avaliacao estetica

8182019 avaliacao estetica

httpslidepdfcomreaderfullavaliacao-estetica 111

Exp Brain Res (2010) 204447ndash456

DOI 101007s00221-010-2283-6

1 3

RESEARCH ARTICLE

Extrastriate body area underlies aesthetic evaluation

of body stimuli

B Calvo-Merino middot C Urgesi middot G Orgs middot S M Aglioti middot

P Haggard

Received 16 April 2009 Accepted 26 April 2010 Published online 16 May 2010983209 Springer-Verlag 2010

Abstract Humans appear to be the only animals to have

developed the practice and culture of art This practice pre-sumably relies on special processing circuits within thehuman brain associated with a distinct subjective experi-ence termed aesthetic experience and preferentiallyevoked by artistic stimuli We assume that positive ornegative aesthetic judgments are an important function of

neuroaesthetic circuits The localisation of these circuits in

the brain remains unclear though neuroimaging studieshave suggested several possible neural correlates of aes-thetic preference We applied repetitive transcranialmagnetic stimulation (rTMS) over candidate brain areas todisrupt aesthetic processing while healthy volunteers madeaesthetic preference judgments between pairs of dance pos-tures or control non-body stimuli Based on evidence fromvisual body perception studies we targeted the ventral pre-motor cortex (vPMC) and extrastriate body area (EBA) inthe left and right hemispheres rTMS over EBA reducedaesthetic sensitivity for body stimuli relative to rTMS overvPMC while no such diV erence was found for non-body

stimuli We interpret our results within the framework of dual routes for visual body processing rTMS over eitherEBA or vPMC reduced the contributions of the stimulatedarea to body processing leaving processing more reliant onthe unaV ected route Disruption of EBA reduces the localprocessing of the stimuli and reduced observersrsquo aestheticsensitivity Conversely disruption of the global route viavPMC increased the relative contribution of the local routevia EBA and thus increased aesthetic sensitivity In thisway we suggest a complementary contribution of bothlocal and global routes to aesthetic processing

Keywords Neuroaesthetic middot Aesthetic perception middotBody perception middot Transcranial magnetic stimulation middotExtrastriate body area middot Premotor cortex

Introduction

The body of a conspeciWc is a salient and powerful stimu-lus Recent studies in both humans and other animals haveshown sensory (Keysers et al 2004 Downing et al 2001)

Electronic supplementary material The online version of thisarticle (doi101007s00221-010-2283-6) contains supplementarymaterial which is available to authorized users

B Calvo-Merino (amp)

Department of Psychology City University LondonNorthampton Square London EC1V 0HB UKe-mail bcalvocityacuk

B Calvo-MerinoDepartamento Psicologiacutea Baacutesica II Procesos CognitivosFacultad de Psicologiacutea Universidad Complutense de MadridCampus de Somosaguas 28223 Madrid Spain

G Orgs middot P HaggardPsychology Department and Institute of Cognitive NeuroscienceUniversity College London 17 Queen squareLondon WC1N 3AR UK

C Urgesi

Dipartimento di FilosoWa Universitagrave di Udine 33100 Udine Italy

C UrgesiIstituto di Ricovero e Cura a Carattere ScientiWco (IRCCS)ldquoE Medeardquo Polo Friuli Venezia Giulia37078 San Vito al Tagliamento Pordenone Italy

S M AgliotiDipartimento di Psicologia Universitagrave di Roma ldquoLa Sapienzardquo00185 Rome Italy

S M AgliotiIRCCS Fondazione Santa Lucia 00179 Rome Italy



448 Exp Brain Res (2010) 204447ndash456

1 3

motor (Di Pellegrino et al 1992 Calvo-Merino et al 20052006) and aV ective responses (Wicker et al 2003) in sev-eral cortical areas triggered by viewing conspeciWcs Theseresponses are often interpreted with reference to a hypothe-sis of lsquothe social brainrsquo (Frith and Frith 2007) This viewemphasises how the brain represents the behaviour andmental states of others in order to learn from them and

interact with them either competitively or co-operatively(Sebanz et al 2005)

Here we focus on a less-studied aspect of visual pro-cessing of bodies namely aesthetics Artistic activity isthought to be a uniquely human behaviour (Cela-Condeet al 2004) associated with development of speciWc corti-cal circuits Art objects may be considered to activate brainnetworks that generate aesthetic experiences Howeverseveral diV erent views exist in the literature SpeciWcallyneural correlates of aesthetic experience have been pro-posed in reward regions of the brain (ie orbitofrontal cor-tex) (Kawabata and Zeki 2004 Vartanian and Goel 2004)

in emotional centres such as the amygdala (Di Dio et al2007) in specialised visual perceptual areas (Zeki andLamb 1994) in dorsolateral prefrontal regions associatedwith higher lsquoexecutiversquo functions such as monitoring (Cela-Conde et al 2004) and with frontomedian regions underly-ing social and moral judgment (Jacobsen et al 2006) (seeNadal et al 2008 for a comprehensive review)

Aesthetic objects are described using speciWc labelsincluding but not limited to lsquobeautyrsquo (Jacobsen et al 2004)In the case of performing arts such as dance the observerrsquosaesthetic experience is presumably grounded in theresponses of their neural sensory motor and aV ective cir-cuits to the expressive actions of the dancerrsquos body Consis-tent with this account viewing dance recruited a network of parietal and premotor areas in a manner dependent on theviewerrsquos previous sensorimotor experience (Calvo-Merinoet al 2005) Importantly dance movements and dance pos-

tures may be judged beautiful or otherwise quite indepen-dently of whether the dancer is judged to be personallyattractive or not (Brown et al 2005) We therefore conjec-tured that brain circuits specialised for representing thebodies and actions of conspeciWcs might also underlie aes-thetic experience associated with dance

