Garcia and Kim Weaver Princeton University Gordon B Moskowitz Lehigh University John M Darley Princeton University Five studies merged the priming methodology with the bystander apathy literature and demonstrate how merely priming a social context a ID: 9201
Download Pdf The PPT/PDF document "INTERPERSONAL RELATIONS AND GROUP PROCES..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
INTERPERSONALRELATIONSANDGROUPPROCESSES CrowdedMinds:TheImplicitBystanderEffectStephenM.GarciaandKimWeaverPrincetonUniversityGordonB.MoskowitzLehighUniversityJohnM.DarleyPrincetonUniversityFivestudiesmergedtheprimingmethodologywiththebystanderapathyliteratureanddemonstratehowmerelyprimingasocialcontextatTime1leadstolesshelpingbehavioronasubsequent,completely StephenM.Garcia,KimWeaver,andJohnM.Darley,DepartmentofPsychology,PrincetonUniversity;GordonB.Moskowitz,DepartmentofPsychology,LehighUniversity.StephenM.GarciawassupportedbyaNationalScienceFoundationgraduatefellowshipandbyagraduatefellowshipfromtheProgramonNegotiationatHarvardLawSchool.WethankDeborahPrentice,SusanFiske,MarkHallahan,andMaxBazermanfortheirinsightfulcommentsandsuggestions. somanypeoplepresent,anindividualmightjustassumethatavictimisreceivinghelporthathelpisalreadyontheway.(Darley&Latane,1968;Darleyetal.,1973)andralisticignorance(Prentice&Miller,1996)explanations,ontheotherhand,reflectthenotionthatpeoplelooktootherstoevaluateanemergencysituation.Iftheotherspresentarejuststandingaroundandappearcalm,thenwould-behelpersinferthatperhapsthesituationreallyisnotanemergency.Smokespewingfromaventmayjustindicateafoggyvent,notaburningfire.Cacioppo,Petty,andLosch(1986)morerecentlyofferedanotheraccountforbystanderapathy,namely,confusionofresponsibility.Thisexpla-nationarguesthatwould-behelpersrefrainfromhelpingavictiminthepresenceofothersbecausetheydonotwanttobeperceivedastheperpetratorofthevictimspainandsuffering.Thatis,insomecases,itmightappeartoobserversthatsomeonehelpingavictimisactuallythecauseofthevictimsharm.Althoughthesesituationalaccountsarecompelling(forare-view,seeLatane&Nida,1981)andarealllikelycontributorstobystanderapathy,ourpurposeinthispaperistoofferaaccountthatcouldprovideanalternativeoradditionalexplanationforthefindings.However,wehastentoaddthatsituationalandnonsituationalaccountsofbystanderapathyneednotbemutuallyexclusive.Aswereview,researchonprimingsuggeststhatmerelyactivatingknowledgestructurescanaffectsocialperceptionandbehavior(e.g.,Bargh,Chen,&Burrows,1996;Higgins,Rholes,&Jones,1977).Inlightofthesefindings,weseektoexaminewhethermerelyactivatingtheconstructofgroupinthemindsofparticipantscouldalsoresultinthebystanderapathyeffect.Inparticular,weareinterestedinexaminingwhetheractivatingtheconstructofagroupofpeoplewhocouldnotpossiblyhelpleadstoanimplicitbystandereffectonasubsequenthelpingbehavior.PrimingAffectsSocialPerceptionandBehaviorBarghetal.(1996)definedprimingastheincidentalactivationofknowledgestructures,suchastraitconceptsandstereotypes230).Researchhasshownthatthecognitiveaccessibilityofthesetypesofknowledgestructurescanhaverealandimportanteffectsonsocialperceptionandbehavior(e.g.,Carver,Ganellen,From-ing,&Chambers,1983;Dijksterhuis,Spears,&Lepinasse,2000;Dijksterhuis&vanKnippenberg,1998;Epley&Gilovich,1999;Higginsetal.,1977;Kawakami,Young,&Dovidio,2000;Macrae&Johnston,1998;Neuberg,1988).Morespecifically,researchonprimingeffectsonsocialbehaviorhasdemonstratedthatsubtlecuesorprimesintheenvironmentcansubsequentlyaffectbehav-ior.Forinstance,inanewbutalreadyclassicstudy,Barghetal.(1996)foundthatprimingtheelderlystereotypecanaffectpeo-ssubsequentbehavior.Intheirstudy,participantswereprimedthroughascrambledsentencetask.Intheelderlycondition,thesentencesincludedwordsrelatedtotheelderlystereotype(e.g.,old,wrinkle,Florida),whereasintheneutralcondition,thescram-bledsentencesincludedage-nonspecificwords.Findingsshowedthatparticipantswhowereprimedwiththeelderlystereotypewalkedoutofthelaboratorysignificantlymoreslowlythandidparticipantsintheneutralcondition.Similarly,otherresearchlookingatprimingandbehaviorhasfoundthatindividualsprimedwiththeconceptofprofessorsperformbetteronageneralknowledgetestthandocontrolpartic-ipants(Dijksterhuis&vanKnippenberg,1998).Themechanismunderlyingtheseprimingeffectsisbasedonthepremisethattriggeringaknowledgestructuresuchasastereotypeactivatesthesemanticknowledgeassociatedwithit.Thissemanticknowledgeincludes,amongotherthings,certaintypesofbehaviors.Forinstance,partofthestoredrepresentationofelderlypersonsisthattheywalkslowly.Hence,whentheconstructorstereotypeofanelderlypersonisprimedinanindividual,thebehavioralrepresen-tationisactivated,andtheindividualwalksmoreslowly.Inadditiontoinfluencingbehaviordirectly,primingcanalsoaffectsocialperceptionand,asaconsequence,influencebehaviorindirectly.Inthesecases,primingaffectspeoplesinternalcues,whichthenspilloverintotheirperceptionsofotherpeopleandthesocialworld.Forinstance,Higginsetal.(1977)demonstratedthatprimingindividualswithconceptsrelatedtohostilityleadsthemtoperceiveanambiguoustargetperson(e.g.,Donald)asmorehostilecomparedwithindividualswhodonotreceivehostilityprimes.Otherresearchon(e.g.,Markus,1977)and(e.g.,Bargh&Thein,1985)speakstotheideathatpeoplecanhavechroniccognitivefiltersthataffectsocialpercep-tion.Researchonrelationalschemas(Baldwin,1994,1995;Bald-win,Fehr,Keedian,Seidel,&Thomson,1993;Baldwin,Keelan,Fehr,Enns,&Koh-Rangarajoo,1996)alsodemonstratesthatprimescanaffectsocialperception.Althoughthemechanismoftheseprimesonsocialperceptionissimilartothatofsocialbehavior,onecriticaldifferenceisthattheseprimedoractivatedsocialrepresentationscoloronesperspectiveoftheworldandsubsequentlyhaveindirecteffectsononesbehavior.Behavioralprimes,ontheotherhand,aredirectlylinkedtoactualbehaviors.Tobetterunderstandhowprimescanaffectbehaviorindirectlythroughinfluencingsocialperception,wecanlookatcompetitiveprimesusedinaPrisonersDilemmaGame(Neuberg,1988).Inthisstudy,participantswereprimedwitheithercompetitiveorneutralprimesandthenbeganroundsofaPrisonersDilemmaGamewithanotherphantomparticipant.Participantsmadetheirfirstmovebeforelearningthemoveofthephantomparticipant.Resultsshowedthatthecompetitiveprimeshadnoeffectonthesfirstmove,indicatingthattheprimesdidnotdirectlyinfluenceparticipantsbehavior.However,participantswithpre-disposedcompetitivedispositionswhoreceivedcompetitiveprimesexhibitedmorecompetitivebehavioronsubsequenttrialsthandidthoseparticipantswithcompetitivepredispositionswhoreceivedneutralprimes.Thesefindingssuggestthatthecompeti-tiveprimesdidnotdirectlyleadtocompetitivebehaviorbutratheraffectedtheparticipantsbehaviorindirectlythroughtheirdis-tortedsocialperceptionofthephantompartner.Inotherwords,thecompetitivelensesoftheseparticipantsledthemtoperceivethephantompartnerasbeingcompetitiveonlyaftertheylearnedthesmove.Theirownbehaviorwasaffectedwhentheyre-spondedinkindtowhattheyperceivedwasacompetitivepartner.PrimingaSocialContextResearchhasclearlydemonstratedthatprimingknowledgestructuressuchastraitconcepts,stereotypes,andbehavior-relatedwordscanhaverealandimportanteffectsonbothsocialpercep-tionandbehavior.Butwhatofmoremacrosituations?IsitGARCIA,WEAVER,MOSKOWITZ,ANDDARLEY