A Pilot Study of Nonconscious Mimicry among Japanese University

APilotStudyofNonconsciousMimicryamong
JapaneseUniversityStudents:
IsaSociallyUndesirableBehaviorMimicked?
YohsukeOhtsubo*
NaraUniversity
Abstract
Thepurposeofthisstudywastoexaminewhetheraculturally-undesirablebehavior
(playingwithapen)wouldbelesslikelytobemimickedbyparticipantsthananeutral
behavior(claspingoneshands).Sixtyparticipantstookpartintheexperiment,in
whichthetypeofbehavior(undesirablevs.neutral)wasawithin-participantfactor.
Duringtwoexperimentalsessions,eachparticipantinteractedwithaconfederatewho
engagedinoneofthetargetbehaviors.Theresultshowedthatnonconsciousmimicry
occurredonlywiththepen-playbehavior.Also,onlymalesshowedmimicryofthepenplaybehavior.Somepossiblereasonsfortheunexpectedresultsarediscussed.
Recently,socialpsychologistshavebeeninterestedinthesocialfunctionsofnonconscious
mimicry‐peoplesautomatictendencytoimitateotherspostures,mannerisms,facial
expressions,andotherbehaviors.Recentexperimentsdemonstratedthatparticipantswho
engagedinabriefinteractionwithanexperimentalconfederatetendedtomimicthe
confederate'srubbingofhis/herface,shakingofhis/herfoot(Chartrand&Bargh,1999;
Cheng&Chartrand,2003;vanBaaren,Maddux,Chartrand,deBouter,&vanKnippenberg,
2003),andplayingwithapen(vanBaarenetal.,2003).Suchtendencytomimicothers'
mannerismsappearsmeaninglessatfirstglance.However,recentresearchhasshownthat
nonconsciousmimicryhasanimportantsocialfunction(Lakin,Jefferis,Cheng,&Chartrand,
2003).InChartrandandBargh's(1999)experiment(Experiment2),forexample,each
participantengagedinanexperimentaltaskwithaconfederate.Duringtheexperimental
session,inoneconditiontheconfederatemimickedtheparticipant'smannerisms(mimicking
condition),whileintheotherconditiontheconfederatebehavedinanon-imitativemanner
(non-mimickingcondition).Thosewhowereinthemimickingconditionratedtheconfederate
morelikablethanthosewhowereinthenon-mimickingcondition.Moreover,thosewhowere
inthemimickingconditionfoundthattheinteractionwiththeconfederatewentmore
平 成18年9月7日
受 理*社
会 学 部 人 間 関係 学 科 助 教 授
一107一
BULLETINOFRESEARCHINSTITUTE
smoothly.Somestudiesalsofoundthatbeingmimickedmakespeoplemorehelpfultoward
others(vanBaaren,Holland,Kawakami,&vanKnippenberg,2004;vanBaaren,Holland,
Steenaert,&vanKnippenberg,2003).Reviewingthesestudies,Lakinetal.pointoutthat
nonconsciousmimicryhelpsustomaintainharmonioussocialrelationshipswithothers.
Giventhesocialfunctionofnonconsciousmimicry,onemightwonderiftherearecertain
typesofbehaviorsthatonemightbebettertorefrainfromimitating.Arecentstudy
concerningdeceptivebehaviorsshowedthatpeopletendtotellliesmorefrequentlytoa
personwhohasliedtothem(Tyler,Feldman,&Reichert,2006).Therefore,mimickinga
deceptivebehaviortriggersasortofconflictspiral.Also,therearecertaintypesofbehaviors
thatareconsideredunacceptableorundesirable.Onemightbedisinclinedtoimitateaperson
whoisengaginginasocially-undesirablebehavior,especiallywhenotherpeopleare
observingit.Inthiscase,amimickerofasocially-undesirablebehaviorisunlikelytodevelop
cordialrelationshipswithobservers.Therefore,itisanticipatedthatnonconsciousmimicry
willserveitssocialfunctionbetterinthepresenceofsomeinhibitorymechanismthan
withoutit.Ekman(1972)showedthatJapaneseundergraduateswhowereexposedto
stressfulstimuliinhibitedthemselvesfrommakingnegativefacialexpressionsinthepresence
ofanauthorityfigure.Similarmechanismsmightapplytononconsciousmimicry;thatis,
peoplemightrefrainfrommimickingculturally-undesirablebehaviors.