Aesthetics has a long history in both psychology andneurology (Fechner 1876) Psychophysical studies aimed atidentifying stimulus features producing positive and nega-tive aesthetic evaluations (Fechner 1876 McManus et al1985) More recent neuroscientiWc studies investigated neu-ral correlates of aesthetic evaluation These studies typi-cally used a lsquosubjectivistrsquo approach presenting a widerange of stimuli and comparing the responses for thoseliked or found lsquobeautifulrsquo to responses for those dislikedor found lsquouglyrsquo while acknowledging that participantsdiV er in their evaluation of any particular stimulus For

example Cela-Conde et al (2004) found that liked pictureselicited stronger prefrontal cortex activations than dislikedpictures Kawabata and Zeki (2004) found stronger activa-tion of orbitofrontal cortex for pictures found beautiful thanfor pictures judged ugly and stronger activation of sensori-motor cortex for the opposite contrast Calvo-Merino et al(2008) applied this approach to dance actions They found

stronger activity in occipital cortex bilaterally and in theright premotor cortex when six subjects viewed short dancepassages that they reported liking in a later evaluationcompared to those they disliked Subjectivist approachesare well suited to small-scale studies of the neural bases of aesthetic experience but cannot explain why particular stim-uli produce particular experiences Only one neuroaestheticstudy relevant to body representation has focussed onobjective stimulus properties to our knowledge Di Dioet al (2007) found stronger neural activity in both occipitalcortex and right anterior insula for images of statues obey-ing the golden section a principle of spatial proportion tra-

ditionally felt to be beautiful than for statues not followingthis principle

Importantly both subjectivist and objectivist neuroim-aging studies have the weakness of being correlationalActivations that correlate with lsquobeautyrsquo or liking could bepurely epiphenomenal and may not indicate the neural cir-cuits that actually underlie aesthetic experience Interven-tion studies in contrast can reveal brain areas or circuitsactively involved in aesthetic evaluation Here followingthe principle of perceptual selectivity (Zeki and Lamb1994) we investigated whether body-sensitive areas alsocontribute to aesthetic experience of dance perception Werecently (Urgesi et al 2007) proposed a dual-route modelof visual body perception by identifying two brain areasinvolved in visual body perception using repetitive trans-cranial magnetic stimulation (rTMS) We suggested thatthe extrastriate body area (EBA) an occipital area specia-lised for bodies (Downing et al 2001) houses a localrepresentation of body parts while the ventral premotorcortex (vPMC) houses a conWgural representation of com-plete body postures SpeciWcally we found impairedperception of body postures presented either upright orinverted (suggesting an analytical or local way of process-ing bodies) following EBA rTMS while rTMS over theleft premotor cortex impaired perception of upright but notinverted bodies (suggesting a global or conWgural type of processing) (Reed and Farah 1995 Reed et al 2003) Innormal function these two routes presumably providecomplementary information which is combined to producea single body percept Here we investigated the contribu-tions of these two routes to aesthetic evaluation of bodypostures by comparing the eV ects of EBA rTMS andvPMC rTMS with sham rTMS in an aesthetic preferencetask



Exp Brain Res (2010) 204447ndash456 449

1 3

Methods

Subjects

Sixteen subjects (8 women) aged 20ndash25 (mean age 217)participated in the experiment Two further participantswere recruited but not included because they could not be

tested in all conditions because of discomfort associatedwith the stimulation of the premotor sites Participants hadno previous experience with dance performance or withany of the body postures used in the experiment A standardhandedness inventory (Briggs and Nebes 1975) revealedthat one participant was left handed while the remaining15 participants were right handed The procedures wereapproved by the ethics committee of the Fondazione SantaLucia (Rome Italy) and were in accordance with the ethicalstandards of the 1964 Declaration of Helsinki None of theparticipants had neurological psychiatric or other medicalproblems or any contraindication for rTMS (Wassermann

1998)

Stimuli

Body posture stimuli were the same as those used in a pre-vious perceptual study (Urgesi et al 2007) BrieXy theyshowed a single male dancer in frontal view in posturestaken from classical ballet and some hybrid postures gener-ated by recombining the upper or lower limb positions fromclassical ballet and other dance styles To investigatewhether any eV ects of rTMS on aesthetic evaluation werespeciWc to body perception as opposed to non-speciWc fac-tors such as the discomfort associated with rTMS at partic-ular scalp locations we also presented scrambled non-body

stimuli These were created by pixellating the original bodyimages into a 4 pound 4 pixel array and swapping two arbi-trarily selected quadrants of the picture to remove informa-tion about limb posture The resulting scrambled imagesresembled abstract patterns rather than bodies A similartechnique was previously used to separate visual processingof biological motion from non-biological motion

(Orgs et al 2008) Sixteen pairs of body stimuli were usedin the experiment together with the 16 pairs of scramblednon-body images derived from them (Fig 1 and Table 1Supplementary Material)

Trial structure

On each trial a pair of pictures was shown The pair con-sisted of two body stimuli or the two scrambled stimuliderived from them Each pair was presented twice in a sin-gle block with the second presentation containing the samestimuli in the reverse order of the Wrst presentation Each

image was therefore judged 10 times across the entireexperiment Each participant was tested in a single experi-mental session lasting 2 h Participants completed a 32 trialpractice block before proceeding to the experimentalblocks During the experimental session two blocks of 32trials were presented for each stimulation site For each par-ticipant each of the four stimulation conditions and shamcondition was repeated twice according to a counterbal-anced sequence Each trial began with presentation of aWxation point After 500 ms this was replaced by the Wrstimage for 100 ms followed by a binary visual noise maskfor 500 ms Then the second image was presented for100 ms followed by a further mask Finally a visualprompt ldquoWhich do you prefer the Wrst or the secondrdquo

Fig 1 Time course and example stimuli for aesthetic preference judgements between pairs of body postures ( left ) and pairs of non-body stimuli(right )

rTMS

Delay150 after

sample onset

500ms 500ms

100ms100ms

500 ms

100ms

500 ms

Prefer

1st or 2nd

500 ms

100ms

500 ms

Prefer

1st or 2nd

Body Trial Example Scrambled Trial Example

rTMS

Delay150 after

sample onset

500ms500ms 500ms500ms

100ms100ms100ms100ms

500 ms500 ms

100ms100ms

500 ms500 ms

Prefer

1st or 2nd

500 ms500 ms

100ms100ms

500 ms500 ms

Prefer

1st or 2nd




1 3

(lsquoQuale ti piace di piugrave il primo o il secondorsquo) asked subjects to indicate which of the two pictures they preferredby an unspeeded keypress response of the right index andmiddle Wngers respectively (Fig 1)