possibletoprimeasocialcontext?Forinstance,cansimplyimag-iningbeinginagroupatTime1affectbehavioratTime2?Tothisend,thepresentstudyexamineswhetheritispossibletoprimeasocialcontextorapsychologicalsituation.Inacleverstudy,EpleyandGilovich(1999)examinedwhetherprimingcouldinfluencehowpeoplereacttoasocialcontext,namelyanAsch-likeconformitysituation.Intheirstudy,partici-pantsfirstcompletedacomputerprimingtaskinwhichtheyviewedeitherconformity(e.g.,comply,conform,obey)ornoncon-formity(e.g.,challenge,defy,deviate)primes.Aftercompletingthecomputerexercise,participantswereledintoanotherroomwithconfederatesinaconformitysituationandwereaskedtorateattitudequestionsthatwereprojectedonascreen.Findingsshowedthatthosewhowereprimedwithconformity-relatedwordsweremorelikelytoconformthanwerethoseprimedwithnon-conformitywords.Thisstudywasanimportantstepinexaminingtheeffectsofprimingonhowpeoplebehaveinasocialcontext.Thepresentresearchseekstobuildonthisstudybyexaminingwhetherasocialcontextitselfcanbebroughttomindand,inturn,influencebehavior.TheImplicitBystanderEffectBystanderapathyresearchshowsthatpeopledonotalwayshelpinemergencysituations,andthelikelihoodofeachpersonsingdecreasesasthephysicalpresenceofotherpeoplebecomesincreased.Diffusionofresponsibilityaccounts(e.g.,Darley&Latane,1968;Latane&Darley,1968)explainthatbeinginagroupleadsonetofeelasifoneislessresponsibleforhelpingortakingaction,andthisfeelingofdiminishedresponsibilityoraccountabilitythenleadstoinaction.Logically,then,anymannerinwhichthisfeelingofdiminishedresponsibilitycanbeintro-ducedshouldleadtosimilarbehaviorinaction.Forinstance,primingnotionsofunaccountabilityshouldleadtheconcepttobecomemorecognitivelyaccessible.Oncesuchincreasedacces-sibility,orperceptualfluency,occurs,itislikelytostructurethemannerinwhichpeoplerespondtosituations.Whenahelpingsituationarises,peoplemaysearchinternalcuestodeterminewhethertheyshouldhelp.Ifunaccountabilityhasjustbeenprimed,theymayhelpless.Weproposethatbeinginorsimplythinkingaboutagroupisenoughtoactivatethisconstruct,becausepartoftheconceptofbeinginagroupisthenotionofbeinglostinacrowd,beingdeindividuated,andhavingaloweredsenseofpersonalaccount-ability.Thus,althoughphysicallybeinginagroupmightleadonetoexperiencethesefeelings,simplyprimingthenotionofbeinginagroupshouldalsotriggeroractivatetheseconcepts.Justastriggeringastereotypeactivatesallthesemanticknowledgeasso-ciatedwithit,triggeringanabstractconceptsuchasbeinginagroupshouldactivatewhateverknowledgeisassociatedwiththatconstruct.Instereotypingresearch,weknowthatthinkingabouttheelderlyleadstonotionsofwalkingslowlybeingtriggereddespitethesenotionsnotbeingdirectlyprimed(Barghetal.,1996).Similarly,thinkingaboutAfricanAmericansleadstono-tionsofhostilitybeingtriggered,despitethesenotionsnotbeingdirectlyprimed.Therefore,primingthepsychologicalsituationofbeinginagroupshouldtriggertheconceptsassociatedwiththatsituation.Ifthisincludesthesenseofdiminishedresponsibilityandlossofaccountability,asresearchsuggests(e.g.,Darley&Latane,1968;Latane&Darley,1968),thenthesenotionsshouldbetriggeredjustaswhenoneisinthephysicalpresenceofagroup.Hence,primingthepsychologicalsituationofbeinginagroupshouldleadtoabystanderapathyeffectwherebypeoplearelesslikelytohelpbecausetheirinternalcuesreflecttheincreasedfluencyofunaccountabilityandsuggesttothemthathelpingisnotrequired.Weproposethatimplicitactivationofconceptsrelatedtounaccountabilityoperatesmuchthesameasexplicitactivationthroughphysicalpresenceinthemidstofmanyothers.Inthefirstthreestudies,weseektoestablishtheparallelthattheimplicitactivationofthegroupleadstothesameeffectasdoesphysicallymanipulatingthepresenceofothers.Hence,ourcentralhypothesisisthatindividualswhomerelyimaginebeinginagroupwillexhibitlesshelpingbehavioronasubsequent,completelyunrelatedtask.Inthelaterstudies,weseektoestablishthemech-anismunderlyingthisproposedeffect.Muchlikepastresearchhasshownthatspecifictraitsandbehaviorsaretriggeredbytheactivationofabroaderstereotype,weassumethatthespecificconceptofunaccountabilityanddiffusedresponsibilityistriggeredbytheactivationoftheconceptofbeingwithagroup.Totestthismechanism,wedirectlyassesswhetherthinkingaboutagroupleadstoheightenedaccessibilityoftheseconcepts.Study1Diningoutstudy.InStudy1,weattempttoidentifyalinearrelation-shipwithnumberofpeopleimaginedandhelpingbehavior.Previousresearchprovidesevidencethatdiffusionofresponsibilitycanaffectchar-itabledonations(e.g.,Wiesenthal,Austrom,&Silverman,1983),sowedecidedtooperationalizehelpingbehaviorinStudy1ashypotheticalcontributionstocharity.Previousresearchhasalsofoundthatpeoplerespondsimilarlytohypotheticalscenariosastheydotorealhelpingsituations(e.g.,J.Baron&Miller,2000;Davis,Mitchell,Hall,Lothert,Snapp,&Meyer,1999;Laner,Benin,&Ventrone,2001).OurpredictionisthatthemorepeopleparticipantsimagineatTime1,thelesshelpingbehaviortheywillexhibitatTime2.Morespecifically,wepredictthatpeoplewhoimaginebeingwithagroupof30peoplewillexhibitlesshelpingbehavioratTime2thanwillthosewhoimaginebeingwithagroupof10people,who,inturn,shouldexhibitlesshelpingbehaviorthanwillthoseinthe1-personandneutralcontrolconditions.Atotalof129undergraduatesfromPrincetonUniversityparticipatedinabetween-subjects(socialcontext:groupof30/groupof10/1person/neutralcontrol)questionnairestudy.ThisquestionnairewaspartofalargerquestionnairepacketforaQuestionnaireDayhostedbythepsychologydepartment,andthekeypageswereimbeddedinthepacketitself.Participantswerepaid$8forcompletingtheentirequestionnaireParticipantscompletedatwo-pagequestionnaire.OnPage1,thegroupof30conditionreadasfollows:Imagineyouwonadinnerforyourselfand30ofyourfriendsatyourfavoriterestaurant.groupof10conditionwasidenticalbutreferenced10friends.Thepersoncontrolconditionwassimilarbutfocusedononlyonefriend:ImagineyouwonadinnerforyourselfandafriendatyourfavoriteNext,allparticipantsansweredthefillerquestion:Whattimeofdaywouldyoumostlikelymakeyourreservation?ThechoicesetCROWDEDMINDS includedthefollowingtimes:5,or10Becausethisstudywasconductedwithinalargerquestionnaire,ourcontrolconditionwassimplythehelpingbehaviordependentmeasure.OnPage2,allparticipantsreadthehelpingbehaviordependentmeasure,entitledAnnualCharityContribution.Itreadasfollows:Imagineyouhavelongsincegraduatedfromcollege.Whatpercentageofyourannualafter-taxearningswouldyoubewillingtodonatetocharity?