Automatic and Controlled

logisticslilacAI and Robotics

Feb 23, 2014 (3 years and 6 months ago)

117 views

292 SOCIAL PSYCHOLOGY AND THE UNCONSCIOUS
personalitymeasures.EuropeanJournal of PsychologicalAssessment,21(2),
100-107.
Sherman,S.J.,Presson,C.C.,Chassin,L.,Rose,J.S.,& Koch,K.(2003).Implicitand
explicitattitudestowardcigarettesmoking:Theeffectsof contextandmotivation.
Journalof SocialandClinicalPsychology,22,13-39.
Simon,H.A.(1955).Abehavioralmodelof rationalchoice.QuarterlyJournalofEconom-
ics,69,99-118.
Sriram,N.,Greenwald,A.G.,& Nosek,B.A.(2006).Scaleinvariantcontrastsofresponse
latencydistributions.Unpublishedmanuscript.
Steffens,M.C..(2004).Is theImplicitAssociationTestimmunetofaking?Experimental
Psychology,51(3),165-179.
Steffens,M.C.,&Buchner,A.
(2003).
ImplicitAssociationTest:Separatingtranssituation-
allystablesandvariablecomponentsof attitudestowardgaymen.Experimental
Psychology,50(1),33-48.
Steffens,M.C.,& Plewe,I.(2001).Items'cross-categoryassociationsasaconfounding
factorintheImplicitAssociationTest.Zeitschriftfur ExperimentellePsychologie,
48,123-134.
Teachman,B.A.,Gregg,A.P.,&Woody,S.R.(2001).Implicitassociationsforfear-relevant
stimuliamongindividualswithsnakeandspiderfears.Journalof AbnonnalPsy-
chology,110(2),226-235.
Teachman,B.A.,&Woody,S.R.(2003).Automaticprocessinginspiderphobia:Implicit
fearassociationsoverthecourseof treatment.Journal of AbnonnalPsychology,
112(1),100-109.
Teachman,B.A,Gapinski,K.D.,Brownell,K.D.,Rawlins,M.,& Jeyaram,S.(2003).
Demonstrationsof implicitanti-fatbias:Theimpactof providingcausalinforma-
tionandevokingempathy.HealthPsychology.22(1),68-78.
Teige,S.,Schnabel,K.,Banse,R.,& Asendorpf,J.B.(2004).Assessmentof multiple
implicitself-conceptdimensionsusingtheExtrinsicAffectiveSimonTask(EAST).
Unpublishedmanuscript.
Wegner,D.M.(2002).Theillusionofconsciouswill.Cambridge,MA:BradfordBooks.
Wigboldus,D.(2004).Single-targetimplicitassociations.Unpublishedmanuscript.
Wilson,T.D.,Lindsey,S.,& Schooler,T.Y.
(2000).
A modelof dual attitudes.Psychological
Review,107(1),101-126.
Wittenbrink,B.,Judd,C.M.,& Park,B.(1997).Evidencefor racialprejudiceat the
implicitlevelanditsrelationshipwithquestionnairemeasures.JournalofPersonal-
ityandSocialPsychology,72(2),262-274.
Wittenbrink,B.,Judd,C.M.,& Park,B.(2001).Spontaneousprejudicein context:
Variabilityin automaticallyactivatedattitudes.Journalof PersonalityandSocial
Psychology,81(5),815-827.
AutomaticandControlled
ComponentsofSocialCognition:
AProcessDissociationApproach
B.KEITHPAYNEandBRANDON D.STEWART
I
n afamousdescriptionof unintendedbehavior,WilliamJames(1890)noted
that,"Veryabsent-mindedpersonsin goingto theirbedroomtodressfor
dinnerhavebeenknowntotakeoff onegarmentafteranotherandfinallyto
getintobed,merelybecausethatwasthehabitualissueofthefirstfewmovements
whenperformedat alaterhour,"(p.115).Thiskindof absent-mindedness,or
somethinglikeit,willprobablystrikemostreadersasalltoofamiliar.Comparethis
witha descriptionbytheneurologistFrancoisLhermitteof apatientnearlya
centurylater.Whenshownintoabedroom,thepatient"immediatelybegantoget
undressed.Hegotintobed,pulledthesheetuptohisneck,andpreparedtogoto
sleep,"(Lhermitte,1986,p.338).We canbe surethatthisbehavioris more
exceptionalthanJames'absent"mindedness,becausethebedroombelongedto
Lhermitte.
Thepatienthadamassivesurgicallesionoftheleftfrontallobe,abrainregion
criticalforstrategicplanningandcontrolofactions.Thisandothersimilarpatients
sufferedfromwhatLhermittetermed"environmentaldependencysyndrome."
Althoughtheirbehaviorswerecoordinatedandcomplex(notsimplereflexes),they
wereunderthecontrolof theenvironmenttoastrikingextent.Theybehavedin
accordwithwhateverenvironmentalcuescaughttheireye.In onedemonstration,
uponnoticingthatasyringehadbeenlaidout,apatientpickedit upandbeganto
givetheneurologistaninjection.Thereaderisnottoldwhethertheinjectionwas
carriedout.In another,evenmoremacabreinvestigation,thepatientwasledtoa
tablewhereapistolandsomebulletshadbeenplaced.Withoutpauseor com-
ment,hepickedupthepistol,pulledbackthemagazine,andloadedit.At this
pointweread,"Theexperimentwasthenstopped,"(p.338).Thesewerenotthe
momentarylapsesof voluntarycontrolthatwe all experience,but profound
absences.Lhermittedescribedit asadisorderof autonomy.
No lessintriguingarethestrugglesof individualswithalienhandsyndrome.
294 SOCIAL PSYCHOLOGY AND THE UNCONSCIOUS
Becauseofdamagetothefrontallobesorthecorpuscallosumconnectingleftand
righthemispheres,thesepatientsexperience"autonomous"actionsby oneor
morelimbs.The actionsareautonomousin thesensethatthepatientcannot
voluntarilycontrolthem,nordotheyexperiencethebehaviors(noreventhelimb
itself)astheirmVll.LikeLhermitte'spatients,the"alien"actionsareusuallytrig-
geredbyenvironmentalcues.For example,patientsmayexperiencea"struggle
betweenthehandsaseachattemptsto answerthetelephone,"or evenmore
drastically,"onehandtriedtoturnleftwhentheotherhandtriedtoturnright
whiledrivingacar,"(Doody&Jankovic,1992;p.807).MarchettiandDellaSala
(1998)reportapatientwho,"atdinner,muchtoherdismaysawherlefthand
takingsomefishbonesfromtheleftoversandputtingthemintoher mouth,"
(p.196).
Thesebizarreconditionsseemso strangebecausetheyare dissociations
betweenfunctionsthatusuallyworkseamlesslytogether,withintentionsreiningin
theautomaticwhenit goesastray.Dissociationsaresoinformativebecauseso
muchcanbelearnedaboutthestructureof asystemfromwhereitsfaultlineslie.
Whenacrystallineicicleshattersonthefloorit comesapartinremarkablyregular
patterns,verydifferentfromtheshatteringof anegg.Their innerorganization
revealsitselfinthewaystheycomeapart.Neurologistshavebeenstudyingdis-
sociationscausedbyanatomicallesionsfor morethana century.Neurological
dissociationshavemorerecentlyattractedtheattentionof socialpsychologistsfor
theirpotentialto shedlightonthecontrolof complexthoughtprocessessuch
asthoseinvolvedinsocialbehavior(e.g.Bargh,2005;BeeretaI.,2003;Wegner,
2002).
Althoughanatomicaldissociationsmapout thesefaultlinesin vividdetail,
dissociationsinthebehaviorofhealthymenandwomencanbejustasinformative.
One way to find the seamsbetweenthe intentionallycontrolledand the
uncontrolledis to compareexplicitandimplicittasks.Amnesiacpatients,for
example,showprofounddeficitswhentestedusingexplicitmemorytests,which
askthepersontointentionally'retrieveamemory.Yetwhentested\vithimplicit
memorytests,theirperformanceshowseffectsof pastexperience\vithoutthe
intenttorememberor thefeelingof remembering(Shimamura,1986).Butit is
notjustamnesiacpatientswhoshowthisdifference.Normalhealthycollegestu-
dentsalsoshowdissociationsbetweenimplicitandexplicitmemorytests(Jacoby
&Dallas,1981).Factorsthataffectperformanceononekindoftestoftenhaveno
impactontheother(seeRoediger&McDermott,1993).Withthehelpofimplicit
andexplicittasks,researcherscanprobethedistinctionsbetweenmentalpro-
cesseswithoutrelyingonrare,unfortunatecasesof braindamage.Theycaninfer
theinnerstructureswithoutwaitingforthemtocrack.Thefactthathealthymen
andwomenshowdissociationsbetweenintentionalandunintentionalaspectsof
behaviorforcesustoaskaboutthefaultlinesinordinarythought.Doweallhave
littledisordersofautonomy?In someways,weshallsee,theanswerisyes.
The casesreviewedhere are all dissociationsbetweenintentionaland
unintentionalprocesses,eachat differentlevelsof analysis.The dissociations
becomeincreasinglysubtle,butnolessintriguingaswezoominfromthemacro
scaletothemicro.Lhermitte'sfrontalpatientsshowedadissociationatthelevelof
METHODS:PROCESS DISSOCIATIONPROCEDURE 295
thewholeperson.Mostpeopleactwithintentionalcontrolmostof thetime,but
thesepatientsdidnot.Casesofalienhandrepresentdissociationswithinaperson,
butbetweenlimbs.Onehandisundervoluntarycontrol,andtheotherisnot.At a
stillfinergrainaredissociationsbetweentasks.Thesamepersonactsdifferently
onthetwotasksbecausethetasksdrawondifferentprocesses.