In addition at the end of the experiment each participantviewed each stimulus body or scrambled body alone onthe screen in random order and used a visual analogue

scale (VAS) to rate how much they liked it (ranging from 0lsquoI do not like it at allrsquo to 100 lsquoI like it very very muchrsquo)

TMS

For each participant the resting motor threshold for the Wrstdorsal interosseous muscle of the right hand was deter-mined Surface AgAgCl electrodes were placed in a bellyndashtendon montage An electromyographic signal was ampli-Wed at a gain of 1000 by a Digitimer (Hertfordshire UK)D360 ampliWer bandpass Wltered (20 Hzndash25 kHz) and dig-itised (sampling rate 5 kHz) by means of a CED Power

1401 controlled with Spike 2 software (Cambridge Elec-tronic Design Cambridge UK) The resting motor thresh-old deWned as the lowest intensity able to evoke 5 of 10motor-evoked potentials with an amplitude of at least50 uV was determined by holding the stimulation coil overthe optimal contralateral scalp position

rTMS was applied by connecting two Magstim Model200 stimulators with Bistim module (The Magstim Com-pany Carmarthenshire Wales UK)) with a 70-mm Wgure-eight stimulation coil (Magstim polyhurethane-coated coil)in separate blocked conditions with each subject having adiV erent random block order For sham stimulation the coilwas placed over the vertex and oriented perpendicular tothe scalp with the border of one wing placed against thehead This ensured that no magnetic stimulation reached thebrain during sham stimulation and controlled for noise andthe sensation of the coil against the head The same stimu-lation intensity and timing were used for magnetic andsham stimulation In the experimental conditions the coilwas held over PMC or EBA of the left or right hemisphere

with the handle pointing posteriorly These areas werelocated on each participantrsquos scalp with the SofTaxicNavigator system (EMS Bologna Italy) Coordinates inTalairach space (Talairach and Tournoux 1988) were auto-matically estimated by the SofTaxic Navigator from a mag-netic resonance imaging-constructed stereotaxic templateThese were vPMC corresponding to Brodmannrsquos area 44

in the pars opercularis of the inferior frontal gyrus (LH ( x iexcl57 y 11 z 22) RH ( x 57 y 11 z 22)) and EBA corre-sponding to Brodmannrsquos area 37 in the posterior part of themiddle temporal gyrus (LH ( x iexcl52 y iexcl72 z 4) RH ( x 52 yiexcl72 z 4)) Participants wore a tightly Wtting bathing cap onwhich the stimulation points of the scalp were marked(Fig 2)

During stimulation of the four sites the coil was held byhand tangential to the scalp with the handle pointing back-ward and medially at a 45deg angle from the middle sagitalaxis of the participantsrsquo head The contours of the handleand of the coil placed over each stimulation site were

marked on the bathing cap to check continuously the posi-tion of the coil with respect to the marks and its orientationin the axial plane

The same pulse delay and stimulation intensity was usedfor the four stimulation sites and for sham stimulationStimulation intensity was 120 of the resting motor thresh-old for both pulses This ranged from 42 to 66 (mean52) of the maximum stimulator output A train of 2 rTMSpulses was delivered at 150 and 250 ms after the onset of the Wrst image Pulses were timed to interfere with the cor-tical processing of the Wrst image

During stimulation participants wore commercial ear-plugs to protect their hearing None of the participantsreported limb muscle twitches or phosphenes due to rTMSsuggesting that we did not inadvertently allow stimulationto spread to either primary motor or visual cortex Stimula-tion occasionally induced peripheral activation of facialmuscles and some jaw movements or blink responses wereobserved in most participants as a result of stimulationBlinking would not prevent the participants from seeing the

Fig 2 Stimulation sites plotted on the lateral views of a standardbrain Mean Talairach coordinates of the stimulation sites were as fol-lows vPMC corresponding to Brodmannrsquos area 44 in the pars operc-ularis of the inferior frontal gyrus (LH ( x iexcl57 y 11 z 22) RH ( x 57

y 11 z 22) and EBA corresponding to Brodmannrsquos area 37 in the pos-terior part of the lateral occipitotemporal (LH x iexcl52 y iexcl72 z 4 RH

x 52 y iexcl72 z 4) L Left Hemisphere R right hemisphere






1 3

be more likely to be preferred following rTMS That isreduced sensitivity would make aesthetic preference judge-ments depart more frequently from what VAS ratings of each stimulus would predict Enhanced sensitivity wouldproduce the converse pattern Thus the precise pattern of aesthetic preference modulation at each rTMS site mayreveal its role in aesthetic evaluation

The percentage modulations of aesthetic preferencechange were subjected to a 3-way repeated measuresANOVA with factors of rTMS site (EBA vPMC) hemi-sphere (Left Right) and image type (Body Scrambled)These data are shown in Table 1 The main eV ects of rTMSsite and image type were non-signiWcant (all F s115 lt 1)The main eV ect of hemisphere showed a trend for left hemi-sphere stimulation to decrease the relation between aes-thetic preference judgement and VAS relative to its valueat sham but for right hemisphere stimulation to increasethe strength of this relation (F 115 = 374 P = 0072mean (sectsem) left 10469 sect 35 of sham right

9692 sect 351 of sham) The two-way interaction of hemisphere with rTMS site and image type was not signiW-cant (all F s115 lt 1) However a signiWcant interactionbetween rTMS site and image type was found (F 115 = 576P = 003 Fig 3) Duncan post-hoc tests showed that theinteraction was explained by preference judgements forbody stimuli departing more strongly from VAS predic-tions after EBA stimulation than after vPMC stimulation(means (sectsem) EBA 10747 sect 818 of shamvPMC 953 sect 456 of sham P = 0042) No diV erencewas obtained between EBA and vPMC stimulation condi-tions on the aesthetic judgement changes for scrambledstimuli (means (sectsem) EBA 9767 sect 431 of shamvPMC 10277 sect 49 of sham P = 033) No other pair-wise comparisons were signiWcant (all Ps gt 008) and noneof these conditions were signiWcantly diV erent from 100(all Ps gt 01) (Fig 3)