Thepurposeofthisstudywastotestifaculturally-undesirablebehaviorislesslikelyto
bemimickedthananeutralbehavior.Participantsofthestudyengagedinanalleged
inspirationexperiment,inwhicheachparticipantwaspairedwithasame-sexconfederateand
askedtoproducemanyadjectivesapplyingtogivennouns.Duringtheinspiration
experiment,theconfederateeitherplayedwithapen(aculturally-undesirablebehavior)or
claspedhisorherhandsontopofadesk(neutralbehavior).InJapan,playingwithapenisnot
taboo,butkeepingstillisconsideredmoredesirableinsuchacontext.Therefore,playingwith
apenisconsideredamoderately-undesirablebehavior.Oneofthefemaleresearchassistants
whoservedasaconfederateprofessedthatshewasafraidthatparticipantswouldthinkher
atomboywhensheplayedwithapen.Thestudyemployedawithin-participantdesignin
whichallparticipantswereexposedtobothtypesofbehaviors.Theinspirationexperiment
consistedoftwosessions,andhalfoftheparticipantswereexposedtoaconfederateplaying
withapeninthefirstsession(pen-handcondition),whiletheotherhalfoftheparticipants
wereexposedtoaconfederateclaspinghisorherhandsinthefirstsession(hand-pen
condition).Itwasanticipatedthattheconfederatewillbemimickedmorefrequentlyby
participantswhenheorsheisclaspinghisorherhandsthanwhenheorsheisplayingwith
apen.
一108一
Ohtsubo:NONCONSCIOUSMIMICRY
Methods
ParticipantsandDesign
Participantswere60undergraduates(30malesand30females)atNaraUniversity,Japan.
Theywerepaid500yenfortheirparticipation.Thirtyparticipants(15malesand15females)
wereassignedtothepen-handconditionandtheremainingthirtyparticipants(15malesand
15females)wereassignedtothehand-pencondition.Therewasabetween-participant
condition:Halfoftheparticipantsweresupraliminallyprimedwiththeword"cooperation,"
whiletheotherhalfoftheparticipantsweresupraliminallyprimedwiththeword
"competition
."Thisbetween-participantconditiondidnothaveanyeffectontheresultsofthis
studyandwillnotbementionedfurther.
Procedure
Uponarrival,eachparticipantwasaskedtostayinawaitingroom.Afterarrivalofthe
confederate,whoshowedupwithinafewminutesaftertheparticipant'sarrival,the
experimentertooktheparticipantandtheconfederatetothelaboratory,inwhichtwochairs
werearrangedinsuchawaythatthechairswerehalf-facingeachotherandhalf-facingthe
experimenter'sdesk.Eachchairwasequippedwithawritingarm.Afterbeingseated,each
participantfirstcompletedabriefquestionnaireaskinghisorhersexandage.This
questionnairewasintroducedsothatapenwouldnecessarilybeplacedonthewritingarm.
Aftercompletionofthequestionnaire,theexperimenterexplainedthenatureofthe
inspirationexperiment,givinganexamplenoun,"hand,"totheparticipantandthe
confederate.Theexperimenterpromptedtheconfederatetoproduceafewadjectivesto
modifythenoun.Whentheconfederateproducedsomeexamples,suchas"small"and"leaf
like,"theexperimenteraffirmedthattheconfederateunderstoodthetask.Afterconfirming
thattheparticipantsunderstoodthetask,theexperimenterexplainedthattheparticipant
andtheconfederatewouldtaketurnsinthinkingupadjectives:Firsttheparticipantwould
compileadjectivesapplyingtoanounforoneminute,andthentheconfederatewouldcompile
adjectivesapplyingtoadifferentnounforoneminute,andsoon.Beforestartingthe
experiment,theexperimenteraskedtheparticipantandtheconfederateiftheywouldmind
beingvideorecorded.Noparticipantsrefusedtoberecorded.Thevideowasmanipulatedby
anexperimentassistantwhostoodbehindtheparticipantduringtheexperimentalsessionsso
thattheassistantsbehaviorswerenotvisibletotheparticipant.
