Table1AverageaccuracyandRTmsforthesecondstimulusinconcordantpairsfromExperiment1Standarddeviationsareinparentheses PairConcordantRandomSigni cance 1ststim2ndstimAccuRTAccuRTAccuRT ID: 281204
Download Pdf The PPT/PDF document "The rsttwostimuliin(1)aretermedcongruent..." 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.
Thersttwostimuliin(1)aretermedcongruent,thelasttwoincongruent.Wewillcalltwosuccessivestimuliinthe ankertaskconcordantiftheyarematchedforcongruence,thatis,eithereachisdrawnfromthetoptworowsof(1),oreachisdrawnfromthebottomtworows.Twosuccessivestimuliarediscordantiftheyarenotconcordant,thatis,oneisdrawnfromthetoptworowsof(1)andonefromthebottom.Thus,congruenceandincongruencearepropertiesofindividualstimuliwhereasconcordanceanddiscordancearepropertiesofpairs.Notethatthetwomembersofaconcordantpairmayormaynotrequirethesameanswer,andlikewisefordiscordance.Itiswelldocumentedthatresponsetimes(RTs)arelowerforcongruentcomparedtoincon-gruentstimuli(EriksenandEriksen,1974;Grattonetal.,1992).IthasalsobeenfoundthatRTsarelowerforthesecondstimulusofconcordantpairscomparedtothesecondstimulusofdiscordantpairs(theGrattone ect,Grattonetal.,1992).Onepossiblemechanismforthelatterphenomenonisthatcongruentstimuliincreaseattentiontosurrounding ankersinthesubsequentstimulus,therebyo eringamoreextendedvisualtargetincaseofcongruencebutincreasinginterferenceincaseofincongruence.Likewise,incongruentstimuliwoulddrawat-tentionawayfrom ankers,therebyslowingtheresponsetoafollowingcongruentstimulusbutlimitinginterferenceincaseofincongruence.Concordancewouldthusenhanceperformanceinbothsituations,comparedtodiscordance.Inanotherversionoftheexperiment(Grattonetal.,1992;Ghinescuetal.,2010),anexplicitcuesignaledthecongruence/incongruenceofsubsequentstimuli.RTswerelowerwhencuespredictedcongruentstimuli,butnodi erenceinRTwasobservedforcuespredictingincongruentstimuli.RecentevidencesuggeststhattheGrattone ecthingesonconcordantpairswiththesamecorrectanswer,thatis,onsuccessivestimulithatareidentical.RTappearsnottodecreaseforthesecondmemberofaconcordantpairthatrequiresadi erentanswerthantherst(DavelaarandStevens,2009;Mayretal.,2003;Nieuwenhuisetal.,2006).TheGrattone ectmaythusre ectmererepetitionprimingratherthanprimingforthemoreabstractpropertyofstimuluscongruenceorincongruence.PerhapsamorerobustGrattone ectcanbeachievedthroughlearningthestatisticalstructureofsuccessive ankerstimuli,insteadofrelyingonexplicitcuing.Thisspeculationismotivatedbyndingsonpreparatorycontrolintaskswitching.Inapredictablealternating-runsparadigm,participantsareabletolearntopreparefortheupcomingstimulusandreduceswitchcost(see,forexample,RogersandMonsell,1995,andKieseletal.,2010forareview).Inthersttwoexperimentsreportedbelow,statisticalregularitiesareimplicitlyembeddedinthestimuli.Specically,thecongruencyofthesecondmemberofapairiscontingentonthatoftherst,unbeknownsttotheparticipants.InExperiment1,thecongruencyofthesecondstimulustendstoremainthesameasfortherst.InExperiment2,thecongruencyofthesecondstimulustendstodi er.2 Table1:Averageaccuracy(%)andRT(ms.)forthesecondstimulusinconcordantpairsfromExperiment1.