estimates for physical abilities. Memory abilities. Memory & Cognition, Lang La ng,, P. (1995). The em emot otio ion n pr prob obe: e: St Stru ruct ctur uree of mo moti tiva vati tion on and at atte tent ntio ion. n. American  American Psychologist,, 50 Psychologist  50,, 372–85.  Image, word, and symbol. symbol. Berlin: de Gruyter. Langacker, R. (1991). ). Image, Larsen, R., Kasimatis, M., & Frey, K. (1992). Facilitating the furrowed brow: An unobtrusive test of the facial feedback hypothesis applied to unpleasant affect. Cognition and Emotion, Emotion, 6  6,, 321–38. Lazarus, R. (1991). ). Emotion  Emotion and adaptation. adaptation. New York: Oxford University Press. Leander, K. (2002). Silencing in classroom interaction: Drawing and relating social  Discourse Processes, Processes, 34  34,, 193–235. space. Discourse space. Leder, D. (1990). ). The  The absent body. body. Chicago: University of Chicago Press. Lede Le derm rman an,, S. S.,, & Kl Klat atzk zky y, R. (1990). Ha Hapt ptic ic ex explo plora rati tion on and ob obje ject ct re repr prese esent ntat atio ion. n. In M. Go Good odale ale (Ed (Ed.) .),, Vis isio ion n an andd act actio ion: n: Th Thee co contr ntrol ol of gr gras aspi ping ng (pp. 98–109). Nor Norwoo wood, d, NJ: Ablex. Lederman, S., & Klatzky, R. ( 2003). Feelng surfaces and objects remotely. In R. Nelson (Ed.), The (Ed.),  The somatosensory system: system:   Deciphering the brain’ brain’s own body image (pp. 103–20). New York: C Press.  The emotional brain. brain. London: Weidenfeld & Nicolson. LeDoux, J. (1998). ). The Leslie, A. (1982). The perception of causality in infants. Perception infants.  Perception,, 11  11,, 173–86.

 

References

 

307

Lesl ie,, A. (1988). Th Thee ne nece cess ssit ity y of il illu lusi sion on:: Pe Perrce cept ptio ion n an and d th thou ough ghtt in in in-Leslie language  (pp.   185–210). Oxford: fancy. In L. Weiskrantz (Ed.), Thought (Ed.),  Thought without language (pp. Clarendon.

Leslie, A. (1994). ToMM, ToBY, and agency: Core architecture and domain specimind (pp. 119–48). New ficity. In L. Hirschfeld & S. Gelman (Eds.), Mapping (Eds.),  Mapping the mind (pp. York: Cambridge University Press. Leslie, A., & Keeble, S. (1987). Do six-month-old infants perceive causality? Cognicausality?  Cognition,, 25 tion  25,, 265–85. Levenson, R., Ekman, P., Heider, K., & Friesen, W. (1992). Emotion and autonomic nervous system activity in the Minangkabau of West Sumatra. Journal Sumatra. Journal of Personality & Social Psychology, 62, 62, 972–88. Levin, D., & Simons, D. (1997). Failure to detect changes to attended objects in 4, 501–506. motion pictures. Psychonomic pictures. Psychonomic Bulletin & Review, 4, Levins, J., & Lewontin, R. (1985). ). The  The dialectical biologist. biologist. Cambridge, MA: Harvard University Press. Levinson, S. (1994). Vision, shape, and linguistic description: Tzeltal body-part terminology and object description. Linguistics description.  Linguistics,, 32  32,, 791–855. Lew,, A. Lew A.,, & But Butte terw rwor orth th,, G. (1997). The dev develop elopmen mentt of hand hand-mo -mouth uth coo coordi rdinat nation ion in Behaviorr 2- to 5-mo -month nth-ol -old d infa infants nts:: Sim Similar ilariti ities es wit with h rea reachin ching g and gra graspin sping. g. Infant Behavio and Development, Development, 20  20,, 59–69. Lewko Lew kowi wicz cz,, D. (1996). Per Percept ception ion of audi auditor tory y -vi -visual sual tem tempor poral al syn synchr chrony ony in hum human an  Journal of Experimental Psychology: Human Perception and Performance, Performance, 22  22,, infants. Journal infants. 1094–106. 1992

Leyton, M.ar (ren, ). Symmetry, ).  Symmetry, causality, mind .ficie Cambridge, MA: Press. Li,, L. Li L.,, & W arre n, W . (2002). Re Reti tina nall flo flow w mind. is suf suffic ient nt for st stee eerin ring g MIT during dur ing obser ob serva vati tion on.. Psychological Science, Science, 13  13,, 485–97. Libby, L. (2003). Imagining perspective and source monitoring in imagination inflation. Memory flation.  Memory & Cognition, Cognition, 31  31,, 1072–81. Liberman, A. (1970). The grammars of speech and language. Cognitive language.  Cognitive Psychology, Psychology, 1, 301–23. Liberman, A. (1991). Speech: A special code. In A. Liberman (Ed.), Learning, (Ed.),  Learning, development and conceptual change (pp. change (pp. 121–145). Cambridge, MA: MIT Press. Liberman, A., Cooper Cooper,, F., Shankweiler, Shankweiler, D., & Studdert-Kennedy Studdert-Ke nnedy,, M. (1967). Perception of the speech code. Psychological code.  Psychological Review, Review, 74  74,, 431–61. Liberman, A., & Mattingly, Mattingly, I. ( 1985). The motor theory of speech perception revis Cognition,, 21  21,, 1–36. ited. Cognition ited. Libe Li berm rman, an, A., & Ma Matt ttin ingl gly y, I. (1989). A sp speci ecial aliza izati tion on fo forr sp speec eech h per percep cepti tion on.. Science Science,, 243,, 489–94. 243 Liberman, A., & Whalen, D. (2000). On the relation relation of speech to language. Trends language. Trends in Cognitive Science, Science, 4  4,, 187–96. Libet, B. (1985). Unconscious cerebral initiative and the role of conscious will in voluntary action. Behavioral action. Behavioral and Brain Sciences, Sciences, 8  8,, 529–66. Liljedahl, P. (2001). Embodied experience of velocity and acceleration: A narrative.  Journal of Mathematical Behavior Behavior,, 20  20,, 439–45. Lloyd, S., Sinha, C., & Freeman, N. (1981). Spatial references systems and the canonicality canonic ality effect in infant search. Journal search. Journal of Experimental Child Psychology, Psychology ,  3  322, 1–10. Logie, R. (1995). ). Visuo-spatial  Visuo-spatial working memory. memory. Hillsdale, NJ: Erlbaum.

 

308

 

References

Logi gie, e, R. R.,, & Ma Marc rche hett tti, i, C. (1991). Visuo isuo-sp -spatia atiall wor workin king g mem memory ory:: Visua isual, l, spat spatial, ial, or Lo cent ce ntra rall ex exec ecut utiv ive? e? In R. Lo Logi giee & M. De Deni niss (E (Eds ds.) .),, Mental images in human cognition (pp. 105–15). Oxford: North-Holland. Logie, R., & Pearson, D. (1997). The inner eye and the inner scribe of visuo-spatial workin wor king g mem memory ory:: Evi Evidenc dencee for deve develop lopmen mental tal frac fractio tionat nation ion.. European Journal of  Cognitive Psychology, Psychology, 9  9,, 241–57. Lori Lo ring ng,, D. D.,, Me Mead ador or,, K. K.,, Al Alli liso son, n, J. J.,, & Wri righ ght, t, J. (2000). Rel Relati ationsh onship ip bet betwee ween n mot motion ion

and linguistic activation using fMRI. Neurology fMRI.  Neurology,, 54  54,, 981–3. Low, S. (1994). Embodied metaphors: Nerves as lived experience. In T. Csordas  Embodiment and experience (pp. experience (pp. 139–62). New York: Cambridge University (Ed.), Embodiment (Ed.), U niversity Press. Luff, P., P., Heath, C., Kuzuoka, Hi., Hindmarsh, Hindmar sh, J., Yamazaki, K., & Oyama, S. (2003). Fractured ecologies: Creating environments for collaboration. Human-Computer collaboration.  Human-Computer Interaction, 18, 18, 51–84. Maalej, Z. (2004). Figurative language in anger expressions in Tunisian Arabic: An Symbol, 19  19,, 51–75. extended view of embodiment. Metaphor embodiment.  Metaphor and Symbol, Mack, A., & Rock, I. (1998). ). Inattentional  Inattentional blindness. blindness. Cambridge, MA: MIT Press. MacNeilage, P. (1975). Preliminaries to the study of single motor unit activity in Phonetics, 1  1,, 55–71. speech musculature. Journal musculature. Journal of Phonetics, MacWhinney, B. (1998). The emergence of language from embodiment. In B. emer erge genc ncee of la langu nguage age   (pp.   213–56). Ma MacWhinney MacWhi nney (Ed.),   The em Mahw hwah, ah, NJ: Erlbaum. Mana Ma nast ster er,, G. G.,, Cl Clel elan and, d, C. C.,, & Br Broo ooks ks,, J. (1978). Em Emot otio ions ns as mo move veme ment nt in rel elat atio ion n to  Journal of Individual Psychology, Psychology, 34  34,, 244–53. others. Journal others. 1975 Mandler, Mandler ). Mind emotion New York: of Wiley. Mandler,, G. G.((1984). ). Mind   Mindand andemotion. body: .Psychology emotion and stress. stress . New York:

Norton. Mandler,, J. (1992). How to build a baby – 2. Psychological Review, Mandler Review, 99  99,, 587–604. Mand Ma ndle lerr, J. (1998). Ba Babi bies es th thin ink k be befo forre th they ey sp spea eak. k. Human  Human Development, Development, 41 41,, 116–26.  The foundations of mind: Origins of conceptual thoughts. thoughts . New York: Mandler,, J. (2004). Mandler ). The Oxford University Press. Mandl Ma ndler er,, J., Bau Bauer er,, P., & Mc McDo Dono nough ugh,, L. (1991). Se Sepa parat ratin ing g th thee sh sheep eep fr from om th thee go goat ats: s: Psychology, 23  23,, 263–98. Differentiating global categories. Cognitive categories. Cognitive Psychology, Manusov,, V., Manusov V., & Rodriguez, M. ( 1989). Intentionally based nonverbal messages: A Behavior, 13  13,, 15–24. perceiver’s perspective. Journal perspective. Journal of Nonverbal Behavior, Markman, A., & Deitrich, E. ( 2000). In defense of representation. Cognitive representation. Cognitive Psychology,, 40 ogy  40,, 135–71. Marks, D. (1999). Co Cons nsci cious ousnes ness, s, me ment ntal al im imag agery ery,, and ac acti tion on.. British Journal of Psychology,, 90 chology  90,, 567–85. Marmor Marm or,, G. G.,, & Za Zaba back ck,, L. (1976). Me Ment ntal al ro rota tati tion on by th thee bl blin ind: d: Do Does es me ment ntal al ro rota tati tion on depend on visual imagery? Journal imagery?  Journal of Experimental Psychology: Human Perception and Performance, Performance, 2  2,, 515–21. Marrone, R. (1990).   Body of knowledge: An introduction to body/mind psychology. psychology . Albany,, NY: SUNY Press. Albany Marschark, M. (1994). Gesture and sign. Applied sign.  Applied Psycholinguistics, Psycholinguistics, 15  15,, 209–36. Martin, A., Ungerleider, L., & Haxby, J. ( 2000). Category specificity and the brain: The se senso nsory ry-m -mot otor or mo mode dell of sem semant antic ic re repr prese esent ntat atio ion n of ob obje ject cts. s. In M. Ga Gazza zzanig nigaa (Ed.), The new cogn cogniti itive ve neu neuro rosci scienc ences es (2nd edi editi tion on)) (p (pp. p. 1023–36). Cam Cambri bridge, dge, MA: MIT Press.

 

References

 

309

Martin, A., Wiggs, C., Ungerleider, L., & Haxby, J. (1996). Neural correlates of   379,, 649–52. category-specific knowledge. Nature knowledge. Nature,, 379 Martin, E. (1994). ). Flexible  Flexible bodies: Tracking immunity in American culture from the days of polio to the age of AIDS. AIDS. Boston: Beacon.  Speech perception by ear and by eye: A paradigm for psychological Massaro, Massar o, D. (1987). ). Speech inquiry.. Hillsdale, NJ: Erlbaum. inquiry  Variational semantics in Tibeto-Burman: The organic approach to Matisoff, J. (1978). ). Variational linguistic comparison. comparison. Occasional paper of the Wolfenden Society. Philadelphia: Institute for the Study of Human Issues. Cognition, 32  32,, Matlock, T. (2004). Fictive motion as cognitive simulation. Memory simulation.  Memory & Cognition, 1389–1400. Maturana, H. (1980). Biology of cognition. In H. Maturana & F. Varela (Eds.),

 Autopoiesis and cognition: The Th e realization of the living (pp. living  (pp. 5–58). Boston: Reidel. Maturana,H.(1983).Whatitistosee? ).Whatitistosee? Archivos  Archivos de Biologica y Medicina Experimentales Experimentales,, 16,, 255–69. 16 Mayberry,, R., & Jacques, Mayberry Ja cques, J. ( 2000). Gesture production during stuttered speech: Insigh si ghts ts in into to th thee nat natur uree of ge gest stur ure-s e-spe peech ech in inte tegr grat atio ion. n. In D. Mc McNei Neill (Ed (Ed.) .),, Language and gestures (pp. gestures (pp. 199–214). New York: Cambridge University Press. McCl Mc Clos oske key y, M. M.,, & Ko Kohl hl,, D. (1983). Na Naiv ivee ph phys ysic ics: s: The cur curvi vili linea nearr im impet petus us pri princ ncipl iplee and it itss ro role le in int intera eract ctio ions ns wi with th mo movi ving ng ob obje ject cts. s. Journal  Journal of Experimental Psychology: Learning, Memory, & Cognition, Cognition, 9  9,, 146–56. McDonnell, P. (1988). Developmental Developmental responses to limb deficiencies deficiencies and limb replacement. Canadian placement.  Canadian Journal of Psychology, Psychology , 42  42,, 120–43. McGlone, M., & Harding, J. (1998). Back (or forward) to the future: The role of perspectiv spec tivee in tem tempor poral al lang language uage com compre prehens hension. ion. Journal  Journal of Experimental Psychology:  1211 23

Learning, and Cognition, Cognition , 24  24, – . lips and seeing voices. Nature  264,, McGurk, H.,Memory, & MacDonald, J. ( 1976 )., Hearing voices.  Nature,,  264 746–8. McNeil, D. (1992). ). Hand  Hand and gesture: What gestures reveal about thought. thought. Chicago: University of Chicago Press. Medin, D., Lynch, E., Coley, J., & Atran, S. ( 1997). Catego Categorizatio rization n and reaso reasoning ning among tree experts: Do all roads lead to Rome? Cognitive Rome?  Cognitive Psychology, Psychology, 32  32,, 49–96. Meier,, B., & Robinson, M. (2004). Why the sunny side is up: Associations between Meier affect and vertical position. Psychological position.  Psychological Science, Science, 15  15,, 243–7. Melt Me ltzo zoff ff,, A. (1990). Fo Foun unda dati tion onss fo forr de deve velo lopi ping ng a co conc ncep eptt of se self lf:: Th Thee rol olee of im imit itaationinrelatingselftootherandthevalueofsocialmirroring,socialmodeling,and self practice in infancy. In D. Cicchetti & M. Beeghly (Eds.), The (Eds.),  The self in trans transition ition (pp. 139–64). Chicago: University of Chicago Press. Meltzoff, A. (1995). Understanding the intentions of others: Re-enactment of intended acts by 18-month-old children. Developmental children. Developmental Psychology, Psychology, 31  31,, 838–50. Nature,, Meltzo Melt zoff ff,, A. A.,, & Bo Bort rton on,, R. (1979). In Inte term rmod odal al ma matc tchi hing ng by hu huma man n no noena enate tes. s. Nature 282,, 403–4. 282 Meltzoff, A., & Brooks, R. ( 2001). “Like me” as building block for understanding other minds: Bodily acts, attention, and intention. In B. Malle, L. Moses, & D. Baldwin (Eds.),   Intentions and intentionality: Foundations of social cognition (pp. 171–91). Cambridge, MA: MIT Press. Meltzoff, A., & Moore, M. (1992). Early imitation within a functional framework: The im impo port rtan ance ce of pe pers rson on ide ident ntit ity y, mo movem vement ent,, and dev develo elopm pmen ent. t. Infant Behavior and Development, Development, 15  15,, 479–505.

