addition, however, there is a need to consider the

ModulationofWord-ReadingProcessesinTaskSwitchingMichaelE.J.MassonandDanielN.BubUniversityofVictoriaToddS.WoodwardUniversityofBritishColumbiaJasonC.K.ChanUniversityofVictoriaTheauthorsexaminedmodulationofthesimpleactofwordnaminginducedbytheconflictarisingwhenthattaskcompeteswithcolornaminginatask-switchingparadigm.Subjectsalternatedbetweennamingawordprintedinblackandnamingthecolorofastimulusin2conditions.Intheincongruentcondition,thecoloredstimuluswasanirrelevantwordgeneratingconflict,andintheneutralcondition,colorwascarriedbyarowofasterisks.Subjectstooksubstantiallylongertonameawordprintedinblackintheincongruentcondition,implyingaformofsuppression.Thismodulationoftheword-namingresponsewasadaptiveinthatitledtomoreefficientcolornaming.Themodulationeffectwasreplicatedusingphonemedetectioninsteadofwordnamingbutnotwithlexicaldecisionorvisualcomparison,impli-catingaphonologicalencodingprocess.Respondingtostimulithatinvitetheapplicationofdifferent,competingprocessesrequiresspecializedcontrolmechanismstoresolveconflict.Forexample,namingthecolorofawrittenwordentailsavoidingthehabitualtendencytoreadtheword,andsubjectsaresubstantiallyslowertonamecolorsunderthiscondi-tioncomparedwithaconditioninwhichtheyarerequiredtonamethecolorofaneutralstimulus,suchasarowofXs(e.g.,Klein,1964).Thistypeofconflictisparticularlyprevalentinsituationswheresubjectsrapidlyswitchfromonetasktoanotherandthestimulusoneachtrialaffordsboththecurrentlydemandedtaskandthepreviouslyexecutedone(e.g.,Allport,Styles,&Hsieh,1994;Meiran,2000;Meiran,Chorev,&Sapir,2000).Switchingbetweentasksunderthesecircumstancesundoubtedlyinvolvesaformofexecutivecontroloperatingthroughacentralexecutiveorasu-pervisoryattentionalsystem(Baddeley,Chincotta,&Adlam,2001;Logan&Gordon,2001;Norman&Shallice,1986).Inaddition,however,thereisaneedtoconsidertheconsequencesofresolvingtheconflictthatisgeneratedwhenastimulushasat-tributesrelevanttothetwocurrentlyactivetasksetsbetweenwhichthesubjectisswitching.Forexample,resolvingtheconflictinherentintheactofnamingthecolorinwhichawordisprintedmayleadtothesuppressionofword-readingprocesses,whichmaythenhaveconsequencesonasubsequenttrialwhenwordreadingisdemanded.Inthisarticle,weexaminetheideathatthisformofsuppressionmayconstituteanessentialcomponentofthecostassociatedwithswitchingbetweentasks.Task-switchingcostsarisewhensubjectsarerequiredtoshiftbetweenatleasttwodifferenttasksonasequenceoftrialsandareparticularlyapparentwheneachtargetstimulusisbivalent;namely,itinvitestheapplicationofeitherofthetwotasks.Incomparisontoperformanceonaseriesoftrialsinvolvingasingletask,performanceintrialsthatarepartofasequenceoftwoalternatingtasksissignificantlyworse(e.g.,Allportetal.,1994;Jersild,1927;Spector&Biederman,1976).Thecostofswitchingbetweentaskscanbereducedbutnoteliminatedbyallowingsubjectssubstantialtime(ontheorderofseconds)toprepareforanupcomingtaskswitch(e.g.,Meiran,1996;Rogers&Monsell,1995).Thereductioninswitchcostachievedbyprovidingsubjectswithpreparationtimeisgenerallyacceptedasevidenceforexec-utivecontrolprocesses.Buttheresidualswitchcost––thecostremainingafteradequatepreparatorytimeisgranted––isacom-ponentofswitchcostthatisthetargetofsomecontroversy.Inoneview,eventheresidualswitchcostisascribedtoanaspectofexecutivecontrol,wherebyanewtaskcannotbefullyengageduntilitisexternallycuedbyatask-relevantstimulus(Rogers&Monsell,1995).Arivalaccountofresidualswitchcostisthatdisengagementofacompletedtaskinvokesinhibitionofthattasksettoenabletheswitchtoanewtask(Mayr&Keele,2000).Thisbackwardinhibitionmakesitmoredifficulttoreturntotheinhibitedtasksetwhenrequiredbyasubsequenttaskswitch.Yetanotheraccountisthatswitchcostistheresultofthecom-binedeffectsof(a)inadvertentandinappropriatecuingoftheMichaelE.J.Masson,DanielN.Bub,andJasonC.K.Chan,Depart-mentofPsychology,UniversityofVictoria,Victoria,BritishColumbia,Canada;ToddS.Woodward,DepartmentofPsychology,UniversityofBritishColumbia,Vancouver,BritishColumbia,Canada.JasonC.K.ChanisnowattheDepartmentofPsychology,WashingtonUniversity.ThisresearchwassupportedbyresearchgrantsfromtheNaturalSciencesandEngineeringResearchCouncilofCanadatoMichaelE.J.MassonandtoDanielN.Bub.Orderofauthorshipforthefirsttwoauthorsisarbitrary.ThanksareextendedtoSethBloom,ShainaHood,EricaMcCollum,CameliaPakzad-Vaezi,NeilPegram,andLucasRiedlforassistancewithdatacollec-tion.WealsothankIringKoch,UlrichMayr,andNachshonMeiranforhelpfulcommentsonanearlierversionofthisarticle.CorrespondenceconcerningthisarticleshouldbeaddressedtoeitherMichaelE.J.MassonorDanielN.Bub,DepartmentofPsychology,UniversityofVictoria,P.O.Box3050STNCSC,Victoria,BritishColum-biaV8W3P5,Canada.E-mail:[e-mail]@uvic.caJournalofExperimentalPsychology:GeneralCopyright2003bytheAmericanPsychologicalAssociation,Inc.2003,Vol.132,No.3,400–4180096-3445/03/$12.00DOI:10.1037/0096-3445.132.3.400400previoustaskbythecurrentstimulusand(b)suppressionofanirrelevanttaskontrialNthatnowbecomesrelevantontrialN
1(Allportetal.,1994;Allport&Wylie,2000).Thecontroversyregardinghowbesttointerpretresi