We canzoomin further.This chapterfocuseson a processdissociation
approach
-
atechniquefor separatingintentionalandunintentionalcontributions
tothesamebehaviorperformedbythesamepersonatthesametime.Imagine
thatamaniswalkingtowardyouonthestreet.Althoughyoudonotrecognizehim,
youhaveabadfeelingabouthimanddecidetocrossthestreet.If youhadbeen
abletointentionallyretrievethefactthatyousawhimyesterdayon awanted
poster,youwouldhavehadanevenbetterbasisforyourdecision,andmighthave
hurriedfaster,or calledthepoliceafterward.But evenwithoutbeingableto
remembertheposter,thevaguesenseofthreatcanalsoguideyourresponse.Here
intentionalandunintentionalformsoflearningcouldbothfeedintotheresponse,
invaryingdegrees.Thefactthatyoucanhaveonewithouttheotherillustratesthat
theyareseparable.Just aslesionstudiesallowdramaticdissociationsbasedon
anatomy,theprocessdissociationapproachseeksto separateintentionaland
unintentionalinfluences,eventhoughtheynormallyoperatetogether.
Thegoalofthischapteristooverviewthelogicbehindtheprocessdissociation
approach,whatit measures,andwhatit doesnot.Wewill describesomeof the
manydifferenttopicswhereprocessdissociationhasbeenused,andthekindsof
insightsit canprovidefor socialpsychology.Alongtheway,wewill discussthe
assumptionsthatmustbemettoproperlyusetheprocedure,andwewillseehow
thiswayof thinkingabouttheautomatic-controlleddistinctioncomparesand
contrastswith other prominentapproaches.Althoughthe procedurewas
developedin thecontextof memoryresearch(Jacoby,1991;Jacoby,Toth,&
Yonelinas,1993)we\villfocusonapplicationsoutsideof purememoryresearch,
emphasizinginsteadsocialcognitionandbehavior.Interestedreadersarereferred
to Jacoby(1998)for an overviewof theprocedureasdevelopedin memory
researchandYonelinas
(2002)
foratheoreticalreviewofdual-processtheoriesof
memoryusingprocessdissociationandrelatedmethods.
PROCESSDISSOCIATIONINSOCIALCOGNITION
A studybyHense,Penner,andNelson
(1995)
markedapointof departurefrom
pure memoryresearchtosocialmemory.We\vill usethisexperimenttoillustrate
howprocessdissociationcanbe usedto studysocialmemorydistortions.Partici-
pantswere askedto remembera list of traitsthat describedelderlyor young
individuals.Each trait was stereotypicalfor either old or youngpeople.After
studyingthe traits,participantswere askedto recallthe traitsthatdescribedthe
older andyoungertargetpersonsundertwo setsof instructions.In the inclusion
condition,participantswere askedto respondwith thetrait theyhadstudiedor,
if theycould not rememberthe trait,respondwith the first word that cameto
mind.In thiscondition,responsescanbedrivenby eitherintentionallyretrieved
j
296 SOCIAL PSYCHOLOGY AND THE UNCONSCIOUS
memoriesorbyautomaticformsofmemorythatcausecertainthoughtstocometo
mindmorereadily.In theinclusionconditionintentionalandunintentionalforms
of memoryworkinconcert.In theexclusioncondition,participantswereaskedto
respondwithanewtraitthatwasnotstudied.If theyrememberedhavingstudied
a trait,theycouldsuccessfullyavoidreportingit.However,if theyfailedto
consciouslyrememberatrait,butit unintentionallycametomind,theywould
belikelytoreportit.Thisconditionpitsintentionalandunintentionalformsof
memoryagainsteachother.
By comparingperformancein inclusionandexclusionconditions,theinflu-
enceof intentionalandunintentionalusesof memorycanbeestimated.Tothe
extentthatapersonrespondswiththecorrecttraitwhentheytryto,butalso
\vithholdsitwhentheytryto,memoryisunderintentionalcontrol.Becauseinthis
paradigmsubjectiveawarenessofthememoryisthebasisforcontrollingmemory
reports,consciousnessof thememorycanalsobeinferred.Buttotheextentthat
pastexperienceinfluencesperformanceregardlessofwhatparticipantsaretrying
todo,theyarebeingunintentionallyinfluencedbymemory.Jacobyandcolleagues
havetermedthisthe"logicof opposition."Theequationsfor estimatingthese
influenceswillbediscussedinalatersection.
Henseandcolleaguesfoundthatstereotypicaltraitshadaselectiveinfluence
on theunintentionaluseof memorywithoutaffectingcontrolledrecollection.
Stereotype-consistenttraitssuchasslowandfrailcametomindeasilyandbiased
memoryreportswhethertheyweretryingtoretrievethemortryingnottoretrieve
them.Consciouslycontrolledmemorywasaffectedbyadividedattentiontask,but
thiswasindependentof thestereotypingeffect.
This studyrevealedanimportantdissociation.Withinthesingleactivityof
rememberingtraits,intentionalandunintentionalformsof memorybothfedinto
responses.Automaticinfluencesreflectedparticipants'ownstereotypes.In con-
trast,consciouslycontrolledmemorywasaffectedbydividedattention.Thisstudy
helpedtocharacterizethemechanismsbehindstereotypicalmemorybiasesthat
havebeenknownforyears(forareviewseeStangor& McMillan,1992).Social
psychologyhasatraditionof emphasizingdissociationsbetweenunderlyingpro-
cessesthatwouldseeminglygotogether.Devine's(1989)importantdemonstra-
tionthattheautomaticactivationof stereotypescouldbedissociatedfromtheir
useprovidesa clearexamplein thedomainof stereotyping.WhereasDevine
contrastedresultsfromaprimingtaskwithresultsfromself-reportmeasures,the
studybyHenseandcolleagues(1995)contrastedtwoaspectsof asingleactof
remembering.
Evenwithinthesametask,intentionalandunintentionalusesof memory
operateverydifferently.Usinga memoryparadigmin whichracestereotypes
couldbiasmemory,wehaveexploredthewaysthatcomponentsof memoryare
relatedtosubjectiveexperience(Payne,Jacoby,&Lambert,2004).Wefoundthat
people'ssubjectivesenseof confidencein theirmemorieswaswell-attunedto
recollection.Whentheywereconsciouslyrecollectingthepast,theyexpectedto
beright,andtheyusuallywere.Whentheyhadnorecollection,theyexpected
tobewrong,andtheyusuallywere.Butconfidencewasnotatalltunedintothe
automaticinfluencesof stereotypesonmemory.Whenconsciousmemoryfailed,
METHODS:PROCESS DISSOCIATIONPROCEDURE 297
feelingsofconfidencegavenocluesaboutwhetheranautomaticbiaswasatwork.
Thisasymmetryof subjectiveawarenesshasimportantconsequencesforavoiding
stereotypicalbiases.
In oneconditionof thestudywerequiredparticipantstoanswereverymem-
oryquestion,whethertheyrememberedthecorrectansweror not.In theother
condition,weinstructedparticipantstoansweronlyif theybelievedtheiranswer
tobecorrect.Whateffectdidthefreedomtochoosehaveonmemoryreports?
Thatdependedonwhataspectof memoryislookedat.Whenit cametooverall
accuracy,thefreedomtochoosehelpedmemory.Becausesubjectiveconfidence
waswell-tunedtorecollection,participantswereabletoavoidansweringquestions
theywouldgetwrong,andtheychosetoanswerthoseitemstheywerelikelytoget
correct.
Butthestorywasverydifferentwhenit cametotheinfluenceof stereotypes.
Memoryreportswerebiasedtowardstereotype-consistentmemoryerrors.But
critically,thepatternof stereotype-consistentbiaswasjustasstrongwhenpartici-
pantswereallowedtochooseaswhentheywererequiredtoanswereveryques-
tion.Whenit cametorecollection,asinmanyareasof life,awarenessbestowed
control.Butwhenstereotypescamereadilytomind,Jamalwaslikelytobecomean
athlete,andWalterwaslikelyto becomea politicianindependentof intent,
independentof subjectiveexperience,andindependentof thechoiceto keep
quiet.
Thesestudiesmovedfrombasicmemoryresearchtosocialmemorybiases.
But theprocessdissociationprocedurecanalsobeusedin contextscompletely
unrelatedto memory.Imaginenowthatthepersonapproachingonthestreet
lookslikeasuspectandyouareapoliceofficer.Thesuspectpullsanobjectfrom
hispocket.Whatshouldyoudo?Thiswasthedecisionfacingfourpoliceofficers
onFebruary4,1999whentheyconfrontedAmadouDiallooutsidehisapartment
inNewYorkCity.Theirdecisionturnedouttobewrong.Diallowaskilleddespite
beingunarmedandhavingnothingtodowiththecrimetheofficerswereinvesti-
gating.Thecasesparkedpublicoutrageandchargesof racismbecauseDiallowas
Black.However,asinsomanyinstancesindailylife,therewasno"controlgroup"
togaugetheimpactof race.Sincethatincident,manysimilarc;aseshavebeen
reportedinthenationalpress.Manyinsideandoutsideof lawenforcementhave
askedthemselves,"WhatwouldI dointhatsituation?"Wehaveusedtheprocess
dissociationproceduretostudywhatpeopleactuallydointhatkindofsplit-second
decisionsituation.
Toseehowprocessdissociationcanbeinformativehere,considerthediffer-
entpossiblescenariosthatmightconfronttheofficer.In onescenario,a Black
suspectpullsagun.Here,the"correct"response(atleastforthepurposesof our
thoughtexperiment)is to"shoot."Thatresponsemightcomeaboutfromtwo
routes.Oneis anintentionallycontrolledresponse,inwhichyoushootbecause
youmeantoshoot.Thesecondrouteis anunintentionalor automaticresponse
evokedby race stereotypesabout the suspect.Becauseintentionaland
unintentionalprocessesareworkingtogetherhere,wecannottell themapart.
Nowconsideradifferentscenario,inwhichtheBlacksuspectholdsonlyawallet
(as,infact,Dialloheld).Hereanintentionallycontrolledresponsewouldbetonot
298 SOCIAL PSYCHOLOGY AND THE UNCONSCIOUS
shoot.Butanautomaticresponsebasedonracialstereotypesmightstillleadone
topull thetrigger.To theextentthatapersonsystematicallyshootsdespitenot
intendingto,weknowthatanunintentionalprocessisatwork.Buttotheextent
thatapersonshootswhentheymeanto,andnototherwise,weknowthattheyare
incontrol.
