ContentslistsavailableatScienceDirect
JournalofExperimentalSocialPsychology
journalhomepage:www.elsevier.com/locate/jesp
FlashReport
Thetwosidesofspontaneity:Movementonsetasymmetriesinfacialexpressionsinfluencesocialjudgments☆
EvanW.Carra,⁎,SebastianKorbb,c,PaulaM.Niedenthalb,c,PiotrWinkielmana,daUniversityofCalifornia—SanDiego,DepartmentofPsychology,9500GilmanDrive0109,LaJolla,CA92093,USA
UniversityofWisconsin—Madison,DepartmentofPsychology,1202W.JohnsonStreet,Madison,WI53706-1969,USAcSwissCenterforAffectiveSciences,UniversityofGeneva,7ruedesBattoirs,1205Geneva,SwitzerlanddUniversityofSocialSciencesandHumanities,Warsaw,Poland
bHIGHLIGHTS•••••
Videosdisplayedonsetasymmetries(OASs)inavatars’emotionalfacialexpressions.WealsovariedOAStimingforthelefthemi-face(LHF)andrighthemi-face(RHF).PerceiversmorequicklyandaccuratelydetectedOASsintargets’LHFexpressions.PerceiversjudgedexpressionswithearlierLHFonsetsasmorespontaneous.
LHFangerwasmoreaccuratelydetectedat20msOAS(withnobiasforhappiness).
articleinfoabstract
Whenformingbasicsocialimpressions,itisimportanttoquicklyandaccuratelyclassifyfacialexpressions(includingtheirspontaneity).Earlystudiesonemotionperception,employingstaticpicturesinthechimeric-faceparadigm,demonstratedthatexpressionsshownonthelefthemi-face(LHF)wereratedasmoreintense,comparedtotherighthemi-face(RHF).Interestingly,recentstudiesonemotionproduction,usinghigh-speedvideorecordings,discoveredanonsetasymmetry(OAS)suchthatspontaneousexpressionsstartearlierintheLHF,whileposedexpressionsstartintheRHF.Here,usinghighlycontrolledanddynamicallydevelopingvideostimuliofavatarfaces,wetestedwhetherOASsinperceivedfacesinfluencetheefficiencywithwhichanexpres-sionisclassified,aswellasjudgmentsofexpressionintensity,spontaneity,andtrustworthiness.Videosofavatarsmakinghappyandangryexpressions,withOASsofeither20or400ms,werejudgedonseveralsocialdimensionsby68participants.TheresultshighlighttheimportanceoftheLHFforemotionclassificationsandsocialjudg-ments:ExpressionswithearlierLHFonsetswerenotonlyjudgedtobemorespontaneousbutwerealsodetectedmorequicklyandaccurately(adifferencethatwasmostevidentforangryexpressionswithabrieflypresentedOAS,butnotforhappyexpressions).Generally,thesefindingsunderscorehowadaptivesocialpercep-tionreliesonsubtlecuesinthedynamicsofemotionalfacialexpressions.
©2014ElsevierInc.Allrightsreserved.
Articlehistory:
Received6March2014Revised8May2014Accepted9May2014
Availableonline17May2014Keywords:Emotions
FacialexpressionsSpontaneity
Socialjudgments
Introduction
Emotionalfacialexpressionsareinformativesynthesesofindivid-uals'feelings,motivations,andintentions,thusmakingtheefficientdecodingofthosedisplaysanimportantsocialskill(Ekman&Friesen,1974;Wagner,MacDonald,&Manstead,1986).Onecomponentof
☆EvanW.CarrconductedthisresearchwithgovernmentsupportunderandawardedbytheUnitedStatesDepartmentofDefense(DoD),AirForceOfficeofScientificResearch,NationalDefenseScienceandEngineeringGraduate(NDSEG)Fellowship,32CFR168a.SebastianKorbwassupportedbyanEarlyPostdocMobilityscholarshipbytheSwissNationalScienceFoundation(PBGEP1-139870).⁎Correspondingauthor.
E-mailaddress:Ewcarr@ucsd.edu(E.W.Carr).
decodingaccuracyistheabilitytodistinguishbetweenspontaneous(i.e.,genuinelyfelt)andposed(i.e.,intentionallyevoked)expressions.Suchdifferentiationisrequiredforsmoothsocialinteractions,giventhatmistakingaposedexpressionforaspontaneousone(orviceversa)canhaveseriousinterpersonalconsequences.Accordingly,muchresearchinsocialpsychologyandneurosciencefocusesonthisessentialability(Bernstein,Young,Brown,Sacco,&Claypool,2008;Maringer,Krumhuber,Fischer,&Niedenthal,2011;McLellan,Johnston,Dalrymple-Alford&Porter,2010;Zuckerman,Hall,DeFrank,&Rosenthal,1976).Thepresentresearchtestedtheideathatspontaneitycueslieinleft–rightasymmetriesintheonsetoffacialexpressions.
Considerableevidencesuggeststhatemotionalfacialexpressionsareproducedandperceivedasymmetrically(Borodetal.,1998).For
http://dx.doi.org/10.1016/j.jesp.2014.05.008
0022-1031/©2014ElsevierInc.Allrightsreserved.