In summary EBA rTMS blunted aesthetic judgementsabout body postures relative to vPMC rTMS making theaesthetic preference performance of each subject less pre-dictable from hisher VAS judgements Of the potentialeV ects of rTMS on aesthetic judgement described abovethis corresponds to a change in aesthetic sensitivity Thenon-signiWcant three-way interaction between rTMS site

hemisphere and image type (F 115 lt 1) shows that rTMS of both left and right EBA disrupted body aesthetic judge-ments relative to the ipsilateral vPMC

Discussion

We applied brief trains of rTMS over each hemisphere totwo brain areas selective for diV erent aspects of visual bodyprocessing (EBA vPMC) to investigate the neural mecha-nisms of aesthetic of body perception For each participantwe classiWed each presented image in a pair according to its

individual VAS aesthetic judgment and whether or not itreceived higher or lower scores with respect to the otherimage in the pair We used this index to predict each indi-vidual subjectrsquos responses to each pair of stimuli presentedto them The analysis of the modulations of aesthetic pref-erences showed that body aesthetic judgements in the VAShad a reduced role in predicting preference choices follow-ing stimulation of EBA when compared to stimulation of the ipsilateral vPMC That is EBA stimulation tended toblunt aesthetic sensitivity (reduce congruency with VASaesthetic scores) and vPMC stimulation tended to enhanceit (increase congruency with VAS aesthetic scores) Nosimilar changes were observed during aesthetic judgementof the scrambled control stimuli and no other interactionsor main eV ects were signiWcant This result allows us toconclude that diV erent routes within the two-route model of body processing (Urgesi et al 2007) play a complementaryrole on aesthetic perception In this way rTMS over EBAblunted aesthetic responding to body postures relative tothe eV ect in the opposite direction to rTMS over vPMCand in contrast to the pattern of results for control non-bodyimages Our results suggest that the EBA and vPMC maybe two complementary components of the aesthetic percep-tion network for bodies

It is important to remember that our study used a lsquosubjectivistrsquo rather than an lsquoobjectivistrsquo design We do notclaim that rTMS altered the evaluation of all images butonly that each subjectrsquos idiosyncratic pattern of aestheticevaluation for body postures was attenuated after stimula-tion over EBA and enhanced after stimulation over vPMCwhile no similar eV ect was found for stimulation followingpresentation of non-body stimuli Moreover no diV erencesbetween bodies and non-body stimuli were found in thesham condition Although our data does not allow any

Fig 3 EV ects of rTMS on normalised aesthetic preference judge-ments as a function of image type and rTMS site (EBA vPMC)Results are shown averaged over left and right hemispheres stimula-tion (see text) Error bars indicate standard errors over participants P lt 005




1 3

general claim about what stimuli produce particular aes-thetic experiences it allows us to identify the regions thatnecessarily participate in the neural processes that underlieeach individualrsquos aesthetic evaluations of body stimuli

To our knowledge this is the Wrst interventive asopposed to correlative study of aesthetic evaluation andthe Wrst neuroaesthetic study of visual perception of static

body postures Moreover the pattern of results rules outsome possible artefactual explanations rTMS did notinduce any simple bias either towards preferring or not pre-ferring the Wrst image of the pair so the results cannot beexplained in terms of general rTMS-induced discomfort ordiscomfort conWned to particular stimulation sites In addi-tion we found no main eV ect of stimulus type nor anyinteraction between stimulus type and preference at shamThis suggests that body and scrambled stimuli werebalanced and were both equally amenable to aestheticevaluation

Two-route model of body processing

It has been suggested that aesthetic judgements are madein frontal regions (Jacobsen et al 2006) that evaluateinformation provided by earlier stages of visual process-ing streams By applying rTMS to two regions of the net-work that contribute diV erently to body perception (EBAand vPMC for local and conWgural processing respec-tively Urgesi et al 2007) we could investigate how thesetwo regions contribute to aesthetic perception Here weshow a reduction in aesthetic sensitivity for body stimulifollowing EBA stimulation relative to vPMC stimulationwhile no such eV ect was found for non-body stimuli Thisresult suggests that the early perceptual analyticalprocessing of body form by EBA contributes to the Wnal

aesthetic evaluation of body stimuli ConWgural process-ing in vPMC may also contribute to aesthetic evaluationbut intervention here produced a less conclusive eV ect(Fig 4) In this way both EBA and vPMC play a role inaesthetic processing and information from both paths isintegrated in a Wnal body percept to be evaluated in fron-tal decision-making regions However our results show

that their individual contributions inXuence Wnal aesthetic judgment in a diV erent manner

Psychology has widely investigated the role of globalconWgurations on aesthetics from artworks (McManuset al 1985) to simple geometrical shapes (ie the goldenratio Livio 2002) and complex biological conWgurationssuch as faces (Abbas and Duchaine 2008) In painting glo-bal conWgurations are suggested to underlie high aestheticquality (Vartanian et al 2005) Therefore it may seem sur-prising that the local processing of the EBA rather than theglobal conWgural processing of the vPMC seems to under-lie strong aesthetic sensitivity in our study However modi-

fying the balance or composition of a painting does notnecessarily change its aesthetic value suggesting that itsvalue is not only in the structural conWguration but also inthe local content (Shaw 1962) Therefore both local andglobal components may play a role for aesthetic judge-ments The present study employed a forced-choice of aes-thetic judgments between two meaningless body posturesthat had diV erent leg and arm positions In this way partici-pantsrsquo attention might have been driven to the detail of thebody parts rather than to the global conWguration Purelylocal changes could be crucial for aesthetic values of theseparticular stimuli Accordingly disruption of the conWguralpath would enhance local processing and therefore facili-tate aesthetic sensitivity Further studies should explorewhether driving attention to conWgural features of visual

Fig 4 Aesthetic processing inthe two-route body model

Aesthetic preferencechange from VAS

Control site(sham)

No change from VAS (100)

Configuralsystem

Local

s stem

V1

Configuralprocessing

Reduced change from VAS (953)