Our laboratoryhasconductedanumberof studiesexaminingthesesortsof
scenarios.WehaveusedasimpleprocedureinwhichthefacesofBlackandWhite
individualsareflashedonacomputerscreenjustbeforepicturesofhandgunsand
handtoolsareshown.All ofthepicturesarepresentedlongenoughtoseeclearly
andalloftheitemsareeasytoidentifY.Weshowpairsofobjectsonthescreenand
askparticipantstorespondbypressinga"gun"keyor a"tool"keyasquicklyas
possible.Acrossmanystudies,wehavefoundaclearandconsistenttendencyto
mistakenlyrespond"gun"whenaBlackfaceisflashed.Thattendencyincreasesas
participantsarerushedtorespondfasterandfaster(Payne,2001;Payne,Lambert,
&Jacoby,2002).
Thesimilarityofthistasktootherimplicitmeasuresmakesittemptingtothink
ofthebiasasan"automaticeffect."Thatwayofthinkingaboutit isconsistent\vith
thetaskdissociationapproach,inwhichimplicittasksareidentified\vithautomatic
processesandexplicittasksareidentified\vithcontrolledprocessing.Butthatway
of thinkingoverlooksthepossibilityof bothautomaticandcontrolledprocesses
feedingintoresponses.
To revealthedistinctprocessesseamlesslyguidingresponses,wecompared
responseswhenautomaticandcontrolledprocesseswereactinginconcertversus
whentheywereopposed.Whenbothautomaticandcontrolledprocessespointed
tothe"gun"response(aninclusioncondition),theprobabilityof agunresponse
was.75.Thiscanbeformalizedasthesumofcontrolledprocessingandautomatic
processingwhencontrolfails:Control+Automaticx (1- Control).Becausethe
valueshereareprobabilities,theabsenceof aneventcanbeeasilyexpressedas
(1
-
theprobabilityof thatevent).OnBlack-tooltrialswhereautomaticstereo-
typingwouldleadtoa"gun"responsebutcontrolledrespondingwouldleadtoa
"tool"response,participantsstillresponded"gun"\vithaprobabilityof.37.This
representsthetendencyforanautomaticbiastodriveresponsesintheabsenceof
control:Automaticx (1- Control).
Toestimatehowmuchof thisbehaviorwasduetointentionalcontrol,wetook
thedifferencein performancewhenbothautomaticandcontrolledprocesses
favoredaresponse,versuswhenautomaticbutnotcontrolledprocessesfavoredit.
Thisgivesacontrolestimateof.38(=.75-.37).1It isimportanttonoticethat
control,asconceptualizedhere,isnotareactiontoastereotypicalthought.It is
notthoughtsuppressionoranafter-the-facteditingofresponses.It istheabilityto
focusattention,thought,andactionongoal-relevantbehaviorsindependentof
automaticdistractions.If controlwereaperfect1.0,actionswouldbedetermined
completelybyintentions.In thisexperimentthevaluewasmuchlower,allmving
forotherfactorstounintentionallyinfluencebehavior.Withsimpleconstraintslike
speededresponding,it is notdifficulttoseemomentarydisordersof autonomy
cripplinggoodintentions.
To estimatetheautomaticeffectof stereotypingwe lookedat howoften
METHODS:PROCESS DISSOCIATIONPROCEDURE 299
participantsrespondedin line\viththestereotypeevenwhentheyintendedto
respondotherwise(.37).Althoughthisvaluereflectsanunintentionalprocess,it is
anunderestimatebecauseit isthejointprobabilitythatanautomaticprocesswas
atworkandthatcontrolfailed.Themorecontrolapersonexerts,themorethis
valueunderestimatestheautomaticstereotypingeffect.Undertheassumption
thatautomaticandcontrolledprocessesareindependent,wecancorrectforthis
underestimation,dividingbytheprobabilitythatcontrolhasfailed(1
-.38
=
.62).
Thisyieldsanautomaticestimateof.60.
Usingthisprocedure,we foundthatrequiringfastresponsesdramatically
reducedcontrol,comparedtoagroupthatrespondedattheirownpace.Butfast
respondinghadnoeffectontheautomaticprocess,aswewouldexpectbecause
automaticprocessesoperatequicklyandrequirelittlecapacity.Ontheotherhand,
theautomaticestimatewasaffectedbytheBlackversusWhitefaces.Theauto-
matictendencytoward"gun"responseswashigheronBlackcomparedtoWhite
trials.
Thesedissociationsareimportantbecausetheyenhanceourabilitytoanswer
questionsabouthowunintendedracialbiasesinfluencepeople.Whenstudying
howstereotypeshavetheireffects,acommonlyaskedquestionis,"isit anauto-
maticoracontrolledeffect?"Andacommonwaytotestthatquestionistoimpose
acognitiveload,torushresponding,orlookformotivationaldifferences.If cogni-
tiveload,rushedresponding,or lowmotivationinterferes\viththeeffect,it is
inferredtoberesourcedependentandthereforelikelycontrolled.In contrast,if
thesevariableshavenoeffect(orincreasethestereotypingeffect),it isinferred
thattheeffectis automatic,becauseit is not dependentontheinvestmentof
cognitiveresources.
Evena relatively"simple"behavioris complex.It is difficultto find any
behaviorthatdoesnotincludesomeamalgamofprocesses\vithautomaticfeatures
andcontrolledfeatures(Bargh,1989).It thereforebecomesimportanttoseparate
complexbehaviorsinto mote basiccomponents.The processdissociation
approachshiftsthequestionfrom"automaticor controlled?"to"whatcombin-
ationof automaticandcontrolled?"Thefocuschangesfromlabelingaphenom-
.enontotakingapartitscomponentprocesses.Importantly,processdissociation
alsoprovidesa measurementmodelfor quantifYingandsummarizingthose
components.
Thisnewquestionbecomesparticularlyinterestingwhenautomaticandcon-
trolledcomponentsbehaveindifferentways.Wehaveuncoveredseveralvariables
thatallimpactorcorrelate\vithpeople'sjudgmentsintheweaponscenarios.Each
of theseselectivelyaffectseitherautomaticstereotypingor intentionalcontrol.
For instance,werecentlyinvestigatedtheeffectsof self-regulationdepletionon
stereotyping(Govorun&Payne,2006).Basedonpriorfindingsthatexertingself-
controlinonedomainreducesself-regulationinasubsequentcontext(Muraven&
Baumeister,2000),wepredictedthatthedepletiongroupwouldshowreduced
intentionalcontrol.Weassignedonegrouptoperforma boringbutattention-
demandingStroopcolornamingtaskforacontinuous15minuteperiod.Follow-
ingthistedioustask,participantscompletedtheweaponidentificationtaskas
describedabove.ThecontrolgroupperformedtheStrooptaskforonly30seconds,
300 SOCIAL PSYCHOLOGY AND THE UNCONSCIOUS
andthenwentonto theweapontaskunfatigued.As predicted,thedepleted
groupshowedpoorercontrol,but no differencesin automaticstereotyping.
For individualswitha stereotypicalautomaticbias,this reductionin control
resultedin morestereotypicalfalse"gun"responses.For theseindividuals,
automaticstereotypeswereleftunopposedbyintentionalcontrol.
Comparethatmanipulationwithanotheroneintendedtoinfluencestereotyp-
ing,ablatantwarningthattheweaponidentificationtaskmeasuresracialstereo-
typingandthattheyshouldbecarefultoavoidstereotyping.Ironically,warned
participantsshowedamorestereotypicalpatternof mistakesthanacontrolgroup
(Payne,Lambert,& Jacoby,2002).Was this becausethe abilityto control
responseswasdiminished,or becausethewarningmaderaceaccessibletothe
pointthatitincreasestheautomaticimpactofstereotypes?Ourresultssuggestthe
latter- a blatantwarningincreasedstereotypingviaincreasingautomaticbias,
havingnoeffectoncontrol.
Fromadistance,thewarningstudyandtheegodepletionstudyappearvery
similar.Comparedtocontrolgroups,bothwarningaboutracestereotypesand
depletingself-regulatorystrengthincreasedstereotypingasmeasuredbyerrors.
But thesetwosuperficiallysimilareffectsweredrivenbydifferentmechanisms
(seealsoLambert,Payne,Jacoby,Shaffer,Chasteen,& Khan,2003).In the
warningstudy,stereotypingincreasedby boostingtheautomaticactivationof
stereotypicassociations.In the depletionstudy,stereotypingincreasedby
reducingcontroloverbehaviors.
The two studieswejust describedsuggesta ratherpessimisticoutlook,
becausebothmanipulationsincreasedstereotypingratherthandecreasingit.It is
worthconsidering,frombothpracticalandscientificpointsof view,howstereo-
typingcanbedecreasedbelowthebaselinelevelof ourcontrolparticipants.Pro-
cessdissociationallowsustotrackhowtheautomaticandcontrolledcomponents
of behaviormediatethatchange.Stereotypicalrespondingcouldbe reduced
eitherbyreducingautomaticbiasorbyincreasingintentionalcontrol.
A recentstudytestedtheutilityof concreteactionplansin overcomingthe
automaticimpactofracestereotypes(Stewart&Payne,
2006).
Previousworkhas
shownthatconcreteactionplanslinkingaspecificenvironmentalcuetoanaction
canhelppeoplecarryout theirintentionsmoreeffectively(Gollwitzer,1999).
Whenappliedtostereotypicalweaponjudgments,thatideamighttaketheformof
"whenI seeaBlackperson,I\villrespond'tool'."However,thiskindofplanwould
justreplaceonebiaswithanother.Wewantedtofindoutwhethertheactionphase
couldbeusedtogenerateathoughtthatwouldcounteracttheinfluenceof the
stereotypewithoutcreatinganewbias.Tothatend,weaskedparticipantsinone
conditiontoformaplansothatwhenevertheysawaBlackperson,theywould
"thinksafe."In fact,thatsimpleplanreducedtheeffectof stereotypes,compared
to a controlgroupwhowereaskedto"thinkquickly"whentheysawa Black
person.
Did this simpletreatmentreducestereotypingby increasingcontrolor
reducingautomaticinfluences?Acrossthreestudies,wefoundthatthethought-
planreducedtheautomaticinfluenceof race,withoutalteringintentionalcontrol.