(PleaseAtthispoint,participantscouldcheckoneofthefollowingpercentageranges:1%orless,2%3%,4%5%,6%10%,11%15%,16%20%,21%25%,over25%.Oncompletionoftheentirequestion-nairepacket,participantswerepaidfortheirtime.ResultsandDiscussionTotestouraprioripredictionthatalinearpatternwouldemergesuchthatparticipantsinthegroupof30conditionwouldpledgelessoftheirincometocharitythanwouldthoseinthegroupof10condition,whointurnwouldpledgelessthanthoseinthe1personandneutralcontrolconditions,weassignedthefollowingcontrastweightstotheconditions:19.5(groupof30condition),0.5(groupof10condition),9.5(1personcontrolcondition),and9.5(neutralcontrolcondition).Acontrastanalysissuggeststhatourdatasignificantlyfitthislinearpattern,(1,125)andtheresidualwasnotsignificant,(2,125).5.Thatis,thepatternsuggeststhatparticipantswhoimaginedagroupof30peoplepledgedlesstocharitythandidthosewhoimaginedagroupof10people,whointurnpledgedlessthandidthosewhoimaginedonepersonorthoseinthecontrolcondition.SeeTable1forthemeans.Theseresultsareconsistentwiththeliteratureondiffusionofresponsibilityinthathelpingbehaviordecreasedasthenumbersofothersimaginedincreased.However,theseresultsshednewlightonthediffusionofresponsibilityliteratureandsuggestthatimag-inedothersneednotbeimmediatelypresentorconnectedtoahelpingbehaviorsituationforthediffusioneffecttooccur.Indeed,merelyimaginingbeingwithothersatTime1wassufficienttocreatethisdiffusioneffect.Study2Themoviesstudy.WhereasparticipantsinStudy1imaginedbeingwithfriends,weattempttoshowinStudy2thatwhenparticipantsareprimedwithasituationthatincludesagroupofstrangersorbeinginthepresenceofacrowdtheywilllikewisedemonstratelesshelpingbehavior.Wecompareresultsfromthegroupprimeconditionwithresultsfromtwocontrolconditions,oneinwhichparticipantswereprimedwithoneotherperson,andoneinwhichtheywerenotprimed.Asforourprediction,wehypothesizedthatindividualswhoimaginebeinginacrowdedsituationwithmanyotherpeoplewouldpledgesignificantlyfewerdollarstoasannualgivingcampaign,comparedwiththoseindividualsinourcontrolconditions,whichshouldnotdiffersignificantlyfromeachAtotalof38HarvardUniversityandPrincetonUniver-sityundergraduatesparticipatedinabetween-subjects(socialcontext:group/oneperson/neutralcontrol)questionnairestudy.DatafromHarvardUniversityparticipantswerecollectedaspartofalargerquestionnairepacketduringQuestionnaireDay,andthekeypageswereimbeddedwithinthepacket.DatafromthePrincetonUniversityparticipantswerecollectedatthecampusstudentcenter;onlystudentswhoweresittingalonewereapproachedtoensurethatparticipantswouldnottalkorconferwitheachotherwhilecompletingtheshortquestionnaire.Participantsreadoneofthreepossiblescenarios,twoofwhichwereaboutbeingatamovietheater.Wechosethesettingofamovietheaterbecausewewantedtheimaginedsettingtobeawayfromthecollegecampus.Thesescenariosmanipulatedwhetherparticipantsthoughtaboutbeingwithafriendinacrowdedmovietheater(groupcondition)orwhethertheythoughtaboutbeingalonewithafriendinamovietheateronepersoncontrolcondition).Thegroupconditionreadasfollows:Imaginethatyouandafriendaresittinginacrowdedmovietheater.Therearepeopleinfrontofyou,behindyou,andtoyoursides.Althoughtherearesomechildren,theaudienceismostlyadults,andyouarejustwatchingthemoviepreviews.Theonepersoncontrolconditionwasidenticalexceptthatparticipantsreadaboutbeingalonewithafriendatamovietheater:Imaginethatyouandafriendaresittingaloneinamovietheater.Youandyourfriendhavetheentiretheatertoyourself,andyouarejustwatchingthemoviepreviews.Afterreadingthescenarios,allpartic-ipantsansweredafillerquestionona7-pointscale:Whatroomtemper-aturewouldyoupreferforthetheater?verycoolverywarmForthePrincetonsample,theneutralcontrolconditionwassimply,roomtemperaturedoyoupreferingeneral?verycool).TheneutralcontrolconditionfortheHarvardsamplewassimplythedependentvariablebelow,asthisdependentmeasurewasalreadypartofalargerquestionnairepacket.OnPage2,participantscompletedthedependentvariable.Helpingbehaviorwasoperationalizedthroughanannualgivingquestionnaire.Allparticipantsreadthefollowing:AnnualGivingistheyearlyappeal(Harvard/Princeton)makestoallalumni,parentsandfriendsforunrestrictedfundswhichcanbeusedimmediatelytomeettheUniversitysmostimportantneedsandop-portunities.Nowimagineyouhavelongsincegraduatedfrom(Har-vard/Princeton).Ofthefollowing,whichlevelofcontributiondoyouforeseeasyourparticipation?(PleaseCheck).Atthispoint,participantscheckedoffoneofthefollowingdollarranges:$0,$149,$5099,$100249,$250499,$500999,$1,0002,499, Toruleoutthepossibilitythatmakingadinnerreservationentailsmoreresponsibilityinourgroupprimethanitdoesinouronepersonprime,wepretestedtheseprimesbyasking40undergraduatestoreadthefollowinginabetween-subjectsquestionnaire:Imagineyouwonadinnerforyour-selfand(1or10)ofyourfriendsatyourfavoriterestaurant.Youarecallingtomakethereservationforatimebetween5and10WethenaskedHowimportantisyourchoiceofreservationtime?importantatallextremelyimportant).Resultsshowedthattheimportanceratingsofthosewhoreadtheonepersonprime(1.87)andthegroupprime(1.09)didnotsignificantlydifferfromeachother(.68),whichsuggeststhatthedecisiontomakeadinnerreservationtimewasnomoreimportantinthegroupconditionthanintheonepersoncondition.Table1MeanCharityContributionsbyCondition CharitylevelMSDGroupof303.61.3Groupof103.91.5Onepersonandneutralcontrols4.21.6Responseoptionsrangedfrom11%orlesstocharityto8over25%.GARCIA,WEAVER,MOSKOWITZ,ANDDARLEY 