Configuralsystem

V1 S s i t e

area

Local

Heightened aesthetic sensitivity

system

Configural

r T M

processingarea

Increased change from VAS (10749)Blunted aesthetic sensitivity

sys em

Localsystem

V1




1 3

stimuli might change the direction of this eV ect betweenboth body processing paths

The present results also extend our view of these twobody processing regions Previous studies identiWed rolesof EBA and vPMC in perception of body parts and whole-body conWgurations respectively (Downing et al 2001Taylor et al 2007 Urgesi et al 2007) EBA has been tradi-

tionally considered merely an early category-selectiveregion for the visual processing of static images of thehuman body and not to pictures of other stimulus categoriessuch as objects (eg a chair a spoon) (Downing et al 2001Moro et al 2008 Urgesi et al 2004 Peelen and Downing2005) However recent studies suggest that EBA alsoperforms more cognitive functions and forms part of anetwork with frontal sensorimotor regions (AstaWev et al2004 Helmich et al 2007 David et al 2007) Recent stud-ies have shown functional connectivity of EBA with highercortical areas such as the PMC and posterior parietal cortex(PPC) during diV erent cognitive tasks such as mental hand

rotation (Helmich et al 2007) or self-other attribution(David et al 2007) In general these studies showed func-tional interactions between EBA and PMC and PPC andextend EBA basic body perception function to one moreintegrative and dynamic that may include visual integra-tion spatial attention and sensorimotor signals involved inthe representation of the observerrsquos body (David et al2007) The present study supports the idea of further andmore extensive roles of EBA within the body perceptioncircuit We propose its contribution to a particularly humanway of seeing such as aesthetic perception

Multiple visual areas for aesthetic processing

While we did not test low-level perceptual processing of these stimuli directly the pattern of our results makes itunlikely that blunting of aesthetic sensitivity was simplydue to participants failing to see the stimuli following stim-ulation over EBA regions First reports of rTMS-inducedvisual masking are conWned to disruption of early visualareas (Kammer 1999) Secondary visual areas such asEBA which is situated in the posterior part of the inferiortemporal sulcus are not classical loci for TMS-inducedmasking Second a purely low-level perceptual eV ectwould apply equally to both types of stimuli presented inthe experiment (bodies and scrambled bodies) Howeverour pattern of results suggests that rTMS over EBA andPMC disrupted aesthetic processing only for body stimuliin contrast to non-body scrambled stimuli

In general an initial perceptual analysis is inevitableduring the aesthetic processing of visual stimuli Most psy-chological work on aesthetic has focussed on investigatingwhat perceptual features related to artworks are generallypreferred by the observer (Berlyne 1974 Zeki 1999 Leder

et al 2004) Most designs vary only one perceptual featureat a time (ie complexity contrast colour symmetry bal-ance) and evaluate how people tend to prefer one stimulusover another Occipital visual processing areas are sup-posed to be involved at this level In the present study wehave not sought for particular stimulus feature that are moreor less preferred Rather we show that the early analytical

visual processing of body stimuli has a signiWcant role inthe later aesthetic evaluation This type of result strengthensthe role of multiple perceptual regions in aesthetic percep-tion (Zeki and Lamb 1994)

Hemispheric lateralisation of aesthetic processing

The lack of hemisphere eV ects suggests that both hemi-spheres contributed in a similar manner to aesthetic pro-cessing This is consistent with other rTMS studiesinvestigating other aspects of body processing such as formand action discrimination on the right and left EBA and

PMC where no modulation by hemisphere stimulated wasfound (Urgesi et al 2007) Although previous studies of body perception reported stronger activation of the right

EBA (Downing et al 2001) for body parts most reportedactivations are clearly bilateral (Chan et al 2004 Peelenand Downing 2007) A recent meta-analysis and review of premotor cortex activation in action observation reportedno convincing lateralisation (Morin and Gregravezes 2008)Moreover lateralisation in aesthetic processing has notbeen clearly addressed The few published neuroaestheticstudies using visual stimuli have not achieved consensus onthe lateralisation of aesthetic responses For example twofMRI studies report left sensorimotor cortex activationsassociated with explicit subjective judgements of ugliness(Di Dio et al 2007 Kawabata and Zeki 2004) In contrastCalvo-Merino et alrsquos (2008) participants Wrst viewed dancemovements and later gave them aesthetic evaluations in aseparate session They found that right but not left PMCactivity correlated with aesthetic evaluation Although weobserved a general trend for left hemisphere rTMS todecrease the aesthetic sensitivity in the present study nosigniWcant hemisphere eV ect or interaction was found Wetherefore suggest that both hemispheres contribute to theperceptual analysis necessary for aesthetic evaluation inEBA in relation to the processing of the ipsilateral PMC

Interpretative and methodological issues

Several previous neuroaesthetic studies identiWed neuralcorrelates of positive and negative aesthetic judgements bycomparing activation for stimuli found beautiful and those

judged ugly Such studies cannot identify the neural pro-

cesses of aesthetic evaluation and discrimination since theevaluation process presumably occurs whether its output is




1 3

positive or negative In contrast the pattern of our data sug-gests that rTMS impaired the aesthetic sensitivity necessaryto perform an aesthetic evaluation judgement itself (whether this is positive or negative) These results implyloss of aesthetic sensitivity or aesthetic discriminationrather than simple bias towards positive or negative evalua-tions In general we suggest that neuroaesthetic studies

should distinguish the process of aesthetic evaluation fromthe stimulus properties that cause speciWc evaluation out-puts on the one hand and from the subjective experiencesassociated with speciWc evaluation outputs on the other Weknow of only one other neuroscientiWc study focussing onaesthetic processing as opposed to aesthetic evaluationJacobsen et al (2006) contrasted activation during aesthetic

judgement with activations during symmetry judgements of the same stimuli However this study faces the same criti-cism as other correlative fMRI designs The activationcould reXect an epiphenomenal correlate of aesthetic pro-cessing rather than aesthetic processing For example

making aesthetic judgements may be more engaging andarousing than making symmetry judgements which couldexplain the greater activation in attentional and limbicregions of cortex Future studies might combine our TMSapproach to aesthetic processing with the approach takenby Jacobsen et al (2006) A study applying TMS to targetareas during both aesthetic and non-aesthetic control judge-ments about the same stimuli might distinguish brain areascontributing to visual perception in general from those con-tributing to aesthetic processing in particular