32E.W.Carretal./JournalofExperimentalSocialPsychology55(2014)31–36
example,researchusingthechimeric-faceparadigmfoundthatindivid-ualsrateleft–leftcompositepicturesoffacialexpressionsasmorein-tensethanright–rightcomposites(Lindell,2013;Nicholls,Wolfgang,Clode,&Lindell,2002;Sackeim,Gur,&Saucy,1978).Suchfindingssup-porttherighthemispherehypothesisfortheencodinganddecodingofemotionalstimuli(Dimberg&Petterson,2000;Sackeimetal.,1982;Schwartz,Davidson,&Maer,1975),andonthishypothesis,greaterfacialexpressivityoccursoverthelefthemi-face(LHF),comparedtotherighthemi-face(RHF)(Korb&Sander,2009;Rinn,1984).Phyloge-neticrootsofsuchproductionandperceptionasymmetriesaresug-gestedbysimilarresultsinnon-humanprimates(Fernández-Carriba,Loeches,Morcillo,&Hopkins,2002;Wallez&Vauclair,2013).
Newerfindingssuggestthattheonsetoffacialexpressionsisalsolateralizedandthatthesideofthisonsetasymmetry(OAS)dependsonwhethertheexpressionisspontaneousorposed.Usinghigh-speedvideographytofilmparticipantsduringtheproductionofdifferentfacialexpressions,RossandPulusu(2013)foundthatspontaneousexpressionsbeginpredominantlyintheLHF(approximately20msear-lier),whileposedexpressionsbeginintheRHF(approximately10msearlier)—apatternthatwasmostrobustforupper-faceexpressions,suchasanger.ThecurrentresearchtestedthehypothesisthatshortOASsintheproductionoffacialexpressions(approximately10–20ms)arerelieduponintheperceptionofthevolitionalnatureofthoseexpressions.Morespecifically,wetestedthepredictionthatOASdifferencesintheproductionofspontaneousandposedexpressionsarerecruitedinjudgingthespontaneityofperceivedexpressions,aswell.
Cruciallythough,investigatingthesesubtlemarkers(suchasOAS)isimportant,especiallygiventhatboththeproductionandperceptionoffacialexpressionsareassociatedwithnumerousdifferencesinprocessingandbehavior:Asexamples,“true”and“fake”smilesdifferinbothdynamicalandmorphologicalfeatures(e.g.,Ambadar,Cohn,&Reed,2009;Hess&Kleck,1994),specificbrainstatesareassociatedwithboththeirproductionandperception(Ekman,Davidson,&Friesen,1990;McLellan,Wilcke,Johnston,Watts,&Miles,2012),andourabilitytodistinguishbetweentheseexpressionsseemstoberelativelyautomatic(McLellan,Johnston,Dalrymple-Alford,&Porter,2010;Miles&Johnston,2007).Inturn,perceiversmayquicklyandeffi-cientlyuseOASsinfacialexpressionstoinferthespontaneityofthedisplayer,whichcouldthenhavedownstreamconsequencesontheresultantsocialinteraction.
Withthisframework,wecreatedhighlycontrolled,preciselytimed,anddynamicallydevelopingavatarvideostimulitotesttheeffectsoftwoOASdurationsonparticipants'perceptionsofhappyandangryex-pressions.Wewereinterestedinevaluatingtheseeffectsinconjunctionwiththreemajorsocialdimensions:First,theprimaryjudgmentofin-terestwasspontaneity:Aswithproduction(Ross&Pulusu,2013),weexpectedthattheperceptionofemotionalexpressionswithearlierLHFonsetswouldresultinhigherspontaneityratings.Second,wealsomeasuredperceivedintensity(ortheperceivedlevelofexpressionactivationinthestimulus)toexaminewhetherOAScuesamplifythesubjective“strength”ofanemotionalexpression,andtoassesswhetherornotexpressionOASsandintensitieshavedissociableeffects.Sinceweobjectivelycontrolledforexpressionintensityacrosshemi-fieldsinourhigh-resolutionvideostimuli(therebyonlymanipulatingframeratestovarydynamicasymmetries),itwascriticaltoevaluatewhetherornotparticipantssubjectivelyviewedtheexpressionsasmoreorlessintense,dependingonOAS—animportantconsideration,givenrecentfindingsusingthechimeric-faceparadigm,whichnecessarilyvariesintensityacrosshemi-fields(e.g.,Indersmitten&Gur,2003).Finally,wealsogatheredtrustworthinessratingstogaugetheimpactofOAScuesonhigher-leveldispositionaljudgmentsofcharacter,whichhavebeenreportedwithmodernchimeric-facestudies(e.g.,Okubo,Ishikawa,&Kobayashi,2013).
