stillunclear.Accordingtoacompensationhypothesisetal.,1997;Cabeza,2002) - PDF document

stillunclear.Accordingtoacompensationhypothesisetal.,1997;Cabeza,2002),increasedbilater-alityinoldadultscouldhelpcounteractage-relatedneurocognitivedecits.Consistentwiththishypothe-sis,inthePETstudybyReuter-Lorenzetal.olderadultswhodisplayedabilateralpatternofPFCactivitywereinaverbalworkingmemorytaskthanthosewhodidnot.Thecompensationhypothesisisalsosupportedbyevidencethathemisphericasym-metryreductionsmayfacilitaterecoveryfrombraindamage.Numerousstudieshavedemonstratedthatrecoveryofmotorandlanguagefunctionsafterunilat-eralbraindamagemayinvolvetherecruitmentoftheunaffectednondominanthemisphere.Thisndinghasbeenobservedwithavarietyoftechniques,includingelectrophysiologicalmeasures(Hondaetal.,etal.,1997);transcranialmagneticstimula-tion(Cicinellietal.,1997;Netzetal.,1997),Dopplerultrasonography(Silvestrinietal.,1993;Silvestrini1998),Xenon-133imaging(Brionetal.,1989;De-meurisseandCapon,1991),PET(DiPieroetal.,etal.,1995;Weilleretal.,1995;Buckner1996;Ohyamaetal.,1996;Hondaetal.,1997),andfunctionalMRI(Caoetal.,1999;Thulbornetal.,Asaresultofcontralateralrecruitment,cognitivefunc-tionsthatarestronglylateralizedinthehealthybrainmaybecomemorebilateralfollowingbraindamage,andseveralstudieshavedirectlylinkedsuccessfulre-coveryoffunctiontobihemisphericinvolvement.Forexample,alongitudinalstudyusingDopplerultra-sonographyfoundthatafteraperiodofspeechtherapy,uencyinagroupofaphasicpatientswasasso-ciatedwithabilateralincreaseinowvelocity(Silves-etal.,1993).AnfMRIstudyfoundasimilarresult:severalmonthsafteraleft-hemispherestroke,betterlanguagerecoverywasobservedinaphasicpatientswhoshowedbilateralactivations(Caoetal.,Becausebihemisphericinvolvementcanfacilitatere-coveryfrombraindamage,itisreasonabletoassumethatitmayalsoplayacompensatoryfunctionintheagingbrain.However,thereisanalternativeaccountofage-re-latedasymmetryreductions:accordingtoadedifferen-tiationhypothesis,reducedhemisphericasymmetryinoldadultsmayreectanage-relateddifcultyinre-cruitingspecializedneuralmechanisms(LiandLin-denberger,1999).Thenotionofage-relateddedifferen-tiationissupportedbyevidencethatcorrelationsamongdifferentcognitivemeasurestendtoincreasewithage(e.g.,MitrushinaandSatz,1991;Babcock1997;BaltesandLindenberger,1997)andbycom-putersimulationslinkingage-relateddeclinesincate-cholaminefunctiontoanincreaseinneuralnoiseandLindenberger,1999;Lietal.,2000).Accordingtothededifferentiationhypothesis,age-relatedasymme-tryreductionsarejustanotherexampleofthedelete-riouseffectsofagingonbrainfunction.Themaingoalofthepresentstudywastocontrastpredictionsofthecompensationanddedifferentiationhypotheses.Beforescanning,wetested33olderadultsTABLE1PET/fMRIActivityinLeftandRightPFCinYoungerandOlderAdults CognitivedomainImagingtechnique:Materials/taskYoungerOlderLeftRightLeftRightEpisodicretrievalPET:Paircued-recall(Cabezaetal.,
PET:Stemcued-recall(Baetal.,PET:Wordrecognition(Maddenetal.,
PET:Facerecognition(Gradyetal.,
Episodicencoding/semanticretrievalfMRI:Wordincidental(Stebbinsetal.,fMRI:Wordintentional(Loganetal.,fMRI:Wordincidental(Loganetal.,fMRI:Wordsubsequentmemory(Morcometal.,WorkingmemoryPET:LetterDR(Reuter-Lorenzetal.,PET:LocationDR(Reuter-Lorenzetal.,PET:NumberN-back:(Dixitetal.,PET:Facematching(Gradyetal.,1994,Expt.2)