Moreover,theplantookeffectinthefirstseveraltrials,suggestingthatitwasvery
METHODS:PROCESS DISSOCIATIONPROCEDURE 301
efficientanddid not requireextensivepracticeasothermethodsof alteringauto-
maticbiasesmay(e.g.Kawakamiet aI.,2000).
SOME WAYSTO THINK ABOUT AUTOMATIC AND
CONTROLLED ESTIMATES
Themeaningsof theestimatesgeneratedbyprocessdissociationdependonthe
kindof behaviorthatisbeinganalyzed.In allof thestudiesreviewedsofar,the
automaticestimatesrepresentedunintentionalinfluencesof stereotypes.But
aswemovedfrommemorystudiestoperceptualjudgments,themeaningsof
thecontrolledestimatechangedmoredramatically.In thememorystudies,the
controlledcomponentwasrecollection:aconsciouslycontrolleduseof memory.
Havingfullaccesstothecontextandthedetailsofaneventallowedparticipantsto
useor not to usewhatevercameto mind.Theseshiftsin theinterpretations
of processestimatesareunavoidable,becausepeoplecontroltheirbehaviorin
differentwaysdependingonwhattheyaredoing.
In theweaponjudgmentstudies,whatis thebestwaytocharacterizethe
processesunderlyingintentionalcontrol?Concretely,thecontrolledcomponent
reflectedtheabilitytorespondbasedononesetofinformation(thefeaturesofthe
targetitems)andnotanother(racialstereotypes).Seeninthislight,theweapon
taskissimilartoothercompatibilitytaskssuchastheStroopcolornamingtask.In
thattask,subjectstrytonametheinkcolorsof wordswhileignoringtheword
itself.Whenthewordisacolorword,it becomesverydifficulttonameaninkif it
isincompatiblewiththewordmeaning(e.g.thewordredingreenink).In fact,in
theGovorunandPaynestudy,performanceontheStrooptaskwassignificantly
correlatedwiththecontrolled(butnottheautomatic)estimatefromtheweapons
task.
ExecutiveFunctionandtheControlofBias
TheStrooptaskandsimilarinterferencetasksareusuallyunderstoodasmeasures
of executivecontrol
-
thewayspeopledirecttheirinformationprocessingand
actionstokeepthemconsistentwiththeirgoals.Executivecontrolisbelievedto
includesubprocessesincludingselectionof relevantinformation,inhibitionof
interferinginformation,andmaintenanceofthecurrentlypursuedgoal(Baddeley,
1986;Kane,Bleckley,Conway,&Engle,2001).Theanalogybetweentheweapon
judgmentscenarioandtheStrooptasksuggeststhatcontrolintheweaponstask
hassomethingtodowithdevotingselectiveattentionto onestreamof inputs,
whileblockingoutanother.
Arecentstudysupportsthissuggestion.Payne(2005)usedanantisaccadetask
totesttheideathatcontrolintheweaponstaskreliesontheexecutiveprocessesof
selectiveattention.Theantisaccadeis awell-establishedmeasureof attentional
controloftenusedin cognitiveandneurosciencestudies(Everling& Fischer,
1998).Thinkbacktothelasttimeyouwereinthemiddleofatalkorlecture,and
someonemistakenlyopenedthedooronlytosheepishlyrealizethattheywerein
302 SOCIAL PSYCHOLOGY AND THE UNCONSCIOUS
thewrongplace.Try aswe mighttoconcentrateonthemattersat hand,it is
incrediblydifficultnottoturnandlookattheinterloper.Thereasonisthatthe
orientingreflexwhichcompelsustoattendtonewitemsin oursurroundingsis
highlyautomatic.Theabilitytointentionallyoverridesuchautomaticreflexeshas
beenusedtomeasureexecutivecontrol.Theantisaccadetaskasksparticipantsto
avoidlookingatanitemthatflashesabruptlyonacomputerscreen.A distracting
item(e.g.aredcircle)appearsononesideofthescreen,andatarget(e.g.aletter
tobeidentified)flashesontheothersideafractionofasecondlater.Lookingaway
fromthedistracterwill enhanceidentificationof thetarget,but lookingat the
distracterwillinterfere.
Researchfromourlabshowedthatperformanceonanantisaccadetaskwas
correlatedwiththecontrolledcomponent,butnottheautomaticcomponent,in
theweaponjudgmenttask(Payne,2005).In contrast,theautomaticestimatewas
correlatedwithtwomeasuresofimplicitraceattitudes,theimplicitassociationtest
andevaluativepriming.Thisdissociationrevealstwoverydifferentkindsof pro-
cessesthatnormallyblendimperceptiblytogetheraspeoplemakeasingledeci-
sion.Theanswertothequestion"howwouldI actinthatsituation?"hasatleast
twoparts.Thefirstdependsonaperson'sautomaticreactionstoBlackindividuals.
Theseconddependsontheperson'sabilitytoengageexecutivecontrol,thatis,to
keeptheirthoughtsandactionsontrackratherthanbeingswayedbyaccessible
butinappropriateinformation.
Thesestudiesillustratea rangeof processesthatmightbe measuredusing
processdissociationmethods.Otherresearchhasfoundcreativeapplications
of theprocedureoutsideof bothmemoryandstereotyping.As oneillustration,
FitzsimonsandWilliams
(2000)
usedamodificationof theprocessdissociation
proceduretoinvestigatethemeremeasurementeffect.Themeremeasurement
effectis thefindingthatsimplyaskinga personabouthowlikelytheyareto
performabehaviorin thefutureactuallyincreasesthelikelihoodthattheywill
performthatbehavior(Morwitz,Johnson,& Schmittlein,1993;Sherman,1980).
FitzsimonsandWilliams(2000,stupy1)askedonegroupofparticipantshowlikely
theywouldbetochooseanewbrandofcandybar,whereasthecontrolgroupwas
notaskedaboutthecandybar.Themeremeasurementeffectsuggeststhatthe
groupwhowasaskedwouldbe morelikelytochoosethecandybar thanthe
controlgroup.In additiontotheintentquestion,participantsweregiveninforma-
tionsuggestingthattheyweremoreorlesslikelytoactuallyreceivethecandybar
if theychoseit.Thismanipulationwasintendedtomanipulatetheself-interestof
participants.Arational(inthesenseofself-interested)analysiswouldsuggestthat
participantsshouldbemorelikelytochoosethecandybarwheninformedthat
theywerelikelytogetthecandybar.Bycrossingthemeasurementofintent\vith
theself-interestinformation,thisstudycreatedconditionsin whichthemere
measurementeffectwascongruentwithself-interest,andconditionsinwhichit
wasincongruentwithself-interest.
Thisstudyshowedthat,indeed,participantsaskedabouttheirlikelychoice
moreoftenchosethecandybar.Usingamodifiedmodelbasedonthelogicof
processdissociation,theseresearchersseparatedtwocomponentsof theeffect.
One componentreflectedhowstronglytheintentquestioninfluencedchoice
METHODS:PROCESS DISSOCIATIONPROCEDURE 303
regardlessof self-interest(akintotheautomaticcomponentin theotherstudies
discussedhere).Theothercomponentreflectedtheextenttowhichchoiceswere
guidedbytheself-interest.Thisisrelatedtothecontro11edcomponentwehave
discussed,butbecausetheprocedurepittedself-interestagainstthemeremeas-
urementeffect,thiscomponentreflectedself-interesteddecision-making.The
intentquestioninfluencedchoiceslargelyirrespectiveofwhetheritwasconsistent
withself-interestornot,aneffectdrivenbytheautomaticcomponent.
Fromthestudiesreviewedhere,it is dearthattheprocessdissociationpro-
cedureis notlimitedtomeasuringreco11ectionandfamiliarityin thecontextof
memoryresearchwhereit wasdeveloped.Althoughextensionsbeyondmemory
researcharea relativelynewendeavor,theprocedureis flexible,andcanbe
adaptedtoanynumberof topics.Thekeytothisflexibilityisthatprocessdisso-
ciationrepresentsa generalframeworkfor thinkingaboutintentionaland
unintentionalprocesses.The basiclogicof placingintendedandunintended
influencesin concertandin oppositionin orderto disentanglethemcanbe
implementedacrossmany,manydomains.
It isimportanttonotethatthemeaningof theprocessesmeasureddepends
entirelyonthetaskbeingstudied.Phrasedanotherway,themeaningof theesti-
matesdependsonwhatprocessesareplacedin concertandin oppositionwith
eachother.If implicitandexplicitinfluencesof memoryarearrangedinthisway,
theprocedurecanyieldestimatesof implicitandexplicitmemory.If automatic
stereotypingandexecutivecontrolarearrangedin thisway,theprocedurecan
estimatetheseprocesses,andsoon.As thesecontrastsillustrate,thelogicof
oppositioncreatesa conceptualandmethodologicalwayto thinkaboutmany
differentkindsof factorsthatmightinfluencepeopleeitherintentiona11yor
regardlessof intent.Nevertheless,it isoftentemptingtothinkof theautomatic
andcontro11edcomponentsashavingfixedmeanings,basedonotherprominent
processdistinctionsthathavebeenmadein socialcognition.In thefo11owing
sectionwecomparetheprocessdissociationapproachtosomeofthesecommonly
invokeddimensionsof automaticandcontro11edprocessing,highlightingthe
similaritiesanddifferences.
InvariantsandParticulars
Givena11thedifferentusestowhichonemightputtheprocedure,onemight
wonderif thereis anythingthatthedifferentautomaticor contro11edprocesses
haveincommon(seeMoors& DeHouwer,Chapter1of thisvolume).Yetthere
appeartobecertainpropertiesconsistentlyattachedtocontro11edcomponents,
andcertainpropertiesattachedtoautomaticcomponents,acrossthemanydiffer-
entdomains.For example,thecontro11edcomponentinmemorystudiesrequires
attention,andis disruptedby distraction(Henseet aI.,1995;Shermanet aI.,
2003).Thecontro11edcomponentinweaponidentificationstudiesrequirestime,
andisdisruptedbyrushedresponding(PayneetaI.,2002).Controlintheweapon
studiesisalsoinfluencedbymotivations.Highermotivationtocontrolprejudiceis
associatedwithmoreintentionalcontrolduringjudgments(Payne,2005;Amodio
etaI.,2004).In contrasttotheseresultsforintentionalcontrol,thesesamestudies
304 SOCIAL PSYCHOLOGY AND THE UNCONSCIOUS
showthattheautomaticestimateappearsunaHectedbydividedattention,occurs
rapidly,andis notassociatedwithmotivations.Readersmayhavenoticedthat
thesearethedefiningcharacteristicsusuallyinvokedtodistinguishautomaticand
controlledprocessingingeneral(Bargh,1989;Posner& Snyder,1975;Shiffrin&
Schneider,1977).