 

310

 

References

Meltzoff, A., & Moore, M. (1994). Imitation, memory, and the representation of   Infant Behavior and Development, Development, 17  17,, 83–99. persons. Infant persons. Meltzoff, A., & Moore, M. (1997). Explaining facial imitation: A theoretical model. Early Development and Parenting, Parenting, 6  6,, 179–92. Meltzoff, A., & Moore, M. (2000). (a) Imitation of facial and manual gestures by human neonates (b) Resolving the debate about early imitation. In D. Muir & A. Slater (Eds.), Infant (Eds.),  Infant development: The essential readings. Essential readings in development psychology (pp. psychology (pp. 167–81). Malden, MA: Blackwell. Melt Me ltzo zoff ff,, A. A.,, & Mo Moor ore, e, M. (2001). Im Imit itat atio ion n of fac facial ial and ma manu nual al ge gest stur ures es by hum human an neona neo nate tes: s: Re Reso solv lvin ing g th thee de deba bate te ab about out ear early ly im imit itat atio ion. n. In D. Mu Muir ir & A. Sl Slat ater er (E (Ed.) d.),, Infant development: The essential readings (pp. readings (pp. 167–81). Malden, MA: Blackwell. Merleau Mer leau-Po -Ponty nty,, M. (1962). Phenom Phenomenolog enologyy of perce perception ption.. Lo Londo ndon: n: Ro Rout utle ledg dgee & Keg Kegan an Paul. Merzenich, M., Kaas, J., Wall, J., Nelson, R., Sur, M., & Felleman, D. ( 1983). Topographic gra phic reo reorga rganiza nizatio tion n of som somato atosens sensory ory cort cortical ical Are Areaa 3Band 1 in adul adultt mon monkey keyss  8,, 33–56. following restricted deafferentation. Neuroscience deafferentation.  Neuroscience,, 8

Metzinger, T. (Ed.) (2000). ). Neural  Neural correlation of consciousness. Cambridge, consciousness. Cambridge, MA: MIT Press. Michel, G., Camras, L., & Sullivan, J. ( 1992). Infant interest expressions as coordinative motor structures. Infant structures. Infant Behavior and Development, Development , 15  15,, 347–58.  The perception of causality. causality. New York: Basic Books. Michotte, A. (1963). ). The Mill Mi ller er,, J. J.,, & St Stig igle lerr, J. (1991). Me Mean anin ings gs of sk skil ill: l: Ef Effe fect ct of ab abac acus us ex expe pert rtis isee in nu numb mber er representation. Cognition representation.  Cognition & Instruction, Instruction, 8  8,, 29–67. Milner, D., & Dyde, R. (2003). Why do some perceptual illusions affect visually guided action, when others don’t? Trends don’t? Trends in Cognitive Sciences, Sciences, 7  7,, 10–11. Milner, D., & Goodale, M. (1995).  The visual brain in action. action . New York: Oxford University Mitchell, R., &Press. Gallaher, M. ( 2001). Embodying music: Matching music and dance as movement. Music movement. Music Perception, Perception, 19  19,, 65–85. Moller, ¨ R. (1999). Perception through anticipation: A behavior-based approach to visual perception. In A. Riegler, M., Peschl, & A. von Stein (Eds.),  Understanding representation in the cognitive sciences (pp. sciences  (pp.  16  169 9–76). New York: Kluwer Academic. Montessori, M. (1914). ). Dr.  Dr. Montessori’ Montessori’s own handbook . London: Heineman. Montpare, J., Goldstein, S., & Clausen, A. ( 1987). The identification of emotions Behavior, 1  1,, 33–42. from gait information. Journal information.  Journal of Nonverbal Behavior, Morrow, D., Bower, G., & Greenspan, S. ( 1989). Updating situation models durBehavior ,   28 28,, ing narrative comprehension. Journal comprehension.  Journal of Verbal Learning and Verbal Behavior, 292–312. Mull Mu llig igan an,, N. N.,, & Ho Horn rnst stei ein, n, S. (2003). Me Memo mory ry fo forr act actio ions: ns: Se Self lf-pe -perfo rform rmed ed ta task skss and the reenactment effect. Memory effect.  Memory & Cognition, Cognition, 31  31,, 412–21. Munakata, Y., McClelland, J., Johnson, M., & Siegler, R. (1997). Rethinking infant know kn owle ledg dge: e: Tow owar ard d an ad adapt aptiv ivee pr proc oces esss ac acco count unt of su succ cces esss and fa failu ilure re on ob obje ject ct  Psychological Review, Review, 104  104,, 686–713. permanence. Psychological permanence. Murphy, G. (1996). On metaphoric representations. Cognition representations. Cognition,, 60  60,, 173–204. Murphy, G. (2002). ). The  The big book of concept conceptss. Cambridge, MA: MIT Press. Imagery, Murph Mu rphy y, S. (1990). Mo Mode dels ls of im imag ager ery y in sp spor orts ts:: A rev evie iew w. Journal of Mental Imagery, 89,, 216–23. 89

 

References

 

311

Murray y, C. ( 2001). The experience of body boundaries by Siamese twins. New twins.  New Ideas Murra in Psychology, Psychology, 19  19,, 117–30. Murray, I., & Trevarthen, C. (1986). The infant’s role in mother-infant communica Journal of Child Language, 13, 13, 15–29. tion. Journal tion. Murray,, J., Klin, C., & Myers, J. (1993). Forward inference in narrative text. Journal Murray text.  Journal of Memory and Language, Language, 32  32,, 464–73. Nairne, J., & Walters, V. (1983). Silent mouthing produces modality and suffix-like effects. Journal effects.  Journal of Verbal Learning and Verbal Behavior, 22, 22 , 475–483. Nakamura, R., & Mishkin, M. (1980). Chronic blindness following non-visual corResearch, 188  188,, 572–7. tical lesions. Brain lesions. Brain Research, Narayanan,, S. (1997). Moving right along: A computational model of metaphoric Narayanan reasoni rea soning ng abo about ut eve events nts.. Unpu Unpubli blished shed Ph.D Ph.D.. diss dissert ertati ation, on, Int Interna ernatio tional nal Com Compute puterr Science Institute, University of California, Berkeley Berkele y. Needham, A. (2001). Object recognition recognition and objec objectt segre segregation gation in  4 .5  month-old infants. Journal infants.  Journal of Experimental Child Psychology, Psychology, 78  78,, 3–24. Need Ne edha ham, m, A. A.,, & Ba Bail illa larg rgeo eon, n, R. (1993). In Intu tuit itio ions ns ab abou outt su supp pport ort in 4.5-month-olds. Cognition,, 47 Cognition  47,, 121–48. Needham, A., Barrett, T., & Peterman, K. ( 2002). A pick-me-up for infants’ exploratory plorat ory skills: Early simulated experiences reaching for object objectss using ‘sticky mittens’ enhances young infants’ object exploration skills.   Infant Behavior & Development,, 25 Development  25,, 279–95.

Neisser, U. (1993). The self perceived. In U. Neisser (Ed.), The (Ed.),  The perceived self: Ecological and interpersonal sources of self-knowledge (pp. self-knowledge  (pp.  3 –21). New York: Cambridge University Press. Nelson, K., Skwerer, D., Goldman, S., Henseler, S., Presler, N., & Walkenfeld, F. (2003). Entering a community of minds: An experientialist approach to a theory Development, 46  46,, 24–46. of mind. Human mind. Human Development, 1972

Neruda, ).   The capt captain ain’’s verses. verses. Evanston, IL: Northwestern University Press. P. ( Neum Ne uman ann, n, C. (2001). Is met metapho aphorr univ universa ersal? l? Cro Cross-l ss-ling inguist uistic ic evi evidenc dencee fro from m German and Japanese. Metaphor Japanese. Metaphor and Symbol, Symbol, 16  16,, 123–42. Newcombe, N. (2002). The nativist-empiricist controversy in the context of reScience,  13  13,, cent research in spatial and quantitative development. Psychological development.  Psychological Science, 395–401.  Foundations of understanding. understanding. Amsterdam: Benjamins. Newton, N. (1996). ). Foundations Newton, N. (2000). Conscious emotion in a dynamic system: How I can know how I feel. In R. Ellis & N. Newton (Eds.), Caldrons (Eds.),  Caldrons of consciousness (pp. consciousness (pp. 91–108). Amsterdam: Benjamins. Nicolelis, M., & Fanselow, E. (2002). Thalamocortical optimization of tactile processing according to behavioral state. Nature state.  Nature Neuroscience, 5, 5, 517–523. Nielsen, T. (1963). Volition: Volition: A new n ew experimental approach approach.. Scandanavian Journal of  Psychology, 4, 4, 225–30. Nieuwenhuyse, B., Offenberg, L., & Frijda, N. (1987). Subjective Subjective emotio emotion n and reEmotion, 11  11,, 169–82. ported body experience. Motivation experience. Motivation and Emotion, Noe, A. (2004). ). Action  Action in perception. perception. Cambridge, MA: MIT Press. Noe, A., & O’Reagan, K. (2002). On the brain-basis of visual consciousness: A sensorimotor account. In A. Noe & E. Thompson (Eds.),  Vision and mind (pp. 567–98). Cambridge, MA: MIT Press.

 

312

 

References

Noe, A., & Thompson, E. ( 2004). Are there neural correlates of consciousness?  Journal of Consciousness Studies, 11 11,, 3–28. Nolfi, S., & Floreano, D. (2000). ). Evolutionary  Evolutionary robotics: The biology, intelligence, and technology of self-organizing machines. machines. Cambridge, MA: MIT Press. Nybe Ny berg rg,, L. L.,, Ha Habi bib, b, R. R.,, Mc McIn Into tosh sh,, A. A.,, & Tul ulvi ving ng,, E. (2000). Re Reac acti tiva vati tion on of enc encod odin inggrelate rel ated d bra brain in act activi ivity ty dur during ing mem memory ory ret retriev rieval. al. Pr Proce oceedi edings ngs of the Nati Nationa onall Aca Academy demy of Sciences, Sciences, 97  97,, 11,120–4. Nygaard, L., Sommers, M., Mitchell, S., & Pisoni, D. ( 1994). Speech perception as talker-contingent process. Psychological process. Psychological Science, Science, 5  5,, 42–6. best st la laid id sc sche heme mes: s: The ps psych ychol olog ogyy of em emot otio ion n. New York: Oatley, K. (1992).   The be Cambridge University Press. O’Br O’ Brie ien, n, E. E.,, & Al Albr brec echt ht,, J. (1992). Co Comp mpre rehen hensio sion n st strat rateg egie iess in th thee dev develo elopm pmen entt of a Cognition , mental model. Journal model. Journal of Experimental Psychology: Learning, Memory, & Cognition, 18,, 777–84. 18 Ochs, E., Jacoby, S., & Gonzales, P. ( 1994). Interpretive journeys: How physicists talk and travel through graphic space. Configurations space.  Configurations,, 2  2,, 157–72. OdlingOdli ng-Sme Smee, e, F. (1988). Nic Niche-c he-cons onstru tructi cting ng phen phenoty otypes. pes. In H. Plot Plotkin kin (Ed (Ed.), .), Th Thee ro role le of behavior in evolution (pp. evolution (pp. 73–132). Cambridge, MA: MIT Press. Ojemann, G. (1994). Cortical stimulation and recording in language. In A. Kertesz (Ed.). Localization (Ed.).  Localization and Neuroimaging in Neuropsychology (pp. Neuropsychology  (pp.  3  35 5–55). San Diego, CA: Academic Press. Ojemann, G., & Mateer, C. (1979). Human language cortex: Localization of memory, syntax, and sequential motor-phoneme identification systems. Science systems.  Science,,  205  205,, 1401–3. Olson, E. (2003). Personal identity. In S. Stich & T. Warfield (Eds.),  The Blackwell

 guide to philosophy of mind (pp. mind  (pp. 352–68). New York: Blackwell. O’Re O’ Rega gan, n, K. (1992). So Solv lvin ing g th thee “r “rea eal” l” my myst ster erie iess of vi visu sual al pe perc rcep epti tion on:: Th Thee wo worl rld d as an outside memory. Canadian memory.  Canadian Journal of Psychology, Psychology, 46  46,, 461–88. O’Regan, K., & Noe, A. (2001). A sensori sensorimotor motor account account of vision and visual con Behavioral and Brain Sciences, Sciences, 24  24,, 939–1031. sciousness. Behavioral sciousness. O’Re O’ Rega gan, n, K. K.,, Res esni nick ck,, R., & Cl Clar ark k, J. (1997). Pi Pict ctur uree ch chan ange gess du duri ring ng bl blin inks ks:: No Nott se seeeing in g wh wher eree yo you u lo look ok an and d se seei eing ng wh wher eree yo you u do don’ n’tt lo look ok.. Investi Investigative gative Ophthalm Ophthalmology ology and Visual Science, Science, 38  38,, S707. Oudejams, R., Michaels, C., Bakker, F., F., & Dolne, M. (1996). The relevance of action in perceiving affordances: Perception of the catchableness of fly balls. Journal balls.  Journal of  Experimental Psychology: Human Perception and Performance, Performance, 22  22,, 879–91. Paillard, J. (1987). Cognitive versus sensorimotor encoding of spatial information. processes and spatial orien orientation tation in In P. Ellen & C. Thinus-Blanc (Eds.), Cognitive (Eds.),  Cognitive processes animals and man (pp. man (pp. 35–54). Dordrecht, Netherlands: Martinus Nijhoff. Paivio, A. (1986). ). Mental  Mental representations: A dual-coding approach. approach. Oxford: Oxford University Press. Pandya, V. (1993). ). Above  Above the forest: A study of Andamanese ethnoemology, cosmology, and the power of ritual ritual.. New York: Oxford University Press.  Affective neuroscience. neuroscience. New York: Oxford University Press. Panksepp, J. (1998). ). Affective Papez, J. (1937). A proposed mechanism of emotion.   Archives of Neurology and Psychiatry,, 38 Psychiatry  38,, 725–43.  Ideas and realities of emotion. emotion. London: Routledge. Parkinson, B. (1995). ). Ideas Parsons, L. (1987a.) Imagined spatial transformations of one’s body. Journal body.  Journal of Ex perimental Psychology: Psychology : General, General , 116  116,, 172–91.