PET:Facematching(Gradyetal.,InhibitorycontrolfMRI:No-gotrails(Nielsonetal.,PlussignsindicatesignicantactivityintheleftorrightPFC,andminussignsindicatenonsignicantactivity.ThenumberofplussesisanapproximateindexoftherelativeamountofactivityinleftandrightPFCineachstudy,anditcannotbecomparedacrossstudies.DR,delayedresponsetask.COMPENSATORYACTIVITYINHIGH-PERFORMINGELDERLY and12youngadults(Youngparticipants)onabatteryofmemorytests.AsillustratedinFig.1,weselected8olderadultsthatperformedaswellyoungadults(Old-Highparticipants)and8olderadultsthatperformedcantlybelowthanyoungadults(Old-Lowpartic-ipants).ThethreegroupsofparticipantswerePETscannedwhileperformingtwomemorytests,recallandsourcememory.ThesetestswereselectedbecausetheyyieldedacleardissociationinthelateralizationofPFCactivityinapreviousPETstudywithyoungadults:whereasleftPFCwasmoreactivatedforrecallthanforsourcememory,rightPFCshowedtheconversepat-tern(Cabezaetal.,2001).GivenasymmetricPFCac-tivityinyoungadults,thecompensationhypothesispredictsthathemisphericasymmetryreductionsinPFCactivityshouldbegreaterinOld-Highpartici-pantsthaninOld-Lowparticipants.Moreover,thecompensationhypothesispredictsthatthesehemi-sphericasymmetryreductionsshouldoccurinthemostdemandingofthetwotasks,whichinthepresentstudywasthesourcetask.Incontrast,thededifferentiationhypothesispredictsthathemisphericasymmetryshouldoccurinthegroupofolderadultsshowingmorepronouncedage-relatedcognitivedecline,thatis,theOld-Lowparticipants.Insummary,ndingtheHA-ROLDpatternduringsourcememoryinOld-Highpar-ticipantswouldsupportthecompensationhypothesis,ndingtheHAROLDpatterninOld-Lowpar-ticipantswouldsupportthededifferentiationhypothe-sis.IftheHAROLDpatternisnotfoundineithergroup,thevalidityoftheHAROLDmodelcouldbeTheparticipantsofthePETstudywere12Youngparticipants(5female,7male;agerange:2035),8Old-Highparticipants(4female,4male;agerange:78),and8Old-Lowparticipants(4female,4male;agerange:6374).Theageofthetwooldgroups(seeTable2)wassimilar(0.3)andgreaterthantheageoftheyounggroup.Old-HighandOld-Lowpartici-pantswereselectedusingacompositememoryscorebasedontheresultsoffourmemorymeasures:LogicalMemoryI,VerbalPairedAssociatesI,andVisualPairedAssociatesIIfromtheWechslerMemoryScale-Revised(Wechsler,1987)andtheLong-DelayCuedRecallfromtheCaliforniaVerbalLearningTest(Delisetal.,1987).Thesetaskswereselectedbecausetheypreviouslyhavebeenshowntodistinguisholderadultswithlowandhighmnemonicfunctioning(Gliskyetal.,1995).TherawandstandardizedscoresofthesescreeningtestsarelistedinTable2.Theaveragestan-dardizedscores,whichwereusedforselectingpartici-pants,arelistedinTable2andFig.1.Averagestan-dardizedscoresweresimilarforYoungandOld-Highparticipants(0.6),butsignicantlylowerforOld-LowthanforYoungandOld-Highparticipants(both0.001).Despitesignicantdifferencesinmemoryperformance,thethreegroupswereequivalent(all0.3)infourtestsassumedtoreectexecutivefunctionandgeneralintellectualperformance(seeTable2):WisconsinCardSortingTest(WCST),orthographicuencytest(FAS),mentalarithmetictest(WAIS-R),andmentalcontroltest(WMS-R).Allparticipantswererighthandedandhadnohistoryofneurologicalorpsychiatricillness.Noneoftheparticipantswastakingmedicationorhadamedicalconditionthatcouldaffectcerebralbloodow(e.g.,highbloodpressure).ThestudywasapprovedbythejointBaycrestCentre/Uni-versityofTorontoResearchEthicsandScienticRe-viewCommittee.BehavioralMethodsThestudyconsistedoftwosessions,screeningandPET.Duringthescreeningsession,participantsoutdemographicandhealthquestionnairesandcom-pletedseveralneuropsychologicaltestsincludingtheonesusedforscreening(seeabove).Attheendofthescreeningsession,theypracticedthetaskstobeper-formedinthescanner.ThePETsessiontookplaceseveraldaysafterthescreeningsessionandincludedparticipantsselectedaccordingtotheirlevelofmemoryperformance(seeabove).Inboththerecallandthe