ORIGINALRESEARCHpublished:30April2015doi:10.3389/fnhum.2015.00242 ... - PDF document

ORIGINALRESEARCHpublished:30April2015doi:10.3389/fnhum.2015.00242 Tooeasy?TheinuenceoftaskdemandsconveyedtacitlyonprospectivememoryJoanaS.LourençoJohnathanH.HillElizabethA.MaylorDepartmentofPsychology,UniversityofWarwick,Coventry,UK Lourençoetal.Tacitdemandsandprospectivememoryparticipantsrelyonmetacognitivebeliefsaboutthecognitivedemandsoftheentiretaskset(i.e.,ongoingandPMactivities)andtheirabilitytoperformtheupcomingtasks(Marshetal.,2005;Meeksetal.,2007;EinsteinandMcDaniel,2008;Rummeletal.,2013).Theseevaluationscaninfluence,forexample,whetherexternalremindersareused(Gilbert,2015).Theycanalsoaffecttheattentionallocationpolicyestablishedbyparticipantsattheoutsetofthetask,whichspecifiestherelativeweightingofattentiontotheongoingandPMtasks(Marshetal.,2003,2005;Hicksetal.,2005;seealsoSmith,2003;Einsteinetal.,2005).ThekeyquestioninthepresentstudywaswhetheravailableinformationaboutpotentialtargeteventshasanybearingonhowattentionalresourcesaredevotedtothePMtask.Studieshaveshownanincreaseintaskinterference(i.e.,slowingtotheongoingtask)whenPMtaskdifficultyisincreasedbychangingobjectivetaskdemandssuchasnumberoftargetsorspecificityofintentions(e.g.,Hicksetal.,2005;Cohenetal.,2008;Lourençoetal.,2013).Also,whenPMtasksarenonfocal(i.e.,ongoingtaskprocessingdoesnotdirectattentiontowardprocessingtherelevantfeaturesofthetarget),individualsdevoteextraresourcestorememberingtheintention(e.g.,Einsteinetal.,2005;Scullinetal.,2010).Anotherapproachhasinvolvedmanipulatinganticipatedtaskdemandsthroughexplicitinstructionswhileleavingobjectivetaskdemandsintact.Forinstance,instructingparticipantsthatthePMtaskismoreimportantthantheongoingtaskaffectsattentionallocationasevidencedbyanincreaseinbothtaskinterferenceandPMperformance(e.g.,Kliegeletal.,2004;Einsteinetal.,2005).Similarly,RummelandMeiser(2013;Experiment2)showedthatexplicitinformationaboutthecognitiveeffortnecessaryforfulfillingaPMtaskinfluencesattention-allocationstrategies.TheirmanipulationofanticipatedPMtaskdemandsbyinstructingparticipantsthatdetectionofthetargetswillberatherhard(vs.quiteeasy)significantlyincreasedongoingtaskcosts.Moreover,BoywittandRummel(2012;Experiment1)manipulatedanticipatedtaskdemandsbyinstructingparticipantsthattargetswouldbepresentedforonly10%ofallparticipants(or90%inanothercondition).Usingadiffusionmodelanalysis,theauthorsshowedthatparticipantswhoexpectedtheprobabilityoftargetpresentationtobelowwerelesscautiousintheirresponding(i.e.,theirresponsethresholdsintheongoingtaskwerelowerasrevealedbythediffusionmodel'sresponsecriterionparameter).Thus,participants''strategicapproachtoperformingtheongoingtaskdependedonanticipatedPMtaskdemands.WeaddressedthequestionofwhethertacitPMtaskdemandscanalsoaffectparticipants'effortandsuccessinanonfocalPMtask.RatherthanmanipulatingexpectedPMdemandsusingexplicitinstructions—asinallpriorwork—weusedacategorical(nonfocal)PMtaskandvariedtheparticulartargetexemplars(typicalvs.atypical)presentedpriortotheexperimentaltrials.WepredictedthatwhenaskedtogiveaPMresponsetoanimalwordsduringanongoinglexicaldecisiontask(LDT),participantsinstructedusingtypicalexemplarsofthetargetcategory(i.e.,exemplarsthatarefluentlyprocessedandeasilyaccessibleinmemory;Koriatetal.,2004)wouldexpecttosuccessfullyaccomplishthePMtaskwithloweffortandthusdisplaysmallerongoingtaskcoststhanthoseinstructedusingatypicalexemplars.Critically,objectivetaskdemandswerekeptconstantsuchthatallPMtargetspresentedduringtheongoingtaskwereatypicalanimals.Hence,wealsopredictedthatparticipantspresentedwithtypicalexemplarsatencodingwouldperformworseonthePMtaskbecausesuccessfulPMperformanceinnonfocaltasksrequirestheengagementofattention-demandingprocesses(e.g.,Einsteinetal.,2005).Furthermore,weinvestigatedwhethertheattentionallocationpolicysetatthebeginningisflexiblesuchthatitcanbeadjustedonthebasisofexperiencewiththePMtargetsorwhetheritisrelativelyimmutable.Specifically,weexaminedwhetherincongruencebetweenexpectedandactualPMtaskdemandscanleadtolocalchangesinparticipants'attentionallocationpolicy.Wepredictedthatparticipantsgiventypicalexemplars(lowtacitPMdemands)wouldadapttothenewdemandsandshowincreasedtaskinterferenceafterrealizingthattargetscouldbeatypicalinstances.ThiswouldprovideevidencethatindividualscanusetheirexperiencewiththePMtasktoadjusttheirstrategieswhenexpectationsregardingthePMtargetsarebiased.MethodDesignandParticipantsThedesignwasa32mixedfactorial,withtacitPMdemands(high,low,none)asthebetween-subjectsfactor,andblock(baseline,PM)asthewithin-subjectsfactor.Participantswere90undergraduatestudents(39female)aged18–23years(M=20.8,SD=1.0).Thirtyparticipantswererandomlyassignedtoeachofthethreeconditions.Testingtookplaceindividuallyinsessionslastingapproximately25min.EthicalapprovalwasgrantedbytheWarwickPsychologyDepartment'sResearchEthicsCommitteeandallparticipantsprovidedtheirinformedconsent.MaterialsandProcedureParticipantswerefirsttoldabouttheLDT.Instructionsstatedthattheyhadtodecideasquicklyandaccuratelyaspossiblewhetherastringofletterswasaword(``J''keypresswithrightindexfinger)ornot(``F''keypresswithleftindexfinger).Followingtheopportunitytoaskquestions,participantsperformed20practicetrialsandthenabaselineblock(seeTable1)consistingof10buffertrialsand100lexicaldecisiontrials(50wordsand50nonwords).Next,participantsweretoldthattheywouldperformasecondblockoftheLDT,andadditionallygiventhePMinstructions.ThoseinthehighandlowtacitPMdemandsconditionsweregiventhesamePMtaskofrespondingtoanimalwords,butwerepresentedwithdifferentanimalexemplarsatbothinstructionsandpractice.Thesewereatypicalexemplars(walrusandraccoon)ortypicalexemplars(dogandmouse)inthehighvs.lowtacitPMdemandsconditions,respectively.Specifically,participantswereinstructedthatiftheyeversawananimalword(e.g.,WALRUSorDOG,includedinbracketsaccordingtocondition)theyshouldFrontiersinHumanNeuroscience|www.frontiersin.org2April2015|Volume9|Article242 Lourençoetal.TacitdemandsandprospectivememoryTABLE1|Illustrationofthemaindesignandprocedureforparticipantswithhigh,lowandnonetacitprospectivememorydemands,withtypicalityofanimalsindicatedinitalics. TacitPMDemands HighLowNone BaselineBlockLexicalDecisionTask(LDT) PMInstructionsPress“Y”toanimalPress“Y”toanimal—words(e.g.,WALRUS)words(e.g.,DOG)AtypicalTypical Practice(1target)LDT+PMtaskLDT+PMtaskLDT::::::raccoon::::::::::::mouse::::::AtypicalTypical DelayProcessingSpeedTest+Questionnaire PMBlock(4targets)LDT+PMtaskLDT:::.:::pufn:::.:::gazelle:::.:::boar:::.:::hyena:::.:::Atypical