Finally we consider some methodological issues Firstthe results obtained here are partial rather than total rTMSover EBA reduced aesthetic sensitivity to body posturesrelative to rTMS stimulation over PMC and in contrast tonon-body stimuli However EBA rTMS did not entirelyabolish aesthetic evaluation This could reXect the rela-tively mild rTMS intervention that we applied Alterna-tively the aesthetic preferences expressed in our study maybe quite subtle since they are caused by changes in limbposition of a single male dancer More extreme aestheticvariations between diV erent dancers or diV erent danceforms might produce stronger results Second our studycannot clarify what body postures if any involve absolutebeauty due to its subjectivist approach We recorded eachpersonrsquos individual evaluation of each stimulus withoutassuming that evaluations generalise across people Theseresults cannot therefore be used for lsquoneuromarketingrsquo(McClure et al 2004) or predicting the aesthetic impact of an object on the population in general Instead subjectivistdesigns focus on identifying the neural correlates of aes-thetic processing Third our study cannot reveal why a sub-

ject prefers a particular stimulus on any particular trialbecause we recorded preferences but not the reasons under-lying preferences Preference judgements have the merit of

face validity (Samuelson 1938) They also have the greatadvantage of not requiring any explicit deWnition of lsquobeautyrsquo either in the instructions given to the participantor in the interpretation of results Rather each individualrsquospreference judgements are taken to reveal their personal yetimplicit concept of beauty However preference judge-ments are not very informative about which aspects of a

stimulus lead subjects to like or dislike it Future studiesmight combine a large-scale psychometric approach toidentify speciWc stimulus features associated with speciWcaesthetic evaluations and then translate these to quantitativeneuroscientiWc studies

Acknowledgments We are grateful to Deborah Bull Tom SapsfordMavin Khoo and Matteo Candidi for advice and assistance This workwas supported by grants from the Economics and Social ScienceResearch Council (ESRCmdashPTA-026-27-1587) City UniversityLondon Fellowship and Ramon y Cajal Fellowship to BC-M aresearch fellowship of the German Academic Exchange Service(DAAD) to GO grants from the IRCCS ldquoE Medeardquo (Ricerca Corrente

2009 Italian Ministry of Health) to CU a BBSRC project grantLeverhulme Trust project grant and Leverhulme Trust Major ResearchFellowship to PH a BBSRC-ISIS grant to PH and SMA and theMinistero Italiano Universitagrave e Ricerca Italy University of RomeldquoLa Sapienzardquo and the IRCCS Fondazione Santa Lucia to SMA

References

Abbas ZA Duchaine B (2008) The role of holistic processing in judg-ment of facial attractiveness Perception 37(8)1187ndash1196

AstaWev SV Stanley CM Shulman GL Corbetta M (2004) Extrastri-ate body area in human occipital cortex responds to the perfor-mance of motor actions Nat Neurosci 7542ndash548

Berlyne DE (1974) Studies in the new experimental aesthetics stepstoward an objective psychology of aesthetic appreciation Hemi-sphere Publishing Washington DC

Briggs GG Nebes RD (1975) Patterns of hand preference in a studentpopulation Cortex 11230ndash238

Brown WM Cronk L Grochow K Jacobson A Liu CK Popovic ZTrivers R (2005) Dance reveals symmetry especially in youngmen Nature 4381148ndash1150

Calvo-Merino B Glaser DE Gregravezes J Passingham RE Haggard P(2005) Action observation and acquired motor skills an FMRIstudy with expert dancers Cereb Cortex 151243ndash1249

Calvo-Merino B Gregravezes J Glaser DE Passingham RE Haggard P(2006) Seeing or doing InXuence of visual and motor familiarityin action observation Curr Biol 161905ndash1910

Calvo-Merino B Jola C Glaser DE Haggard P (2008) Towards a sen-

sorimotor aesthetics of performing art Conscious Cogn 17911ndash922Cela-Conde CJ Marty G Maestu F Ortiz T Munar E Fernandez A

Roca M Rossello J Quesney F (2004) Activation of the prefron-tal cortex in the human visual aesthetic perception Proc NatlAcad Sci USA 1016321ndash6325

Chan AW Peelen MV Downing PE (2004) The eV ect of viewpoint onbody representation in the extrastriate body area Neuroreport152407ndash2410

David N Cohen MX Newen A Bewernick BH Shah NJ Fink GRVogeley K (2007) The extrastriate cortex distinguishes betweenthe consequences of onersquos own and othersrsquo behavior Neuroimage136(3)1004ndash1014




1 3

Di Dio C Macaluso E Rizzolatti G (2007) The golden beauty brainresponse to classical and renaissance sculptures PLoS ONE2e1201

di Pellegrino G Fadiga L Fogassi L et al (1992) Understanding motorevents a neurophysiological study Exp Brain Res 91(1)176ndash180

Downing PE Jiang Y Shuman M Kanwisher N (2001) A cortical areaselective for visual processing of the human body Science2932470ndash2473

Fechner GT (1876) Vorschule der Aumlsthetik Teil 1 Breitkopf amp HaumlrtelLeipzig

Frith CD Frith U (2007) Social cognition in humans Curr Biol17R724ndashR732

Helmich RC de Lange FP Bloem BR Toni I (2007) Cerebral compen-sation during motor imagery in parkinsonrsquos disease Neuropsych-ologia 1145(10)2201ndash2215

Jacobsen T Buchta K Koumlhler M Schroumlger E (2004) The primacy of beauty in judging the aesthetics of objects Psychol Rep94(3)1253ndash1260

Jacobsen T Schubotz RI Hofel L Cramon DY (2006) Brain correlatesof aesthetic judgment of beauty Neuroimage 29276ndash285

Kammer T (1999) Phosphenes and transient scotomas induced bymagnetic stimulation of the occipital lobe their topographic rela-tionship Neuropsychologia 37191ndash198

Kawabata H Zeki S (2004) Neural correlates of beauty J Neurophys-iol 911699ndash1705Keysers C Wicker B Gazzola V Anton JL Fogassi L Gallese V