 

References

 

313

Parsons, L. (1987 b). Imagined spatial transformations of one’s hands and feet. Cognitive Psychology, Psychology, 19  19,, 178–241. Parsons, L. (1994). Temporal Temporal and kinematic properties of motor behavior reflected in men mental tally ly sim simulat ulated ed act action ion.. Journal of Experimental Psychology: Human Perception and Performance, Performance, 20  20,, 709–30. Pars Pa rson ons, s, L. L.,, & Fo Fox, x, P. (1998). Th Thee ne neur ural al ba basi siss of im impl plic icit it mo move veme ment ntss us used ed in re reco cogg15, 583–615. nising hand shape. Cognitive shape. Cognitive Neuropsychology, 15, Parsons, L., Fox, P., Downs, J., Glass, T., Hirsch, T., Martin, C., Jerabek, P., & Lancaster, J. (1995). Use of implicit motor imagery for visual shape discrimi 375,, 54–58. nation as revealed by PET. Nature PET. Nature,, 375 Pavlenko, A. (2002). Emotions and the body in Russian and English. Pragmatics English.  Pragmatics & Cognition,, 10 Cognition  10,, 207–41. Pazzani, M. (1997). Influence of prior knowledge on concept acquisition: Experimental and computational results. Journal results.  Journal of Experi Experimental mental Psychology: Learning,  Memory, and Cognition Cognition,, 17  17,, 416–32. Pech Pe cher er,, D. D.,, Ze Zeel elen enbe berg rg,, R. R.,, & Ba Bars rsal alou ou,, L. (2003). Verif erifying ying dif differ ferent ent mod modalit ality y pro propperties for concepts produces switching costs. Psychological costs.  Psychological Science, Science, 14  14,, 119–24. Pecher, D., Zeelenberg, R., & Raaijmakers, J. (1998). Does pizza prime coin? Perceptual processing in lexical decision and pronunciation. Journal pronunciation.  Journal of Memory and Language,, 38 Language  38,, 407–18. symptoms. New York: SpringerPennybaker, J. (1982).   The psychology of physical symptoms. Verlag. Philippot, P., P., Rime, B. (1997). The perception of bodily sensations during emotion: Bulletin, 28  28,, 175–88. A cross-cultural perspective. Polish perspective. Polish Psychological Bulletin, Phillips, R., Wagner agner,, S., Fell, C., & Lynch, M. ( 1990). Do infants recognize emotion in fac facial ial expr expressi essions ons?? Cat Catego egorica ricall and met metapho aphoric rical al evid evidenc ence. e. Infant Behavior and Development,, 13 Development  13,, 71–84. Phillips, W., Baron-Cohen, S., & Rutter, M. ( 1992). The role of eye contact in goal detection: Evidence from normal infants and children with autism or mental  Development & Psychopathology, Psychopathology, 4  4,, 375–83. handicaps. Development handicaps.

Piaget, J. (1952). ). The  The origins of intelligence in childhood. childhood . New York: International Universities Press.  The construction of reality in the child. child . New York: Basic Books. Piaget, J. (1954). ). The Piaget, J. (1975). ). The  The equilibrium of cognitive structures. structures. Paris: Presses Universitairse de France.  The psychology of the child. child. London: Routledge & Piaget, J., & Inhelder, B. (1969). ). The Kegan Paul. Pine Pi ne,, K. K.,, & Mes esse serr, D. (2003). Th Thee dev develo elopm pment ent of re repr pres esent entat atio ions ns as chi child ldre ren n le learn arn Psychology, 21  21,, 285–301. about balancing. British balancing. British Journal of Developmental Psychology, Pink Pi nker er,, S. S.,, & Ul Ullm lman an,, M. (2002). Th Thee pa past st an and d fu futu ture re of pa past st te tens nse. e. Tren rends ds in Cognit Cognitive ive Sciences, 6, 6, 456–63. Plaut, D. (1995). Double-dissociation without modularity: Evidence from connectionist neuropsychology neuropsychology.. Journal of Clinical and Experimental Neuropsychology, Neuropsychology, 17  17,, 291–321. Pollio, H., Henley, T., & Thompson, C. (1997). ). The  The phenomenology of everyday life. life . New York: York: Cambridge University Un iversity Press.  Mind as motion: Explorations Explorations in the dynamics Port, R., & van Gelder, T (Eds.) (1995). ). Mind of cognition. cognition. Cambridge, MA: MIT Press. Prem Pr emac ack, k, D. (1990). The inf infant ant’s ’s th theo eory ry of se selflf-pr prop opel elled led ob obje ject cts. s. Cognition Cognition,, 36 36,, 1–16.

 

314

 

References

Presson, C., & Montello, D. (1994). Updating of rotational and translational body  23,, 1447–55. movements: Coordinate structures of perceptual spaces. Perception spaces.  Perception,, 23 Pres Pr estton on,, S. S.,, & de Waa aal, l, F. (2002). Em Empat pathy: hy: It Itss ul ulti tima mate te and pr prox oxim imat atee ba bases ses.. Behavioral and Brain Sciences, Sciences, 25  25,, 1–25.  Brain and perception: Holonomy and studies in figural processing. processing . Pribram, K. (1991). ). Brain Hillsdale, NJ: Erlbaum. Prinz, W. (1997). Perception and action planning. European planning.  European Journal of Cognitive Psychology,, 9 chology  9,, 129. Prinz, J. (2002). ). Furnishing  Furnishing the mind: Concepts and their perceptual basis. basis. Cambridge, MA: MIT Press. Prin Pr inz, z, J. J.,, & Ba Bars rsal alou ou,, L. (2000). St Steer eerin ing g a co cour urse se fo forr em embo bodi died ed re repr prese esent ntat atio ions ns.. In E. Deit De itri rich ch & A, Ma Mark rkma man n (E (Eds ds.), .), Cognitive dynamics: Conceptual and representational changes in humans and machines (pp. machines (pp. 51–78). Mahwah, NJ: Erlbaum. Prof Pr offit fitt, t, D. D.,, Cr Cree eem, m, S. S.,, & Zo Zosh sh,, W. (2001). Se Seein eing g mo moun unta tains ins in mo mole le hil hills ls:: Geographical-slant perception. Psychological perception. Psychological Science, Science, 12  12,, 418–23. Behavo avoria riall and Bra Brain in Sci Scienc ences es,, Pulvermueller Pulverm ueller,, F. (1999). Wor ords ds in th thee br brain ain’s ’s la langu nguag age. e. Beh 22,, 253–336. 22 Putn Pu tnam am,, H. (1975). ). Mind,  Mind, language, and reality: Philosophical papers, Vol. Vol. 2. 2. Ne New w Yor ork: k: Cambridge University Press. Quinn, J. (1994). Toward a clarification of spatial processing.  Quarterly Journal of  Experimental Psychology, 47A, 47A, 465–80. Quinn, N. (1991). The cultural basis of metaphor. In J. Fernandez (Ed.),  Beyond metaphor: The theory of tropes in anthropology (pp. anthropology (pp. 56–93). Stanford, CA: Stanford University Press. Radcliff-Brown, A. (1964). ). The  The Andaman islanders. islanders. New York: Free Press. Ramachandran, V., & Blakeslee, S. ( 1998). ). Phantoms  Phantoms in the brain. brain. London: Fourth Estate. Ramachandran, V., & Hirstein, W. (1997). Three laws of qualia: What neurology tells us about the biological functions of consciousness. Journal consciousness.  Journal of Consciousness Studies,, 4 Studies  4,, 429–57. Redd Re ddin ing, g, G. G.,, & Wal alla lace ce,, B. (1997). ). Adaptive  Adaptive spatial alignment alignment.. Ma Mahw hwah, ah, NJ: Er Erlb lbau aum. m. Reed, C., & Farah, M. ( 1995). The psychological reality of the body schema: A test with normal partic participants. ipants. Journal  Journal of Experimental Psychology: Human Perception and Performance, Performance, 21  21,, 334–43.

human n sem semant antic ic pote potenti ntial al.. Ch Regier,, T. (1996). The huma Regier Chica icago go:: Uni Unive vers rsit ity y of Ch Chica icago go Pr Press ess.. Reisberg, D., Rappaport, I., & O’Shaughnessy, M. ( 1984). Limits of working memory: The digit span.  Journal of Experimental Psychology: Learning, Memory and Cognition, Cognition, 10  10,, 203–21. Reiser, J., Doxey, P., McCarrell, N., & Brooks, P. (1982). Wayfinding and toddlers’ Psychology,  18  18,, use of information from an aerial view of a maze. Developmental maze.  Developmental Psychology, 714–20. Reis Re iser er,, J. J.,, Lo Lock ckma man, n, J. J.,, & Pi Pick ck,, H. (1980). Th Thee rol olee of vi visu sual al ex expe peri rien ence ce in kn know owle ledg dgee Psychophysics, 28  28,, 185–90. of spatial layout. Perception layout. Perception and Psychophysics, Reiser, J., & Rider, E. (1991). Young children’s spatial orientation with respect to multiple Psychology,   27 27,, 97–107. targets when walking without vision.  Developmental Psychology, Remez, R., Fellowes, J., & Rubin, P. (1997). Talker Talker identification based on phonetic information. Journal information.  Journal of Experimental Psychology: Human Perception & Performance, 23,, 651–661. 23

 

References

 

315

Repp, B. (1998). Musical motion in perception and performance. In D. Rosenbaum & C. Collyer (Eds.),  Timing of behavior: Neural, psychological, and computational  perspectives (pp.  perspectives  (pp. 125–41). Cambridge, MA: MIT Press. Resnick, A., O’Regan, K., & Clark, J. ( 1997). To see or not to see: The need for attention to perceive changes in scenes. Psychological scenes.  Psychological Science, Science, 8  8,, 368–73. Rich Ri char ardso dson, n, D. D.,, & Sp Spiv ivey ey,, M. (2000). Rep Repres resent entati ation, on, spac space, e, and Hol Hollyw lywood ood  76,, 269–75. Squares: Looking at things that tha t aren’t there anymore. Cognition anymore. Cognition,, 76 Richardson, D., Spivey, M., Barsalou, L., & McRae, K. ( 2003). Spatial representations activated during realreal-time time comprehension comprehension of verbs. verbs. Cognitive  Cognitive Science, Science,   27 27,, 767–80. Rieser, J., & Rider, E. (1991). Young children’s spatial orientation with respect to multiple targets while walking without vision. Developmental vision.  Developmental Psychology, Psychology, 27  27,, 97– 107. Rime Ri me,, B., Phi Phili lippo ppot, t, P., & Ci Cisa samo molo lo,, D. (1990). So Soci cial al sc sche hema mata ta of pe perip ripher heral al ch chang anges es in emotion. Journal emotion. Journal of Personality and Social Psychology, Psychology, 59  59,, 38–49. Sciences, 17  17,, Rizzolatti, G. (1994). Nonconscious motor images. Behavioral images.  Behavioral and Brain Sciences, 220. Tren rends ds in Neur Neuroscien oscience ce,, Rizzolatti,G.,&Arbib,M.( 1998).Languagewithinourgrasp. ).Languagewithinourgrasp.T 21,, 188–194. 21 Rizz Ri zzol olat atti ti,, G. G.,, Fo Foga gass ssi, i, L. L.,, & Ga Gall lles ese, e, V. (1997). Pa Pari riet etal al co cort rtex ex:: Fr From om si sigh ghtt to ac acti tion on.. Current Opinion in Neurobiology, 7, 7, 562–67. Rizzolatti, G., Riggio, L., & Sheliga, B. (1994). Space and selective attention. In C. Umilta & M. Moscovitch (Eds.), Attention (Eds.),  Attention and performance 15: Conscious and nonconscious information processing (pp. processing (pp. 232–65). Cambridge, MA: MIT Press. Roch Ro chat at,, P. (1989). Ob Obje ject ct ma mani nipul pulat atio ion n and exp explo lorat ratio ion n in 2- to 5-month -month-old -old infants infants.. Developmental Psychology, Psychology, 28  28,, 871–84.  The infant’ infant’s world. world. Cambridge, MA: Harvard University Press. Rochat, P. (2001). ). The Rosch, E. (1975). Cognitive reference points. Cognitive points. Cognitive Psychology, Psychology, 7  7,, 532–57. Rosch, E. (1999). Reclaiming concepts. Journal concepts. Journal of Consciousness Studies, Studies , 6  6,, 61–77. Rosch,E.,&Mervis,C.(1975). Fam Family ily res resemb emblanc lances: es: Stu Studie diess in the int interna ernall str struct ucture ure of categories. Cognitive categories. Cognitive Psychology, Psychology, 7  7,, 573–605. Rosc Ro sch, h, E. E.,, Me Merv rvis is,, C. C.,, Gr Gray ay,, W., Jo John hnso son, n, M. M.,, & Bo Boye yess-Br Brae aem, m, P. (1976). Bas Basic ic ob obje ject ctss Psychology, 8  8,, 382–439. in natural categories. Cognitive categories.  Cognitive Psychology, Roseblad, B., & von Hofsten, C. (1994). Repetitive goal-directed arm movements in children with development coordination: Role of visual information. Adapted information.  Adapted Physical Activity, Activity, 11  11,, 190–202. Rosenbaum, D. (1991). ). Human  Human motor control. control. New York: Academic Press. Rosler,, F., Rosler F., Heil, M., & Hennighausen, E. (1995). Distinct cortical activation patterns during long-term memory retrieval of verbal, spatial, and color information.