pressthe``Y''keyaftertheymadetheirlexicaldecisionorassoonthereafterastheycould.Participantsexplainedtheinstructionstotheexperimenter(toconfirmtheirunderstanding)beforecompleting20practicetrials,whichincludedthepresentationofananimalword(raccoonormouse,accordingtocondition)onTrial15.TocreateadelaybetweenPMtaskinstructionsandthestartofthePMblock,participantscompleteda2-mintestofprocessingspeedandademographicquestionnaire.Thoseinthe``none''PMdemandsconditionwentthroughthesameprocedureexceptthattheydidnotreceivethePMtaskinstructions.ThePMblockcomprised10buffertrialsand260lexicaldecisiontrials,ofwhich256werefillertrials(128wordsand128nonwords)and4werePMtrials.PMtargets(allatypicalanimals)werepresentedonTrials101,152,203and254(puffin,gazelle,boar,andhyena,1respectively,forallparticipants).Eachtrialconsistedofafixationcrosspresentedfor250ms,followedbytheletterstringinlowercase(30-ptfont)untilclassifiedasaword/nonword,andfinallyawaitingmessageuntilthespacebarwaspressed.FillerwordsintheLDT,matchedwiththePMtargetsonmeanlength,syllables,andfrequency,were4–7letters,1–3syllables,andHALfrequency5.5–7.5accordingtoBalotaetal.(2007;nonwordswith4–7letterswereselectedfromthesamesource).AttheendofthePMblock,participantscompletedaquestionnairetotesttheirrecalloftheintendedaction.Recallwasperfectforallparticipants.ResultsDataScreeningTwoparticipantsinthehightacitPMdemandsconditionwhoweremorethan2.5SDsfromtheirgroup'smean 1Beforebeingdebriefed,participantsinthePMgroupswereaskedtofillintypicalityratingsonascalefrom1(verytypicalanimal)to5(veryatypicalanimal).Asexpected,dogandmouse(M=1.5,SD=0.4)wereratedasmoretypicalthanwalrusandraccoon(M=3.9,SD=0.6)andpuffin,gazelle,boar,andhyena(M=3.2,SD=0.4),andthetwoPMgroupsdidnotdifferintheirratings.responsetime(RT)intheongoingtaskwereexcluded.AsiscommonlyobservedinLDTs,performancewashighlyaccuratewith93%ofwordsidentifiedcorrectlyandnosignificantdifferencesacrossconditions.BasedonpreviousPMresearch(e.g.,Knightetal.,2011),wordRTsweretrimmedtoincludeonlycorrectresponsestowordsthatwerelessthan2.5SDsawayfromeachparticipant'smean.TrimmingwasdoneseparatelyforthebaselineandPMblocks(PMtargetsandthetrialimmediatelyfollowingeachofthetargetswereexcluded)andresultedintheeliminationof2.6%ofcorrectRTs.OngoingTaskPerformanceOurmainquestionwaswhethertacitinformationaboutthePMtargetsatinstructions/practicecaninfluenceexpectationsaboutPMtaskdemandsasreflectedbytaskinterference.MeanRTsonfillerwordtrialswereincludedina32mixedANOVAwithtacitPMdemands(high,low,none)asthebetween-subjectsfactorandblock(baseline,PM)asthewithin-subjectsfactor(seeFigure1formeans).Neithermaineffectwassignificant(ps�0.2)buttheinteractionwassignificant,F.2,85/=17.48,MSE=3,731.21,p0.001,p2=0.29.Wethereforeconductedtwofurthermixed22(tacitPMdemandsblock)ANOVAsforhighvs.lowPMdemandsconditions,andlowvs.nonePMdemandsconditions,bothyieldingsignificantinteractions(p=0.006and0.003,respectively).WealsodividedthePMblockintofoursubsets(i.e.,correctwordtrialsprecedingeachPMtarget;seeFigure2)andexaminedongoingtaskcostforthefirstsubset,namely,trialsoccurringbeforethefirsttargetpresentation.Thepatternofresultswassimilartothatfromtheoveralltaskinterferenceanalysis,withasignificantinteractionforthe32ANOVA(p0.001),asignificantinteractionforthehighvs.lowdemandsANOVA(p0.002),butthistimeonlyamarginallysignificantinteractionforthelowvs.nonedemandsANOVA(p=0.074).Therefore,inlinewithourpredictions,ongoingtaskcostwasinfluencedbyourmanipulationoftacitPMtaskdemandssuchthattaskinterferenceinthelowtacitdemandsconditionwasFrontiersinHumanNeuroscience|www.frontiersin.org3April2015|Volume9|Article242 