4,999,$5,000$10,000andabove.Thesedollarrangesweretreatedasacontinuousvariable(1;10$10,000andaboveinouranalyses.Aftercompletingthequestionnaire,participantswerethankedfortheirtimeanddebriefed.ResultsandDiscussionTotestouraprioripredictionthatindividualsinthegroupconditionwouldpledgesignificantlyfewerdollarsthanwouldthoseintheonepersonandneutralcontrolconditions,wecom-paredthemeanofthegroupcondition(2.28)withthatoftheonepersonandcontrolconditions(2.62)andfoundasignificantdifferenceinthepredicted(1,35).05.Afollowupaprioricompar-isonalsoconfirmedthattheoneperson(2.43)andneutralcontrolconditions(2.84)didnotsignif-icantlydifferfromeachother,(1,35).4.TheSocialContextinteractionwasnotsignificant(whichsuggeststhattheHarvardandPrincetonsamplesyieldedsimilardatapatterns.Theseresultssuggestthatparticipantsprimedwiththepresenceofmanyotherpeopleexhibitedlesshelpingbehaviorthandidthosewhoimaginedbeingwithoneotherpersonorthosewhowerenotprimed.Thepatternofresultsinthisstudygivescredencetotheargumentthatdiffusionofresponsibility,notself-awareness,isthebestexplanationfortheeffect.Thatis,apossibleexplanationforadifferencebetweentheonepersonandgroupconditionscouldbethatthereisanincreaseinhelpingbehaviorintheonepersoncondition,asopposedtoadecreaseinthegroupcondition.Thisalternativeexplanationmaybeplausibleifweinadvertentlymanipulatedself-awareness(e.g.,Duval&Wicklund,1972)intheonepersoncondition.However,becauseparticipantsinthegroupconditionpledgedfewerdollarsthandidthoseintheonepersonandneutralcontrolconditions,whichdidnotsignificantlydifferfromeachother,theimplicationisthatdiffusionofresponsibilityunderliesthiseffect,asopposedtoincreasedself-awareness.Moreover,thisstudydemonstrateshowdiffusionofresponsi-bilitycanspilloverintoahelpingtaskforwhichtheimaginedothersarenotlikelycontributorsorwould-behelpers.Thatis,imaginingacrowdedmovietheaterarguablyevokesthinkingaboutabunchofstrangers.So,whenaskedaboutcontributingtoauniversitysannualgivingcampaign,participantsshouldnotlogicallyexpectstrangersinamovietheatertohaveanymotiva-tionorcompunctiontodonatetoaHarvardorPrincetonannualgivingcampaign.However,visualizingtheseimaginedstrangersstillaffecteddonationpledgelevels.Hence,itseemsthatprimingthepresenceofotherscancreateanimplicitbystandereffectthatspillsoverintoasubsequenthelpingtask.Study3Diningoutstudyrealhelping.Inthenextstudy,wewantedtoreplicatethefindingthatmerelyimaginingthepresenceofothersatTime1affectshelpingbehavioratTime2.Wealsowantedtogeneralizethiseffecttorealhelpingbehavior.Thatis,whereasStudies1and2demonstratedtheeffectwithhypotheticalgiving,itisimportanttodemonstrateitwithrealhelpingbehavior.Again,ourcentralhypothesisisthatthoseprimedwiththepresenceofotherswillofferlesshelpingbehavior.Weagaincomparedresultsfromthegroupconditionwithresultsfromtwocontrolconditions,oneinwhichparticipantsimaginedoneotherperson,andoneinwhichtheywerenotprimed.Atotalof129undergraduatesfromHarvard,Princeton,andRutgersparticipatedinabetween-subjects(socialcontext:group/oneperson/neutralcontrol)questionnairedesign.Participantswererecruitedatcampusstudentcenters.Again,onlythosewhoweresittingbythemselveswereapproachedtopreventcommunicationamongparticipants.Theparticipantscompletedatwo-pagequestionnaire,whichincludedthediningoutprimesofStudy1.OnPage1,thereadasfollows:Imagineyouwonadinnerforyourselfand10ofyourfriendsatyourfavoriterestaurant.onepersoncontrolwasidenticalbutfocusedononefriend:Imagineyouwonadinnerforyourselfandafriendatyourfavoriterestaurant.Next,allparticipantsansweredthefillerquestion:Whattimeofdaywouldyoumostlikelymakeyourreservation?Thechoicesetincludedthefollowingtimes:5,or10.Theneutralcontrolconditionparticipantsonlyreadthefillerquestion,whichwasslightlymodifiedtoWhattimeofdaywouldyoumakeadinnerreservation?OnPage2,weoperationalizedrealhelpingbehaviorasvolunteeringtohelpoutwithanexperimentbeingconductedinanotherroom.Morespecifically,allparticipantsreadthefollowingonPage2:Inadditiontothissurvey,weareconductingabriefexperimentinanotherroom.Howmuchtimeareyouwillingtospendonthisotherexperiment?Atthispoint,participantscheckedoffoneofthefollowingminuteintervals:minutes,2minutes,5minutes,10minutes,15minutes,20minutes,2530minutes.Afterthequestionnaireswerecollected,participantswhohadvolun-teeredtheirtimewereready(andsomeeager)toparticipateinanotherexperiment.However,atthatpoint,weexplainedtothemourhypothesisandthattherewasactuallynootherexperiment.ResultsandDiscussionAsinthepreviousstudies,wemadetheaprioripredictionthatthegroupconditionwoulddiffersignificantlyfromtheonepersonandneutralcontrolconditionsandthattheonepersonandneutralcontrolconditionswouldnotdiffersignificantlyfromeachother.Acomparisonofthemeanofthegroupcondition(2.99)withthemeanoftheonepersonandcontrolconditions6.15)showedasignificantdifferenceinthepredicteddirection,(1,126).05(seeFigure1).Afollow-upcomparisonconfirmedthattheonepersoncontrolcon-dition(7.07)andtheneutralcontrolcondition5.09)werenotsignificantlydifferentfromeach(1,126) Figure1.Study3helpingbehaviorbycondition.CROWDEDMINDS Theseresultssuggestthatparticipantswhoimaginedagroupof10peopleatTime1offeredlesshelpingassistanceatTime2thandidcontrolparticipants,whoeitherimagined1otherpersonorsawnoprime.Again,thisdatapatternisconsistentwithadiffusionofresponsibilityinterpretation,asopposedtoaself-awarenessone,asthe1personandneutralcontrolconditionsarenotsignificantlydifferentfromeachother.Anotherinterestingaspectofthisstudyis,again,thedisconnectbetweenthoseimag-inedandthosewhocanhelp.Thatis,eventhoughtheparticipantsinthegroupconditionimaginedtheirfriends,theseimaginedfriendswerenotintheimmediatevicinitytoofferhelpingbehav-iorassistance.Hence,theseresultssuggestthatothersneednotbephysicallypresentfordiffusionofresponsibilitytobecreated.Althoughwehavedemonstratedthisimplicitbystandereffectwithbothrealandhypotheticalformsofhelpingbehavior,inthenextsetofstudies,weexploreinmoredetailthemechanismorpsy-chologicalprocessesbehindtheeffect.Study4Asdiscussedearlier,althoughtheconceptofreducedhelpingperseisnotlikelytobeapartofthestoredmentalrepresentationofagrouporcrowd,itispossiblethattheconceptualconstructofbeingwithagrouporinthepresenceofothersislinkedtospecificnotionsofdeindividuation,beinglostinacrowd,andaloweredsenseofpersonalaccountability.Ratherthanprimingtheseconstructsdirectlyinthefirststudies,weassumedthattheyweretriggeredbythinkingaboutbeinginagroup.Muchlikepastresearchhasshownthatspecifictraitsandbehaviorsaretriggeredbytheactivationofabroaderstereotype,weassumethatthespecificconceptofdiffusedresponsibilityistriggeredbytheactivationoftheconceptofbeingwithagroup.Totestthisassumption,inStudy4wedirectlyassesswhetherthinkingaboutacrowdleadstotheheightenedaccessibilityofconceptsrelatedtounaccountability.Hence,weseekevi-dencethatconceptsrelatedtounaccountabilityandlackofresponsibilityaremoreaccessibleafterparticipantsimaginebeingwithagroupthanaftertheyimaginebeingwithoneotherperson.Additionally,anotherinterestingpossibilityisthatthinkingaboutbeinginagrouporcrowdtriggersfeelingsofcomfortandsafetyinnumbers,andthismightbewhatisdisruptinghelpingbehavior.Feelingasenseofcomfortmightleadonetointerpretahelpingsituationaslessinneedofhelp.Thatis,whenonefeelscomfortableandsecure,onemaylooktoanambiguoushelpingsituation,projectonesownfeelings,andseeasituationthatismoresecureandcomfortablethanonemighthaveotherwisesur-mised.Ifthisexplanationiscorrect,thenparticipantsinthegroupcondi-tioninStudy3helpedusoutlesswithourstudyintheotherroomnotbecausetheyfeltlessaccountablebutinsteadbecausetheyperceivedusasbeinglessinneedthandidthoseinthecontrolconditions.