 Journal of Cognitive Neuroscience, Neuroscience, 7  7,, 51–65. Cognitive Psychol Psychology ogy,,   22 22,, Ross, B. (1990). Remind Reminding-bas ing-based ed catego category ry learning learning..   Cognitive 460–92. Ross, B. (1999). Postclassification category use: The effects of learning to use categories after learning to classify.   Journal of Experimental Psychology: Learning,  Memory, and Cognition Cognition,, 25  25,, 743–57. Ross, B., Perkins, S., & Tenpenny, P. ( 1990). Remindi Reminding-bas ng-based ed catego category ry learning learning.. Cognitive Psychology, Psychology, 22  22,, 460–92. Ross, E., & Mesulam, M-M. (1979). Damaged language functions of the righthemisphere for prosody and emotional feeling. Archives feeling.  Archives of Neurology, Neurology, 36  36,, 144–8.

 

316

 

References

Rossi, si, S., Tecc ecchio, hio, F., Pasq Pasquale ualetti tti,, P., Uliv Ulivell elli, i, M., Pizz Pizzella ella,, V., Rom Romani, ani, G., Pass Passer ero, o, S., Ros Battistini, N., & Rossini, P. P. (2002). Somatosensory processing during movement observation in humans. Clincal humans. Clincal Neurophysiology, Neurophysiology, 113  113,, 16–24. Rotenberg, K., & Sullivan, C. ( 2003). Children’s use of gaze and limb movement Psychology, 164  164,, 175–87. cues to infer deception. Journal deception. Journal of Genetic Psychology, Roth, E., & Shoben, E. (1983). The effect of context on the structure of categories. Cognitive Psychology, Psychology, 15  15,, 346–78. Roth Ro thma man, n, G. (1987). Unde Underst rstandi anding ng or order der of mov moveme ement nt in you youngs ngster terss wit with h cer cerebra ebrall palsy. Perceptual palsy.  Perceptual and Motor Skills, Skills, 65  65,, 391–7. Rovee-Collier,, C., & Hayne, H. (2000). Memory in infancy and early childhood. In Rovee-Collier E. Tulving & F. Craik (Eds.), The (Eds.),  The Oxford handbook of memory (pp. memory  (pp.  267 –82). New York: Oxford University Un iversity Press. Rumelhart, D. (1980). Schemata: The building blocks of cognition. In R. Spiro, B. Bruce, & W. Brewer (Eds.), Theoretical (Eds.), Theoretical issues in reading comprehension (pp. comprehension (pp. 33–58). Hillsdale, NJ: Erlbaum. p rocessing: Volume Volume Rume Ru melh lhar art, t, D. D.,, & Mc McCl Clel ella land nd,, J. (1986). (E (Eds ds.), .), Parallel distributed processing: 1 – Foundations. Foundations. Cambridge, MA: MIT Press. Runson, S., & Frykolm, G. (1981). Visual perception of lifted weights. Journal weights.  Journal of  Experimental Psychology: Human Perception and Performance, Performance, 7  7,, 733–40. Russell, J. (1995). Fac Facia iall ex expr pres essio sions ns of em emot otio ion: n: Wha Whatt li lies es be beyo yond nd mi mini nima mall  Psychological Bulletin, Bulletin, 118  118,, 379–91. universality? Psychological universality? Salway, A., & Logie, R. ( 1995). Visuospatial working memory, movement control, and executive demands. British demands.  British Journal of Psychology, Psychology , 86  86,, 253–69.  Being and nothingness. nothingness. New York: Philosophical Library. Sartre, J.-P. (1956). ). Being Schachter,, S., & Singer Schachter Singer,, J. ( 1962). Cognitive, social, and physiological determinants of emotional state. Psychological state. Psychological Review, Review, 69  69,, 379–99.  Dynamic memory. memory. New York: Cambridge University Press. Schank, R. (1982). ). Dynamic Schank, R., & Abelson, R. (1976). ). Scripts,  Scripts, plans, goals, and understanding. understanding. Hillsdale, NJ: Erlbaum. Scheper-Hughs, N., & Lock, M. (1987). The mindful body: A prolegomenon to future work in medical anthropology anthropology.. Medical Anthropology Quarterly, Quarterly, 1  1,, 6–41. Scherer, K., & Wallbott, H. (1994). Evidence for universality and cultural variaPersonality nality and Social tion of differential emotion response patterning. Journal patterning.  Journal of Perso Psychology,, 66 Psychology  66,, 310–28. Schladt, M. (1997).   Kognitive Strukturen von Korerteilvokabularien in kenianischen Sprachen.. Cologne: Institut fur Afrikanistik, University of Cologne. Sprachen Schreu Sch reuder der,, R., Flo Flore ress D’A D’Arca rcais, is, G., & Gla Glazenb zenbor org, g, G. (1984). Ef Effec fects ts of pe perc rcep eptu tual al and 45, 339–354. conceptual similarity in semantic priming. Psychological priming. Psychological Research, 45, Schwartz, D. (1999). Physical imagery: Kinematic versus dynamical models. Cogmodels.  Cognitive Psychology, Psychology, 38  38,, 433–64. Schwartz, D., & Black, T. (1996). Analog imagery in mental model reasoning: Depictive models. Cognitive models. Cognitive Psychology, Psychology, 30  30,, 154–219. Schwartz, D., & Black, T. (1999). Infer Inferences ences through imagin imagined ed actions actions:: Knowin Knowing g

 Journal of Experimental Psychology: Learning, Memory, and  by simulated doing. doing. Journal Cognition,, 25 Cognition  25,, 116–36. Schwartz, D., & Holton, D. ( 2000). Tool use and the effect of action on the imagination.  Journal of Experimental Psychology, Learning, Memory, and Cognition, ination. Journal Cognition ,   26 26,, 1655–65.

 

References

 

317

Scott, Scot t, C. C.,, Ha Harr rris is,, R. R.,, & Ro Roth the, e, A. (2001). Emb Embodi odied ed cog cognit nition ion thr throug ough h imp improv rovisa isatio tion n Processes, 31  31,, 293–305. improves memory for dramatic monologue. Discourse monologue. Discourse Processes, Scruton, R. (1997). ). The  The aesthetics of music. music. Oxford: Clarendon Press. Segal,S.,&Fusella,S.(1970). In Influ fluenc ences es of im imag aged ed pi pict ctur ures es and so soun unds ds on de dete tect ctio ion n Psychology, 83  83,, 458–64. of visual and auditory signals. Journal signals. Journal of Experimental Psychology, Segrin,C.(1998). Int Interpe erperson rsonal al com commun municat ication ionpat patter terns ns ass associ ociated atedwit with h depr depressi ession on and loneliness. In P. Anderson & L. Guerrero (Eds.), Handbook (Eds.),  Handbook of communication and emotion (pp. emotion (pp. 215–42). San Diego: Academic Press. Seifert,, C., Rober Seifert Robertson, tson, S., & Black, J. ( 1985). Types of inferences generated during  Journal of Memory and Language, Language, 24  24,, 405–22. reading. Journal reading. Shallice, T. (1988). ). From  From neuropsychology to mental structure. structure. New York: Cambridge University Press.  Journal of Consciousness Shan Sh anno non, n, B. (1997). Wh What at is th thee fu func ncti tion onss of co cons nsci ciou ousn snes ess? s? Journal Studies,, 5 Studies  5,, 295–308. Shannon, B. (2002).   The antipodes of the mind: Charting the phenomenology of the  Ayahuasca experience. experience. New York: Oxford University Press. Sharkey, N., & Sharkey, A. ( 1987). What is the point of integration? The loci of  knowledge-based facilitation in sentence processing. Journal processing. Journal of Memory and Lan guage,, 26  guage  26,, 255–76. Sharkey,N.,&Ziemke,T.(1998).Aconsiderationofthebiologicalandpsychological foundations of autonomous robotics. Connection robotics. Connection Science, Science, 10  10,, 361–91. Shaw, R. (2001). Processes, acts, and experiences: Three stances on the problem of  intentionality. Ecological intentionality.  Ecological Psychology, Psychology, 13  13,, 636–51. Shaw, R., & Turvey, M. (1999). Ecological foundations foundations of cognition II: Degre Degrees es of  freedom and conserved quantities in animal-environment systems. In R. Nunez & W. Freeman (Eds.),  Reclaiming cognition (pp. cognition  (pp.   11 111 1–23). Bowling Green, OH: Imprint Academic. Sheb Sh ebils ilske ke,, W. (1977). Vis isuom uomot otor or co coor ordi dinat natio ion n in vi visu sual al dir direct ectio ion n and po posi siti tion on co connstancie sta ncies. s. In W. Eps Epstein tein(Ed. (Ed.), ), Sta Stabil bility ity and con consta stancy ncy in vis visual ual per percep ception tion:: Mec Mechani hanisms sms and processes (pp. processes (pp. 89–112). New York: Wiley. SheetsShee ts-Joh Johnst nstone, one, M. (1998). Con Consci sciousn ousness ess:: A nat natural ural hist history ory.. Journal of Consciousness Studies, Studies, 5  5,, 260–94.  The primacy of movement. movement. Amsterdam: Benjamins. Sheets-Johnstone, M. (1999). ). The Shepard, R. (1984). Ecological constraints on internal representations: Resonant kinemat kin ematics ics of per perceiv ceiving ing,, ima imagin gining ing,, thi thinkin nking, g, and dr dreami eaming. ng.   Psychological Review,, 91 Review  91,, 417–47. Shepard, R., & Metzler, J. (1971). Mental rotation of three-dimensional objects. Science,, 171 Science  171,, 701–3. system . New York: C. Sherrington, C. (1906). ).   The integrative action of the nervous system. Scribner’s Sons. Shiffrar, M., & Pinto, J. (2002). The visual analysis of bodily motion. In W. Prinz and an d B. Ho Homm mmel el (E (Eds ds.) .),, Common mechanisms in perception and action (pp. 381–400). New York: York: Oxford University Press. Shirouzu, H., Miyake, N., & Masukawa, H. ( 2002). Cognitively active externalization for situated reflection. Cognitive reflection. Cognitive Science, Science, 26  26,, 469–501. Shontz, F. (1969).   Perceptual and cognitive aspects of bodily experience. experience . New York: Academic Press.

Shore, B. (1996). ). Culture  Culture in mind. mind. New York: Oxford University Press.

 

318

 

References

Shore, P., & Repp, B. ( 1995). Musical motion and performance: Theoretical and performance  (pp.   55–83). empirical perspectives. In J. Link (Ed.),  The practice of performance (pp. New York: York: Cambridge University Un iversity Press. psychology. Shweder, R. (1991).   Thinking through cultures: Explorations in cultural psychology. Cambridge, MA: Harvard University Press. Simo Si mon, n, J. (1969). Re Reac acti tion onss to towa ward rd th thee so sour urce ce of st stim imula ulati tion on.. Journal of Experimental Psychology,, 81 Psychology  81,, 174–6. Simo Si mon, n, J. J.,, & Ru Rudd ddel ell, l, A. (1967). Au Audit ditory ory SS-R R co comp mpat atib ibil ilit ity: y: Th Thee ef effec fectt of irr irrele eleva vant nt cue on information processing. Journal processing.  Journal of Applied Psychology, Psychology, 51  51,, 433–5. Simons, D., & Chabris, C. ( 1999). Gorillas in our midst: Sustained inattentional  blindness for dynamic events. Perception, events.  Perception, 28, 28, 1059–74. Sinclair, H. (1971). Sensorimotor activity patterns as a condition for the acquisition of syntax. In R. Huxley & E. Ingram (Eds.), Language (Eds.),  Language acquisition: Models and methods (pp. methods  (pp. 121–45). Oxford: Academic Press. Sitskoorn, M., & Smitsman, A. (1991). ). Infants  Infants’’  visual perception of relative size in and containment and support events. events. Paper presented at the Biennial Meeting of the International Society for the Study of Behavioral Development, Minneapolis. Sitskoorn, M., & Smitsman, A. (1995). Infants’ perception of dynamic relations Developmental al Psychol Psychology ogy,, 31 31,,  between objects: Passing through or support?   Development 437–47. Slamecka, N., & Graf, P. (1978). The generation effect: Delineation of a phenomenon.   Journal of Experimental Psychology: Human Learning and Memory, Memory,   4, 592–604. Smee Sm eets ts,, J. J.,, & Br Bren enne nerr, E. (1995). Pe Perrce cept ptio ion n an and d ac acttio ion n ar aree ba base sed d on th thee sa same me vi visu sual al information. Journal information.  Journal of Experimental Psychology: Human Perception and Performance Performance,, 21,, 19–31. 21 Smets, G., Strappers, P., & Overbeeke, L., & van der Mast, C. ( 1995). Designing in virtual reality: Perception-action coupling and affordances. In K. Carr & R. England Eng land (Ed (Eds.), s.), Simulated and virtual realities: Elements of perception (pp. 189–208). Philadelphia: Taylor & Francis. Smith, C. (1981). ). A  A search for structure: Selected essays on science, art, and history. history . Cambridge, MA: MIT Press. Smit Sm ith, h, L. L.,, Th Thel elen en,, E. E.,, Tit itze zerr, R. R.,, & Mc McLi Lin, n, D. (1999). Kno Knowi wing ng in th thee co cont ntex extt of ac acti ting ng:: The task dynamics of the A-not-B error. Psychological error.  Psychological Review, Review, 106  106,, 235–60. Smyt Sm yth, h, M. M.,, & Wal alle lerr, A. (1998). Mo Move veme ment nt im imag agery ery in ro rock ck cl clim imbi bing ng:: Pat Patte tern rnss of interference from visual, spatial and kinesthetic secondary tasks. Applied tasks.  Applied Cognitive Psychology,, 12 Psychology  12,, 145–57. Solomon, K. (1997). The spontaneous use of perceptual representations during conceptual processing. Doctoral dissertation. University of Chicago. Solomon, K., & Barsalou, L. (2001). Representing properties locally. locally. Cognitive Psychology,, 43 chology  43,, 129–69. Spelke, E. (1976). Infants’ intermodal perception of events. Cognitive events. Cognitive Psychology, Psychology, 8  8,, 626–36. Spelke, E. (1988). When perceiving ends and thinking begins: The apprehension of objects in infancy. In Albert Yonas (Ed.),   Perceptual development in infancy (pp. 197–234). Hillsdale, NJ: Erlbaum. Spelke, E. (1990). Origins of visual knowledge. In D. Osherson & S. Kosslyn (Eds.), Visual (Eds.),  Visual cognition and action: An invitation to cognitive science  (pp.  92 –127). Cambridge, MA: MIT Press.