Lourençoetal.Tacitdemandsandprospectivememory FIGURE1|Meancorrectresponsetime(RT)inmilliseconds(ms)forlexicaldecisionstollerwordsforhigh,lowandnonetacitprospectivememory(PM)demandsconditionsacrossblocks.Errorbarsrepresent1standarderror. FIGURE2|MeancorrectRTinmilliseconds(ms)forlexicaldecisionstollerwordsacrossconditionsandsubsetsinthePMblock.Errorbarsrepresent1standarderror. significantlylowerthaninthehighdemandsconditionandthiswasevidentoverallandbeforethefirstPMtargetoccurrence.WhileactualPMtaskdemandsdidnotdifferbetweenconditions,targetswereconsistentwithexpectationsinthehighbutnotinthelowdemandscondition.ThusanimportantquestioniswhetherparticipantsinthelowdemandsconditionadjustedtheirallocationofattentionwhenPMtaskdemandsturnedouthigherthanexpected.ToexamineiftaskinterferencechangedfromthefirsttothefourthPMblocksubset,weincludedfillerwordRTsina34mixedANOVAwithtacitPMdemands(high,low,none)asthebetween-subjectsfactorandPMblocksubset(1–4)asthewithin-subjectsfactor(seeFigure2).TherewasasignificanteffectoftacitPMdemands,F.2,85/=7.89,MSE=71,628.20,p0.001,p2=0.157,butnoeffectofPMblocksubset,F.3,255/=1.90,p=0.13,andnointeraction,F.6,255/=1.72,p=0.117.Thus,RTsremainedrelativelystablethroughoutthePMblockinallconditions,suggestingthatparticipantsinthelowtacitdemandsconditionallocatedfewerresourcestothePMtaskandalsofailedtoadapttothehigherthanexpectedattentionaldemandsposedbythetask(althoughitcanbeseenfromFigure2thatthetrendtowardaninteractionreflectstheresemblanceofthelowPMdemandsconditiontothenonePMdemandsconditioninthefirstPMblocksubsetbuttothehighPMdemandsconditionbythefourthPMblocksubset).However,ofparticularinteresthereisexaminationoftaskinterferenceaccordingtosuccesstothefirsttargetpresentationinthelowtacitdemandscondition.Didparticipantswhosuccessfullydetectedthefirsttarget(n=12)showsubsequentlyincreasedtaskinterferenceincomparisontothosewhofailed(n=18)?WeincludedRTsinthelowtacitdemandsconditionina22mixedANOVAwithfirsttarget(success,failure)asthebetween-subjectsfactorandPMblocksubset(1vs.2)asthewithin-subjectsfactor.Resultsrevealedasignificantinteraction,F.1,28/=8.60,MSE=64,211.91,p=0.007,p2=0.24,suchthattherewasslowingfromtrialsprecedingtothosesucceedingthefirsttargetwhentargetdetectionwasasuccess(Ms=711and765ms,SDs=93and121,respectively;t.11/=�2.84,p=0.016),butnotwhenitwasafailure(Ms=722and701ms,SDs=120and112,respectively;t.17/=1.25,p=0.230).2PMTaskPerformanceHavingshownthatmanipulationoftacitPMdemandsaffectedattentionallocationpolicies,wenextconsiderwhetheritalsoaffectedPMtaskperformance.PMresponseswerescoredascorrectifparticipantspressedthe``Y''keyduringthetargettrialorwithinthenexttrial(seeFigure3formeans).A24mixedANOVAwithtacitPMdemands(high,low)andPMtarget(1–4)asbetween-andwithin-subjectsfactorsrevealedaneffectoftacitPMdemands,F.1,56/=8.34,MSE=0.44,p=0.006,p2=0.13,suchthatPMperformancewassignificantlybetterwithhighthanwithlowtacitdemands(0.64vs.0.39)withnoothersignificanteffects(bothps&#x]TJ/;༠ ; .46;E T; 12;&#x.135;&#x 0 