(Standarddeviationsareinparentheses) PairConcordantRandomSignicance 1ststim.2ndstim.Accu.RTAccu.RTAccu.RT �����95:0(8:1)416:2(71:5)97:7(6:3)500:3(90:7)p=:16p:001&#x-120;�&#x-120;�&#x-120;�&#x-120;�&#x-120;�97:1(5:2)417:5(64:2)93:6(11:7)492:2(93:2)p=:14p:00199:6(2:3)388:7(68:0)100(0)449:0(77:1)p=:32p:01&#x-278;&#x-278;&#x-278;&#x-278;&#x-278;&#x-278;&#x-278;&#x-278;&#x-278;&#x-278;99:3(2:7)380:8(62:6)98:9(6:1)445:3(80:5)p=:74p:001�&#x-120;�&#x-120;���95:4(7:5)474:2(79:6)89:8(19:5)543:4(92:5)p=:39p:01&#x-278;&#x-278;���94:5(10:5)471:7(78:6)91:0(13:7)548:9(103:0)p=:94p:01&#x-278;&#x-278;������94:7(8:0)455:4(79:9)93:6(8:8)524:9(95:7)p=:30p:01��94:8(7:1)438:7(60:9)92:6(8:8)498:5(87:1)p=:57p:01 trialsmayre ecttheadvantageaccruingtospreadingattentionacrossmultiplearrowheadswiththesamemessage.Experiment2Experiment1documentstheabilitytoexploitconcordanceinsequentialstimulibutleavesopenthesamequestionaboutdiscordance.Inadiscordantpairiftherststimulusiscon-gruentthenthesecondisincongruent,andviceversa.Experiment2wasisomorphictotherstexceptthatitinvolvedadiscordantconditioninplaceoftheoriginalconcordantcon-dition.Thediscordantconditionwascomposedof200trials,80%ofwhichwerediscordantpairs,20%concordant.Halfofthediscordanttrialsinvolvedacongruentstimulusfollowedbyincongruent,andthereversefortheotherhalf.Fortherandomcondition,thedatafromExperiment1wereusedagain.ThirtynewparticipantswererecruitedforExperiment2,drawnfromthesamepoolasbefore(21female,meanage=23:8yrs,SD=4:1).Wecomparedperformanceonmatchingsecondstimuliinthediscordantversusrandomconditions,takingintoaccountjusttrials101�200.Asbefore,trialswithRTsbeyond2.5standarddeviationsoftheparticipant'smeanwereexcluded.AccuracyandRTforthesecondstimulusofdiscordantpairsinthetwoconditionsarepresentedinTable2.ItcanbeseenthatfornoneoftheeighttypesofdiscordanttrialswasRTreliablylowerinthediscordantcomparedtorandomcondition;therewerealsonoreliabledi erencesinaccuracy.Notice,however,thatforalltypesoftrials,theRTswere(non-signicantly)lowerinthediscordantcomparedtorandomcondition.Onlyfourofthe30participantsinthediscordantconditionnoticedthestatisticalrelationship.Comparisonofthetwoexperimentssuggeststhatitismorediculttolearndiscordant5 Table4:Averageaccuracy(%)andRT(ms.)forthesecondstimulusindiscordantpairsinExperiment3comparedtotherandomconditioninExperiment1.(Standarddeviationsareinparentheses) PairDiscordantRandomSignicance 1ststim.2ndstim.Accu.RTAccu.RTAccu.RT ���������95:5(6:3)478:6(144:4)95:3(12:1)543:0(82:1)p=:94p=:04&#x-120;�&#x-120;���89:9(14:9)492:5(145:0)83:9(23:8)538:5(85:0)p=:25p=:14�95:6(6:5)480:6(151:2)89:2(18:8)541:2(103:4)p=:08p=:07&#x]TJ/;ø 9;&#x.962; Tf;&#x 103;&#x.667;&#x 0 T; [0;&#x]TJ/;ø 9;&#x.962; Tf;&#x 103;&#x.667;&#x 0 T; [0;&#x]TJ/;ø 9;&#x.962; Tf;&#x 103;&#x.667;&#x 0 T; [0;&#x]TJ/;ø 9;&#x.962; Tf;&#x 103;&#x.667;&#x 0 T; [0;&#x]TJ/;ø 9;&#x.962; Tf;&#x 103;&#x.667;&#x 0 T; [0;�84:8(15:1)478:6(142:5)82:6(21:1)545:3(80:2)p=:64p=:03�98:6(3:3)429:4(135:9)98:7(3:9)495:4(68:7)p=:85p=:02&#x]TJ/;ø 9;&#x.962; Tf;&#x 