(2004) A touching sight SIIPV activation during the observationand experience of touch Neuron 42335ndash346

Leder H Benno B Andries O Dorothee A (2004) A model of aestheticappreciation and aesthetic judgments Br J Psychol 95(4)489ndash508

Livio M (2002) The golden ratio The story of Phi the extraordinarynumber of nature art and beauty Headline Book PublishingLondon

McClure SM Li J Tomlin D Cypert KS Montague LM MontaguePR (2004) Neural correlates of behavioral preference for cultur-ally familiar drinks Neuron 44379ndash387

McManus IC Edmondson D Rodger J (1985) Balance in pictures BrJ Psychol 76311ndash324

Morin O Gregravezes J (2008) What is ldquomirrorrdquo in the premotor cortexA review Neurophysiol Clin 38189ndash195Moro V Urgesi C Pernigo S Lanteri P Pazzaglia M Aglioti SM

(2008) The neural basis of body form and body action agnosiaNeuron 23235ndash246

Nadal M Munar E Capoacute MA Rosselloacute J Cela-Conde CJ (2008)Towards a framework for the study of the neural correlates of aesthetic preference Spat Vis 21379ndash396

Orgs G Dombrowski JH Heil M Jansen-Osmann P (2008) Expertisein dance modulates alphabeta event-related desynchronizationduring action observation Eur J Neurosci 273380ndash3384

Peelen MV Downing PE (2005) Selectivity for the human body in thefusiform gyrus J Neurophysiol 93603ndash608

Peelen MV Downing PE (2007) The neural basis of visual body per-ception Nat Rev Neurosci 8(8)636ndash648

Reed CL Farah MJ (1995) The psychological reality of the body sche-ma a test with normal participants J Exp Psychol Hum PerceptPerform 21334ndash343

Reed CL Stone VE Bozova S Tanaka J (2003) The body-inversioneV ect Psychol Sci 14302ndash308

Samuelson PA (1938) A note on the pure theory of consumersrsquo behav-ior Economica NS 561ndash71

Sebanz N Knoblich G Prinz W (2005) How two share a task corep-resenting stimulus-response mappings J Exp Psychol Hum Per-cept Perform 311234ndash1246

Shaw TL (1962) Hypocrisy about art Stuart Publications BostonTalairach J Tournoux P (1988) Co-planar stereotaxic atlas of the

human brain 3-dimensional proportional systemmdashan approach tocerebral imaging Thieme New York

Taylor JC Wiggett AJ Downing PE (2007) Functional MRI analysisof body and body part representations in the extrastriate and fusi-

form body areas J Neurophysiol 981626ndash1633Urgesi C Berlucchi G Aglioti SM (2004) Magnetic stimulation of extrastriate body area impairs visual processing of nonfacial bodyparts Curr Biol 142130ndash2134

Urgesi C Calvo-Merino B Haggard P Aglioti SM (2007) Transcra-nial magnetic stimulation reveals two cortical pathways for visualbody processing J Neurosci 278023ndash8030

Vartanian O Goel V (2004) Neuroanatomical correlates of aestheticpreference for paintings Neuroreport 15893ndash897

Vartanian O Martindale C Podsiadlo J Overbay S Borkum J (2005)The link between composition and balance in masterworks vspaintings of lower artistic quality Br J Psychol 96493ndash503

Wassermann EM (1998) Risk and safety of repetitive transcranial mag-netic stimulation report and suggested guidelines from the Inter-national Workshop on the Safety of Repetitive Transcranial

Magnetic Stimulation June 5ndash7 1996 Electroencephalogr ClinNeurophysiol 1081ndash16Wicker B Keysers C Plailly J Royet JP Gallese V Rizzolatti G

(2003) Both of us disgusted in My insula the common neuralbasis of seeing and feeling disgust Neuron 40655ndash664

Zeki S (1999) Inner vision Oxford University Press OxfordZeki S Lamb M (1994) The neurology of kinetic art Brain 117607ndash

636



Copyright of Experimental Brain Research is the property of Springer Science amp Business Media BV and its

content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holders

express written permission However users may print download or email articles for individual use




1 3

motor (Di Pellegrino et al 1992 Calvo-Merino et al 20052006) and aV ective responses (Wicker et al 2003) in sev-eral cortical areas triggered by viewing conspeciWcs Theseresponses are often interpreted with reference to a hypothe-sis of lsquothe social brainrsquo (Frith and Frith 2007) This viewemphasises how the brain represents the behaviour andmental states of others in order to learn from them and

interact with them either competitively or co-operatively(Sebanz et al 2005)

Here we focus on a less-studied aspect of visual pro-cessing of bodies namely aesthetics Artistic activity isthought to be a uniquely human behaviour (Cela-Condeet al 2004) associated with development of speciWc corti-cal circuits Art objects may be considered to activate brainnetworks that generate aesthetic experiences Howeverseveral diV erent views exist in the literature SpeciWcallyneural correlates of aesthetic experience have been pro-posed in reward regions of the brain (ie orbitofrontal cor-tex) (Kawabata and Zeki 2004 Vartanian and Goel 2004)

in emotional centres such as the amygdala (Di Dio et al2007) in specialised visual perceptual areas (Zeki andLamb 1994) in dorsolateral prefrontal regions associatedwith higher lsquoexecutiversquo functions such as monitoring (Cela-Conde et al 2004) and with frontomedian regions underly-ing social and moral judgment (Jacobsen et al 2006) (seeNadal et al 2008 for a comprehensive review)

Aesthetic objects are described using speciWc labelsincluding but not limited to lsquobeautyrsquo (Jacobsen et al 2004)In the case of performing arts such as dance the observerrsquosaesthetic experience is presumably grounded in theresponses of their neural sensory motor and aV ective cir-cuits to the expressive actions of the dancerrsquos body Consis-tent with this account viewing dance recruited a network of parietal and premotor areas in a manner dependent on theviewerrsquos previous sensorimotor experience (Calvo-Merinoet al 2005) Importantly dance movements and dance pos-

tures may be judged beautiful or otherwise quite indepen-dently of whether the dancer is judged to be personallyattractive or not (Brown et al 2005) We therefore conjec-tured that brain circuits specialised for representing thebodies and actions of conspeciWcs might also underlie aes-thetic experience associated with dance