Presentingatargetnoun,theexperimenteralwaysfirstaskedtheparticipanttoproduce
asmanyadjectivesaspossiblewithinthespanofoneminute.Afteroneminutehadelapsed,
theexperimenterpresentedadifferentnounandaskedtheconfederatetoproduceasmany
adjectivesaspossible.Twentysecondsaftertheconfederatebeganhisorherturn,the
一109一
BULLETINOFRESEARCHINSTITUTE
experimenterliftedupastopwatchtosignaltheconfederatetoengageinatargetbehavior
(i.e.,playingwithapenorclaspinghisorherhands).Afteroneminutehadelapsed,the
experimenterpresentedanewnounandaskedtheparticipanttoproduceasmanyadjectives
foritaspossible.Duringtheparticipantsturn,theexperimentersignaledtheconfederateto
stopthetargetbehavior.Afterpresentingsixnounsinthismanner,theexperimenter
announcedthattheywouldhaveashortbreak.Theconfederateengagedinthetarget
behaviorthreetimesduringthefirstsession,for90seconds,60seconds,and40seconds,
respectively.Afterashortbreak,theexperimenterstartedthesecondsession.Theprocedure
forthesecondsessionwasthesameasthatforthefirstsessionexceptthatadifferentsetof
nounswasusedandthetargetbehaviorwasaltered.
Aftercompletionoftheexperiment,theexperimenterexplainedfullythepurposeofthe
experimentandaskedtheparticipantifheorshehadanysuspicionsabouttheexperiment.
Noneoftheparticipantsreportedsuspicionsrelevanttothepurposeoftheexperiment.The
participantwasthenaskedtosignaformallowingtheexperimentertousethevideo
recording.Thentheparticipantwaspaidandthankedforhisorherparticipationinthestudy.
Results
Thevideorecordingswerecodedbyfourindependentjudgeswhowereblindtothe
conditionsofparticipants.Eachvideorecordingwascodedbytwoofthejudges.Thejudges
countedthenumberoftimestheparticipantclaspedhisorherhandsandplayedwithapen.
Theinterraterreliabilitiesareasfollows:forthehand-claspingbehavior,.64(firstsession)and
.74(secondsession),andforthepen-playbehavior,.74(firstsession)and.97(secondsession).
Althoughsomeinterraterreliabilitiesareonlymoderatelyhigh,thetwojudgesalwaysagreed
onwhetherornotaparticularparticipanthadengagedineachofthetargetbehaviorsatleast
onceinagivensession.Thedatafromthetwojudgeswereaveragedandusedasdependent
variales.
Itturnedoutthatalargeportionoftheparticipantsdidnotengageineithertypeof
behavior.Duringthesessioninwhichtheconfederateclaspedhisorherhandsanddidnot
playwithapen,19participantsclaspedtheirhandsatleastonce,andsixparticipantsplayed
withapenatleastonce.Duringthesessioninwhichtheconfederateplayedwithapenand
didnotclasphisorherhands,toparticipantsplayedwithapenatleastonce,and22
participantsclaspedtheirhandsatleastonce.Therefore,distributionsofthedependent
variableswerehighlypositively-skewed.Accordingly,eachdatapointwasgivenaconstant,
1.00,andthensubmittedtothelog10transformation.Twoseparate2(condition:hand-penvs.
pen-hand)x2(confederatesbehavior:clasping-handsvs.pen-play)ANOVAswiththelatter
一110一
Ohtsubo:NONCONSCIOUSMIMICRY
factorasrepeatedmeasureswereconductedforeachtypeofbehavior(seeFigurelfor
untransformedmeans).Forthehand-claspingbehavior,amaineffectoftheconfederate's
behaviorwasnotsignificant,F(1,58)=2.12,ns.Othereffectswerealsonotsignificant.Infact,
participantsclaspedtheirhandsslightlymoreoftenduringthesessioninwhichthe
confederatedidnotclasphisorherhands(untransformedmean=.94,SD=1.55)thanduring
thesessioninwhichtheconfederateclaspedhisorherhands(untransformedmean=.58,SD
=1
.19).Nonconsciousmimicrywasthusnotobservedwiththehand-claspingbehavior.For
thepen-playbehavior,amaineffectofconfederate'sbehaviorwasmarginallysignificant,F(1,
58)=3.24,ρ=.08,η2=.053,andtheinteractioneffectbetweentheconditionandthe
confederate'sbehaviorwasmarginallysignificant,F(1,58)=3.36,ρ=.07,η2=.055.Amain
effectofconditionwasnotsignificant.Participantsplayedwithapenmorefrequentlyduring
thesessioninwhichtheconfederateplayedwithapen(untransformedmean=.18,SD=.44)
thanduringthesessioninwhichtheconfederatedidnotplaywithapen(untransformed
mean=.13,SD=.48).Thispatternsuggeststhatnonconsciousmimicryoccurred,although
theeffectwasweak.Themarginally-significantinteractioneffectwasduetothefactthatthis
patternwasobservedonlyinthehand-pencondition.