T; [0;0.3).DiscussionOurdatashowthattacitPMtaskdemandscanaffectparticipants'effortandsuccessinaPMtask.Specifically,thosegiventypicalexemplarsatencoding(lowtacitdemandscondition)showedlesstaskinterferenceandworsePMperformancethanthosegivenatypicalones,demonstratingthatindirectlyconveyedinformationaboutPMtaskdifficulty(i.e.,examplesoftargetsatencoding)affectsexpectationsaboutPMdemandsandtherebytheamountofattentionallocatedtotheintention.Crucially,ourresultsalsodemonstratethat,althoughbiasedexpectations 2Somereadersmightbeconcernedthatthispatterncouldalsobeexplainedbymonitoringbeingreinstatedfollowingdetectionofthefirsttarget(cf.Scullinetal.,2010,2013).WethereforerepeatedthisanalysiswhilealsoincludingtacitPMdemands(high,low)asabetween-subjectsfactor.Althoughtherewasareducednumberofparticipants(six)missingthefirsttargetinthehighdemandscondition,thethree-wayinteractionwasmarginallysignificant,F.1,54/=2.83,MSE=3,108.34,p=0.098,p2=0.05.Inthehightacitdemandscondition(contrarytothelow)therewasnogreaterslowingfromtrialsprecedingtothosesucceedingthefirsttargetwhentargetdetectionwassuccessful(791to799ms)relativetowhenitwasunsuccessful(847to859ms).Thesedataareconsistentwithourinterpretationoftheresultsasreflectinganadjustmentoftheattentionallocationpolicyfollowingrealizationthattargetscouldbeatypicalexemplars.FrontiersinHumanNeuroscience|www.frontiersin.org4April2015|Volume9|Article242 Lourençoetal.Tacitdemandsandprospectivememory FIGURE3|MeanproportioncorrectforthePMtaskacrossconditionsforeachofthefourPMtargets.Errorbarsrepresent1standarderror. canharmPMperformancewhenactualdemandsturnouthigherthanexpected,participantscanadaptfollowingtargetexperience.Inotherwords,thepresentresultsshowthatindividualscanadjustattention-allocationstrategiesfollowingsuccessfuldetectionofatargetthatisincongruentwiththeir(biased)metacognitiveexpectations.ThepresentstudyprovidesnovelevidencethatdetectingPMtargetsthatareinconsistentwithtacitdemandscanelicitlocalchangesinattentionallocation.Specifically,participantsinthelowtacitdemandsconditionwhodetectedthefirstPMtarget(hencerealizingthattargetscouldbeatypicalexemplars)showedanincreaseinongoingtaskRTsfollowingthisfirsttargetandalsowentontoperformsimilarlytothoseinthehightacitdemandscondition(0.67vs.0.64successtoTarget2).ThisextendspreviousresearchshowingthatindividualscanadjusttheamountofattentiondevotedtotheintentionwhenmonitoringgoesunreinforcedduetothelackofPMtargetoccurrences(e.g.,Loftetal.,2008;Scullinetal.,2010;cf.BoywittandRummel,2012),andalsothattrial-by-trialchangesintheallocationofattentioncanoccur.Forexample,taskinterferencecanchangeflexiblyasaresultofchangesintheefforttowardanongoingtask(Marshetal.,2005)oranitem'srelevanceforthePMtask(e.g.,Marshetal.,2006;LourençoandMaylor,2014).Thepresentresultsprovidesupportfortheproposalthatattentionallocationisflexible(Scullinetal.,2013),suchthatexperiencewiththeongoingandthePMtaskcanalsochangethepolicyovertime(Hicksetal.,2005;Marshetal.,2005).Interestingly,KuhlmannandRummel(2014)recentlyshowedthatindividualscanalsoflexiblyupdatetheirattention-allocationpolicyafterinitialintentionencodingasaresultoflearningduringtheongoingtaskwhichtrialsarePM-relevant.Thus,theseauthorsshowedthatparticipantswhoknewthatPMcueswouldonlyoccurinasub-setoftrialswereabletolearnwhichtrialswerePM-relevantasdemonstratedbyareductioninongoingcostforPM-irrelevanttrials(seealsoLourençoetal.,2013).Inaddition,wepresentedparticipantswithanatypical/typicalcategoryexemplaratbothinstructionsandpractice.Previousresearchsuggeststhatlearners'metacognitivebeliefsabouthowitemcharacteristicsaffectmemorabilityaresensitivetotaskexperience(e.g.,TullisandBenjamin,2012).Byanalogy,weassumethatintentionretrievalduringpracticemayhavestrengthenedparticipants'beliefsaboutthedifficulty/easeofsuccessfullyfulfillingtheintention.ParticularlyinlaboratorysettingsthismightprovideparticipantswithadditionalinformationonwhichtobasetheirexpectationsaboutPMdemandssincetheylackpriorexperiencewiththePMtask(cf.RummelandMeiser,2013).Futureresearchcouldexaminewhethertargetexemplarspresentedatinstructions(i.e.,atthetimeofintentionformation)guideattentionallocationorwhethermetacognitivebeliefsaboutthedifficultyofcompletingthePMtaskarealsodeterminedbydirecttaskexperienceduringpractice.Althoughthepresentstudydoesnotisolatethelocusoftheeffectmoreprecisely,itdoesdemonstratethatinformationaboutparticulartargetexemplarsinfluencesmetacognitiveexpectationsandtherebytheamountofattentiondevotedtoacategoricalintentionandtaskinterference.WearguedthatworsePMperformanceinthelowrelativetothehightacitdemandsconditionwasduetodifferencesinattentionallocationpolicies.Alternatively,itcouldbeclaimedthatworsePMperformanceinthelowdemandscondition(typicalexemplarsatencoding)wasduetothemismatchbetweenencodingandretrievalcontexts(TulvingandThomson,1973).Participantsmighthavegeneratedanimalexemplarsatencoding(EllisandMilne,1996)and,becauseallPMtargetswereatypicalanimals,doingsowouldfacilitaterecognitionoftargetsforindividualsinthehighdemandsconditiononly.Evenifweassumethatparticipantsspontaneouslygeneratedcategoryexemplarsatencoding,andthatinthehighdemandsconditionthesewerethesameitemslaterpresented,thecontextmatchingaccountisinconsistentwithourobservationthattacitdemandsaffectedtaskinterferencepriortoanytargetoccurrence.IfparticipantsdisregardedtheinformationaboutthetargetexemplarswhenallocatingattentiontothePMtask,thereshouldhavebeennodifferenceintherepresentationoftheintentioninmemoryand,accordingly,nocostdifferencesbetweenlowandhightacitdemandsconditions.Therefore,themostparsimoniousexplanationoftheresultsisthatreducedPMperformanceforthelowdemandsconditionwasprimarilyduetoparticipantsallocatinginsufficientresourcestomeetactualtaskdemands.Finally,notethatbytacitdemandswemeanthatwedidnot,atanypoint,directlyinstructparticipantswithrespecttothedemandsofthePMtask.Instead,demandswereconveyedindirectlybyprovidingparticipantswithparticularexemplarsofPMtargetsbeforeongoingtaskperformance.Wearenotclaimingthattheeffectoftargetexemplarsonattentionallocationoccurredwithoutindividuals'consciousapprehension,althoughweacknowledgethatthisisapossibility.AsproposedbyHicksetal.(2005,p.442)``[t]hesettingofaninitialattentionalallocationpolicyneednotbeconscious,butmayrepresentametacognitivestrategyabouthowtoapproachtheentiretaskset,andtherefore,notnecessarilybeaccessibletoconsciousawareness''.Inconclusion,threekeyaspectsofthepresentstudydistinguishitfrompreviousresearch.First,wemanipulatedtacitdemandsaboutthePMtask.Second,objectivePMtaskdemandswerethesameforallparticipants.Third,forsomeFrontiersinHumanNeuroscience|www.frontiersin.org5April2015|Volume9|Article242 