Aesthetics has a long history in both psychology andneurology (Fechner 1876) Psychophysical studies aimed atidentifying stimulus features producing positive and nega-tive aesthetic evaluations (Fechner 1876 McManus et al1985) More recent neuroscientiWc studies investigated neu-ral correlates of aesthetic evaluation These studies typi-cally used a lsquosubjectivistrsquo approach presenting a widerange of stimuli and comparing the responses for thoseliked or found lsquobeautifulrsquo to responses for those dislikedor found lsquouglyrsquo while acknowledging that participantsdiV er in their evaluation of any particular stimulus For

example Cela-Conde et al (2004) found that liked pictureselicited stronger prefrontal cortex activations than dislikedpictures Kawabata and Zeki (2004) found stronger activa-tion of orbitofrontal cortex for pictures found beautiful thanfor pictures judged ugly and stronger activation of sensori-motor cortex for the opposite contrast Calvo-Merino et al(2008) applied this approach to dance actions They found

stronger activity in occipital cortex bilaterally and in theright premotor cortex when six subjects viewed short dancepassages that they reported liking in a later evaluationcompared to those they disliked Subjectivist approachesare well suited to small-scale studies of the neural bases of aesthetic experience but cannot explain why particular stim-uli produce particular experiences Only one neuroaestheticstudy relevant to body representation has focussed onobjective stimulus properties to our knowledge Di Dioet al (2007) found stronger neural activity in both occipitalcortex and right anterior insula for images of statues obey-ing the golden section a principle of spatial proportion tra-

ditionally felt to be beautiful than for statues not followingthis principle

Importantly both subjectivist and objectivist neuroim-aging studies have the weakness of being correlationalActivations that correlate with lsquobeautyrsquo or liking could bepurely epiphenomenal and may not indicate the neural cir-cuits that actually underlie aesthetic experience Interven-tion studies in contrast can reveal brain areas or circuitsactively involved in aesthetic evaluation Here followingthe principle of perceptual selectivity (Zeki and Lamb1994) we investigated whether body-sensitive areas alsocontribute to aesthetic experience of dance perception Werecently (Urgesi et al 2007) proposed a dual-route modelof visual body perception by identifying two brain areasinvolved in visual body perception using repetitive trans-cranial magnetic stimulation (rTMS) We suggested thatthe extrastriate body area (EBA) an occipital area specia-lised for bodies (Downing et al 2001) houses a localrepresentation of body parts while the ventral premotorcortex (vPMC) houses a conWgural representation of com-plete body postures SpeciWcally we found impairedperception of body postures presented either upright orinverted (suggesting an analytical or local way of process-ing bodies) following EBA rTMS while rTMS over theleft premotor cortex impaired perception of upright but notinverted bodies (suggesting a global or conWgural type of processing) (Reed and Farah 1995 Reed et al 2003) Innormal function these two routes presumably providecomplementary information which is combined to producea single body percept Here we investigated the contribu-tions of these two routes to aesthetic evaluation of bodypostures by comparing the eV ects of EBA rTMS andvPMC rTMS with sham rTMS in an aesthetic preferencetask




1 3

Methods

Subjects



1998)

Stimuli




Trial structure




rTMS

Delay150 after

sample onset

500ms 500ms

100ms100ms

500 ms

100ms

500 ms

Prefer

1st or 2nd

500 ms

100ms

500 ms

Prefer

1st or 2nd


rTMS

Delay150 after

sample onset

500ms500ms 500ms500ms


500 ms500 ms

100ms100ms

500 ms500 ms

Prefer

1st or 2nd

500 ms500 ms

100ms100ms

500 ms500 ms

Prefer

1st or 2nd




1 3




TMS


















1 3






Discussion








1 3












Control site(sham)


Configuralsystem

Local

s stem

V1



Configuralsystem

V1 S s i t e

area

Local


system

Configural

r T M

processingarea


sys em

Localsystem

V1




1 3





















1 3











References
















1 3





































636









1 3

Methods

Subjects



1998)

Stimuli




Trial structure




rTMS

Delay150 after

sample onset

500ms 500ms

100ms100ms

500 ms

100ms

500 ms

Prefer

1st or 2nd

500 ms

100ms

500 ms

Prefer

1st or 2nd


rTMS

Delay150 after

sample onset

500ms500ms 500ms500ms


500 ms500 ms

100ms100ms

500 ms500 ms

Prefer

1st or 2nd

500 ms500 ms

100ms100ms

500 ms500 ms

Prefer

1st or 2nd




1 3




TMS


















1 3






Discussion








1 3












Control site(sham)


Configuralsystem

Local

s stem

V1



Configuralsystem

V1 S s i t e

area

Local


system

Configural

r T M

processingarea


sys em

Localsystem

V1




1 3





















1 3











References
















1 3





































636









1 3




TMS


















1 3






Discussion








1 3












Control site(sham)


Configuralsystem

Local

s stem

V1



Configuralsystem

V1 S s i t e

area

Local


system

Configural

r T M

processingarea


sys em

Localsystem

V1




1 3





















1 3











References
















1 3





































636











1 3






Discussion








1 3












Control site(sham)


Configuralsystem

Local

s stem

V1



Configuralsystem

V1 S s i t e

area

Local


system

Configural

r T M

processingarea


sys em

Localsystem

V1




1 3





















1 3











References
















1 3





































636









1 3






Discussion








1 3












Control site(sham)


Configuralsystem

Local

s stem

V1



Configuralsystem

V1 S s i t e

area

Local


system

Configural

r T M

processingarea


sys em

Localsystem

V1




1 3





















1 3











References
















1 3





































636









1 3












Control site(sham)


Configuralsystem

Local

s stem

V1



Configuralsystem

V1 S s i t e

area

Local


system

Configural

r T M

processingarea


sys em

Localsystem

V1




1 3





















1 3











References
















1 3





































636









1 3





















1 3











References
















1 3





































636









1 3











References
















1 3





































636









1 3





































636











Documents

avaliacao estetica