Closescrutinyofthepen-playdatashowedthatonlyonefemaleplayedwithapen,and
shedidsothroughouttheexperiment(i.e.,inbothsessionsregardlessofwhetherthe
confederateplayedwithapenornot).Thesexdifferenceinnonconsciousmimicryofthepen-
playbehaviorwassignificantbyFisher'sexacttest.Nineof30malesplayedwithapenduring
thesessioninwhichtheconfederateplayedwithapen,whileonlyoneof30femalesdidso.
Theprobabilityofobservingthispatternbychancewas.ol.Itisnoteworthythatthiswasnot
ハU
0
8
ほ
0
2
の
(U
0
4
2 0欄
詫 慧 笛 も ①隻 ト 転 醤
4雛
勤o 台 器 尋 2 匹 § o謹
O
0
0
の
ClaspingHandsPen-Play
Participant'sE3ehavior
Figure1.Meanfrequencyofeachtypeofbehaviorasafunction
oftheconfederatebehavior(clasping-handsvs.pen-play).
一111一
BULLETINOFRESEARCHINSTITUTE
蛍0。40C。
酬erate癖sBehavi。r
1
$
y..0・30a
§020
塁側 。
峯 α。
。
MaleFemale
Figure2.Meanfrequencyofpen-playbehaviorasafunctionoftheconfederatebehavior
(clasping-handsvs.pen-play)andparticipantssex.
solelyduetothefactthatmalesweresimplymorepronetoplaywithapenthanwere
females.Althoughtherewassuchtendency(i.e.,50f30malesplayedwithapenduringthe
sessioninwhichtheconfederatedidnotplaywithapen,whileonlyoneof30femalesdidso),it
wasnotsignificantbyFischer'sexacttest.Giventhesexdifferenceinnonconsciousmimicry,
the2(condition)x2(confederate'sbehavior)ANOVAwasconductedincludingonlythemales'
data.Themaineffectoftheconfederatesbehaviorreachedthestatisticallysignificantlevel,
F(1,28)=4.65,ρ<.05,η2=.14(seeFigure2foruntransformedmeans).
Discussion
Theexperimentwasaimedattestingthehypothesisthatnonconsciousmimicrywould
bemorefrequentlyobservedforaculturally-neutralbehavior(i.e.,hand-claspingbehavior)
thanaculturally-undesirablebehavior(i.e.,pen-playbehavior).Thehand-claspingbehavior
wasmorefrequentlyengagedinthanthepen-playbehavior.Ofthe60participants,24
participantsclaspedtheirhandsatleastonceduringtheexperiment,whilellparticipants
playedwithapenatleastonceduringtheexperiment.Thispatternseemsconsistentwiththe
assumptionthatthepen-playbehaviorwaslessculturallydesirable.Thereisaquestionasto
whynonconsciousmimicrydidnotoccurwiththehand-claspingbehavior.Accordingtothe
confederates,thehand-claspingbehaviorwaslessnoticeablethanthepen-playbehavior.
Theybelievedthatamajorityoftheparticipantsweresoconcentratedonthetaskthatthey
didnotlookattheconfederateatall.Therefore,totestthehypothesis,morenoticeabletarget
behaviorsareneeded.