GiventhattheRossandPulusu(2013)findingsshowthemostsubstantialproductioneffectsforupper-faceemotionalexpressions
(i.e.,anger),weexpectedthattheperceptionofangerwouldalsobemostinfluencedbythelateralizationofOASs.EventhoughthisRossandPulusu(2013)productioneffectmightbeconsideredsurprisinginlightofthefactthattheupper-faceismorebilaterallyinnervated(i.e.,bysupplementarymotorarea[SMA]andtherostralcingulatemotorcortex;Korb&Sander,2009;Morecraft,Louie,Herrick,&Stilwell-Morecraft,2001)thanthelowerface,otherworkhashighlight-edtheimportanceoftheupper-faceinoverallexpressionproduction(Ross,Prodan,&Monnot,2007;Ross,Reddy,Nair,Mikawa,&Prodan,2007)andinconveyingmore“true”orgenuineemotion(Ross,Shayya,Champlain,Monnot,&Prodan,2013).Assuch,anyOASeffectsonemo-tionperceptionshoulddependonwhethertheexpressionstartsintheupper-face(i.e.,anger)versuslower-face(i.e.,happiness),sincetheseregionsdifferinbothneuroanatomicalorganizationandbehavioralfunc-tion(Matsumoto&Lee,1993;Ross,Prodan,&Monnot,2007).Thesedemonstrationsofproductiondifferencesforangersuggestthatper-ceiversshouldpayspecialattentiontotheupper-face(alongwithanyOASs)whenmakingspontaneityjudgmentsoftargetexpressions.
Moreover,wealsoexpectedtheseangereffectstobedependentonthetimingoftheOAS,wherebymoredifferenceswouldemergeforangerexpressionswithashorterOAS.ThispredictionfollowsdirectlyfromtheoriginalproductionfindinginRossandPulusu(2013),inwhichOASsusuallyrangedfromonly10to20ms.Further,recentfindingshighlightthatangerexpressionscaptureandguideattention,especiallywhenpresentedatshorterdurations(Blagrove&Watson,2010,2014).Suchfindingsareconsistentwiththenotionofmoreefficientprocessingofangerexpressions,thuscorroboratingmoderndemonstrationsoftheangersuperiorityeffect(orASE;Calvo,Avero,&Lundqvist,2006;Dickins&Lipp,2014;Öhman,Lundqvist,&Esteves,2001;Schubö,Gendolla,Meinecke,&Abele,2006),whichemergesearlyindevelopment(LoBue,2009)andoccursrobustlywhenusingactualphotos,multipleidentities,andrealisticdynamicallydevelopingexpressions,suchasthoseusedinthecurrentwork(Ceccarini&Caudek,2013;Horstmann&Bauland,2006;Pinkham,Griffin,Baron,Sasson,&Gur,2010).
Insum,weexpectedgreaterspontaneityratingsandmoreefficientOASdetectionforemotionalexpressionswithearlierLHFonsets,andwepredictedthattheseeffectswouldbeamplifiedforanger(especiallywhenpresentedforshorterOASdurations).MethodsParticipants
Sixty-eightundergraduatestudentsattheUniversityofCalifornia,SanDiego(Mage=21.10years,SDage=2.25years;sixmales)partici-patedforcoursecredit.1Stimuli
Fourfullysymmetricalimagesofavatarfaces(twomaleandtwofemale)wereconstructedusingFaceGenModeller3.5(SingularInver-sions,Inc.).Wecreated32differentdynamicstimulibycrossingthefactorsAvatar(4),Emotion(2:happy,angry),OAS_Side(2:right,left),andOAS_Time(2:20ms,400ms),usingFacsGensoftware(Roeschetal.,2011),basedontheFacialActionCodingSystem(FACS;Ekman&Friesen,1978).Theavatars'facialdynamicsweremanipulatedbymodifyingcompositesofthealternatehemi-faceatdifferentframes(eachcontaining150frames,totalingthreesecondsofvideoat50fps),accordingtowhethertheexpressionexhibitedanOASof20msor400ms(seeSupplementaryMaterials).Allvideosstartedwithaneutralexpression,whichthenconvertedintoexpressionsofhappiness(AU12=100%,AU6=70%)oranger(AU4=100%,AU5/AU7=50%).
1AllstatisticspresentedintheResultssectionremainsignificantwhenexcludingthe6maleparticipantstogetanall-femalesample.
E.W.Carretal./JournalofExperimentalSocialPsychology55(2014)31–3633
Fig.1.Theexperimentaldesignconsistedoftwophases.Tominimizedemandcharacteristics,asymmetryjudgmentswereonlyaskedinthesecondphase.Theorderofphaseswasidenticalforallparticipants.
Procedure
Theexperimentconsistedoftwophases,completedinthesameorderbyallparticipants(seeFig.1).
Phase1containedthreecounterbalancedblocksof32trialseach,to-taling96trials.Oneachblock,participantsratedtheintensity,spontane-ity,ortrustworthinessofthepresentedexpressionwithhorizontallypresentedscalesfrom1(notatall)to7(verymuch),usingthenumberkeysonthekeyboard.Eachtrialwasprecededbyafixationcross(durationbetween1500and2000ms),followedbythe3000msvideostimulus,andtheratingscreen(seeSupplementaryMaterialsfortaskinstructions).2Phase2included32trialsoftwodifferenttwo-alternativeforced-choicedetectiontasks:Participantsfirstreportedwhichemotionthefaceexpressed(“happy/angry,”withsideofresponsescounterbalanced)andthenindicatedthesideofthefaceonwhichtheexpressionbegan.ParticipantsmadetheemotionandOASjudgmentswiththeleft-and
right-arrowkeys(usingtheirleft-andright-indexfingers,respectively).TheOASmeasureswerenotincludedinPhase1andalwaystakenafteremotionclassificationsinPhase2,inordertominimizedemandcharac-teristics.Videoswithineachphaseandblockwererandomizedandcounterbalanced.Results
Socialjudgments(phase1)
Toevaluateparticipants'socialjudgments,weconductedrepeated-measuresANOVAsforeachofthethreeratings,accordingtoanEmotion(happy,angry)×OAS_Side(left,right)×OAS_Time(20ms,400ms)within-subjectsmodel.Notethatallthreesocialdimensions(i.e.,spon-taneity,intensity,andtrustworthiness)wereanalyzedusingseparatestatisticalmodels,eachfollowingthesameaforementionedfactorstructure.