Thisis noaccident.Thefactthatthesepropertiesremainattachedtoauto-
maticandcontrolledestimatesregardlessof thespecifictasksuggestssomething
interesting.It suggeststhatregardlessofwhatoneintendstodo,it istheconstrain-
ingof behaviorto intentthatrequiresresources,time,andmotivation.On the
otherhand,it isbeingpushedalongregardlessof intentthatisquick,effortless,
andeasy,perhapsregardlessofwhatforceisdoingthepushing.
In thislight,therelationshipbetweentheprocessdissociationframeworkand
othercommonlystudiedformsof automaticandcontrolledbehaviorbecomes
clearer.To bringthisrelationshipmoresharplyintofocus,we\vill consideras
examplestwodual-processdistinctionsprominentinsocialpsychology.Thefirstis
deliberativereasoningversusshallowheuristic-basedreasoning.Thesecondisthe
distinctionbetweenimplicitandexplicitattitudes.Thesedistinctionsformthe
cruxof morethanoneprominentdual-processtheoryaimedatexplainingwhy,
when,or howautomaticandcontrolledaspectsof cognitioncombineto drive
,behavior.Eachdistinctioncouldpotentiallybeilluminatedbyaprocessdissoci-
ationapproach,onceit isunderstoodwhatassumptionsmustbemadeandhow
processestimatescouldmapontothesedistinctions.
Thecontrastbetweendeliberativereasoningandshallowinferencesbasedon
heuristicsis at theheartof manydual-processtheories(Chaiken,1980;Fazio,
1990;Petty&Cacioppo,1986).Theprocessdissociationestimatesdiscussedthus
farinthecontextof memoryandweaponidentificationstudiesseemfarremoved
fromthisdistinction.However,theconceptualapproachcanbeappliedherejust
asintheseotherareasof study.Perhapsthemostcloselyrelatedresearchisthe
workof FitzsimonsandWilliams
(2000)
describedabove.In theirstudiesthey
placedself-interestandthemeremeasurementeffectinconcertandinopposition
toteaseaparttheirseparatecontributions.It is easytoimagineothersituations
wherecarefulthoughtwouldleadtoonekindof outcome,andheuristicswould
leadtoanother.Herethetwoprocesseswouldbesetinopposition.It isjustaseasy
toimagineanexperimentalarrangementinwhichcarefulthoughtandheuristics
wouldleadtothesameresponse,therebysettingupanin-concertcondition.
In fact,thiskindof paradigmis routinelyusedin studiesof attitudesand
persuasion,althoughit maynotbeframedintheseterms.A commonprocedure
forstudyingtheprocessesof attitudechangeistomanipulateargumentstrength.
The assumptionis thatwhenpeoplearethinkingcarefullythey\vill be more
persuadedby strongthanweakarguments.This manipulationis sometimes
crossed\vithaheuristicor cue,suchasthenumberof argumentsusedor the
attractivenessof thesourcemakingthearguments.Theassumptionisthatwhen
peopleareprocessingshallowly,theyaremorelikelytobepersuadedbythese
simplecues.
Whenargumentstrengthandheuristiccuesarefullycrossed,thedesign
createssomeconditionsinwhichbothdeliberatethinkingandheuristicthinking
METHODS:PROCESS DISSOCIATIONPROCEDURE 305
wouldleadtoagreementor disagreementwiththemessage(Le.thestrongargu-
ment,positivecuecellandtheweakargument,negativecuecell).Alsocreatedare
conditionsin whichdeliberativeandheuristicthinkingwouldleadtodWerent
outcomes(Le.,thestrongargument,negativecuecell andtheweakargument,
positivecuecell).Becausedeliberativethinkingandheuristic-basedthinkingcan
bearrangedinthisway,theycanpotentiallybeseparatedandquantifiedusingthe
processdissociationapproach.Oneestimatewouldrepresentthecontributionof
deliberatereasoningandtheotherwouldrepresentthecontributionof heuristic
reasoning.Asinanynewapplicationofamodel,validationtestswouldneedtobe
performedtotestwhethertheassumptionsof theprocessdissociationprocedure
matchthepropertiesof theattitudechangeparadigm.Thebroaderpointisthat
processdissociationdoesnot fixthemeaningsof theautomaticandcontrolled
estimates.It is insteadawayto thinkaboutandquantifythecontributionsof
differentprocesses,asneededtoanswerspecificquestions.
In the studiesof race biasesin memoryand perceptualidentification
describedin theprevioussection,theautomaticestimatecanbesaidtoreflect
implicitattitudesorstereotypes.Implicitattitudesarecommonlycontrastedwith
explicitattitudes,whicharetheattitudespeopleovertlyexpresswhendirectly
asked.However,in thestudiesdescribed,automaticbiaseswerecontrastednot
withexplicitattitudes,butwiththeabilitytointentionallycontrolresponses.By
thisapproachanimplicitattitudeis theevaluationthatdrivesresponseswhen
intentionalcontrolfails.Forsomepurposes,however,researcherswishtoseparate
implicitattitudesfromexplicitattitudes.
Thetypicalwayofseparatingimplicitandexplicitattitudesisbycomparingan
implicitmeasureandanexplicitmeasure.If anoutcomebehaviorcorrelateswith
animplicitmeasurebutnotanexplicitmeasure,it is saidtobetheproductof
implicitorautomaticprocesses.If abehaviorcorrelates\vithanexplicitmeasure
butnotanimplicitmeasure,it issaidtobetheproductof controlledprocesses.
However,manydifferentfactorsvarybetweenimplicitandexplicitmeasures
beyondautomaticityandcontrol.For example,howshouldonecomparereaction
timesinwordpairingstoLikertscales?Evenif theyarestandardizedtothesame
scale,shouldtheybeinterpretedinthesameways?Whatif onemeasureismore
reliableorsensitivethantheother?Earlyfindingsthatimplicitandexplicitmeas-
uresofattitudestowardthesametopicstendedtocorrelateweakly,if atall,ledto
agreatdealof theorizingaboutwhetherimplicitandexplicitattitudesrepresent
separateconstructs(e.g.,Fazio&Olson,2003;Wilson,Lindsey,&Schooler,2000).
In ourview,therearemanyreasonstoexpecttherelationshipbetweenimplicit
andexplicitmeasurestobeweakened,evenif theyaretappingthesameconstruct.
Beforeweareabletofullyaddressthisissueofsingleversusseparateconstructs,it
\villbeimportanttodealwithotherfactorssuchasreliability(whichtendstobe
lowerforimplicitmeasures;seeCunningham,Preacher,& Banaji,2001)andthe
difficultiesof comparingvastlydifferenttypesof behavioron implicitversus
explicittasks.
Becauseof theselimitationsto the taskdissociationapproach,we have
pursuedacomplementaryapproach\vithintheprocessdissociationframework.It
maybepossibletodesignexperimentsin whichintentionalandunintentional
306 SOCIAL PSYCHOLOGY AND THE UNCONSCIOUS
contributionsestimatedfromthesamebehaviorrepresentexplicitandimplicit
attitudes.If so,thiswouldallowintentionalandunintentionalaspectsof attitudes
to becomparedon thesamescaleandwithinthesametask,overcomingthe
problemsof comparingacrossradicallydifferentmeasures.We nextdescribea
newlydevelopedmethodformeasuringattitudesimplicitly\vithinthetaskdissoci-
ationframework,andthenexplorehowthatmethodmightbeexpandedfor a
processdissociationanalysis.
Wehaveknownforyearsthatpeoplesometimesmisattributetheirevaluative
reactionsfromonesourcetoanothersource(Dutton& Aron,1974;Schwarz&
Clore,1983).Forexample,MurphyandZajonc(1993)showedthatflashingpleas-
antandunpleasantimagesbeforepresentinganambiguousChinesepictograph
influencedthewaypeopleevaluatedthepictographs.Inthisstudy,theoutcomeof
interestwasnotresponsetimes,buthowpleasantorunpleasantparticipantsfound
thepictographs.Whentheprimewaspleasant,participantsfoundthepictograph
morepleasant;whentheprimewasunpleasant,participantsfoundthepictograph
lesspleasant.
Althoughthiseffecthasbeen\videlyknownformorethanadecade,thereisan
importantimplicationthathasgoneunnoticed:Thismisattributionproducesan
indirectmeasureof individuals'attitudestowardtheprimes.If aparticularprime
itemsystematicallycausesparticipantstoevaluateanambibl"Uouspictographposi-
tively,it suggestsapositiveattitudetowardtheprimeitem.IntheoriginalMurphy
andZajonc(1993)procedure,theprimesonlyaffectedjudgmentsof thepicto-
graphswhentheywereflashedtoobrieflytobeconsciouslyidentified.However,
withsomemodificationstotheprocedure,wewereabletoproducestrongmis-
attributionsevenwhentheprimeswereplainlyvisible,andevenwhenpartici-
pantswereblatantlywarnedagainstbeinginfluencedbythem(Payne,Cheng,
Govorun,& Stewart,2005).
Usingthismodifiedprocedure(whichwerefertoasanaffectmisattribution
procedure),wefoundthatthekindsof misattributionspeoplemadewhenprimed
withpicturesofGeorgeW.BushandJohnKerrystronglypredictedtheirexplicitly
ratedattitudestowardthecandidates(r
=
.65)andwhotheyintendedtovotefor
(r
=
.58).We expectedhighimplicit-explicitcorrespondencein thisdomain
becausepeoplearewell awareof theirpoliticalattitudesandquite\villingto
expressthem.In thiscase,whatevervarianceinthetwomeasuresdidnotoverlap
mayhavebeencausedbytheproblemsof comparingacrosstwoverydifferent
kindsof measures.