 

References

 

319

Spelke, E. (1991). Physical knowledge in infancy: Reflections on Piaget’s theory. theory. In S. Ca Carrey & R. Ge Gelm lman an (E (Eds ds.) .),, The epigenesis of mind: Essays on biology and cognition (pp. 133–64). Hillsdale, NJ: Erlbaum.  50,, 431–45. Spelke, E. (1994). Initial knowledge: Six suggestions. Cognition suggestions.  Cognition,, 50 Spelke, E. (1998). Nativism, empiricism, empiricism, and the origins of knowledge. Infant knowledge. Infant Behavior & Development, 21, 21, 181–200. Spelke, E., Breinlinger, K., Macomber, S., & Jacobson, K. (1992). Origins of knowledge. Psychological edge.  Psychological Review, Review, 99  99,, 605–32. Spelke, E., & Newport, E. (1998). Nativism, empiricism, empiricism, and the develo development pment of  knowledge. In R. Lerner (Ed.),  (Ed.),   Handbook of child psychology: Vol. 1. Theoretical models of human development ( development (5th Edition) (pp. 275–340). New York: Wiley. Spelke, E., Philip, A., & Woodward, Woodward, A. (1995). Infants’ knowledge of object motion Causall cogniti cognition on (pp. 44–78). andhumanaction.InD.Sperber&D.Premack(Eds.), Causa New York: York: Oxford University Press. Spence, J., Smith, L., & Thelen, E. ( 2001). Tests Tests of a dynamic d ynamic systems account of the A-not-B error: The influence of prior experience in the spatial memory abilities of two-year-olds. Child two-year-olds. Child Development, Development, 72  72,, 1327–46. Sperber, D. (2001). In defense of massive modularity. In E. Dupoux (Ed.),  Lan guage, brain, and cognitive development: Essays in honor of Jacques Mehler (pp. 47–57). Cambridge, MA: MIT Press. Sperr Sp erry y, R. (1939). Act Action ion curr current ent stu study dy in mov movemen ementt coo coordi rdinat nation ion.. Journal of General Psychology,, 20 Psychology  20,, 295–313. Spivey, M., & Geng, J. (2001). Oculomotor mechanism activated activated by imagery and 65, 235–241. memory: Eye movements to absent objects. Psychology objects.  Psychology Research, 65, Stampe,D.(1976). Car Cardina dinall num numeral eral sys system tems. s. Chicago Lingu Linguistic istic Societ Societyy, 12 12,, 594–609. Stanfield, R., & Zwaan, R. ( 2001). The effect of implied orientation derived from verbal context on picture recognition. Psychological recognition. Psychological Science, Science, 12  12,, 153–6. Steels,L.(1994). Cooper Cooperation ationbetwee between n distri distributed buted agents agentsthrou through gh self-or self-organizati ganization. on. In Y. Demazeau & J.-P. Muller (Eds.), Decentralized (Eds.),  Decentralized AI  (pp.  (pp. 175–96). Amsterdam: North-Holland. Steri, A., Spelke, E., & Rameix, E. ( 1993). Modality-specific and amodal aspects of   47,, 251–79. object perception in infancy: The case of active touch. Cognition touch.  Cognition,, 47 Stern, D. (1985). ). The  The interpersonal world of the infant. infant. Cambridge, MA: Harvard University Press. Stevens, J., Fonlupt, P., Shiffrar, M., & Decety, J. (2000). New aspects of motion perception: Selective neural encoding for apparent human movement. Neuroremovement.  Neurore port,, 11  port  11,, 109–15. Stevens, Steven s, K., & Blumstein, S. (1981). The search for invariant acoustic correlates of  phonetic features. In P. Eimas & J. Miller (Eds.), Perspectives (Eds.),  Perspectives on the study of speech (pp. 1–38). Hillsdale, NJ: Erlbaum. Stewart, J. (1983). Perception of animacy. Dissertation Abstracts International, 43 International,  43,, 2376–7. Stiefehagen, R., Yang, J., & Waibel, A. (2002). Modeling focus of attention for meeting index based on multiple cues. IEEE cues.  IEEE Transactions on Neural Networks, Networks, 13  13,, 928–38. Stig St igle lerr, J. (1984). Me Ment ntal al ab abac acus us:: Th Thee ef effe fect ct of ab abac acus us tr trai aini ning ng on Ch Chin ines esee ch chil ildr dren en’s ’s mental calculation. Cognitive calculation. Cognitive Psychology, Psychology, 16  16,, 145–76. Stigler, J., Lee, S., & Stevenson, H. (1986). Digit memory in Chinese and English: Evidence for a temporally limited store. Cognition store.  Cognition,, 23  23,, 1–20.  

320

 

References

Brain,, 115 115,, Stoe St oeri rig, g, P., & Co Cowe wey y, A. (1992). Wavel avelengt ength h dis discrim criminat ination ion in bli blinds ndsight ight.. Brain 425–44. Stoffregen, T., Gordan, K., Sheng, Y.-Y., & Flynn, S. (1997). Perceiving affordance  Journal of Experimental Psychology: Human Perception forr ano fo anoth ther er per perso son’ n’ss act actio ions. ns. Journal and Performance, Performance, 25  25,, 120–36. Stoller Sto ller,, P. (1989).   The ta tast stee of et ethn hnog ogra raphi phicc thi things ngs:: The se sens nses es in an anthr throp opol ology ogy.. Philadelphia: University of Pennsylvania Press. Stolz, T. (1994). ). Sprachdynamik   Sprachdynamik . Bochum: Brockmeyer Brockmeyer.. Strack, F., Martin, L., & Stepper, S. (1998). Inhibiting and facilitating conditions of  the human smile: A nonobtrusive test of the facial feedback hypothesis.  Journal of Personality and Social Psychology, Psychology, 54  54,, 768–77. Stra St raus uss, s, C. C.,, & Qu Quin inn, n, N. (1997). ). A  A cognitive theory of meaning meaning.. New Yor ork: k: Ca Camb mbrid ridge ge University Press. Strathern, A. (1996). ). Body  Body thoughts. thoughts. Ann Arbor: University of Michigan Press. Studdert-Kennedy,, M. (1981). The emergence of phonetic structure. Cognition, Studdert-Kennedy structure.  Cognition, 10, 10, 301–306. Studdert-Kennedy, M. (1983). The phoneme as perceptuomotor stimulus. In A. Allport & D. MacKay (Eds.), Language (Eds.), Language perception and production: Relationships between listening, speaking, reading, and writing (pp. writing  (pp.  67 –84). San Diego: Academic Press.  Plans and situated action: The problem of human-machine commuSuchman, L. (1987). ). Plans nication.. New York: Cambridge University Press. nication Sudno Su dnow w, D. (1978).   Ways Ways of the han hand: d: The or orga gani niza zati tion on of imp impro rovi vise sedd co condu nduct ct.. Cambridge, MA: Harvard University Press. Supr Su pren enan ant, t, A. A.,, Pi Pitt tt,, M. M.,, & Cr Crow owde derr, R. (1993). Aud Audit itor ory y re rece cenc ncy y in im imme medi diat atee me memmory. Quarterly ory.  Quarterly Journal of Experimental Psychology, 46A, 46A , 193–223. Sweetser, E. (1986). Polysemy vs. abstraction: Mutually exclusive or complementary. In D. Feder, M. Niepokuj, V. Nikforidou, & M. Van Clay (Eds.),  Papers from the Twelfth Meeting of the Berkeley Linguistics Society (pp. Society (pp.  528 –38). Berkeley, CA: Berkeley Linguistic Society. Society. Sweet Sw eetser ser,, E. (1990). From etymology to pragmatics: The mind-body metaphor in semantic structure and semantic change. change. New York: Cambridge University Press. Science,  12  12,, Talmy, L. (1988). Force dynamics in language and cognition. Cognitive cognition.  Cognitive Science, 49–100. Talm almy y, L. (1996). Fi Fict ctiv ivee mo moti tion on in la langu nguag agee and co cogn gnit itio ion. n. In P. Bl Bloo oom m & M. Peterson (Eds.),   Language and space   (pp.   211–276). Cambridge, MA: MIT Press.  Toward a cognitive semantics. semantics. Cambridge, MA: MIT Press. Talmy, L. (2000). ). Toward Tanaka, S., & Inui, T. ( 2002). Cortical movement for action imitation of hand/ arm positions versus finger configurations: An fMRI study.  NeuroReport, 13, 13, 1599–1602. Taub, S. (2001). ). Language  Language from the body. body. New York: Cambridge University Press. Thagard, P., & Nerb, J. (2002). Emotional gestalts: Appraisal, change, and the dyReview , 6  6,, 274–82. namics of affect. Personality affect. Personality and Social Psychology Review, Thelen, E. (2000). Grounded in the world: Developmental origins of the embodied mind. Infancy mind.  Infancy,, 1  1,, 3–28. Applications. Thel Th elen en,, E. E.,, & Sm Smit ith, h, L. (1994). Dynamic systems approach to development: Applications. Cambridge, MA: MIT Press.

 

References

 

321

Thel en,, E. E.,, Sc Scho hoen ener er,, G. G.,, Sc Sche heie ierr, C. C.,, & Sm Smit ith, h, L. (2001). Th Thee dy dyna nami mics cs of  Thelen embodiment: A field theory of infant perservative reaching. Behavorial reaching.  Behavorial and Brain Science,, 24 Science  24,, 1–86. Thomas, A., Bulevich, J., & Loftus, E. ( 2003). Exploring the role of repetitive and

Cognition,  31  31,, sensory exploration in the imagination inflation effect. Memory effect.  Memory & Cognition, 630–40. Thomas, N. (1999). Are theories of imagery theories of imagination? An active perception approach to conscious mental content. Cognitive content.  Cognitive Science, Science, 23  23,, 207–45. Thompson, E., & Varela, F. ( 2001). Radical embodiment: Neural dynamics and  Trends in Cognitive Science, Science, 5  5,, 418–25. consciousness. Trends consciousness. Thompson, E., Palacios, A., & Varela, F. (2002). Ways of coloring: Comparative vision as a case study. In A. Noe & E. Thompson (Eds.),   Visi Vision on and min mindd (pp. 351–418). Cambridge, MA: MIT Press. Tilley, C. (1994). ). A  A phenomenology of landscape. landscape. Oxford: Oxford University Press. Tippe ipperr, S. (1985). The nega negativ tivee prim priming ing eff effect ect:: Inhi Inhibit biting ing prim priming ing by igno ignored red obj object ects. s. Quarterly Journal of Experimental Psychology, 37A, 37A, 571–91. Titchner, E. (1909). ). Lectures  Lectures on the experimental psychology of the thought processes. processes . New York: Macmillan. Todd, N. (1999). Motion in music: A neurobiological perspective. Music perspective. Music Perception, 17,, 115–126. 17 Tooby, J., & Cosmides, L. (1995). Mapping the evolved functional organization of  neurosciences (pp. 1185–95). mind and brain. In M. Gazzaniga (Ed.), The (Ed.),  The cognitive neurosciences (pp. Cambridge, MA: MIT Press. Tranel, D., Damasio, H., & Damasio, A. ( 1997). A neural basis for the retrieval of  conceptual knowledge. Neuropsychologia knowledge. Neuropsychologia,, 35  35,, 1319–27. Traugott,E.,&Dasher,R.(2002). Regularityinsemanticchange Regularityinsemanticchange.NewYork:Cambridge .NewYork:Cambridge University Press. Trevarthen, C. (1977). ). The  The interpersonal world of the infant. infant . New York: Basic Books. Tucker, M., & Ellis, R. (1998). On the relations between seen objects and compo Journal of Experimental Psychology: Human Perception nents of potent potential ial action actions. s. Journal and Performance, Performance, 24  24,, 830–46. Turing, A. (1950). Computing machinery and intelligence. Mind intelligence.  Mind,, 59  59,, 433–63. Turvey, M., Solomon, H., & Burton, G. ( 1989). An ecolog ecological ical analysis of knowing  by wielding. Journal wielding. Journal of the Experimental Analysis of Behavior, Behavior , 52  52,, 387–407. Ungerleider, L., & Miskin, M. ( 1982). Two cortical visual systems. In D. Ingle, M.  Analysis of visual behavior (pp. 549–86). Cam Good Go odal ale, e, &. Ma Mansfi nsfield eld (E (Eds ds.), .), Analysis Cambri bridge dge,, MA: MIT Press. Valenti, S., & Costall, A. (1997). Visual perception of lifted weight from kinematic Psychology:   Human and static (photographic) displays.  displays.   Journal of Experimental Psychology: Perception and Performance, Performance, 23  23,, 181–98. Vallee-Tourangeau, F., Anthony, S., & Austin, N. (1998). Strategies for generating multiple instances of common and ad hoc categories. Memory categories.  Memory,, 6  6,, 555–92. Van der Heijden, A., Mussler, J., & Bridgeman, B. ( 1999). On the perception of  position. positi on. In G. Aschersleben, Aschersleben, T. Bachmann, & J. Musseler (Eds.), Cognitive (Eds.),  Cognitive contributions to the perception of spatial and temporal events (pp. events  (pp.  19 –37). Amsterdam: Elsevier. van der Meer, A., van der Weel, R., & Lee, D. (1995). The functional significance of  arm movements in neonates. Science neonates.  Science,, 267  267,, 693–5.