Spontaneity
Aspredicted,participantsjudgedexpressionsstartingontheavatars'LHFtobemorespontaneous,asrevealedbyamaineffectofOAS_Side(F(1,67)=6.26,p=.02,d=.43);seeFig.2,panela).WealsoobservedanEmotion×OAS_Timeinteraction(F(1,67)=8.21,pb.01),
Notethatweseparatedeachsocialratingdimension(i.e.,intensity,spontaneity,andtrustworthiness)intoindividualcounterbalancedblocksof32trialseach,ratherthancollectingalltheratingsoneachtrial.Thiswasdonetoensurethatparticipantscouldeas-ilycompletethetask,putfullattentiontowardeachspecificrating,andtoallowforphase1instructionstobereiteratedbeforeandin-betweenblocks(alsoseeSupplementaryMaterials).
234E.W.Carretal./JournalofExperimentalSocialPsychology55(2014)31–36
Fig.2.Expressionsstartingontheleftsideoftheface(leftOAS)wereratedasmorespontaneous(a),andtheironsetwasdetectedmorerapidly(b),comparedtoexpressionsstartingontherightsideoftheface(rightOAS).Attheshorter20msOAS,participantsweremostaccurateindetectingonsetsforangerexpressionswithaleftOAS—bothcomparedtoangerexpres-sionswithrightOAS,andhappyexpressionswithleftOAS(c).Nosuchdifferenceswerefoundatthelonger400msOAS(**pb.01,*pb.05).
whichdemonstratedthatangryexpressionswiththe20msOASwerejudgedtobemorespontaneousthanexpressionswiththe400msOAS(t(67)=2.61,p=0.02[Bonferroni-corrected],d=.37).Thiswasnotthecaseforhappyexpressions(t(67)=1.44,ns).
Intensity
AnalysisoftheintensityjudgmentsrevealedamaineffectofEmotion(F(1,67)=73.89,pb.001,d=1.48),suchthatparticipantsratedangryexpressionsasmoreintensethanhappyexpressions.Noothereffectsforintensitywerefound.
Trustworthiness
WealsoobservedamaineffectofEmotionfortrustworthinessjudg-ments(F(1,67)=150.02,pb.001,d=2.10),suchthatparticipantsratedhappyexpressionsasmoretrustworthythanangryexpressions.Noothereffectsfortrustworthinesswerefound.EmotionandOAScategorizations(phase2)
WeanalyzedtheRTs(log10-transformed,toreducetheimpactofoutliers)andaccuracyscoresofparticipants'emotionandOAScategori-zationsinseparaterepeated-measuresANOVAsbycrossingthefactorsEmotion(happy,angry)×OAS_Side(left,right)×OAS_Time(20ms,400ms).
Emotioncategorizations
Overall,participants'emotioncategorizationswerebothhighlyac-curate(MACC=97.21%,SEACC=.40%)andrapid(MRT=791.50ms,SERT=13.18ms).Specifically,onthesimplequestionabouttheavatar'semotion(seeFig.1,bottompanel),noeffectswerefoundforlog10-transformedRTs(Fsb1.40,psN.24)oraccuracyofemotioncategoriza-tionscores(Fsb2.08,psN.15).OASdetection
Importantly,ouranalysesofparticipants'speedandaccuracytode-tectthesideofexpressiononsetrevealedseveralmorespecificeffects.Unsurprisingly,whentheasymmetrylasted400ms,participantswerebothfaster(MRT=633.74ms,SERT=23.29ms)andmoreaccurate(MACC=96.60%,SEACC=1.23%)inindicatingthesideoftheexpressiononset,comparedto20ms(MRT=789.59ms,SERT=23.29ms;MACC=56.20%,SEACC=1.23%).ThiswasrevealedbymaineffectsofOAS_Timeforbothlog10-RTs(F(1,67)=64.71,pb.001,d=1.41)andOASdetec-tionaccuracy(F(1,67)=540.47,pb.001,d=4.21).
Interestingly,analysisofOASdetectionRTsrevealedthatpartici-pantswerefasterinrecognizingLHFexpressiononset(Fig.2,panelb).ThisisshownbyamaineffectofOAS_Sideonlog10-transformedOASdetectionRTs(F(1,67)=6.10,p=.02,d=.43).NootherrelevanteffectsonOASdetectionRTswereobserved.