In anotherstudy,wereplacedthepicturesof BushandKerry\vithpicturesof
WhiteandBlackyoungmenwhowerejudgedasappearingprototypicalof their
respectivegroups.For peoplewhohadanegativeaffectivereactionwhenpre-
sented\vitha Blackperson,weexpectedthisreactiontobe reflectedin their
judgmentsofthepictographs.Asexpected,wefoundapatternofin-groupfavorit-
ismonthetask.Whitesubjectsshowedastrongpatternof misattributionsimply-
ingmorefavorableattitudestowardtheWhitephotosthantheBlackphotos.In
contrast,Blackparticipantsshowedtheoppositepattern.Thesepatternsof mis-
attributionspersisteddespiteblatantwarningsagainstbeinginfluencedbythe
primes.Theyalsocorrelatedwithexplicitratingsof attitudestowardBlacksand
METHODS:PROCESS DISSOCIATION PROCEDURE 307
Whitesasgroups(r
=
.58).However,thisrelationshipwasmoderatedbypeople's
motivationstorespondwithoutprejudice.Amongthosewholackedsuchmotiv-
ation,theirimplicitandexplicitmeasurescorrelatedverystrongly.However,
amongindividualswhowerehighlymotivatedtoavoidprejudice,therelationship
wasmuchweaker.Thiswasbecausehighlymotivatedpeopleexpressedhighly
positiveattitudestowardBlackson theself-reportmeasure,but still showed
negativitytowardBlacksontheimplicitmeasure.
Sofar,theseresultsareexactlywhatonewouldexpectbasedonprevious
researchcomparingimplicitandexplicitattitudemeasures,exceptintworespects.
Thefirstisthatthecorrelationsarelargerthanmostfindingsreported.Thesecond
is thatthemisattributionmeasureproducedmuchhigherreliabilitythanmost
implicitmeasures,andequaltomanyexplicitmeasures(averageCronbach'salpha
=
.88).Althoughthekindsof responsescomparedin thesestudies(pleasantness
judgmentsaboutpictographsandfavorabilityratingsofindividualsorgroups)are
notasdiscordantascomparingreactiontimesandratingscales,theywerestillnot
directlycomparable.
Considernowamodificationoftheaffectmisattributionprocedureconsisting
of twophases.Onephaseisidenticaltotheprocedurealreadydescribed:partici-
pantsjudgepictographsaspleasantor unpleasantandareinstructedto avoid
beinginfluencedbytheprimesof BlackandWhitephotos.In thesecondphase,
participantsseethesameitempairs,butthistimetheyareinstructedtoignorethe
pictographsandevaluatetheirreactionstotheprimephotosthemselvesaspleas-
antor unpleasant.The firstphasemeasuresattitudestowardtheprimeitems
indirectly,throughtheirunintendedeffectsonjudgmentsof thepictographs.The
secondphasemeasuresattitudestowardtheprimeitemsdirectly.In thefirst
phase,participantsaretryingnottoexpressanyevaluationof theprimes.In the
secondphase,theyaretryingtoexpresstheirevaluationoftheprimes.Thestimuli
andjudgmentscaleareheldconstant;theonlyfactorthatvariesisparticipants'
intentions.Withthestimuliandjudgmentscaleheldconstant,weareinaposition
tomakedirectcomparisonsbetweenthetwoconditions.Wecouldthenusethe
firstphaseasanimplicitmeasureof racialattitudes,andthesecondphaseasan
explicitmeasure.
Wedonotneedtostop\viththiscomparison.Thesecondphaseof thetaskis
notonlyanexplicitmeasureof attitudestowardtheprimes
-
it isalsoaninclusion
condition.The wayspeoplejudgethepleasantnessmaybea productof both
intentionalandunintentionalevaluativeinfluences.Thefirstphaseisnotonlyan
implicitmeasure,butit is alsoanexclusioncondition.Judgmentsof thepicto-
graphsareonlyinfluencedbyevaluationsof theprimeswhenpeople'sintentions
toavoidtheirinfluencefail.Bycomparingthesetwoconditions,wecanestimate
howmuchcontroleachpersonhasoverwhethertheyexpressaparticularevalu-
ation,andwecanestimatewhatevaluationisrevealedwhencontrolfails.
Aspartofvalidatingthisprocedure,wehavetovalidatesomeassumptionsthat
aremadewhencarryingouttheprocessdissociationanalysis.In thenextsection,
weoutlinethoseassumptionsthataremadebyanyapplicationofprocessdissoci-
ation,andplacetheminthecontextof othermodelsandalternativeassumptions
thatmaybemade.
308 SOCIAL PSYCHOLOGY AND THE UNCONSCIOUS
Assumptions and Alternatives
All mathematicalmodelsmakeassumptionsinordertorelateactualdatatoformal
equations.Whenapplyingtheprocessdissociationframeworkinanewcontext,it
is importantto be clearaboutthoseassumptions.Oneassumptionof process
dissociationis thatthecontrolledandautomaticprocessesatworkexertsimilar
influencesininclusionandexclusionconditions.Inotherwords,thetwoprocesses
shouldexertasmuchinfluencetogetherin theinclusionconditionastheyexert
againsteachotherin theexclusioncondition.It is importanttoavoidthemis-
interpretationthatautomaticandcontrolledestimatesshouldbe numerically
equalacrossexperimentalconditions,or toeachother.If theestimateslogically
hadtoequalsomeparticularvalue,wewouldnotneedtodotheexperimentor
computetheestimatesfromdata.Instead,it meansthattheautomaticandcon-
trolledprocessesin questionplaythe samerolesin inclusionandexclusion
conditions.
Thesecondassumption,whichhasbeendiscussedmorewidely,isthatauto-
maticandcontrolledprocessesareindependentof eachother(fordiscussionsof
theindependenceassumptionin memoryresearch,seeCurranandHintzman,
1995;1997;Jacoby,Begg,& Toth,1997;Jacoby& Shrout,1997).Whetherthis
assumptionis metdependsontheexperimentalparadigmthatisbeingused.In
somecases,automaticandcontrolledprocessescouldbepositivelyor negatively
correlatedwitheachother,whichwouldviolatetheindependenceassumption.
Becausewecannotdirectlyobservetheprocesses,wemustindirectlytestwhether
theprocessesarelikelytobeindependentordependent.Themostcommonway
todothisistolookfordissociations,orselectiveeffectsononeorbothestimates.
Thelogicis thatif automaticandcontrolledprocessesareindependent,thenit
shouldberelativelyeasytofindvariablesthataffectonebutnottheother.If the
independenceassumptionisbadlyviolated,thenautomaticandcontrolledpro-
cesseswouldstronglycovarywitheachother.Asaresult,it wouldbedifficultto
findvariablesthataffectoneprocess\vithoutaffectingtheother.
Muchof ourworkhasbeenfocusedonexaminingselectiveeffectsonauto-
maticandcontrolledcomponentsin theweaponidentificationprocedure.For
example,Payne(2001)foundadoubledissociationbetweenthetwoprocesses.
Primepicturesof BlackandWhitefacesaffectedtheautomaticcomponentbut
notthecontrolledcomponent.In contrast,speededrespondingaffectedthecon-
trolledcomponentbut not theautomaticcomponent.Further,racialattitudes
correlatedselectivelywiththeautomaticcomponent.Lambertandcolleagues
(2003)foundthatanxietyoveranimpendingpublicdiscussionreducedthecon-
trolledcomponentwithoutaffectingthe automaticcomponent.The study
describedabovebyGovorunandPayne
(2006)
showedthategodepletioninflu-
encedthecontrolledcomponentbutnottheautomaticcomponent.Finally,the
studybyStewartandPayne(2006)describedaboveshowedthatspecificaction
planscouldaffecttheautomaticcomponentwithoutchangingthecontrolled
component.
Thesedissociationswouldnotbeexpectedif theindependenceassumptioI1
wereviolatedintheweaponidentificationtask.However,forsometasks,orunder
METHODS:PROCESS DISSOCIATIONPROCEDURE 309
someconditions,itisalwayspossibletoviolateone'sassumptions.Researcherswill
bemostfamiliarwiththeseconsiderationsin thecontextof commonstatistical
tests.It iswidelyunderstoodthatdifferentstatisticaltestsmakedifferentassump-
tions.For example,analysisof variance(ANOVA)assumesadependentvariable
thatis atleastanintervalscale,anormallydistributeddependentvariable,and
homogeneityof varianceacrossdifferentconditions,amongotherthings.If an
assumptionis violatedslightly(e.g.,aslightlyskeweddistribution)theresulting
biasesareusuallysmall.If anassumptionisviolatedbadly(forexample,distribu-
tionsareheavilyskewed)it isoftenagoodideatochooseadifferenttestthatdoes
notdependontheproblematicassumption.Justasthefailureofanassumptionin
aparticularstudydoesnotinvalidatetheANOVAtechniqueingeneral,studies
showingthatanassumptionofprocessdissociationhasbeenviolateddonotinvali-
datethegeneralmethod.Instead,othermethodsmaybemoreappropriatein a
particularcontext.
Severalothermethodsaresometimesusedasalternativestoprocessdissoci-
ation.Theseincludethetaskdissociationmethod(comparingexplicitandimplicit
measures),signaldetectiontheory,andmultinomialmodels.Althoughtheymay
notbeexplicitlystated,eachoftheseapproachesalsoinvolvesassumptionswhich
maybeviolated.Let us takefirstthetaskdissociationmethod.Althoughthis
methoddoesnotuseamathematicalmodel,it stillmakessomeassumptions.By
usinganimplicittaskto measureautomaticor unconsciousprocesses,andan
explicittaskto measurecontrolledor consciousprocesses,thetaskdissociation
approachmakesthetacitassumptionthateachmeasureisprocess-pure.Thatis,
oneassumesthatthemeasuresdifferonlyonthedimensionof interesttothe
researcher.If thetwotasksdifferinwaysotherthantheexplicit/implicitdimen-
sion,thenanydifferentresultsonexplicitversusimplicittaskscouldbebecauseof
thoseother(confounded)features.