Hence,toinvestigatethispossibility,wealsoexaminedwhethernotionsofsafetyandcomfortbecomemoreaccessibleasaconsequenceofthinkingaboutaAtotalof52LehighUniversityundergraduatespartici-patedinabetween-subjectsdesign(socialcontext:group/oneperson/neutralcontrol)studyforcoursecredit.Participantswereescortedtoindividualexperimentalroomsequippedwithcomputers.Participantsthenfollowedself-guidedinstructionsthroughthesoftwareprogramMediaLab(Jarvis,2002).MediaLabinstructedparticipantsthattheywouldbecompletingacom-puterexerciseinwhichletterstringswouldflashonthescreen.Theletterstringsthemselveswouldeitherbewordsornonwords.Participantsweretoldtopressthekeyboardkeymarkedifthestringofletterswasawordorthekeymarkedifthestringwasanonword.Atthispoint,allparticipantsperformedalexicaldecisionpracticeexercisethatincluded3015words(e.g.,chair,table,door,knob)and15nonwords.Alltrialsbeganwithanorientingstimulus()thatwaspresentedinthecenterofthescreenfor250msandwasfollowedbyablankscreenfor400ms.Theletterstring(wordornonword)thenappearedinthecenterofthescreenfor50ms.Onceparticipantsindicatedwordornonword,anothertrialAfterthistrialsession,thecomputerintroducedparticipantstoourmanipulation.ParticipantsinthegroupandonepersoncontrolconditionsreadthemoviescenariousedinStudy2.Participantswererandomlyassignedtoreadoneoftheversionsassociatedwiththefollowingcondi-groupconditionImaginethatyouandafriendaresittinginacrowdedmovietheate)andonepersoncontrolconditionthatyouandafriendaresittingaloneinamovietheate).Participantsintheneutralcontrolconditionreadnoversion.Afterreadingthescenario,participantsansweredthefillerquestiononroomtemperaturedescribedinStudy2.Afterthisexperimentalmanipulation,thecomputerinstructedpartici-pantsthattheywoulddoanotherlexicaldecisiontask.Forty-twoletterstringswerethenpresentedinarandomorder.Again,eachwordornonwordwasprecededbyanorientingstimulus()thatappearedfor250msandwasfollowedbyablankscreenfor400ms.Thewordornonwordwasthenpresentedinthecenterofthecomputerscreenfor50ms.Becausemostprimingstudiesshowfacilitationeffects,forourrespon-sibilitycomposite,wechosesinglewordsthatwererelatedtotheconceptofunaccountability.Forourmainunaccountabilitycomposite,weselectedthreesynonymsofunaccountabilityunaccountable,innocent,Thesewordswerematchedtothreeneutralwordsrepresentingthesamedegreeofpositivitytoformamatchedneutralcomposite(e.g.,whimsical,impenetrableSixcomfort-relatedwordswerealsoincluded:protected,safe,unthreat-ened,secure,comforted,Thesewordswerealsomatchedto6neutraltraitwords(e.g.,diligent,intelligent).Finally,5neutralfillernounstable,mountain,flower,program,)and19nonwordswereOncompletingthecomputertask,participantswerethankedfortheirtimeandthendebriefedaboutthepurposeoftheexperiment.Datareduction.Toanalyzethereactiontimedata,wefol-lowedproceduresoutlinedinBarghandChartrand(2000).Datapointsthatwerethreestandarddeviationsaboveorbelowthemeanforeachwordwereconsideredoutliersandweredroppedfromsubsequentanalyses.Thisaffectedlessthan3%ofthedatapoints.Becausereactiontimedataareoftenskewed,weappliedalogtransformationtonormalizethedatatomeettheassumptionsofthestatisticaltests.Althoughthelogtransformedvalueswereusedintheanalyses,wereporttheactualreactiontimemeansinFillernouns.Weanalyzedparticipantsreactiontimestothefillernounsfirsttoensurethatparticipantsinthegroupandcontrolconditionsdidnotshowadifferenceinoverallreactiontimebycondition.Resultsshowedthattherewerenosignificantdiffer-encesinresponsetimestothefillernouns(groupcondition,129.00;controlconditions,(1,49) Weconsideredincludingthetermhowever,thistermhasaslightlydifferentconnotationthannotresponsible.Morespecifically,seemstoevokethoughtsofincompetence,whereasourintendedusageofnotresponsibledoesnot.GARCIA,WEAVER,MOSKOWITZ,ANDDARLEY Unaccountablecomposite.Ouraprioripredictionwasthatindividualsinthegroupconditionwouldrespondmorequicklytotheunaccountablecompositewordsthanwouldparticipantsintheonepersonandneutralcontrolconditions,whereasthetwogroupswouldshowequalresponsetimestothematchedneutralwords.Totestthisprediction,weconductedaplannedcontrastusingstan-dardizedcontrastscores(score)computedforeachparticipant.Wemultipliedthefollowingcontrastweightstoourdependentmeasures:1(unaccountablewords)and1(matchedneutralwords).Thisprocedureresultedinscoresrepresentingthesumofeachparticipantsresidualscoreaftereachscorehadbeenmulti-pliedbytheappropriatecontrastweight(Rosenthal&Rosnow,1985;Rosenthal,Rosnow,&Rubin,2000).Ananalysisofthesescoresbytypeofprimeyieldedasignificantdifferenceinthepredicteddirection,(1,50)Participantsinthegroupconditionrespondedmorequicklytotheunaccountablecompositewords(517.00ms,129.00ms)relativetothoseintheonepersonandneutralcontrolconditions(151.00ms).Incontrast,thoseinthegroup(158.00ms)andcontrol(558.00ms,ms)conditionsshowedsimilarreactiontimestothematchedneutralwords.SeeFigure2formeans.Asimpleeffectstestontheunaccountablecompositeonly,whichdoesnottakeintoaccountbaselinereactiontimestothematchedneutralwords,showedamarginallysignificantdifferencebetweenthegroupconditionandcontrolconditionsinthepredicteddirection,(1,49).09.Asimpleeffectstestonthematchedneutralcompositeshowednosignificantdifferencebycondition,(1,49).81,asexpected.Weconductedafollow-upplannedcomparisononthetoensurethatthetwocontrolconditionsdidnotdifferfromeachother.Thiscomparisonindicatedthatthetwocontrolconditions(onepersonandneutralcontrol)werenotsignificantlydifferentfromeachother,(1,29)Comfortcomposite.Totestthepossibilitythatthinkingaboutagrouphasitseffectonhelpingbehaviorbecausegrouporcrowdprimesthoughtsaboutcomfort,weperformedasimilarcontrastscores.Wemultipliedthefollowingcontrastweightstoourcomfortdependentmeasures:1(comfortwords)and1(matchedneutralwords).Resultsfromthisanalysisdidnotrevealasignif-icantdifferencebyexperimentalcondition,(1,50)Inlightofthesefindings,itappearsthatmerelyimaginingthepresenceofothersoracrowdedsituationleadsparticipantstorespondmorequicklytowordsrelatedtotheconceptofunac-countabilityrelativetomatchedneutralwords.Thefactthatthecrowdedandcontrolparticipantsrespondedequallyquicklytothematchedneutralwordsindicatesthatthiseffectisspecifictowordsrelatedtounaccountabilityandisnotduetoparticipantsinthegroupconditionrespondingmorequicklyacrosstheboard.Hence,itseemsthattheincreasedaccessibilityoftheconceptmaybethemechanismbywhichprimingthepresenceofothersproduceslesshelpingonasubsequent,unrelatedtask.Inaddition,resultsfromStudy4arenotconsistentwiththecomforthypothesis.Participantsprimedwithacrowdedsituationdidnotrespondsignificantlymorequicklytowordsrelatedtocomfortthandidthoseintheonepersonorneutralcontrolconditions.Study5OurpurposeinStudy5wastwofold.First,wewantedtoconceptuallyreplicatetheresultsofStudy4toensurethereliabil-ityofthereactiontimefindings.Second,wewantedtoinvestigatewhetherthecomforthypothesiswouldbesupportedifweusedastrongergroupprime.Forthisreason,wenotonlyhadparticipantsreadthemovietheaterprimesusedinStudy4butalsoaskedthemtolisttheirthoughtsaboutthetheaterfor3min.ThethoughtlistingasaprimestrengthmanipulationwasmodeledafteroneusedbyDijksterhuisandvanKnippenberg(1998),whichshowedthatthelongerindividualsareexposedtoprimes,thestrongertheeffectsAtotalof74PrincetonUniversityundergraduatespar-ticipatedinabetween-subjectsdesign(socialcontext:group/oneperson/neutralcontrol)studyforcoursecredit.Participantswereruninindividualexperimentalrooms.TheentireprocedureofStudy5wasthesameasthatusedinStudy4,exceptfortheprimemanipulation.Afterthepracticelexicaldecisionsession,thecomputerintroducedparticipantstoourmanipulation.ParticipantsinthegroupandonepersoncontrolconditionsreadthemoviescenariousedinStudies1and4groupconditionImaginethatyouandafriendaresittinginacrowdedmovietheate),andonepersoncontrolconditionImaginethatyouandafriendaresittingaloneinamovietheate).Theywerethengivenablanksheetofpaperandwereaskedtolistasmanythoughtsascametomindfor3min.Participantsinneutralcontrolconditionreadnoversionandwentdirectlytothereactiontimetask.ThelexicaldecisiontaskusedthesamewordsandthesamespecificationsasdidStudy4.ResultsandDiscussionDatareduction.Beforeanalyzingthereactiontimedata,weagainfollowedproceduresoutlinedinBarghandChartrand(2000)fordealingwithreactiontimedata.Datapointsthatwerethreestandarddeviationsaboveorbelowthemeanforeachwordwere Notethatweusetheerrortermonlyforthegroupsinvolvedinthebetween-subjectscomparison(group/onepersonandneutralcontrol)whenscoresinvolveawithin-subjectcomparison. Figure2.Study4reactiontime(inms)byconditionandcomposite.CROWDEDMINDS oftenskewed(foradiscussion,seeBargh&Chartrand,2000),weappliedalogtransformationtonormalizethedatatomeettheassumptionsofthestatisticaltests.Again,althoughweusedlogsintheanalyses,wereporttheactualmeansinmilliseconds.Fillernouns.Weanalyzedparticipantsreactiontimestothefillernounsfirsttoensurethatparticipantsinthegroupandcontrolconditionsdidnotshowadifferenceinoverallreactiontimebycondition.Resultsshowedthatthoseinthegroupcondition(73.00)andthecontrolconditions(94.00)respondedequallytothefillernouns,Unaccountablecomposite.Ouraprioripredictionwasthatin-dividualsinthegroupconditionwouldrespondmorequicklytotheunaccountablecompositewordsthanwouldparticipantsintheonepersonandneutralcontrolconditions,whereasthetwogroupswouldrespondequallyquicklytothematchedneutralwords.Totestthisprediction,weagainconductedaplannedcontrastusingastan-dardizedcontrastscore(score)computedforeachparticipant.Wemultipliedthefollowingcontrastweightstoourdependentmeasures:1(unaccountablewords)and1(matchedneutralwords).Thisresultedinscoresrepresentingthesumofeachsresidualscoreaftereachscorehadbeenmultipliedbytheappropriatecontrastweight.Ananalysisofthesescoresbyexperimentalconditionyieldedasignificantdifferenceinthepre-dicteddirection,(1,72).05.Asimpleeffectstestontheunaccountablecompositewordsonlyshowedthatparticipantsinthegroupconditionrespondedmorequicklytotheunaccount-ablecompositewords(429.00ms,99.00ms)thandidthoseinthetwocontrolconditions(476.00ms,(1,71).05.Incontrast,thoseinthegroup(450.00ms,101.00ms)andcontrol(461.00ms,94.00ms)conditionsshowedsimilarreactiontimestothematchedneutralwords,(1,71).57.SeeFigure3forthemeans.Afollow-upcomparisontoensurethatthetwocontrolcondi-tionsdidnotdifferintheirscoresagainshowedthatthetwocontrolconditions(onepersonandneutralcontrol)werenotsig-nificantlydifferent,(1,45)Comfortcomposite.Toexaminethepossibilitythatthinkingaboutagrouphasitseffectonhelpingbehaviorbecausegrouporcrowdprimesthoughtsaboutcomfort,weperformedasimilarcontrastusingscores.Wemultipliedthefollowingcontrastweightstoourcomfortdependentmeasures:1(comfortwords)1(matchedneutralwords).Althoughthepatternofmeanswasinthepredicteddirection,ananalysisofthesescoresbyexperimentalconditiondidnotrevealasignificantdifferenceby(1,72).24.Despitethestrongerprime,resultsfromthisanalysisdidnotrevealasignificanteffectofsocialcontextoncomfortExploringunaccountableandcomfort.Tofurtherexaminetherelationshipbetweenunaccountableandcomfort,weransomeadditionalregressionanalyses.BecauseStudies4and5werevirtuallyidentical(exceptforthedurationoftheprime),wecollapsedthedatafromthetwostudiestoincreasethepowerfortheanalyses.Beforecollapsingthedata,wefirsttransformedthedatafromeachofthetwostudiesintostandardizedscoresandthenconcatenatedthesedatasets.Ourgoalfortheseanalyseswastwofold.First,wewereinterestedinwhetheraboostinpowerwouldlendsupporttothecomforthypothesisand,ifso,whetherthisrelationshipwouldholdwhenparticipantsreactiontimestotheunaccountablecompositewereheldconstant.Second,wewantedtoinvestigatewhethertherelationshipbetweensocialcontext(onepersonandneutralcontrol/group)andunaccountablewouldholdifwecontrolledforparticipantsresponsetimestothecomfortcomposite.Toaddressthesequestionsandthusfurtherinvestigatethemechanismoftheimplicitbystandereffect,wefollowedthepro-ceduresoutlinedinR.M.BaronandKenny(1986)andKenny,Kashy,andBolger(1998).First,wewereinterestedinwhethertherelationshipbetweensocialcontextandthecomfortcompositewouldbecomesignificantwiththeboostinpowerbroughtaboutbycollapsingthetwostudies.Totestthis,wedummycodedsocialcontext(1oneperson/control;2group)anduseditastheindependentvariableintheanalyses.Regressionanalysesindeedshowedthatsocialcontexthadasignificantrelationshipwiththecomfortcomposite(.05).Thenwetestedwhethertherelationshipbetweensocialcontextandthecomfortcompositewouldremainsignificantwhenwecontrolledfortheunaccount-ablecompositebytreatingitasamediator.Socialcontexthadasignificantrelationshipbothwithparticipantsreactiontimestothecomfortcomposite(.05)andwithparticipantsreactiontimestotheunaccountablecomposite(themediatorinthis.01).Theunaccountablecompositealsohadasignificantrelationshipwiththecomfortcompositewhilecon-trollingfortheeffectofsocialcontext(However,whenparticipantsreactiontimestotheunaccountablecompositewerecontrolled,therelationshipbetweensocialcontext Evenbeforestandardizingthescores,weinitiallyconcatenatedtherawdatafromStudies4and5toverifythattherewerenosignificantinterac-tionsbetweensocialcontext(group/onepersonandneutralcontrol)andschool(Lehigh/Princeton)acrossourkeydependentmeasures.Allswerelessthan1.00;unaccountablecomposite,(1,122)comfortcomposite,(1,122).80;unaccountablematched(1,122).90;andcomfortmatchedneutral, Figure3.Study5reactiontime(inms)byconditionandcomposite.GARCIA,WEAVER,MOSKOWITZ,ANDDARLEY