103;&#x.667;&#x 0 T; [0;&#x]TJ/;ø 9;&#x.962; Tf;&#x 103;&#x.667;&#x 0 T; [0;��99:1(3:8)421:7(154:0)98:9(6:1)473:6(74:1)p=:86p=:10&#x]TJ/;ø 9;&#x.962; Tf;&#x 103;&#x.667;&#x 0 T; [0;&#x]TJ/;ø 9;&#x.962; Tf;&#x 103;&#x.667;&#x 0 T; [0;�������97:8(4:3)430:9(157:9)96:8(7:8)501:8(84:8)p=:54p=:04�&#x-120;�&#x-120;�&#x-120;�&#x-120;�&#x-120;�98:7(4:1)418:3(156:7)98:4(6:5)469:2(74:0)p=:85p=:12 Experiment3ThecurrentprocedurewaslikeExperiment2exceptthatparticipantswereexplicitlyinformedthateverypairwasdiscordant,thatis,congruencewas(invariably)followedbyincongruenceandviceversa.Asbefore,halfofthediscordanttrialsinvolvedacongruentstimulusfollowedbyanincongruentstimulus,andthereversefortheotherhalf.ThirtynewparticipantscompletedExperiment3(20female,meanage=21:5yrs,SD=3:2).Todeterminewhetherparticipantsexploiteddiscordanceinthepresentprocedure,wecomparedperformanceonthesecondstimuliwithperformanceonmatchingstimuliintherandomconditionofExperiment1.Asbefore,onlytrials101�200wereincludedandoutlierswereexcluded.TheresultsarepresentedinTable4.ItcanbeseenthatRTsinthepresentexperimentwerereliablylowerformoststimulustypescomparedtotherandomconditionofExperiment1.Theenhancedperformanceseemsnottobeduetodelayingtheresponsetorststimuli.TheaverageRTsoftherststimuliinthepresentconditionandintherandomconditionofExperiment1were513:8(SD=167:3)and571:0(SD=82:6),respectively,notreliablydi erent[t(58)=1:68,p=:10].Wealsonotethattherewasnoreliabledi erenceinaccuracybetweenthetwoconditionsforanytypeofdiscordanttrial.GeneralDiscussionOurrstexperimentdocumentsthecontroloverattentionthatobserverscanexercisewhentheylearnthatthecongruencyofonestimulustendstomatchthatofasuccessor.Presumably,thecontrolisbasedonextendingattentiontothe ankersforupcomingcongruentstimuliand7 ReferencesDavelaar,E.J.andStevens,J.(2009).Sequentialdependenciesintheeriksen ankertask:Adirectcomparisonoftwocompetingaccounts.Psychonomicbulletin&review,16(1):121{126.Eriksen,B.A.andEriksen,C.W.(1974).E ectsofnoiselettersupontheidenticationofatargetletterinanonsearchtask.Perception&psychophysics,16(1):143{149.Ghinescu,R.,Schachtman,T.R.,Stadler,M.A.,Fabiani,M.,andGratton,G.(2010).Strategicbehaviorwithoutawareness?e ectsofimplicitlearningintheeriksen ankerparadigm.Memory&Cognition,32(2):197{205.Gratton,G.,Coles,M.G.,andDonchin,E.(1992).Optimizingtheuseofinformation:Strategiccontrolofactivationofresponses.JournalofExperimentalPsychology:General,121(4):480{506.Kiesel,A.,Steinhauser,M.,Wendt,M.,Falkenstein,M.,Jost,K.,Philipp,Andrea,M.,andKoch,I.(2010).Controlandinterferenceintaskswitching|areview.PsychologicalBulletin,136(5):849{874.Mayr,U.,Awh,E.,andLaurey,P.(2003).Con ictadaptatione ectsintheabsenceofexecutivecontrol.Natureneuroscience,6(5):450{452.Nieuwenhuis,S.,Stins,J.F.,Posthuma,D.,Polderman,T.J.C.,Boomsma,D.I.,anddeGeus,E.J.(2006).Accountingforsequentialtriale ectsinthe ankertask:Con ictadaptationorassociativepriming?Memory&cognition,34(6):1260{1272.Rogers,R.D.andMonsell,S.(1995).Costsofapredictableswitchbetweensimplecognitivetasks.JournalofExperimentalPsychology:General,124:207{231.9