Thepsychologicalprocessesbehindimplicittasks(suchasreactiontimesto
classifywords)andthosebehindexplicittasks(suchasendorsingcomplexprop-
ositionalstatements)areverydifferent.Asaresult,theassumptionthatimplicit
andexplicittasksdifferonlyonthedimensionof interestisnotlikelytobecom-
monlymet.Bothprocessdissociationandtaskdissociationmethodsmakeassump-
tionstorelateobserveddatatounobservedtheoreticalideas.Inthecaseofprocess
dissociation,thoseassumptionsaremadeexplicit,whereasinthetaskdissociation
methodtheyoftenremainunstated.
A secondalternativeapproachis signaldetectiontheory(SDT).Signal
detectiontheoryassumesthatperceiversarenaturalstatisticians,who make
decisionsaboutworldeventsin thewaythatresearchersdecidewhetherto
rejecta null hypothesis(Tanner& Swets,1954).A decisionaboutwhatoneis
perceivingor howtorespondis treatedasaproblemof detectingasignalin a
noisyenvironment.Perceivershavea certainamountof evidence,andthey
selecta criterion(similarto the conventionaluseof p <.05in psychology
research)thatmarksoff howstrongtheevidencehasto be beforetheywill
acceptthatasignalispresent.Givenapatternof correctresponsesanderrors,
signaldetectiontheorycanseparatesensitivity(theabilitytodiscriminatewhen
a signalis actuallypresentor absent)frombias(atendencytorespondasif a
310 SOCIAL PSYCHOLOGY AND THE UNCONSCIOUS
signalispresentwhetherit is or not).Signaldetectiontheoryis muteonissues
of automaticversusintentionallycontrolledbehavior,anditsdevelopmentpre-
datedthecurrentinterestinautomaticity.Nonetheless,signaldetectionanalyses
aresometimespreferredto processdissociationonthebeliefthatSDT makes
fewerassumptions.
That beliefis mistaken.Signaldetectiontheorymakessomeof thesame
assumptionsasprocessdissociation,andsomethataredifferent.For instance,
SDTalsomakesanindependenceassumption.It assumesthatsensitivityandbias
areindependentinthesamewaythatprocessdissociationassumesthatcontrolled
andautomaticcomponentsareindependent.Signaldetectionalsoassumesnormal
distributionsofevidencestrength,andequalvariances.Beyondstatisticalassump-
tions,signaldetectionmakessubstantiveassumptionsaboutthewayhumanspro-
cessinformation.Forinstance,it assumesthatdecisionsaremadeonthebasisofa
singlecontinuumof evidence.Thereis noallowancefor qualitativelydifferent
kindsof evidence.Processdissociation,incontrast,treatsintentionalcontroland
automaticbiasesasqualitativelydifferentprocessesfeedingintobehavior.Like
ANOVA,bothmodelsmakeassumptionsthatmaybemoreor lesssuitablein a
givenconte2't.
Finally,multinomialmodelshaveagreatdealincommonwithprocessdissoci-
ation.A multinomialmodelpositsabranchingtreeof unobservedcognitivepro-
cesses,leadingeventuallytobehavioralresponses(Riefer& Batchelder,1988).
For example,KlauerandWegener(1998)developedamodeltostudytheeffects
of stereotypicexpectationson memory.In thismodel,participantsattemptto
rememberwhethertheywitnessedagivenaction.If so,theyattempttoremember
whoperformedtheaction.If theycannotrememberwhoperformedtheaction,
theymayrememberthesocialcategorytowhichtheactorbelonged.Andif they
cannotrememberthesocialcategory,theymayguessthecategoryonthebasisof
stereotypes,andso on.Usingthe patternof correctandincorrectmemory
responses,acomputeralgorithmisusedtoestimatethebest-fittingvaluesforeach
process.In thisway,thedegreeof memoryandguessingat eachstagecanbe
estimated,andthemodelcanbetestedstatisticallytoseehowwellit fitsthedata
(seealsoConreyetaI.,2005).
Multinomialmodelsaresometimespresentedasalternativesto a process
dissociationapproach.However,it isprobablymoreaccuratetothinkof process
dissociationandmultinomialmodelsastwospecificcasesof a generalfamily
of models.BothmakesimiIa:rassumptions.Bothareaimedat separatingun-
observablepsychologicalprocessesthatgiverisetoobservedbehavior.Process
dissociationusesalgebrato estimatethecognitiveprocessesinvolved,whereas
multinomialmodelsuseacomputeralgorithm.However,therelatednessof the
modelscanbeseeninthefactthattheprocessdissociationmodelcanberepre-
sentedandestimatedas a multinomialmodelwith twoprocessparameters
(automaticandcontrolledcomponents;Jacoby,1998;Payne,Jacoby,& Lambert,
2005).
A multinomialmodelmayhaveanynumberofparameters,inanynumberof
combinations.Thisisbothastrengthandaweakness.It isastrengthbecauseit
allowsflexibilityin exploringvarioustheoreticalmodels.However,the more
METHODS:PROCESS DISSOCIATION PROCEDURE 311
.parametersamodelhas,themoreit islikelytofitanygivensetof data,evenan
incorrectone.Asthenumberofparametersincreases,thepossiblewaystocom-
binethoseparametersincreasesexponentially(dopeopleattempttoremember
socialcategoryonlyaftermemoryforthepersonfails,ordotheyfirstremember
thecategoryandthentheperson?).Thereareoftenmanydifferentmodelsthat
couldfitthedataequallywell.It isimportantwhenusingthisapproachtohave
stronga priori theoreticalpredictionsto avoidchoosinganarbitrarymodelor
capitalizingonchance.
Despitethesecaveats,multinomialmodelsaswellastaskdissociations,signal
detection,andprocessdissociationallprovidevaluabletoolsfortakingapartthe
complexpatternsin socialbehaviorintosimpler,moredigestibleparts.Social
psychologistsknowalot abouthowknowledgestructuresandpriorexperience
guideourreactions.Thoseproceduresweknowalotabouttendtoseemharmless,
whereaslessfamiliaronestendtoseemmoremenacing.As weshiftfromthe
comfortableANOVAandtaskdissociationmethodstowardsignaldetectionthe-
ory,whichislessfrequentlyseeninsocialpsychology,manyreadersmayfeelless
sure.Andaswelookcloselyatmultinomialmodelsandprocessdissociation,there
will probablybemoreunease.Themethodsseemforeign,theassumptionsseem
difficult.All of thesetoolsmakeassumptions,andthereisvariabilityinhowner-
voustheseassumptionsmakeus.Butthedifficultyin manycasesisnotsomuch
withthemethodsandassumptionsaswiththenewnessof thesetoolsfor social
psychologists.Thecostsofmakingassumptionsmustbegaugedagainstthepoten-
tialgainsinknowledgegeneratedbyusingthesetools.In ourview,theevidence
reviewedhereshowsthatpotentialtobehigh.
CONCLUSION
We beganthischapterbyexploringthestartlingwaysthatbraindamagecan
fragmentmentaleventsthatnormallyflowsilentlytogether.At afinergrainof
analysis,comparisonofimplicitandexplicittasksrevealsthesamekindsoffissures
innormalhealthypersons.Socialpsychologicalresearchonautomaticity(Bargh&
Ferguson,2000),willedbehavior(Wegner,2002),andimplicitcognition(Green-
wald& Banaji,1995)ishumbling,becauseit suggeststhatdisordersof autonomy
maynotbesuchrareconditionsafterall.Theycanbecapturedinlittleslipsand
subtlelapsesthataremadeandforgotteneveryday.Wemayhavealittlemorein
commonwithsomelesionpatientsthanwethought.But howmuch?Process
dissociationhelpstallyupthelapses,givinganumbertosomethingasgossameras
goals,intent,andwill.
312 SOCIAL PSYCHOLOGYAND THE UNCONSCIOUS
NOTE
1.
Thevaluesof processestimatesheredifferby.02-.03fromthevaluesreportedin
Payne(2001)becauseastatisticalcorrectionwasappliedtothedatainthatarticleto
correctforextremevaluespriortocalculatingestimates.Herewehaveusedtheraw
dataforthesakeofclarity.
REFERENCES
Amodio,D.M.,Harmon-Jones,E.,Devine,P.G.,Curtin,J.J.,Hartley,S.L.,& Covert,
A.E.(2004).Neuralsignsforthedetectionofunintentionalracebias.Psychologi-
calScience,15,225-232.
Baddeley,A.D.(1986).Workingmenwry.London:OxfordUniversityPress.
Bargh,J.A.(1989).Conditionalautomaticity:Varietiesof automaticinfluencesin social
perceptionandcognition.InJ.S.Uleman&J.A.Bargh(Eds.)Unintendedthought
(pp.3-51).NewYork:Guilford.
Bargh,J.A.(2005).Bypassingthewill:Towardsdemystifyingbehavioralprimingeffects.In
R.Hassin,J.Uleman&J.Bargh(Eds.),Thenewunconscious.Oxford,UK:Oxford
Univ/(rsityPress.
Bargh,J.A.,&Ferguson,M.J.(2000).Beyondbehaviorism:Ontheautomaticityofhigher
mentalprocesses.PsychologicalBulletin,126,925-945.
Beer,J.S.,Heerey,E.H.,Keltner,D.,Scabini,D.,& Knight,R.T.(2003).Theregulatory
functionof self-consciousemotion:Insightsfrompatientswith orbitofrontal
damage.journalofPersonalityandSocialPsychology,85,594-Q04.
Chaiken,S.(1980).Heuristicversussystematicinformationprocessingandtheuseof
sourceversusmessagecuesin persuasion.journal of Personalityand Social
Psychology,39,752-766.
Conrey,F.R.,Sherman,J.W,Gawronski,B.,Hugenberg,K.,& Groom,C.(2005).
Separatingmultipleprocessesin implicitsocialcognition:The Quad-Model
of implicittaskperformance.journal of PersonalityandSocialPsychology,89,
469-487.
Cunningham,W A.,Preacher,K.J.,& Banaji,M.R.(2001).Implicitattitudemeasures:
Consistency,stability,and convergentvalidity.PsychologicalScience,12,
163-170.
Curran,T.,& Hintzman,D.L.(1995).Violationsof theindependenceassumptioninpro-
cessdissociation.journal of ExperimentalPsychology:Learning,Memon),and
Cognition,21,531--547.