Althoughtheoriginalhypothesiswasnottestablebecauseofthelackofnonconscious
一112一
Ohtsubo:NONCONSCIOUSMIMICRY
mimicryofthehand-claspingbehavior,aninterestingpatternrelatedtothehypothesiswas
observed.Therewasatendencyforfemalesnottoexhibitnonconsciousmimicryofthepenplaybehavior.Itisreasonabletoassumethataculturalnorminhibitingrudebehaviorsaffects
femalesmoremarkedlythanmales.Giventhefactthatnosexdifferenceinnonconscious
mimicrywasobservedinChartrandandBargh's(1999)study,bothmalesandfemalessharea
predispositiontononconsciousmimicry.Perhapstheresultsofthisstudyshouldbe
interpretedasillustratingthatmalesarelesssusceptibletoculturalnormsthanfemales,and
thusaremorelikelytoplayoutnonconsciouslyformedbehavioralintentionevenwhenthe
intendedbehaviorisperceivedasslightlyundesirableinagivencontext.Thisinterpretation
isconsistentwiththebasicideaofthestudythatnonconsciousmimicryisaccompaniedby
someinhibitorymechanism.However,thisevidenceisatbestindirectandweak.Moredirect
demonstrationofthepresenceofsomeinhibitorymechanismofnonconsciousmimicryis
needed.
Thisstudyshowedarelativelylowfrequencyofnonconsciousmimicry.Thisresult
seemstocontradictvanBaarenetal.'s(2003)result(Experiment3).VanBaarenetal.
comparedmimickingbehaviorsofJapaneseandAmericans.Consistentwiththehypothesis
thatmimickingbehaviorsaremoreprominentamongJapanese,whohaverelatively
interdependentself-construals,thanamongAmericans,whohaverelativelyindependentself
contruals,Japaneseparticipantsshowedagreateramountofnonconsciousmimicry.Given
vanBaarenetal.'sresult,thelowfrequencyofnonconsciousmimicryinthisstudy,inwhich
Japaneseundergraduatestookpart,seemspuzzling.ItisnoteworthythatvanBaarenetal.'s
JapanesesamplewasdrawnfromJapanesestudentsstudyingatanAmericanuniversity,as
thesestudentsmightbesomewhatdifferentfromordinaryJapanese.Also,thedifferencein
theresultsmightbeattributabletosomemethodologicaldifferences.Forexample,van
Baarenetal.'sarrangementmighthaveallowedparticipantstoseetheconfederatebetter
thanourarrangementdid.AstudythatcloselysimulatesvanBaarenetal.'sproceduremay
beworthconductinginJapaninordertoexaminefullythematterofculturalinfluenceon
nonconsciousmimicry.
Acknowledgment
IamgratefultoKazuhiroKii,EmiNoguchi,SaoriTakeuchi,andTetsuyaYamadafor
theirassistanceinconductingtheexperiment.
一113一
BULLETINOFRESEARCHINSTITUTE
References
Chartrand,T.L.,&Bargh,J.A.(1999).Thechameleoneffect:Theperception-behaviorlinkandsocial
interaction.JournalofPersonalityandSocialPsychology,76,893-910.
Cheng,C.M.,&Chartrand,T.L.(2003).Self-monitoringwithoutawareness:Usingmimicryasa
nonconsciousaffiliationstrategy.JournalofPersonalityandSocia!Psychology,85,1170-1179.
Ekman,P.(1972).Universalsandculturaldifferencesinfacialexpressionsofemotion.InJ.Cole(Ed.),
NebraskaSymposiumonMotivation,1971(pp.207-283).Lincoln,NE,UniversityofNebraskaPress.
Lakin,J.L.,Jefferis,V.E.,Cheng,C.M.,&Chartrand,T.L.(2003).Thechameleoneffectassocialglue:
Evidencefortheevolutionarysignificanceofnonconsciousmimicry.JournalofNonverbalBehavior,
27,145-162.
Tyler,J.M.,Feldman,R.S.,&Reichert,A.(2006).Thepriceofdeceptivebehavior:Dislikingandlyingto
peoplewholietous.Journa!ofExperimentalSocia!Psychology,42,69-77.
vanBaaren,R.B.,Holland,R.W.,Kawakami,K.,&vanKnippenberg,A.(2004).Mimicryandprosocial
behavior.Psycholog.たalScience,1571-74.
vanBaaren,R.B.,Holland,R.W.,Steenaert,B.,&vanKnippenberg,A.(2003).Mimicryformoney:
Behavioralconsequencesofimitation.JournalofExperimentalSocia!Psychology,39,393-398.
vanBaaren,R.B.,Maddux,W.W.,Chartrand,T.L.,deBouter,C.,&vanKnippenberg,A.(2003).Ittakesto
twotomimic:Behavioralconsequencesofself-construals.JournalofPersonalityandSocial
Psychology,84,1093-1102.
一114一