Critically,whenevaluatingaccuracyscoresforOASdetection,partic-ipantsweremoreaccurateindetectingthesideoftheexpressiononsetforangrycomparedtohappyexpressions,asshownbyamaineffectofEmotion(F(1,67)=4.55,p=.04,d=.38).Further,thiseffectdependedonthesideoftheOAS,asdemonstratedbyanEmotion×OAS_Sideinter-action(F(1,67)=8.76,pb.01),whichrevealedthatparticipantsweremoreaccurateindetectingLHFangerasopposedtoRHFanger(t(67)=3.08,p=.01[Bonferroni-corrected],d=.53).Thiseffectwasnotpresentforhappyexpressions(t(67)=1.11,ns).
Additionally,anEmotion×OAS_Side×OAS_Timeinteraction(F(1,67)=6.95,p=.01)forOASdetectionaccuracyrevealedaLHFpri-orityforbrieflypresentedangerexpressions(Fig.2,panelc).Follow-uptestsrevealednodifferencesatthelonger400msOASbetweenLHFandRHFangryexpressions(t(67)=0.01,ns)orbetweenLHFandRHFhappyexpressions(t(67)=0.68,ns).However,participantsshowedgreateraccuracyindetectingLHFangryexpressionsattheshorter20msduration(MACC=65.80%,SEACC=2.66%)—bothcomparedtoRHFangryexpressions(t(67)=4.88,pb.001[Bonferroni-corrected],d=.54)andLHFhappyexpressions,(t(67)=3.22,pb.01
E.W.Carretal./JournalofExperimentalSocialPsychology55(2014)31–3635
[Bonferroni-corrected],d=.43).Nosimilardifferenceswereobservedforhappyexpressions.Discussion
Inthecurrentstudy,weusedpreciselytimedvideostimulitoinves-tigatehowonsetasymmetries(OASs)infacialexpressionsinfluenceparticipants'socialjudgmentsandclassificationsofthosedisplays.Wereportthefirstdemonstrationthatparticipantsarefasterandmoreaccurateindetectingexpressiononsetsstartinginatarget'slefthemi-face(LHF),comparedtotherighthemi-face(RHF),andthattheyratethoseLHFexpressionsasmorespontaneous.Thesefindingsareespeciallyinformativenotonlyfortheempiricalstudyoffacialex-pressionsandsocialperception,butalsoforbroadersocialpsychologicaltheoryinemotion(Feldman-Barrett,Niedenthal,&Winkielman,2005).
Inthefollowing,wereviewanddiscussthemajorfindings.First,perceiversmorequicklyandaccuratelydetectedexpressiononsetsintheLHF.ThisfindingisconsistentwithpastresultsthatindividualsjudgeLHFemotionsasmoreexpressive(e.g.,Failla,Sheppard,&Bradshaw,2003).Importantly,itextendsthecurrentliteraturebyshowingaLHFdecodingprioritywhenjudgingthesideofonsetofemo-tionalfacialexpressions.Recallthatweobservedthesedifferencesevenwhencontrollingforpeakexpressionactivationinourstimuli,suggest-ingthattheeffectsofexpressiononsetsandintensitiesaredissociable(andthatOASscanindependentlyinfluenceemotionclassificationandperception).
Second,wefoundthathigherspontaneityratingsweregiventotargets'expressionswithOASsstartingintheLHFcomparedtotheRHF.Whilerecentworkhasshownthatindividualsproducespontane-ousexpressionsearlierintheLHF(Ross&Pulusu,2013),thisisthefirstdemonstrationthatindividualsalsoperceiveothers'expressionsasmorespontaneouswhentheironsetisintheLHF.Critically,thesere-sultssuggestthatOAScuesmaybeusedtoseamlesslyinferspontaneityoffacialexpressions.Thisisparticularlyessential,sincerecentresearchonspontaneousversusposedexpressionsrelieslargelyontheanalysisofstaticmorphologicalfacialfeatures(andisstilldebated).Forexam-ple,thecrinklingaroundtheeyesinaDuchennesmilehasbeenproposedasamarkerofgenuineness(Ekmanetal.,1990),butthischaracteristichasalsobeenfoundtobepresentin“false”smiles(Krumhuber&Manstead,2009).OurresultssuggestthatOASsareanimportantdynamicfacialfeaturethatcontributesnotonlytothepro-ductionof“true”and“false”expressionsbutalsototheperceptionofthesedisplays(Ambadar,Schooler,&Cohn,2005;Krumhuber&Kappas,2005).
Third,thisLHFpriorityinOASdetectionwasespeciallypronouncedfortheupper-faceemotion(anger),specificallyattheshorterdurationof20ms—inotherwords,thesalienceofanexpression'sOASdependsonbothitstypeandduration.Thisfindingisconsistentwithrecentworkontheangersuperiorityeffect(ASE;e.g.,Pinkhametal.,2010),whichisespeciallypronouncedfordynamicstimuli(Ceccarini&Caudek,2013).Interestingly,participantswereabout20%moreaccu-rateindetectingangeronsetsintheLHFduringshort20msOASs,whilenosucheffectoccurredat400msOASs.Thissuggeststhatangercuesmaybeparticularlyrelevantincapturingandguidingatten-tionwhenperceivedquicklyandperhapsunconsciously(Blagrove&Watson,2010,2014),sinceoverallaccuracyin20msOASdetectionwasapproximatelychance-level.However,itisindeedpossiblethatafailuretoobservehemi-facedifferencesinaccuracyatthelonger400msOASmayhavebeencausedbyceilingeffects,duetonear-perfectperformance.