Curran,T.,&Hintzman,D.L.(1997).Causesandconsequencesofcorrelationsinprocess
dissociation.journalof ExperimentalPsychology:Learning,Menwn),andCogni-
tion,23,496--504.
Devine,P.G.(1989).Stereotypesandprejudice:Their automaticandcontrolledcom-
ponents.journalofPersonalityandSocialPsychology,56,5-18.
Doody,R.S.,&Jankovic,J.(1992).Thealienhandandrelatedsigns.journalofNeurology,
Neurosurgen),andPsychiatn),55,806-810.
Dutton,D.G.,& Aron,A.P.(1974).Someevidencefor heightenedsexualattraction
underconditionsof highanxiety.journalofPersonalityandSocialPsychology,30,
510--517.
Everling,S.,& Fischer,B.(1998).Theantisaccade:A reviewofbasicresearchandclinical
studies.Neuropsychologia,
36,885-899.
METHODS:PROCESSDISSOCIATION PROCEDURE 313
Fazio,R.H.(1990).Multipleprocessesbywhichattitudesguidebehavior:The MODE
modelasanintegrativeframework.In M.P.Zanna(Ed.),Advancesinexperimental
socialpsychology(Vol.23,pp.
75-109).
NewYork:AcademicPress.
Fazio,R.H.,& Olson,M.A.(2003).Implicitmeasuresinsocialcognitionresearch:Their
meaninganduses.AnnualReviewofPsychology,54,297-327.
Fitzsimons,G.J.,&Williams,P.(2000).Askingquestioncanchangechoicebehavior:Does
it dosoautomaticallyoreffortfully?journalofExperimentalPsychology:
Applied,
6,195-206.
Gollwitzer,P.M.(1999).Implementationintentions:Strongeffectsof simpleplans.
AmericanPsychologist,
54,493-503.
Govorun,0.,& Payne,B.K.(2006).Egodepletionandprejudice:Separatingautomatic
andcontrolledcomponents.SocialCognition,24,111-136.
Greenwald,A.G.,&Banaji,M.R.(1995).Implicitsocialcognition:Attitudes,self-esteem,
andstereotypes.PsychologicalReview,102,4-27.
Hense,R.L.,Penner,L.A.,&Nelson,D.L.(1995).Implicitmemoryforagestereotypes.
SocialCognition,13,399-415.
Jacoby,L.L.(1991).A processdissociationframework:Separatingautomaticfrominten-
tionalusesofmemory.journalofMemon)andLanguage,30,513-541.
Jacoby,L.L.(1998).Invarianceinautomaticinfluencesofmemory:Towardauser'sguide
for the process-dissociationprocedure.journal of ExperimentalPsychology:
Learning,Menwry,andCognition.,24,3-26.
Jacoby,L.L.,Begg,I.M.,& Toth,J.P.(1997).In defenseof functionalindependence:
Violationsof assumptionsunderlyingtheprocess-dissociationprocedure?journal
ofExperimentalPsychology:Learning,Memory,andCognition,23,484-495.
Jacoby,L.L.,&Dallas,M.(1981).Ontherelationshipsbetweenautobiographicalmemory
andperceptuallearning.journalof
Experimental Psychology:General,110,
306-340.
Jacoby,L.L.,& Shrout,P.E.(1997).Towarda psychometricanalysisof violationsof
theindependenceassumptionin processdissociation.journal of Experimental
Psychology:Learning,Menwry,andCognition,23,505--510.
Jacoby,L.L.,Toth,J.P.,& Yonelinas,A.P.(I 993).Separatingconsciousandunconscious
influencesof memory:Measuringrecollection.journal of ExperimentalPsy-
chology:General,122,139-154.
James,W (1890).Principlesofpsychology.NewYork:Holt.
Kane,M.J.,Bleckley,M.K.,Conway,A.R.A.,& Engle,R.\V.(2001).A controlled-
attentionviewof working-memorycapacity.journalof ExperimentalPsychology:
General,
130,169-183.
Kawakami,K.,Dovidio,J.F.,Moll,J.,Hermsen,S.,& Russin,
A.(2000).
Justsayno
(tostereotyping):Effectsof trainingin thenegationof stereotypicassociations
on stereotypeactivation.journal of Personalityand Social Psychology,78,
871-888.
Klauer,K.C.,& Wegener,I.(1998).Unravelingsocialcategorizationin the"Whosaid
what?"paradigm.journalofPersonalityandSocialPsychology,75,1155-1178.
Lambert,A.J.,Payne,B.K.,Jacoby,L.L.,Shaffer,L.M.,Chasteen,A.L.,& Khan,
S.K.
(2003).Stereotypesasdominantresponses:Onthe"socialfacilitation"ofprejudice
inanticipatedpubliccontexts.journalof Personality
and
SocialPsychology,
84,
277-295.
Lhermitte,
F.(1986).Humanautonomyandthefrontallobes.PartII:Patientbehaviorin
complexandsocialsituations:The"environmentaldependencysyndrome."Annals
ofNeurology,19,335-343.
j
314 SOCIAL PSYCHOLOGYAND THE UNCONSCIOUS
METHODS:PROCESSDISSOCIATION PROCEDURE 315
Marchetti,C.,& DellaSala,S.(1998).Disentanglingthealienhandandanarchichand.
CognitiveNeuropsychiatry,3,191-207.
Morwitz,V.G.,Johnson,E.,& Schmittlein,D.(1993).Doesmeasuringintentchange
behavior?JournalofConsumerResearch,20,46-61.
Muraven,M.,&Baumeister,R.F.(2000).Self-regulationanddepletionoflimitedresources:
Doesself-controlresembleamuscle?PsychologicalBulletin,126,247-259.
Murphy,S.T.,&Zajonc,R.B.(1993).Affect,cognition,andawareness:Affectivepriming
withoptimalandsuboptimalstimulusexposures.JournalofPersonalityandSocial
Psychology,64,723-739.
Payne,B.K.(2001).Prejudiceandperception:The roleof automaticandcontrolled
processesinmisperceivingaweapon.JournalofPersonalityandSocialPsychology,
81,181-192.
Payne,B.K.(2005).ConceptualizingControlin SocialCognition:Howexecutivecontrol
modulatestheexpressionof automaticstereotyping.Journal of Personalityand
SocialPsychology,89,488-503.
Payne,B.K.,Cheng,C.M.,Govorun,0.,&Stewart,B.D.(2005).Aninkblotforattitudes:
Affectmisattributionasimplicitmeasurement.Journalof PersonalityandSocial
Psychology,89,277-293.
Payne,B.K.,Jacoby,L.L.,&Lambert,A.J.(2004).Memorymonitoringandthecontrolof
stereotyp.edistortion.JournalofExperimentalSocialPsychology,40,52-64.
Payne,B.K.,Jacoby,L.L.,&Lambert,A.J.(2005).Attitudesasaccessibilitybias:Dissoci-
atingautomaticandcontrolledcomponents.In R.Hassin,J.Bargh,&J.Uleman,
(Eds.),Thenewunconscious.Oxford,UK:OxfordUniversityPress.
Payne,B.K.,Lambert,A.J.,&Jacoby,L.L.(2002).Bestlaidplans:Effectsof goalson
accessibilitybiasandcognitivecontrolin race-basedmisperceptionsof weapons.
JournalofExperimentalSocialPsychology,38,384-396.
Petty,R.E.,&Cacioppo,J.T.(1986).TheElaborationLikelihoodMocklofpersuasion.New
York:AcademicPress.
Posner,M.I.,& Snyder,C.R.R.(1975).Attentionandcognitivecontrol.In R.L.Solo
(Ed.),Informationprocessingandcognition:TheLoyolasymposium(pp.55--85).
Hillsdale,NJ:Erlbaum.
Riefer,D.M.,&Batchelder,W.H.(1988).Multinomialmodellingandthemeasurementof
cognitiveprocesses.PsychologicalReview,95,318-339.
Roediger,H.L.,& McDermott,K.B.(1993).Implicitmemoryinnormalhumansubjects.
In H.Spinnler& F.Boller (Eds.),Handbookof neuropsychology(Vol.8,pp.
63-131).Amsterdam:Elsevier..
Schwarz,N.,& Clore,G.L.(1983).Mood,misattribution,andjudgmentsof well-being:
Informativeanddirectivefunctionsof affectivestates.Journalof Personalityand
SocialPsychology,45,513-523.
Sherman,S.J.(1980).Ontheself-erasingnatureoferrorsofprediction.JournalofPerson-
alityandSocialPsychology,39,340-350.
Sherman,J.w.,Groom,C.J.,Ehrenberg,K.,&Klauer,K.C.(2003).Bearingfalsewitness
underpressure:Implicitandexplicitcomponentsof stereotype-drivenmemory
distortions.SocialCognition,21,213-246.
Shiffrin,R.,&Schneider,W.(1977).Controlledandautomatichumaninformationprocess-
ing:II.Perceptualleaming,automaticattending,andageneraltheory.Psycho-
logicalReview,84,127-190.
Shimamura,A.P.(1986).Primingeffectsinamnesia:Evidenceforadissociablememory
function.QuarterlyJournal of ExperimentalPsychology:HumanExperimental
Psychology,38,619-644.
Stangor,C.,& McMillan,D.(1992).Memoryfor expectancy-congruentandexpectancy-
incongruentinformation:A reviewof the social and social developmental
literatures.PsychologicalBulletin,111,42-61.
Stewart,B.D.,& Payne,B.K.(2006).Counterstereotypicalthoughtplansreduceauto-
maticstereotyping.Manuscriptsubmittedforpublication.
Tanner,W.P.,& Swets,J.A.(1954).A decisionmakingtheoryofvisualdetection.Psycho-
logicalReview,61,401--409.
Wegner,D.M.(2002).Theillusionofconsciouswill.Cambridge,MA:MIT Press.
Wilson,T.D.,Lindsey,S.,&Schooler,T.Y.(2000).Amodelofdualattitudes.Psychological
Review,107,101-126.
Yonelinas,A.P.
(2002).
Thenatureofrecollectionandfamiliarity:Areviewof30yearsof
research.JournalofMemoryandLanguage,46,441--451.