 

322

 

References

van Geert, P. (1991). A dynamic systems model of cognitive and language growth. Psychological Review, Review, 98  98,, 3–53. van va n Ge Geld lder er,, T. (1998). Th Thee dy dynam namica icall hyp hypot othes hesis is in co cogn gnit itiv ivee sc scie ience nce.. Beh Behavo avoria riall and Brain Sciences, Sciences, 21  21,, 615–65. van va n Le Leeu euwe wen, n, L. L.,, Sm Smit itsm sman an,, A. A.,, & va van n Le Leeu euwe wen, n, C. (1994). Aff Afford ordanc ances, es, per percept ceptual ual complexity, and the development of tool use. Journal use.  Journal of Experimental Psychology:  Human Perception and Performance, Performance , 20  20,, 174–91. Van Orden, G., Jansen op de Haar, M., & Bosman, A. (1997). Complex dynamic systems also predict dissociations, but they do not reduce to autonomous com-

Neuropsychology, 14  14,, 131–65. ponents. Cognitive Neuropsychology, ponents. Cognitive Van Orden, G., Pennington, B., & Stone, G. (2001). What do double dissoci dissociations ations  111 1–72. prove? Cognitive prove?  Cognitive Science, Science, 25  25,, 11 van Ro van Rooi oij, j, D. D.,, Bo Bong ngers ers,, R. R.,, & Ha Hasel selag ager er,, W. (2002). A nonnon-rep repres resent entatio ational nal appr approac oach h to imagined action. Cognitive action.  Cognitive Science, Science, 26  26,, 345–75. Varela, F. (2002). Upward and downward causation in the brain: Case studies in the emergence and efficacy of consciousness. In Y. Yasue & M. Jibu (Eds.),   No matter, never mind: Proceedings of toward a science of consciousness (pp. consciousness  (pp.   95–117). Amsterdam: Benjamins. Thee vie view w fr from om wi with thin in.. Imp Var arela ela,, F., & Sh Shear ear,, J. (E (Eds ds.) .) (1999). Th Imprint rint Acad Academi emics: cs: Kluw Kluwer er.. Varela, F., Thompson, E., & Rosch, E. ( 1991). ). The  The embodied mind. mind. Cambridge, MA: MIT Press. Vishton, P., Rea, J., Cutting, J., & Nunez, L. ( 1999). Comparing effects of the horizontal-vertical illusion on grip scaling and judgment: Relative versus absolute,, not perception versus action. Journal solute action.  Journal of Experimental Psychology: Human Perception and Performance, Performance, 25  25,, 1659–72. Viviani, iviani,P P. (2002). Mo Moto torr co comp mpet etenc encee in th thee per perce cept ptio ion n of dy dyna nami micc ev event ents: s: A tu tuto tori rial. al. In W. Prinz & B. Hommel (Eds.),  Common mechanisms in perception and action (pp. 406–42). New York: Oxford University Press. Viv ivia iani, ni, P., Ba Baudud-Bo Bovy vy,, G. G.,, & Re Redo dolfi lfi,, M. (1997). Per Perceiv ceiving ing and tra trackin cking g kine kinesth stheti eticc  Journal of Experimental stimuli: stimul i: Further Furtherevidenc evidencee of motor motor-perc -perception eptioninterac interactions. tions. Journal Psychology: Human Perception and Performance, Performance, 23  23,, 1232–52. Viviani, P., & Stucchi, N. (1992a). Biological movements look uniform: Evidence of  motor-perceptual interactions. Journal interactions. Journal of Experimental Psychology: Human Perception & Performance, 18, 18, 603–23. Viviani, P., & Stucchi, N. (1992 b). Motor-perceptual interactions. In G. Stelmach &  Tutorials in motor behavior II  (pp.  J. Requin (Eds.), (Eds.), Tutorials   (pp.  229 –48). Amsterdam: NorthHolland. Vogt, S. (1995). On the relations between perceiving, imagining and performing Psychology , 86  86,, in the learning of cyclical movement sequences. British sequences.  British Journal of Psychology, 191–216. Wagman, J., & Carello, C. (2001). Affordances and inertial constraints in tool use. Ecological Psychology, Psychology, 13  13,, 173–95. Wag agne nerr, S. S.,, Win inne nerr, E. E.,, Ci Cicc cche hett tti, i, D. D.,, & Ga Garrdn dner er,, H. (1981). Met Metapho aphorica ricall mapp mapping ingss Development, 52  52,, 728–31. in human infants. Child infants.  Child Development, Walk alker er,, A. (1982). In Inte term rmod odal al pe perc rcept eptio ion n of exp expre ress ssiv ivee be beha havi vior or by hum human an in infan fants ts..  Journal of Experimental Child Psychology Psychology,, 33  33,, 514–35. Europe opean an Jou Journa rnall of Soc Social ial Psy Psychol chol-Wall allbo bott tt,, H. (1998). Bo Bodil dily y ex expr pres essi sion on of em emot otio ion. n. Eur ogy,, 28 ogy  28,, 879–96.

 

References

 

323

Wang ang,, S. S.-H -H., ., Ba Baill illar arge geon on,, R. R.,, & Br Brue ueck ckner ner,, L. (2004). Youn oung g infa infants nts’’ rea reason soning ing abo about ut hidden hidd en obj object ects: s: Evi Eviden dence ce fr from om vio violati lationon-of-e of-expe xpecta ctatio tion n tas tasks ks wit with h tes testt tri trials als onl only y. Cognition, 93, 93, 167–98.  Journal Warren, W. (1984). Per Percei ceiving ving aff afford ordance ances: s: Visua isuall guid guidanc ancee of sta stair ir clim climbin bing. g. Journal of Experimental Psychology: Human Perception and Performance, Performance, 10  10,, 683–703. Warrington, E., & McCarthy, R. ( 1987). Categories of knowledge: Further fraction 110,, 1273–96. ations and an attempted integration. Brain integration. Brain,, 110 Warr arring ingto ton, n, E. E.,, & Sh Shall allic ice, e, T. (1984). Cat Categor egory-s y-spec pecific ific sem semant antic ic imp impairm airment ent.. Brain Brain,, 107,, 829–54. 107  Movement and mental imagery. imagery. Boston: Houghton Mifflin. Washburn, M. (1916). ). Movement Wegne egnerr, D. (2002). The il illu lusi sion on of con consc scio ious us wi will ll.. Cam Cambri bridge, dge, MA: Har Harvar vard d Univ Universi ersity ty Press. Wegner, D., & Wheatley, T. (1999). Apparent mental causation: Sources of the ex-

perience of will. American will. American Psychologist, 54, 54, 480 492. Weiskrantz, L. (1980). Varieties of residual experience. Quarterly experience.  Quarterly Journal of Experimental Psychology, Psychology, 32  32,, 365–86. Weiskrantz, L., Warrington, E., Sanders, M., & Marshall, J. (1974). Visual capacBrain,   97 97,, ity in the hemianopic field following restricted occipital ablation.   Brain, 709–29. Welch, R. (1978). ). Perceptual  Perceptual modification: Adapting to altered sensory environments. environments . New York: York: Academic Press. Werner, A. (1904). Note on the terms used for “right hand” and “lefthan “lefthand” d” in the Bantu languages. Journal languages. Journal of the African Society, Society, 13  13,, 112–16. Wexler, M., Kosslyn, S., & Berthoz, A. (1998). Motor processes in mental rotation. Cognition,, 68 Cognition  68,, 77–94. Wheel Wh eeler er,, M. M.,, Pet Peters erson on,, S. S.,, & Buc Buckn kner er,, R. (2000). Mem Memory ory’s ’s echo echo:: Vivi ivid d rem rememb emberin ering g reactivates react ivates sensor sensory-speci y-specific fic cortex cortex.. Proceedings of the National Academy of Sciences, Sciences, 97,, 11,125–9. 97 Whit Wh ite, e, P. (1999). Tow owar ard d a ca causa usall re reali alist st acc accou ount nt of ca causa usall und unders ersta tand nding ing.. American  Journal of Psychology Psychology,, 112  112,, 605–42. Wier ierzbic zbicka, ka, A. (1999). Emo Emotio tions ns acr across oss lan langua guages ges and cul cultur tures: es: Dive Diversi rsity ty and uni univer versal sals. s. New York: York: Cambridge University Press.  Metaphor in American sign language. language . Washington, DC: Gallaudet Wilcox, P. (2001). ). Metaphor University Press. Wilson, M. (2001). The case for sensorimotor coding in working memory. Psychomemory.  Psychonomic Bulletin and Review, Review, 8  8,, 49–57. Wilson,M.(2002). Six vie views ws of emb embodi odied ed cog cogniti nition. on.Psy Psychon chonomi omicc Bul Bullet letin in and andRev Review iew,, 9, 625–36. Wilson, M., & Emmorey, K. (1997). A visuospatial phonological loop in working memory: Evidence from American Sign Language. Memory Language.  Memory & Cogniti Cognition on,,  25  25,, 313–20. Wilson, M., Iverson, A., & Emmorey, Emmorey, K. (2000). Further investigation of the phonological similarity effect for sign language: Two effects of spatial similarity. Manuscript submitted for publication. Wilson,N.,&Gibbs,R.( 2005).Realandimaginedbodymovementprimesmetaphor comprehension. Manuscript submitted for publication.  Boundaries of the mind: The individual in the fragile sciences – cogniWilson, R. (2004). ). Boundaries tion.. New York: Cambridge University Press. tion

 

324

 

References

Strangers ers to our oursel selves ves:: Dis Discov coveri ering ng the adap adaptive tive unc uncons onsciou ciouss. ilso son, n, T. (2002).   Strang Wil Cambridge, MA: Harvard University Press. Winkler, C. (1994). Rape trauma: Contexts of meaning. in T. Csordas (Ed.), Embod(Ed.), Embodiment and experience (pp. experience (pp. 248–68). New York: Cambridge University Press.  Journal Wohlschlager, S., & Wohlschlager, A. (1998). Mental and manual rotation. rotation. Journal of Experimental Psychology: Human Perception and Performance, Performance, 24  24,, 397–412. Wolff, P. (1999). Space perception and intended action. In G. Aschersleben, T. Bachmann & J. Musseler (Eds.), Cognitive (Eds.),  Cognitive contributions contributions to the perc perception eption of spatia spatiall and temporal events (pp. events (pp. 43–630. Amsterdam: Elsevier E lsevier.. Wol olfff, P., & Le Levi vin, n, J. (1972). Th Thee ro role le of ov over ertt ac acti tivi vity ty in ch chil ildr dren en’s ’s im imag ager ery y pr prod oduc uc-tion. Child tion.  Child Development, Development, 43  43,, 537–47. Wood oodwar ward, d, A. (1999). In Infan fants ts’’ ab abil ilit ities ies to dis disti tingu nguis ish h be betw tween een pur purpo pose seful ful and non non-Development, 22  22,, 145–60. purposeful behaviors. Infant behaviors. Infant Behavior and Development, Wri righ ght, t, T. (2001). Ka Karren in mo moti tion on:: Th Thee ro role le of ph phys ysic ical al en enac actm tmen entt in de deve velo lopi ping ng an understanding of distance, time, and speed. Mathematical speed.  Mathematical Behavior, Behavior, 20  20,, 145–62. Wu, L., & Barsalou, L. (2001). Grounding concepts in perceptual simulations: I. Evidence from property generation. Manuscript submitted for publication. Xu, F., & Carey, S. (1996). Infants’ metaphysics: The case of numerical identity.

Cognitive Psychology, Psychology, 30  30,, 11  111 1–53. Yarbus, A. (1965). Role of eye movements in the visual process. Oxford: Nauka. The con contemp tempora orary ry theo theory ry of met metapho aphor: r: Per Perspe specti ctives ves fr from om Chi Chines nesee. Yu, N. (1999).   The Amsterdam: Benajmins. Yu, N. (2003). Chinese metaphors of thinking. Cognitive thinking. Cognitive Linguistics, Linguistics, 14  14,, 141–66. Psychologique,, 102 102,, Zaba Za bali lia, a, M. (2002). Ac Acti tion on an and d me ment ntal al im imag ager ery y in ch chil ildr dren en.. Anne Psychologique 409–22. Zajo Za jonc nc,, R. R.,, Mu Murph rphy y, S. S.,, In Ingl gleh ehart art,, M. (1989). Feel Feeling ing and faci facial al eff effere erence nce:: Imp Implica licatio tion n Review, 96  96,, 395–416. of the vascular theory of emotion. Psychological emotion.  Psychological Review, Ziemke, T. (1999). Rethinking grounding. In A. Riegler, M. Peschl, & A. von Stein (Eds.), Unders Understanding tandingrepr represen esentation tation in the cogniti cognitive ve scien sciences ces (pp. 177–90). New York ork:: Kluwer Academic. Zimler, J., & Keenan, J. (1983). Imagery in the congenitally blind: How visual are visual images images?? Journal of Experimental Psychology: Learning, Memory, and Cognition Cognition,, 9, 269–82. Zubin, D., & Cho hoii, S. (1984). Orie Orienta ntatio tion n and ges gestal talt: t: Con Concept ceptual ual or organi ganizing zing pri princinciples pl es in th thee le lexi xica cali liza zati tion on of sp spac ace. e. In D. Tes este ten, n, V. Mi Mish sha, a, & J. Dr Drog ogo o (E (Eds ds.) .),, Papers  from the parasession of lexical semantics (pp. semantics (pp. 333–45). Chicago: Chicago Linguistics Society. Zwaan,R.(1996). Pro Proces cessing singnarr narrati ative ve tim timee shi shifts fts.. Journal of Experimental Psychology: Learning, Memory, & Cognition, 22, 22, 1196–1207. Zwaan, R., Magliano, J., & Graesser, A. (1995). Dimensions of situated model con Journal of Experimental Psychology: Learning, struction struc tion in narrati narrative ve compr comprehensio ehension. n. Journal  Memory, & Cognition Cognition,, 21  21,, 386–97. Zwaan, R., Stanfield, R., & Yaley, R. (2002). Language comprehenders mentally Science, 13  13,, 168–71. represent the shapes of objects. Psychological objects. Psychological Science,

 

Index

AA. See AA. See Action  Action Attribute abstract concepts. See concepts. See also concepts also concepts cognitive linguistics and, 119 concreteness of, 99 creation of, 122 cross-modal connections and, 231 image schemas, metaphors and, 96 metaphoric meaning of, 183

affordance, 21 consciousness and, 265 micro-affordances and, 61 of objects, 50 odor perception and, 48 perception and, 43 agency action and, 22–23

source-path-goal schema and, 91–93 source-path-goal understandings of, 90 accommodation,  208 action adaptation and, 62 agency and, 22–23

concept development of, 8 development of, 223 image schemas and, 224 sense of, 22 AI. See AI.  See artificial  artificial intelligence Akaka, Daniel, 109

cognitive functions and, 124 compensation and, 62 as conscious experience, 273 conscious imagery and, 133 consciousness from, 267 from dynamical systems perspective, 75 hallucination experience and, 269 ideomotor, 125 in imagination, memory, reasoning, 157 memory as, 142–144 mental intentions and, 73 objects and, 50 of others, 52 ownership for, 23 perception and, 42–45, 49, 57–64, 77–78