Notethattheselectivenatureoftheseeffects(basedonemotion,alongwithboththesideanddurationofOAS)demonstratesthatourre-sultscannotbeabyproductofageneralLHF“sidebias.”Corresponding-ly,spontaneityjudgmentsalsofollowedaselectivepattern,withangerbeingratedasmoreauthenticat20msOASscomparedto400msOASs(whereasnodifferencewasfoundforhappiness).Eventhough
theeffectsofthesideofOASondetectionaccuracyseemtobethemostcomplex,thesharedinfluenceoftheLHFonspontaneityratingsandOASdetectionRTs(alongwithOASaccuracy)certainlysuggeststheimportanceofthesedynamicasymmetriesintheperceptionofemotionalexpressions.
Asmentioned,theOASsseemedtomostlyinfluencetheperceptionoftheupper-faceemotion(i.e.,angerexpressions)withlimitedinfluenceonthelower-faceemotion(i.e.,happyexpressions).Suchdif-ferencescanbeunderstoodinthecontextofthedistinctivesocialfunc-tionandneuroanatomicalorganizationforeachoftheseexpressions(Ross,Reddy,etal.,2007;Ross,Prodan,&Monnot,2007;Ross&Pulusu,2013).Forexample,theperceptionofangerisfunctionallyandneurologicallydifferentfromrespondingtosmiles,sinceindivid-ualsencounter“real”angerlessfrequentlyinthesocialenvironment(e.g.,Calder,Keane,Lawrence,&Manes,2004).Moreover,smilesrepre-sentthemostcomplexemotionalexpression,bothintheintentionsthattheycanconveyandthefactorsbywhichtheycanbeinfluenced(e.g.,eyegaze,priorknowledge,socialinhibition,etc.;Niedenthal,Mermillod,Maringer,&Hess,2010).Consequently,OAScuesmorereadilyaffectangerlikelybecausesmilingresponsestakeintoaccountsomanyexternalsocialfactors,especiallywhenstimuliaredynamic,aswasthecasewiththecurrentparadigm(Krumhuber&Kappas,2005).Andfromsocialandevolutionaryperspectives,itisadvantageoustobeabletorapidlydetect“genuine”anger,sincethisemotioncanrepresentanimmediatethreatsituationfortheperceiver(e.g.,Fischer&Roseman,2007).
RecallthatwefoundnodifferencesbyOASfortrustworthinessratingswhencontrollingforpeakintensityinourstimuli,incontrasttorecentresearchusingchimericfaces(Okuboetal.,2013).Thismayindicatethatshiftsinhigher-leveljudgmentsofdispositional“charac-ter”areactuallydrivenbytheexpression'sintensity.Also,judgmentsaboutthevolitionalnatureoftheexpression(i.e.,posedversussponta-neous)shouldbedistinguishedfromjudgmentsaboutthedisplayer'sintent(i.e.,trustworthyversusuntrustworthy)becausemanyposedex-pressions(e.g.,“affiliative”smile)havenodeceitfulintent(Niedenthaletal.,2010).Thus,OASsmaybeparticularlyusefulinresearchinghowindividualsinitiatejudgmentsofgenuineness(e.g.,“true”versus“fake”smiles)comparedtojudgmentsofcharacter(e.g.,“dependable”versus“deceitful”target).
Inconclusion,thisisthefirstdemonstrationthatperceiversmorequicklyandaccuratelydetecttheonsetoffacialexpressionsstartingintheLHFcomparedtotheRHF(especiallyforangeratshortOASs)andthatOASsareusedtoinferexpressions'spontaneity.Generally,thesere-sultsaremostimportantinshowingthatindividualsareadeptatusingsubtlesignals,suchaslateralizedfacialexpressions,asinterpersonalcuesinnavigatingthesocialworld.AppendixA.Supplementarydata
Supplementarymaterialtothisarticlecanbefoundonlineathttp://dx.doi.org/10.1016/j.jesp.2014.05.008.References
Ambadar,Z.,Cohn,J.F.,&Reed,L.I.(2009).Allsmilesarenotcreatedequal:Morphology
andtimingofsmilesperceivedasamused,polite,andembarrassed/nervous.JournalofNonverbalBehavior,33(1),17–34.
Ambadar,Z.,Schooler,J.W.,&Cohn,J.F.(2005).Decipheringtheenigmaticface:Theim-portanceoffacialdynamicsininterpretingsubtlefacialexpressions.PsychologicalScience,16(5),403–410.
Bernstein,M.J.,Young,S.G.,Brown,C.M.,Sacco,D.F.,&Claypool,H.M.(2008).Adaptive
responsestosocialexclusion:Socialrejectionimprovesdetectionofrealandfakesmiles.PsychologicalScience,19,981–983.
Blagrove,E.,&Watson,D.G.(2010).Visualmarkingandfacialaffect:Cananemotional
facebeignored?Emotion,10(2),147.
Blagrove,E.,&Watson,D.G.(2014).Ignoringrealfaces:Effectsofvalence,threat,and
salience.Attention,Perception,&Psychophysics,1–21.