andcomfortdroppeddramatically(AfurtheranalysisusingtheratioofoverthestandarderrorofR.M.Baron&Kenny,1986)revealedthatincludingtheunac-countablecompositeasamediatorproducedasignificantreduc-tionintherelationshipbetweenconditionandcomfort(ra-Thepurposeofoursecondanalysiswastotestwhethertherelationshipbetweensocialcontextandtheunaccountablecom-positewouldremainsignificantwhenwecontrolledforthecom-fortcompositebytreatingitasamediator.Socialcontexthadasignificantimpactbothonparticipantsreactiontimestotheun-accountablecomposite(.01)andonparticipantsreactiontimestothecomfortcomposite(themediatorinthiscase;.05).Thecomfortcompositealsohadasignificantrelationshipwithparticipantsreactiontimestotheunaccountablecompositewhensocialcontextwascontrolledfor(.001).Resultswerenotconsistentwithamediationalinterpreta-tion,however,becauseanalysesshowedthattherelationshipbe-tweensocialcontextandtheunaccountablecompositeremainedstatisticallysignificant,evenwhenthecomfortcompositewascontrolledfor(Theseanalysessuggestanintriguingdynamicbetweenthecon-ceptsofunaccountableandcomfortandsuggestthatthediffusionofresponsibilityexplanationmayprovideatighteraccountoftheimplicitbystandereffectthandoesacomfortexplanation.Thatis,asignificantrelationshipexistsbetweenunaccountablethoughtsandsocialcontext,evenwhenonecontrolsforwordsassociatedwithcomfort.However,comfortnolongerhasasignificantrela-tionshipwithsocialcontextwhenonecontrolsforparticipantsreactiontimestounaccountablewords.Althoughtheseresultsdonotcategoricallydeterminewhetheradiffusionofresponsibilityaccountisbetterthanacomfortone,theydosuggestthatanyrelationshipcomforthaswithsocialcontextmayhavesomethingtodowithresponsibility.WespeculatethatsomepeoplemayfeelmorecomfortedandsecureasaconsequenceoffeelinglessGeneralDiscussionThepresentstudieshelpadvanceourcurrentunderstandingofthebystanderapathyliteraturebysuggestinganalternativeac-countinwhichimaginedothersneednotbeimbeddedorimpliedinthehelpingsituationathandforthebystanderapathyeffecttobeproduced.Thatis,inourstudies,individualswhojustimaginedthepresenceofothersatTime1exhibitedlesshelpingbehavioronasubsequent,completelyunrelatedtaskatTime2,evenwhenimaginingotherswhocouldnotpossiblyhelpout.Forinstance,imaginingstrangersinthemovietheatershouldhavenobearingonannualgivingtoauniversityalumnicampaign,butfromthecurrentresearchitisclearthatitdoes.Althoughwefeelthisresearchprovidesevidencethat,undersomecircumstances,amoresocialcognitiveaccountmaybesufficienttoexplainbystanderapathy,itbynomeanssuggeststhattraditionalaccountssuchasclassicdiffusionofresponsibility(e.g.,Darley&Latane,1968)orconfusionofresponsibility(e.g.,Ca-cioppoetal.,1986)arewrong.Thatis,ouralternativeaccountmayinteractwithtraditionalonesinanadditiveoremergentway.Nevertheless,contrarytoourcurrentunderstandingofbystanderapathy,thepresentresearchsuggeststhatthepresenceofothersneednotbeimbeddedinorassociatedwiththehelpingsituationtoexertdiffusionofresponsibilityeffects.Thepresentresearchalsopresentsevidencethatmerelyimag-iningagroupcanleadtolessenedlevelsofresponsibility.Thatis,fromourreactiontimestudies,wehavelearnedthatparticipantswhoimaginethepresenceofothersrespondmorequicklytowordssuchasrelativetoparticipantsinthecontrolandonepersonconditions.Althoughdiffusionofresponsibilityap-pearstocharacterizethemechanismbehindtheimplicitbystandereffect,futureresearchmustinvestigatemorepreciselyhowthismechanismworks.Theaccountthatweputforthholdsthataparalleloftheexplicittypeofbystanderapathyiswhatisbeingevidenced.Thatis,peoplewhoimaginegroupsituationshavenotionsofunaccountabilitytriggered,andthisleadsthemtoseethesituationasonethatdoesnotcallfortheirpersonalhelp.However,thereisthepossibilitythatthemechanismthroughwhichthegroupprimesexertaninfluenceislesscomplicatedthanwesurmised.Pastresearch(e.g.,Barghetal.,1996;Dijksterhuis&vanKnippenberg,1998)hasshownthatprimescandrivebehaviorunmediatedthroughsuchconstrualprocesses.Itremainswhollypossiblethatbystanderapathycouldalsobecausedbybehavioralmanifestationsofunaccountability,whicharedirectlytriggeredbypriminggroups(orphysicallybeinginagroup),andthesebehav-iorsinterferewithordirectlyopposehelpingbehavior.Thecurrentstudiesdonotdistinguishbetweentheconstrualhypothesisandthedirectprimingofbehavioralmanifestations,and,infact,bothprocessesarecompatiblewithourdata.Whatthedatarevealisthattheprimingofgrouprepresentationsincludesandsimultaneouslyprimesnotionsoflackingresponsibilityandaccountabilitythatappeartodisrupthelpingbehavior.Thatis,evenwhenoneisnotphysicallyinagroup,thereisthepossibilityforanimplicitbystandereffect,inwhichdiffusionofresponsibilityistriggeredimplicitlythroughprimesandspreadingactivation.FutureDirectionsFromthepresentstudy,weknowthatprimingtheconceptofgroupleadstotheimplicitbystandereffect.However,aninterest-ingavenueforfutureresearchistoexaminewhetherprimingtheconceptofbeingwithagroupbelowparticipantslevelofcon-sciousawarenessstillresultsinthiseffect.Thatis,onecanenvisionwordsrelatedtotheconceptofbeingwithagroupflashingonacomputerscreentotestwhethertheimplicitby-standereffectcanbeinducedbyunconsciouspriming.Totheextentthatactivatingtheconceptofagrouporcrowdedsituationmakesaccessiblenotionsofunaccountability,itshouldnotmatterwhetherparticipantsareawareofthegroupprimesorwhethertheyareprimedbelowtheirconsciousawareness.Nevertheless,thisremainsanempiricalquestion.Futureresearchshouldalsoexplorewhetherthetypeofgroupimaginedplaysaroleintheimplicitbystandereffect.Forinstance,theinterconnectednessofindividualsinagroupsituationmayinfluenceresponding.Inclassicbystanderstudies,thebystanders Incidentally,theseanalysesyieldsimilarresultswiththeunstandardizeddatasetcombiningStudies4and5.Again,theseanalysesyieldsimilarresultswiththeunstandardizeddatasetcombiningStudies4and5.CROWDEDMINDS