AL. See artificial AL. See  artificial life American Sign Language (ASL), 169 communication communic ation ideas in, 190 conceptual metaphors in, 191 container schema in, 192 double mappings in, 191 embodied metaphors in, 190–194 metaphors in, 193 time in, 192 amputees  body and, 34  body schema schema and, 35 Anlo-Ewe, 38 appearance, 27 artificial intelligence (AI), 69

perceptual illusions on, 59 robotics and, 72 visual cognition and, 59 Action Attribute (AA), 258 active intermodal mapping (AIM), 232 activity, of body, 26 adaptation, perception, action and, 62

artificial life (AL), 69 criticisms of, 73 environment and, 72 robots with, 69–72 ASL. See ASL.  American Sign Language  See American aspect, 196 semantics of, 196

325

 

 

326 assimilation, 208 attention consciousness and, 267 patterns of, 267 autobiographical self, 21 Baker, Lynne, 15  balance schema, schema, 93–94. See also image also image schemas Gulf War and, 108 linguistic action and, 103 Bateson, Gregory Gregory,, 18  behavioral control, control, perception, perception, action action and, 59

 blindness “change,” 66 “inattentional,” 66 mental imagery and, 131  blindsight, 57  body activity of, 26 amputees and, 34 appearance and, 27 awareness of, 25–27  brain and, 5, 45 connection with, 14 corporeality of, 26 culture and, 36–39 disordered, 33–35 emotion and, 250–259

Index schema, 29 self and, 18–25 self expressions and, 27 vitality and, 26 world and, 16–18  body image, 32  body schema schema and, 32  body movements, movements, discourse discourse and, 170–174  body schema, schema, 29 amputees and, 35  body image and, and, 32 environment and, 32 explanation for, 32 self-recognition and, 30–31 as supramodal, 30  brain  body and, 5, 45 cycles of operation and, 270–271 embodiment and, 9 gesture and, 167–168 interaction with, 24 mental images and, 131–132, 134–138 NTL and, 194–197 schizophrenia and, 267 somatasensory somatasenso ry system and, 136 somatotopic map in, 45 Brown, Hank, 110 Cage, Nicholas, 239 Cartesian dualism, 4

empathy and, 35 engagement with, 26 environment and, 22, 27 experiences of, 25–27 first-person perspective of, 15–16 historical conceptions of, 3 ideological views and, 172 ideomotor action of, 35 information systems of, 28 as instrument, 27 intermodal correlation and, 19 interpersonal meaning for, 27 in knowledge, cognition, 6 metaphorical correspondences to, 80 mind and, 3, 4, 275 movement of, 27, 34 nonorganic material and, 16, 18 as object, 4, 14, 25, 27 other’s, 35–36 personhood and, 14 representations of, 28 satisfaction with, 32

categorization, concepts and, 84–86 causality  babies’ understanding understanding of, 220–222 of consciousness, 271 image schemas and, 221 movement and, 54–55 on others, objects, 55 static pattern perception and, 56 child development  balance and, 237 imitation and, 231–234 of mental imagery, 237 theory of mind and, 234–236 Clark, Andy Andy,, 16 cognition  bodily action and, 226 cognitive linguistic evidence to, 118–121 culture and, 13 dynamical systems theory and, 10–11, 281 embodied action and, 12 embodied “representations” and, 281 embodied view of, 157, 279, 281

 

Index as enaction, 17 image schemas and, 142 mind-body, language-body connections and, 9 personhood and, 17 phenomenological phenomen ological action and, 3 cognitive categories  basic level of, 82–83 concept flexibility and, 84 cognitive developmen development, t, 208–238 early embodied action in, 208 embodied perspective of, 238 guidelines for, 216 of infants, 214 Mandler’s theory of, 219–224 multimodal perception and, 231 of physically handicapped, 227–229 Piaget’s contribution to, 208–210 sensorimotorr experience and, 214–216, sensorimoto 237

cognitive functions action and, 124 image schemas and, 138 Cognitive Grammar, 198 cognitive linguistics, 90 cognitive science and, 118–121 conventional convention al metaphors and, 119 embodied mind, language and, 1  11 1 research methods of, 120 cognitive processes  body and, 14 in cognitive psychology, 6 internal, external processes and, 12 cognitive psychology cognitive processes in, 6

 

327 cognitive systems, functional states of,  2 cognitive unconsciou unconscious, s, 40 color perception enactive view of, 44 environment and, 44 common ground for discourse, 172 physical co-presence and, 172 sources for f or,, 172 communication from bodily experience, 207 language and, 158–207 traditional views of, 159 compensation, perception, action and, 62 concepts  basic level categories categories of, 82–83 categorization and, 84–86 combinationss of, 87 combination context and, 81, 86 counterfactual thought and, 90 “effects   =   structures” fallacy fallacy of, 83, 278 embodied action and, 216–217 embodiment of, 100 flexibility of, 84 grammar, spatial concepts and, 104–107 incorrect assumptions of, 82 new view of, 89, 121 numerical systems and, 104–105 objective properties of, 81 perceptual processes and, 86 perceptual symbols and, 89 processing of, 88 productivity and, 87 prototype theory and, 83 prototypes for, 81, 84

developments in, 7 developments embodied action and, 6 cognitive science “effects   =   structures” fallacy of, of, 83,

rules and, 80 as temporary construction constructions, s, 86 traditional views of, 80–81 traditional view’s problems and, 81 conceptual development, 215 conceptual metaphor theory, 96 conceptual domains and, 116 experiential grounding and, 115 metaphor relationships and, 116 questions about, 115–116 source-to-target domain mapping problem and, 115 conceptual metaphors, as dynamic, 121 conceptual processing, 88 congruence, 229 conjoined twins, self-body relationship and, 23–24

278

embodiment and, 3, 276 emergence of, 5 history of, 5 human cognition in, 9 interrelationship interrelations hip of, 276 kinesthetic action and, 3 methodological issue of, 5 mind, body conception in, 3 modularity theory in, 279–280 neural basis for language and, 194–197 neuropsychological neuropsyc hological findings and, 279 problem of, 277

 

328 conscious experience as actions, 273 embodied nature of, 268 non-ordinary types of, 268, 269–270 ordinary types of, 268 consciousness,  262–273 altered states of, 268–270 animate motion and, 12 attention and, 267  bodily feelings and, 267  brain dynamics and, 270–271 causal effects of, 271 “corporeal,”  267 definition of, 263–265 dynamical model of, 272, 273 embodied nature of, 264 emotion and, 239–274 emotion interaction and, 261–262 enactment and, 265–268 evolutionary advantages of, 264 experience of, 265 extension of, 272 functions of, 264–265 goals and, 273 hallucination experience and, 269 human action and, 273 movement and, 265, 266 neural processes and, 263 non-ordinary types of, 268, 269–270 ordinary types of, 268 processes for f or,, 270 self-organization self-organizatio n and, 270–273 sensory, motor processes and, 265 types of, 265 Construction Grammar, 198 constructions,  198 dynamic knowledge and, 199 containment,  37  bodily experience experience and, 222 cognitive development and, 222–223

 

Index counterfactual thought, 90 creativity  bodily sensations sensations and, 124 in embodied thought, 123 culture  bodies and, 36–39 containmentt and, 37 containmen emotional experience and, 257–258 senses and, 37–39 Damasio, Antonio, 21 deferred imitation, 232 Descartes, Rene, 4 Cartesian dualism of, 4 developmental psychology, 8 Dewey, John, 43 discourse  body movements movements and, 170–174 “common ground” and, 172 “dynamic (depiction) model of imagery,” 132

dynamical systems theory, 10–11 action and, 75 aim of, 10 cognitive development and, 226 cognitive performance and, 281 consciousness and, 272, 273 effect of, emotional expression and, 259–261, 273 global stability and, 260 higher-order higher-or der cognitive behavior and, 128–129 implication of, 74 mental imagery and, 128–129, 138 object permanence and, 225 “representations” v., 281 in robotics, 72 ED. See Evaluative ED. See  Evaluative Descriptions Einstein, Albert, 123

culture and, 37 linguistic action and, 104 support and, 223 containment schema in ASL, 192 mathematical concepts and, 113 political ideas and, 110 context, concepts and, 81, 86 conversation. See conversation.  See embodied  embodied conversation

embodied action. See action. See also action also action emotion and, 243 emotions, consciousness and, 240 infant exploration and, 218 in metaphor processin processing, g, 183–184 of newborns, 216–217 in physical reasonin reasoning, g, 216–219 utterance interpretation and, 190 Embodied construction grammar,

core self, 21 corporeal consciousness, 265 corporeality, experiences of, 26

198–199 embodied conversation, conversation, computer and, 173–174

 

Index embodied experience language in, 277 linguistic communication and, 159 linguistic meaning and, 177 metaphorical meaning and, 187

 

329 experiences of, 1–2 importance of, 13, 79 interpretation and, 2 language change and, 160–161 levels of, 39–40

text comprehension and, 206 word meaning and, 174–175 embodied metaphor abstract ideas and, 170  bodily ideas/experiences ideas/experiences and, and, 160  brain and, 118 case study in, 184–187  118 18 connections and, 1 consistency of, 182–183 emotions and, 242–243 in figurative language interpretation interpretation,, 182–183 new view of, 116–118 semantic change by, 161 time expressions and, 187–190

 111 1 mathematical concepts and, 11 mental imagery and, 124–142 metaphorical projection of, 100–101 narrative of, 1 neural, 39 numerical concepts and, 104–105 on-line v. off-line, 157 in ordinary concepts, 79 phenomenological level of, 40 111 1 of political ideas, 107–11 premise of, 9, 275 in psycholinguistics psycholinguistics,, 7 of spatial orientation, 105–107 subjective experience and, 12 “embodiment premise,” 9

x-schemas and, 195 embodied perception, 42–45 perception-action couplings and, 78 embodied psychology, 25 embodied reasoning, 154 embodied schemes, 198 embodied simulation, fictive motion and,

emotion(s),  240–262. See also emotional also emotional experience;; emotional expression experience AA and, 258 affective processes and, 259 “affective space” and, 244, 245–246 anatomical structure and, 250 animate motion and, 12 assumptions of, 247  bodily awareness awareness in, 250  bodily changes changes and, 250–259  body movements, movements, postures postures and, 251, 253–254 consciousness and, 239–274 consciousness interaction and, 261–262 corporeal experience of, 246

190

embodied text understanding, 199–205 adaptive character of, 204 embodied possibilities and, 200 microworlds and, 201 perspectives and, 199–200 protagonist’s perspective and, 201, 202 scripts and, 202 spatial perspectives and, 201 embodied thought, creativity, imagination and, 123 embodiment AL and, 69  brain and, 9 in cognitive psychology, 6 in cognitive science, 3, 276 cognitive unconscious and, 40 of concepts, 100

ED and, 258 embodied action and, 243 embodied metaphors and, 242–243 enactment of, 262 experience of, 240 facial expressions and, 246–250 as felt movements, 243–246 gait and, 256 goals and, 273 hand, arm postures, movements and, 251

conceptual simulations and, 121 containment and, 37 cultural meaning and, 37 cultural processes and, 13 definition of, 1, 10

human action and, 273 integrated view of, 257 intention and, 259 language of, 240–243 meaningful change and, 244

in developmental psychology, 8 emotional experience and, 243

motor mimicry and, 248 muscular affect on, 255

 

330

 

Index

emotion(s) (cont. (cont.)) other’s bodily cues to, 255 palpitable feeling of, 246 related thought and, 255 removedness and, 245 self, world-focus world-focused, ed, 258

experiences of interpersonal meaning, 27 experiences exploratory procedures (EPs), 50 expressions. See expressions.  See facial  facial expressio expressions ns expressions of self, body and, 27

as self-organize self-organized d process, 261 specificity debate of, 250 spontaneous v. intentional, 248, 249 study difficulties with, 247 emotional experience  body’s role in, 256 cultural differences in, 257–258 facial feedback and, 254 physical handicaps and, 252–253 emotional expression coordinative structures for, 260 dynamical view of, 259–261, 273 global stability and, 260 non-correspondences of, 260 empathy

communicative function of, 249 emotions and, 246–250, 253 intentional, 248 motor mimicry and, 248 as natural, automatic, 248 odor and, 249 study difficulties with, 247 Fazio, Peter, 108 first-person perspective, 15–16 force image schema, linguistic action and,

other’s bodies and, 35 simulation theory and, 36 enactment consciousness and, 265–268 definition of, 266 examples of, 266 engagement, experiences of, 26 environment AL and, 72  body and, 22, 27  body schemas schemas and, 32 cognitive categories and, 82 “complementary “complemen tary strategies” to, 152 conscious experience in, 265 embodied reasoning and, 154

 brain activity and, 167–168 conceptualization and, 166 information packaging hypothesis of,

perception and, 43–45, 77 persons and, 16 robots and, 72 sensorimotorr contingency theory and, 66 sensorimoto EPs. See EPs.  exploratory procedures  See exploratory Evaluative Descriptions (ED), 258 experience(s). See experience(s).  See also embodied also embodied experience embodiment and, 1–2 as “persons,” 40 remembering and, 143–144 senses and, 37–39 shaping of, 2 sharing of, 36

facial expressio expressions ns

104

Freeman, Walter, 47 Geertz, Clifford, 18 gesture(s)

166

lexical retrieval hypothesis of, 166 origin of, 168 sign systems and, 169–170 speech and, 165–170 speech perception and, 165 Gibson, James, 21 Glucksberg, Sam, 119 grounding metaphor, 11  111 1  111 1–113 examples of, 11 Gulf War, image schemas and, 108 haptic perception perception,, 54 Hatch, Orrin, 108 Hemingway, Ernest, 99 Hobbes, Thomas, 107 homunculus,  45 Hussein, Saddam, 109 Husserl, Edmund, 28 Hyperspace Analog to Language (HAL), 158

ideomotor action, 125 ideomotor mimicry, 125–126 idioms consistency of, 182–183

experiences of corporeality, 26

embodied knowledge and, 183

experiences of engagement, 26

mental images for, 182

 