Lourençoetal.TacitdemandsandprospectivememoryparticipantstacitandactualPMdemandswereincongruentbutparticipantswereneverexplicitlywarnedaboutsuchchange.OurfindingssuggestthatinstudyingattentionallocationpoliciesandtheirimpactontaskinterferenceinPMtasksitisimportanttoconsidertheroleoftacitinformationaboutPMtaskdemands(e.g.,asconveyedbyspecifictargetexemplarsatencoding).Suchinformationcaninfluenceindividuals'beliefsabouttheeaseoffulfillingaPMtask,asevidencedbyitseffectonongoingtaskprocessing,andcanharmPMperformancewhenactualdemandsturnouthigherthanexpected.AcknowledgmentsThisresearchwassupportedbythePortugueseFoundationforScienceandTechnology(FCT)throughascholarshiptoJSL.WethankAdamSanbornforhelpfuldiscussion.ReferencesBalota,D.A.,Yap,M.J.,Cortese,M.J.,Hutchison,K.A.,Kessler,B.,Loftis,B.,etal.(2007).TheEnglishlexiconproject.Behav.Res.Methods39,445–459.doi:10.3758/BF03193014Boywitt,C.D.,andRummel,J.(2012).Adiffusionmodelanalysisoftaskinterferenceeffectsinprospectivememory.Mem.Cognit.40,70–82.doi:10.3758/s13421-011-0128-6Cohen,A.L.,Jaudas,A.,andGollwitzer,P.M.(2008).Numberofcuesinfluencesthecostofrememberingtoremember.Mem.Cognit.36,149–156.doi:10.3758/mc.36.1.149Einstein,G.O.,andMcDaniel,M.A.(2005).Prospectivememory:multipleretrievalprocesses.Curr.Dir.Psychol.Sci.14,286–290.doi:10.1111/j.0963-7214.2005.00382.xEinstein,G.O.,andMcDaniel,M.A.(2008).``Prospectivememoryandmetamemory:theskilleduseofbasicattentionalandmemoryprocesses,''inThePsychologyofLearningandMotivation,edsA.S.BenjaminandB.H.Ross(SanDiego:AcademicPress),145–173.Einstein,G.O.,McDaniel,M.A.,Thomas,R.,Mayfield,S.,Shank,H.,Morrisette,N.,etal.(2005).Multipleprocessesinprospectivememoryretrieval:factorsdeterminingmonitoringversusspontaneousretrieval.J.Exp.Psychol.Gen.134,327–342.doi:10.1037/0096-3445.134.3.327Ellis,J.,andMilne,A.(1996).Retrievalcuespecificityandtherealizationofdelayedintentions.Q.J.Exp.Psychol.Sect.A.Hum.Exp.Psychol.49,862–887.doi:10.1080/713755662Gilbert,S.J.(2015).Strategicuseofreminders:influenceofbothdomain-generalandtask-specificmetacognitiveconfidence,independentofobjectivememoryability.Conscious.Cogn.33,245–260.doi:10.1016/j.concog.2015.01.006Hicks,J.L.,Marsh,R.L.,andCook,G.I.(2005).Taskinterferenceintime-based,event-based,anddualintentionprospectivememoryconditions.J.Mem.Lang.53,430–444.doi:10.1016/j.jml.2005.04.001Kliegel,M.,Martin,M.,McDaniel,M.A.,andEinstein,G.O.(2004).Importanceeffectsonperformanceinevent-basedprospectivememorytasks.Memory12,553–561.doi:10.1080/09658210344000099Knight,R.G.,Harnett,M.,andTitov,N.(2005).Theeffectsoftraumaticbraininjuryonthepredictedandactualperformanceofatestofprospectiveremembering.BrainInj.19,19–27.doi:10.1080/02699050410001720022Knight,J.B.,Meeks,J.T.,Marsh,R.L.,Cook,G.I.,Brewer,G.A.,andHicks,J.L.(2011).Anobservationonthespontaneousnoticingofprospectivememoryevent-basedcues.J.Exp.Psychol.Learn.Mem.Cogn.37,298–307.doi:10.1037/a0021969Koriat,A.,Bjork,R.A.,Sheffer,L.,andBar,S.K.(2004).Predictingone'sownforgetting:theroleofexperience-basedandtheory-basedprocesses.J.Exp.Psychol.Gen.133,643–656.doi:10.1037/0096-3445.133.4.643Kuhlmann,B.G.,andRummel,J.(2014).Context-specificprospective-memoryprocessing:evidenceforflexibleattentionallocationadjustmentsafterintentionencoding.Mem.Cognit.42,943–949.doi:10.3758/s13421-014-0405-2Loft,S.,Kearney,R.,andRemington,R.