Borod,J.C.,Cicero,B.A.,Obler,L.K.,Welkowitz,J.,Erhan,H.M.,Santschi,C.,etal.(1998).
Righthemisphereemotionalperception:Evidenceacrossmultiplechannels.Neuropsychology,12(3),446–458.
36E.W.Carretal./JournalofExperimentalSocialPsychology55(2014)31–36
Calder,A.J.,Keane,J.,Lawrence,A.D.,&Manes,F.(2004).Impairedrecognitionofanger
followingdamagetotheventralstriatum.Brain,127(9),1958–1969.
Calvo,M.G.,Avero,P.,&Lundqvist,D.(2006).Facilitateddetectionofangryfaces:Initial
orientingandprocessingefficiency.Cognition&Emotion,20(6),785–811.
Ceccarini,F.,&Caudek,C.(2013).Angersuperiorityeffect:Theimportanceofdynamic
emotionalfacialexpressions.VisualCognition,21(4),498–540.
Dickins,D.S.,&Lipp,O.V.(2014).Visualsearchforschematicemotionalfaces:Angry
facesaremorethancrosses.Cognition&Emotion,28(1),98–114.
Dimberg,U.,&Petterson,M.(2000).Facialreactionstohappyandangryfacialexpres-sions:Evidenceforrighthemispheredominance.Psychophysiology,37(5),693–696.Ekman,P.,Davidson,R.J.,&Friesen,W.V.(1990).TheDuchennesmile:Emotionalexpres-sionandbrainphysiology:II.JournalofPersonalityandSocialPsychology,58(2),342–353.Ekman,P.,&Friesen,W.V.(1974).Detectingdeceptionfromthebodyorface.Journalof
PersonalityandSocialPsychology,29(3),288.
Ekman,P.,&Friesen,W.V.(1978).Facialactioncodingsystem(FACS).Atechniqueforthe
measurementoffacialaction.Consulting,PaloAlto.
Failla,C.V.,Sheppard,D.M.,&Bradshaw,J.L.(2003).Ageandresponding-handrelated
changesinperformanceofneurologicallynormalsubjectsontheline-bisectionandchimeric-facestasks.Brain&Cognition,52(3),353–363.
Feldman-Barrett,L.,Niedenthal,P.M.,&Winkielman,P.(Eds.).(2005).Emotionand
Consciousness:GuilfordPress.
Fernández-Carriba,S.,Loeches,Á.,Morcillo,A.,&Hopkins,W.D.(2002).Asymmetryin
facialexpressionofemotionsbychimpanzees.Neuropsychologia,40(9),1523–1533.Fischer,A.H.,&Roseman,I.J.(2007).Beatthemorbanthem:Thecharacteristicsand
socialfunctionsofangerandcontempt.JournalofPersonalityandSocialPsychology,93(1),103.
Hess,U.,&Kleck,R.E.(1994).Thecuesdecodersuseinattemptingtodifferentiate
emotion-elicitedandposedfacialexpressions.EuropeanJournalofSocialPsychology,24(3),367–381.
Horstmann,G.,&Bauland,A.(2006).Searchasymmetrieswithrealfaces:Testingthe
anger-superiorityeffect.Emotion,6(2),193.
Indersmitten,T.,&Gur,R.C.(2003).Emotionprocessinginchimericfaces:Hemispheric
asymmetriesinexpressionandrecognitionofemotions.JournalofNeuroscience,23(9),3820–3825.
Korb,S.,&Sander,D.(2009).Theneuralarchitectureoffacialexpressions.InD.
Sander,&K.R.Scherer(Eds.),TheOxfordcompaniontoemotionandtheaffectivesciences(pp.173–175).NewYork:OxfordUniversityPress.
Krumhuber,E.G.,&Kappas,A.(2005).Movingsmiles:Theroleofdynamiccomponents
fortheperceptionofthegenuinenessofsmiles.JournalofNonverbalBehavior,29(1),3–24.
Krumhuber,E.G.,&Manstead,A.S.R.(2009).CanDuchennesmilesbefeigned?New
evidenceonfeltandfalsesmiles.Emotion,9(6),807–820.
Lindell,A.K.(2013).Continuitiesinemotionlateralizationinhumanandnon-human
primates.FrontiersinHumanNeuroscience,7,464.
LoBue,V.(2009).Morethanjustanotherfaceinthecrowd:Superiordetectionofthreatening
facialexpressionsinchildrenandadults.DevelopmentalScience,12(2),305–313.
Maringer,M.,Krumhuber,E.,Fischer,A.,&Niedenthal,P.M.(2011).Beyondsmiledynam-ics:Mimicryandbeliefsinjudgmentsofsmiles.Emotion,11,181–187.
Matsumoto,D.,&Lee,M.(1993).Consciousness,volition,andtheneuropsychologyof
facialexpressionsofemotion.ConsciousnessandCognition,2(3),237–254.