athandoftenhavenorealsocialconnectiontoeachotherexceptforbeingatthesameplaceatthesametime.Thatis,inthesecases,thegroupisjustacollectionofindependentindividuals.Bystanderapathystudieshavealsodemonstratedthiseffectwithbystanderswithsocialconnectionstotheothersaroundthem.Inthesecases,anindividualmaybeamemberofthegroupathand.Alongtheselines,ourstudiesprovideevidencefortheimplicitbystandereffectusingbothunconnectedothersandconnectedothers.Namely,ourdiningoutprimeshowsthiseffectwithimaginedacquaintances,whereasourmovietheaterprimeshowsthebystandereffectwhenparticipantsweresimplyimaginingstrangers.Althoughwegrantthatbothtypesofprimesleadtolesshelpingbehavior,itmightbeinterestingtoexplorewhetherimaginingonetypeofgroupbe-comesastrongerprimethananother.Dopeopleexperienceagreatersenseofunaccountabilitywhentheyareprimedwithbeinginaninterconnectedgroupthanwithbeingwithrandomstrangers?Additionally,thereisapossibilitythatsometypesofgroupprimesmayactuallyprimemorehelpingbehaviorratherthanless.Forinstance,leadingpeopletothinkaboutgroupsofprosocialothers,suchasagroupoflifeguardsordoctors,mayincreasehelping.Alongtheselines,manipulationsthatincreaseself-awareness(e.g.,Duval&Wicklund,1972)mayalsointeractwithgroupprimestoincreasehelpingbehavior.Forinstance,ifapersonimaginesthateveryoneinacrowdedplaceisfocusedonhimorherorifapersonthinksofacrowdedsituationwhilestandinginfrontofamirror,helpingwilllikelyincreaseaspeopleattempttomaketheirbehaviorconsistentwiththeirstandards.Furthermore,levelofidentificationwithaparticulargroupmayaffecthowagroupprimeinfluenceshelpingbehavior.AllofthesequestionsarebeyondthescopeofthisarticleanddeservefutureOverall,thisresearchsuggeststhatprimingthepresenceofothersatTime1affectshelpingbehavioronacompletelyunre-latedtaskatTime2.Thisresearchcontributestotheprimingmethodologyinthatitprimesasocialcontext.Secondly,thisresearchadvancesourunderstandingofbystanderapathyandproposesanadditionalaccountforthebystanderapathyeffect.Futureresearchmustcontinuetoexploretheboundaryconditionsoftheimplicitbystandereffectaswellasthemechanismsthatcontributetothisprocess.Baldwin,M.W.(1994).Primedrelationalschemasasasourceofself-evaluativereactions.JournalofSocialandClinicalPsychology,13,Baldwin,M.W.(1995).Relationalschemasandcognitionincloserela-JournalofSocialandPersonalRelationships,12,Baldwin,M.W.,Fehr,B.,Keedian,E.,Seidel,M.,&Thomson,D.W.(1993).Anexplorationoftherelationalschemataunderlyingattachmentstyles:Self-reportandlexicaldecisionapproaches.PersonalityandSocialPsychologyBulletin,19,Baldwin,M.W.,Keelan,J.P.R.,Fehr,B.,Enns,V.,&Koh-Rangarajoo,E.(1996).Socialcognitiveconceptualizationofattachmentworkingmodels:Availabilityandaccessibilityeffects.JournalofPersonalityandSocialPsychology,71,Bargh,J.A.,&Chartrand,T.L.(2000).Themindinthemiddle:Apracticalguidetoprimingandautomaticityresearch.InH.T.Reis&C.M.Judd(Eds.),Handbookofresearchmethodsinsocialandper-sonalitypsychology.Cambridge,England:CambridgeUniversityPress.Bargh,J.A.,Chen,M.,&Burrows,L.(1996).Theautomaticityofsocialbehavior:DirecteffectsoftraitconceptandstereotypeactivationonJournalofPersonalityandSocialPsychology,71,Bargh,J.A.,&Thein,R.D.(1985).Individualconstructaccessibility,personmemory,andtherecalljudgmentlink:ThecaseofinformationJournalofPersonalityandSocialPsychology,49,Baron,J.,&Miller,J.G.(2000).Limitingthescopeofmoralobligationstohelp:Across-culturalinvestigation.JournalofCross-CulturalPsy-chology,31,Baron,R.M.,&Kenny,D.A.(1986).Themoderatormediatorvariabledistinctioninsocialpsychologicalresearch:Conceptual,strategicandstatisticalconsiderations.JournalofPersonalityandSocialPsychol-ogy,51,Bickman,L.(1972).SocialinfluenceanddiffusionofresponsibilityinanJournalofExperimentalSocialPsychology,8,Cacioppo,J.T.,Petty,R.E.,&Losch,M.E.(1986).Attributionsofresponsibilityforhelpinganddoingharm:EvidenceforconfusionofJournalofPersonalityandSocialPsychology,50,Carver,C.S.,Ganellen,R.J.,Froming,W.J.,&Chambers,W.(1983).Modeling:Ananalysisintermsofcategoryaccessibility.JournalofExperimentalSocialPsychology,19,Clark,R.D.,&Word,L.E.(1974).Whereistheapatheticbystander?Situationalcharacteristicsoftheemergency.JournalofPersonalityandSocialPsychology,29,Darley,J.M.,&Latane,B.(1968).Bystanderinterventioninemergencies:Diffusionofresponsibility.JournalofPersonalityandSocialPsychol-ogy,8,Darley,J.M.,Teger,A.L.,&Lewis,L.D.(1973).Dogroupsalwaysinhibitindividualsresponsestopotentialemergencies?JournalofPer-sonalityandSocialPsychology,26,Davis,M.H.,Mitchell,K.V.,Hall,J.A.,Lothert,J.,Snapp,T.,&Meyer,M.(1999).Empathy,expectations,andsituationalpreferences:Person-alityinfluencesonthedecisiontoparticipateinvolunteerhelpingJournalofPersonality,67,Dijksterhuis,A.,Spears,R.,&Lepinasse,V.(2000).Reflectinganddeflectingstereotypes:Assimilationandcontrastinautomaticbehavior.JournalofExperimentalSocialPsychology.Dijksterhuis,A.,&vanKnippenberg,A.(1998).Therelationbetweenperceptionandbehavior,orhowtowinagameoftrivialpursuit.ofPersonalityandSocialPsychology,74,Duval,T.S.,&Wicklund,R.(1972).Thetheoryofobjectiveselfaware-NewYork:AcademicPress.Epley,N.,&Gilovich,T.(1999).Justgoingalong:Nonconsciousprimingandconformitytosocialpressure.JournalofExperimentalSocialPsy-chology,35,Higgins,E.T.,Rholes,W.S.,&Jones,C.R.(1977).Categoryaccessibilityandimpressionformation.JournalofExperimentalSocialPsychol-ogy,13,Jarvis,B.G.(2002).MediaLabResearchSoftware(Version2002)[Com-puterprogram].NewYork:Empirisoft.Kawakami,K.,Young,H.,&Dovidio,J.F.(2000).Automaticstereotyp-ing:Category,traitandbehavioractivations.ManuscriptsubmittedforKenny,D.A.,Kashy,D.A.,&Bolger,N.(1998).Dataanalysisinsocialpsychology.InD.Gilbert,S.Fiske,&G.Lindzey(Eds.),Thehandbookofsocialpsychology(Vol.1,pp.233265).Boston:McGraw-Hill.Korte,C.(1971).Effectsofindividualresponsibilityandgroupcommu-nicationonhelp-givinginanemergency.HumanRelations,24,Laner,M.R.,Benin,M.H.,&Ventrone,N.A.(2001).BystanderattitudesGARCIA,WEAVER,MOSKOWITZ,ANDDARLEY Behavior,22,Latane,B.(1970).Fieldstudiesinaltruisticcompliance.ResearchinSocialPsychology,1,Latane,B.,&Darley,J.M.(1968).Groupinhibitionofbystanderinter-JournalofPersonalityandSocialPsychology,10,Latane,B.,&Darley,J.M.(1970).Theunresponsivebystander:Whythehelp?NewYork:Appleton-Century-Crofts.Latane,B.,&Nida,S.(1981).TenyearsofresearchongroupsizeandPsychologicalBulletin,89,Macrae,C.N.,&Johnston,L.(1998).Help,Ineedsomebody:Automaticactionandinaction.SocialCognition,16,Markus,H.(1977).Self-schemataandprocessinginformationabouttheJournalofPersonalityandSocialPsychology,35,Neuberg,S.L.(1988).Behavioralimplicationsofinformationpresentedoutsideofconsciousawareness:Theeffectofsubliminalpresentationoftraitinformationonbehaviorintheprisonersdilemmagame.Cognition,6,Prentice,D.A.,&Miller,D.T.(1996).Pluralisticignoranceandtheperpetuationofsocialnormsbyunwittingactors.InM.P.Zanna(Ed.),Advancesinexperimentalsocialpsychology(Vol.29,pp.161209).SanDiego,CA:AcademicPress.Rosenthal,R.,&Rosnow,R.(1985).Contrastanalysis:Focusedcompar-isonsintheanalysisofvariance.NewYork:CambridgeUniversityRosenthal,R.,Rosnow,R.,&Rubin,D.(2000).Contrastsandeffectsizesinbehavioralresearch:Acorrelationalapproach.NewYork:Cam-bridgeUniversityPress.Wiesenthal,D.L.,Austrom,D.,&Silverman,I.(1983).Diffusionofresponsibilityincharitabledonations.BasicandAppliedSocialPsychol-ogy,4,ReceivedDecember29,2001RevisionreceivedFebruary9,2002AcceptedFebruary13,2002 CROWDEDMINDS