(2008).Istaskinterferenceinevent-basedprospectivememorydependentoncuepresentation?Mem.Cognit.36,139–148.doi:10.3758/mc.36.1.139Lourenço,J.S.,andMaylor,E.A.(2014).Isitrelevant?Influenceoftrialmanipulationsofprospectivememorycontextontaskinterference.Q.J.Exp.Psychol.(Hove)67,687–702.doi:10.1080/17470218.2013.826257Lourenço,J.S.,White,K.,andMaylor,E.A.(2013).Targetcontextspecificationcanreducecostsinnonfocalprospectivememory.J.Exp.Psychol.Learn.Mem.Cogn.39,1757–1764.doi:10.1037/a0033702Marsh,R.L.,Cook,G.I.,andHicks,J.L.(2006).Taskinterferencefromevent-basedintentionscanbematerialspecific.Mem.Cognit.34,1636–1643.doi:10.3758/bf03195926Marsh,R.L.,Hicks,J.L.,andCook,G.I.(2005).Ontherelationshipbetweenefforttowardanongoingtaskandcuedetectioninevent-basedprospectivememory.J.Exp.Psychol.Learn.Mem.Cogn.31,68–75.doi:10.1037/0278-7393.31.1.68Marsh,R.L.,Hicks,J.L.,Cook,G.I.,Hansen,J.S.,andPallos,A.L.(2003).Interferencetoongoingactivitiescovarieswiththecharacteristicsofanevent-basedintention.J.Exp.Psychol.Learn.Mem.Cogn.29,861–870.doi:10.1037/0278-7393.29.5.861Meeks,J.T.,Hicks,J.L.,andMarsh,R.L.(2007).Metacognitiveawarenessofevent-basedprospectivememory.Conscious.Cogn.16,997–1004.doi:10.1016/j.concog.2006.09.005Rummel,J.,Kuhlmann,B.G.,andTouron,D.R.(2013).Performancepredictionsaffectattentionalprocessesofevent-basedprospectivememory.Conscious.Cogn.22,729–741.doi:10.1016/j.concog.2013.04.012Rummel,J.,andMeiser,T.(2013).Theroleofmetacognitioninprospectivememory:anticipatedtaskdemandsinfluenceattentionallocationstrategies.Conscious.Cogn.22,931–943.doi:10.1016/j.concog.2013.06.006Schnitzspahn,K.M.,Zeintl,M.,Jäeger,T.,andKliegel,M.(2011).Metacognitioninprospectivememory:areperformancepredictionsaccurate?Can.J.Exp.Psychol.65,19–26.doi:10.1037/a0022842Scullin,M.K.,McDaniel,M.A.,andShelton,J.T.(2013).Thedynamicmultiprocessframework:evidencefromprospectivememorywithcontextualvariability.Cogn.Psychol.67,55–71.doi:10.1016/j.cogpsych.2013.07.001Scullin,M.K.,McDaniel,M.A.,Shelton,J.T.,andLee,J.H.(2010).Focal/nonfocalcueeffectsinprospectivememory:monitoringdifficultyordifferentretrievalprocesses?J.Exp.Psychol.Learn.Mem.Cogn.36,736–749.doi:10.1037/a0018971Smith,R.E.(2003).Thecostofrememberingtorememberinevent-basedprospectivememory:investigatingthecapacitydemandsofdelayedintentionperformance.J.Exp.Psychol.Learn.Mem.Cogn.29,347–361.doi:10.1037/0278-7393.29.3.347Tullis,J.G.,andBenjamin,A.S.(2012).Theeffectivenessofupdatingmetacognitiveknowledgeintheelderly:evidencefrommetamnemonicjudgmentsofwordfrequency.Psychol.Aging27,683–690.doi:10.1037/a0025838Tulving,E.,andThomson,D.M.(1973).Encodingspecificityandretrievalprocessesinepisodicmemory.Psychol.Rev.80,352–373.doi:10.1037/h0020071ConictofInterestStatement:Theauthorsdeclarethattheresearchwasconductedintheabsenceofanycommercialorfinancialrelationshipsthatcouldbeconstruedasapotentialconflictofinterest.Copyright©2015Lourenço,HillandMaylor.Thisisanopen-accessarticledistributedunderthetermsoftheCreativeCommonsAttributionLicense(CCBY).Theuse,distributionandreproductioninotherforumsispermitted,providedtheoriginalauthor(s)orlicensorarecreditedandthattheoriginalpublicationinthisjournaliscited,inaccordancewithacceptedacademicpractice.Nouse,distributionorreproductionispermittedwhichdoesnotcomplywiththeseterms.FrontiersinHumanNeuroscience|www.frontiersin.org6April2015|Volume9|Article242