McLellan,T.,Johnston,L.,Dalrymple-Alford,J.,&Porter,R.(2010).Sensitivitytogenuinever-susposedemotionspecifiedinfacialdisplays.Cognition&Emotion,24(8),1277–1292.McLellan,T.,Wilcke,J.C.,Johnston,L.,Watts,R.,&Miles,L.K.(2012).Sensitivitytoposed
andgenuinedisplaysofhappinessandsadness:AnfMRIstudy.NeuroscienceLetters,531(2),149–154.
Miles,L.,&Johnston,L.(2007).Detectinghappiness:Perceiversensitivitytoenjoyment
andnon-enjoymentsmiles.JournalofNonverbalBehavior,31(4),259–275.
Morecraft,R.J.,Louie,J.L.,Herrick,J.L.,&Stilwell-Morecraft,K.S.(2001).Corticalinner-vationofthefacialnucleusinthenon-humanprimate:Anewinterpretationoftheeffectsofstrokeandrelatedsubtotalbraintraumaonthemusclesoffacialexpression.Brain,124(1),176–208.
Nicholls,M.E.,Wolfgang,B.J.,Clode,D.,&Lindell,A.K.(2002).Theeffectofleftandright
posesontheexpressionoffacialemotion.Neuropsychologia,40(10),1662–1665.Niedenthal,P.M.,Mermillod,M.,Maringer,M.,&Hess,U.(2010).TheSimulationofSmiles
(SIMS)model:Embodiedsimulationandthemeaningoffacialexpression.BehavioralandBrainSciences,33(6),417–433.
Öhman,A.,Lundqvist,D.,&Esteves,F.(2001).Thefaceinthecrowdrevisited:Athreatad-vantagewithschematicstimuli.JournalofPersonalityandSocialPsychology,80(3),381.Okubo,M.,Ishikawa,K.,&Kobayashi,A.(2013).Notrustontheleftside:Hemifacial
asymmetriesfortrustworthinessandemotionalexpressions.Brain&Cognition,82(2),181–186.
Pinkham,A.E.,Griffin,M.,Baron,R.,Sasson,N.J.,&Gur,R.C.(2010).Thefaceinthecrowd
effect:Angersuperioritywhenusingrealfacesandmultipleidentities.Emotion,10(1),141.
Rinn,W.E.(1984).Theneuropsychologyoffacialexpression:Areviewoftheneurological
andpsychologicalmechanismsforproducingfacialexpressions.PsychologicalBulletin,95(1),52–77.
Roesch,E.B.,Tamarit,L.,Reveret,L.,Grandjean,D.,Sander,D.,&Scherer,K.R.(2011).
FACSGen:Atooltosynthesizeemotionalfacialexpressionsthroughsystematicma-nipulationoffacialactionunits.JournalofNonverbalBehavior,35(1),1–16.
Ross,E.D.,Prodan,C.I.,&Monnot,M.(2007).Humanfacialexpressionsareorganized
functionallyacrosstheupper–lowerfacialaxis.TheNeuroscientist,13(5),433–446.Ross,E.D.,&Pulusu,V.K.(2013).Posedversusspontaneousfacialexpressionsaremod-ulatedbyoppositecerebralhemispheres.Cortex,49(5),1280–1291.
Ross,E.D.,Reddy,A.L.,Nair,A.,Mikawa,K.,&Prodan,C.I.(2007).Facialexpressionsare
moreeasilyproducedontheupper–lowercomparedtotheright–lefthemiface.PerceptualandMotorSkills,104(1),155–165.
Ross,E.D.,Shayya,L.,Champlain,A.,Monnot,M.,&Prodan,C.I.(2013).Decoding
facialblendsofemotion:Visualfield,attentionalandhemisphericbiases.Brain&Cognition,83(3),252–261.
Sackeim,H.A.,Greenberg,M.S.,Weiman,A.L.,Gur,R.C.,Hungerbuhler,J.P.,&
Geschwind,N.(1982).Hemisphericasymmetryintheexpressionofpositiveandnegativeemotions.ArchivesofNeurology,39(4),210–218.
Sackeim,H.A.,Gur,R.C.,&Saucy,M.C.(1978).Emotionsareexpressedmoreintenselyon
theleftsideoftheface.Science,202(4366),434–436.
Schubö,A.,Gendolla,G.H.,Meinecke,C.,&Abele,A.E.(2006).Detectingemotionalfaces
andfeaturesinavisualsearchparadigm:Arefacesspecial?Emotion,6(2),246.
Schwartz,G.E.,Davidson,R.J.,&Maer,F.(1975).Righthemispherelateralizationforemo-tioninthehumanbrain:Interactionswithcognition.Science,190(4211),286–288.Wagner,H.L.,MacDonald,C.J.,&Manstead,A.S.(1986).Communicationofindividual
emotionsbyspontaneousfacialexpressions.JournalofPersonalityandSocialPsychology,50(4),737.
Wallez,C.,&Vauclair,J.(2013).Human(Homosapiens)andBaboon(Papiopapio)
chimericfaceprocessing:Right-hemisphereinvolvement.JournalofComparativePsychology,127(3),237–244.
Zuckerman,M.,Hall,J.A.,DeFrank,R.S.,&Rosenthal,R.(1976).Encodinganddecodingof
spontaneousandposedfacialexpressions.JournalofPersonalityandSocialPsychology,34(5),966.
因篇幅问题不能全部显示，请点此查看更多更全内容