Index image schemas, 90. See also specific also specific schemas agency and, 224 as attractors, 115  balance schema schema and, 93–94, 103  bodily experience experience and, 114  bodily perception, perception, movement movement and, 91 causality and, 221 containment as, 104 cross-modal connections and, 231 disabled children and, 229 emergent nature of, 114 in everyday thinking, 91 as experiential gestalts, 114 kinesthetic character of, 138 linguistic action and, 103–104 for mathematical concepts, 11  111 1–113 mental imagery and, 138–142 metaphor and, 93–94, 96 momentum schema and, 94, 180–181 perceptual meaning analysis and, 220 political ideas and, 108–110 questions about, 114–115 representational momentum and, 139 source-path-goal source-path -goal schema and, 91–93 of “stand,” 175–177 straight schema and, 94–95 toss schema and, 198 transformations of, 95–96, 141 utterance interpretation and, 180–181 word meaning and, 175–177 of young children, 220–223 imagery. See imagery.  See mental  mental imagery imagination. See imagination.  See also mental also mental imagery  bodily action and, and, 127 embodied action and, 157 inflation of, 147–148 “motor imagery theory” and, 132 imagination inflation, 147–148 imitation, 231–234 analogical transfer and, 234 deferred,  232 of infants, 232–234 indexical hypothesis, 205–207 afforded v. non-afforded sentences and, 205

steps of, 205 instrument, body as, 27 intellectual development, Piaget’s theory of, 208–210 intentional action dynamical model of, 73–77 phenomenological demonstration of, 75

 

331 intentional behavior, causes of, 74 intentions, as dynamic processes, 77 interactions, knowledge acquisition, representation and, 85 interactive vision, computational economy and, support for, interiorization, 233 interpersonal meaning, experiences of, 27

 James, William, William, 19, 250 Kant, Emmanuel, 54 Kelso, Scott, 9 Kerry,, John, Kerry Joh n, 110 kinesthetic experience, of infants, 217 kinesthetic thinking, 124 Knowledge-based Action Representations for Metaphor and Aspect (KARMA), 195–196 language action imagining and, 150 animate motion and, 12 from bodily experience, 207 change of, 160–161 in cognitive linguistics, 11 cognitive unconscious and, 40 communication communic ation and, 158–207 embodied activity in, 207 in embodied experience, 277 embodied mental simulation and, 206 embodied view of, 275 indexical hypothesis of, 205–207 meaning and, 158 memory for, 148–151 mind and, 12  160 modal verbs and, 194 neural theory of, –197 recall of, 150 traditional views of, 159 language interpretation perceptual, embodied information in, 205 perceptual symbols and, 178 priming effects and, 177 language understanding, as embodied simulation, 207 Latent Semantic Analysis (LSA), 158 Leder, Drew, 17 lexical retrieval hypothesis, gesture and, 166

Lieberman, Joseph, 109

 

 

332 linguistic action, 99 image schemas and, 103–104 intensity of, 103 nonverbal communication and, 101 restricted movement and, 103 sensory apparatus and, 101 understanding of, 101 violent physical action and, 102

embodiment and, 124–142 “emulation theory” and, 136 “enactive” approach to, 137 environmental consequences and, 127 eye movements and, 131 ideomotor mimicry and, 125–126 ideomotor movement and, 125 image schemas and, 138–142

walking metaphors and, 103 linguistic communicatio communication n embodied experience and, 159 hypotheses for, 159 time-course of, 159–160 linguistic knowledge, constructions and,

imaging events and, 126 imaging human movement with, 124–129 kinesthetic activity and, 7, 135 mental calculations and, 128 mental rotation effects and, 133 mental simulation and, 135 models of, 132 “motor imagery theory” and, 132 motoric processes in, 129–134 movement planning in, 127 neural dynamics and, 137 neural mechanisms of, 134–135 overt, covert actions and, 127 “perceptual activity theory” of, 133

198

link schemas, causal relations and, 221 listening, movement and, 54 Lott, Trent, 110 mathematical concepts grounding metaphors for, 1  111 11–113 linking metaphors for, 113 Maturana, Humberto, 42  109 McCloskey, memory, 142Frank, –151. See also working also working memory action imagining and, 150  brain activity and, 149 composition of, 143 as embodied action, 142–144 embodied action and, 157 enactment in, 149 environmental information and, 144 false, 147–148 “generation effect” and, 148 for language, 148–151 language recall and, 150 in linguistic expressio expressions, ns, 179

 128 physical overestimation and,  132 as “planning hypothesis,” sensorimotor sensorimo tor simulations and, 12 visual perception and, 124 visual representation and, 131 as whole-person activity, 135 mental state, 3 metaphor(s). See metaphor(s).  See also embodied also embodied metaphor  breathing and, and, 101 “complex,”  117 conceptual processing and, 122 as conceptual prototypes, 120 desire as hunger and, 184–187  111 1 grounding, linking, 11 image schemas and, 93–94, 96

overt movement and, 145–146 RM effect and, 140 sensorimotorr simulations and, 12 sensorimoto memory organization packets (MOPs), 203–204 mental awareness, 265 mental imagery, 237 action performance and, 133  blind and, 131  brain activity and, 131–132 “dynamic (depiction) model” of, 132 dynamical systems account and, 128–129,

 160–101 language and, 100 linguistic change action and, mappings of, 97–99  111 1 for mathematical concepts, 11 “primary,” 1  117 17 speaking and, 100 for thinking, 97–99 thinking and, 96–99 violent physical action as, 102 walking as, 103  Metaphor and Emotion (Koveceses), Emotion (Koveceses), 119 metaphor processing, embodied action in, 183–184 Miller, George, 264

138

embodied movement and, 237

 

Index

Index

 

333

mind abstract concepts and, 12 as body, 97  body and, 3, 4, 275 characteristics of, 12–13 conscious experience and, 265 embodied nature of, 9, 12–13, 41 environmental environm ental interaction and, 12 as material object, 4

music perception events awareness and, 53 movement in, 53–54

mental states of, 262 “representations” and, 282 modularity theory, in cognitive science, 279–280 momentum image schema, 94. See also image schemas inferences of, 180–181 RM and, 141 MOPs. See MOPs.  memory organization packets  See memory motor development, process of, 74 “motor imagery theory,” 132 motor processes in mental imagery, 129–134 mental rotation and, 130 motor theory of speech perception, 162, 164

cycles of operation and, 137 , 270 –271 neural theory of language (NTL),  194 –197 hand motion verbs and, 196–197 KARMA in, 195–196 spatial relations terms in, 194–195 New York Times, Times, 110 Newton, Naitka, 261 non-ordinary consciousness, 268 analogy for, 269 types of, 269–270 NTL. See NTL.  See neural  neural theory of language numerical concepts. See concepts. See also mathematical also mathematical concepts embodiment of, 104–105

 164 revision to, movement. movement. See  See also body movements also body “alien hand” and, 34 of body, 27 causality perception from, 54–55 consciousness and, 265, 266 creative problem-solving and, 152 ideomotor action and, 35 imagery in, 237 listening and, 54 mental imagery and, 7 in music perception, 53–54 onset of motion and, 218 people identification and, 51–53 to perception, 49

 64  65 object perception,  bodily actions and, object permanence, 210–212 adaptive processes approach to, 225 connectionist model of, 224 dynamic systems perspective on, 225–226 image schema development and, 223 newer studies of, 224–227 objects, categorization of, 84–86 “Ode and Burgeonin Burgeonings” gs” (Neruda), 92 odor perception, 47–49 olfactory system, as chaotic dynamical system, 48 Ongee, 38

narrative, of embodimen embodimentt experiences, 1 Neruda, Pablo, 92 neural embodiment, 39 neural processes conscious mental imagery and, 137

 51

perceptual representation of, proprioceptive system and,  29 static patterns and, 55–57 Moynihan, Patrick, 1  110 10 multimodal perception perception,, 229–231 amodal v. arbitrary relations and, 231  basic principles of, 231 cognitive development and, 231 cross-modal connections and, 229 global relations and, 231 nested relations and, 231 synesthesia studies and, 230 of young children, 230

path image schema linguistic action and, 104 RM and, 142 Penfield, Wilfred, 45 perception. See perception.  See also embodied also embodied perception; perception; perceptual experience; experience; perceptual processes action and, 49, 58–64, 77–78 action v., 57–64 affordances and, 43 as anticipated embodied interaction, 64–65

 

334 perception (cont. (cont.)) “binding” in, 67

 

Index perceptual simulations and, 88 situation simulation and, 88–89

perceptual systems theory, 86. See also perceptual symbols theory assumption of, 87 support for, 87 personhood  body and, 14 cognition and, 17 couplings and, 40

 body representation representation and, and, 51  body-world interactions interactions and, and, 45 “change-blindness” “change-blindn ess” and, 66 color, 44 compensation and, 62 computational economy and, “corporeal disappearance” and, 14 definition of, 42  43

 15 16

Dewey on, dynamic couplings and, 49 embodied action and, 12, 42–45 environment and, 43–45, 77 haptic, 54 of human movement, 52 image schemas and, 142 “inattentional blindness” and, 66 interactive vision and, kinesthetic activity and, 12 mirror neurons and, 52 motor movements and, 60 motor planning and, 61 movement to, 49 of music, 53–54 of object properties, 56, 64 of odor, 47–49 preattentive, 48 robotics and, 72 sensory consequences and, 69 sensory/neural sensory/ne ural map and, 47 as skill-based activity, 67 of speech, 161–165 subjunctive thought processes and, 64 tactile, 68 visual, 54 perception-action perception -action couplings, dynamical account of, 77–78 “perceptual activity theory,” 133 perceptual experience, real, anticipated movement in, 65 perceptual grounding, 265 perceptual meaning analysis, 219–224 perceptual processes, concept construction and, 86 perceptual symbols, 86–90 abstract concepts and, 89, 90 counterfactual thought and, 90 as multimodal, 86 perceptual symbols theory. See theory. See also perceptual systems theory conceptual knowledge and, 89

first-person perspective and, – in philosophy phi losophy, , 16 persons environment and, 16 self-definition of, 16 phenomenological phenome nological primitives (p-prims), 154–155, 156 phenomenology, environment and, 17 philosophy personhood and, 16 physical body and, 6 physical handicaps children with, 227–229 degree of impairment and, 228 emotional experience and, 252–253 physical reasoning Baillargeon’s view of, 215–216 embodied action in, 216–219 habituation-dishabituation habituation-dishab ituation studies of, 212–213 object permanence and, 210–212, 213

perspectives on, 215 recent studies on, 210–214 sensorimotor behavior, experience and, 212

Piaget, Jean, 8, 208–210 criticism of, 21  211 1 development mechanisms and, 208 disequilibrium and, 210 imitation and, 231 interiorization and, 233 sensorimotor sensorimo tor stage and, 209 Plains Sign Talk (PST), 169 Plato, 3 political ideas 111 1 embodiment of, 107–11 image schemas for, 108–110 posture, emotion and, 254 p-prims. See p-prims.  See  phenome phenomenological nological primitives preattentive perception perception,, 48 “premotor theory of attention,” 61

 

Index problem-solving  body movement movement and, 152 “complementary “complemen tary strategies” to, 152 environmental resources and, 152 sensorimotorr simulations and, 12 sensorimoto tools for, 154 proprioception,  29

 

335 self  autobiographical, 21 autositic-parasitic twins and, 24  body and, 18–25  boundaries of, of, 40 conjoined twins and, 23–24 core, 21

function of, 29 loss of, 33 proto self, 21 prototype theory, concepts and, 83 prototypes, formation of, 83 PST. See PST.  See Plains  Plains Sign Talk psycholinguistics, psycholinguistic s, embodimen embodimentt and, 7 psychology. See psychology.  See also embodied also embodied psychology  body and, 6 embodied, 25

dynamic system of, 24 fragmentation of, 20 future action and, 23 as indivisible, 18 kinds of, 21 levels of, 21 physical/cultural world and, 21 proto, 21 sensory experience and, 19 tactile-kinesthetic tactile-kinesthe tic activity and, 12, 20 Western view of, 18 world and, 17 self-concept  body and, 18 embodied interactions and, 22 kinesthetic-tactile kinesthetic-tac tile experience and, 15 self-identity, 20 senses culture and, 37–39 linguistic action and, 101 sensorimotor contingency of, 67 sensorimotor contingencies, 65 sensorimotor contingency theory, 65–69 “change-blindness” “change-blind ness” and, 66 consequence of, 67 implication of, 66 “inattentional blindness” and, 66 senses and, 67 support for, 68 sensorimotor experience cognitive development and, 227 motion patterns and, 219 object understanding, behavior and,

qualia, 6 reasoning,  151–157 as computational skill, 151 embodied, 154 embodied action and, 157 embodied knowledge and, 156 embodied simulation and, 151 environmental information and, 153, 154

physical interaction and, 156 reflection, 265 Reid, Thomas, 18 representational momentum (RM), 139 auditory stimuli and, 140 image schemas and, 140–142 RM effect and, 139–140 RM. See RM.  representational momentum  See representational robotics,  69–73 AL and, 69–72 dynamical systems theory in, 72 environment and, 72 “robot grounding” problem of, 73 Roelofs effect, 57 Rowles, Jimmy Jimmy,, 76

218

physically handicapped and, 227 seselalame, 38 Sherrington,, Charles, 29 Sherrington short-term memory memory.. See  See working  working memory Simon effect, 60 Smith, Cyril Stanley, 123 somatasensory somatasenso ry system, “somatic marker hypothesis” and, 136 somatotopic map, 45 multiple, 47 Songhay, 37

Sarbanes, Paul, 109 schemas. See schemas.  See also image also image schemas TOPs and, 204 schizophrenia, schizophr enia, brain and, 267 Schroeder, Patricia, 109 scripts, 202 creation of, 203 person-specific prototypes and, 204

 

336 source-path-goal source-path-goal schema, 91–93. See also image schemas RM and, 141 spatial orientation, embodiment of, 105–107 speaking “generation effect” and, 148 working memory and, 144 speech gesture and, 165–170 intensity of, 103 recency effect and, 163

 

Index mentalistic assumption and, 234 simulation theory and, 234 “theory theory” of, 234 thinking as embodied, 97, 99 metaphor and, 96–99 thought. See thought.  See also embodied also embodied thought embodied view of, 275 sensorimotor origins of, 169 time distinct schemas for, 187–188 embodied view of, 190