FIG.1.Standardizedcompositememoryscoresforyoungandolderadultsthatcompletedabatteryofmemorytestsduringascreeningsession.Amongolderadults,eightwhoperformedaswellasyoungadults(Old-High)andeightwhoperformedsignibelowyoungadults(Old-Low)wereselectedtobepartofthePETCABEZAETAL. contextmemoryconditions,subjectsstudiedalistofitemsbeforethescanandtheirmemoryfortheseitemswastestedduringthescan.Atstudy,itemswerepre-sentedata3s/itemrate,andsubjectswereinstructedtorememberthemforasubsequentmemorytest.Inbothconditions,thestudylistwaspresentedonceorfourtimes,butforthepresentanalysesweaveragedacrossthismanipulation.Intherecallcondition,sub-jectsstudiedavisuallypresentedlistof24unrelatedword-pairs(e.g.,lawyerwindow),andduringthescan,theywerepresentedtherstwordofeachstud-iedpair(e.g.,lawyer)andtriedtorecallthesecondword(e.g.,window).Inthesourcerecognitiontest,subjectsstudiedtwolistsof12singlewordsbeforethescan,1presentedauditorilyand1visually,andduringthescan,theyreadeachstudiedwordanddecidedwhetherthewordwaseitherheardorreadduringthestudyphase.Attest,wordswerepresentedfor4sandfollowedbyxationfor1s.Thetestliststarted30sbeforeandcontinued30safterthe60-sscanwindow.Inallconditions,subjectsrespondedtoeachitembysaying1wordaloud:thewordrecalledintherecalltaskandinCRN.Intherecalltask,ifsubjectscouldnotrecallawordbeforexationap-peared,theysaidsothat1wordwasspokenineverytrial.PETMethodsTwoPETscanswereconductedforeachofthetwotasks,andtheorderoftwotaskswascounterbalancedacrosssubjects.ThePETsessionincludedotherscansthatarenotreportedhere.PETscanswereobtainedwithaGEMS-ScanditronixPC2048-15Bheadscannerusingabolusinjectionof35.5mCiof15OO.Imageprocessingandstatisticalanalyseswereperformedus-ingSPM99b(WellcomeDepartmentofCognitiveNeu-rology,London,UK)implementedinMatlab(Math-worksInc.,Sherborn,MA).First,thedifferentimagesfromeachsubjectwererealignedtotherstimage.Second,therealignedimagesfromeachsubjectweretransformedintoastandardspace(TalairachandTournoux,1988)andsmoothedusinga10-mmisotro-picGaussiankernel.Third,theeffectsoftheconditionsontheregionalcerebralbloodowateachvoxelwereestimatedusingagenerallinearmodel,whereinthechangesinglobalcountsareconsideredasacovariate.Theeffectsofeachcomparisonwereestimatedusinglinearcontrasts,whichyieldastatistic(expressedas-score)foragivencomparisonateachvoxel.Statis-ticalcomparisonswereperformedintwosteps.(1)PFCregionsshowingsignicanttaskeffects(recallvssource)wereidentiedatathresholdof3.09(0.001uncorrected).(2)ForthehandfulofPFCregionsedinStep1,taskgroupinteractionswereedwithathresholdof2.66(uncorrected).Sinceinalltheseregionsonegroupwasdifferentthantheothertwo,taskgroupinteractionscomparedonegrouptotheaverageoftheothertwogroups.Table3reportstheresultsforthegroupingthatyieldedthesignicantinteraction,suchasYoungTABLE2BehavioralData YoungOld-HighOld-LowMSDMSDMSDAge25.34.168.04.469.93.7ExecutivetasksWCSTtest6.00.06.00.05.61.1FAS46.811.948.311.149.49.4Mentalcontrol(WAIS-R)5.31.35.11.05.31.2Mentalarithmetic13.03.915.32.714.03.9ScreeningtasksRawscoresLogicalMemoryI29.55.932.36.425.45.0PairAssociatesI22.61.121.91.517.32.8PairAssociatesII6.00.06.00.04.61.6CVLT14.11.414.31.69.81.7StandardizedscoresLogicalMemoryI0.380.920.810.990.260.77PairAssociatesI0.810.280.620.380.570.72PairAssociatesII0.340.000.340.001.151.73CVLT0.630.580.700.661.190.70Standardizedmean0.540.330.620.330.790.30ScannedtasksRecall0.750.140.860.070.640.19Source0.730.170.730.160.550.23CVLT,CaliforniaVerbalLearningTest.COMPENSATORYACTIVITYINHIGH-PERFORMINGELDERLY vsOld-High/Old-Low(e.g.,SPMcontrast:2,1,1,1,1).Inbothsteps,theriskoffalse-positiveactiva-tionswasfurtherreducedbysettingaminimumacti-vationsizeof10contiguousvoxels.BehavioralDataMemoryperformanceduringscanningwasconsis-tentwiththescreeningdata.RecallaccuracyandsourcememoryaccuracyadjustedbychanceareshowninTable2.A3(group)2(task)ANOVAyieldedreliablemaineffectsofgroup(0.002)andtask(0.04)andanonsignicantgrouptaskinteraction(1).Themaineffectoftaskre-ectedloweraccuracyforsourcememorythanforre-call.Althoughthegrouptaskinteractionwasnotreliable,therecallsourcedifferencewasnumericallygreaterintheolderadultsgroupsthanintheyoungadultsgroup.Toinvestigatesignicantmaineffectofgroup,FisherPLSDcontrastswereperformedbetweeneachpairofgroups.Thesecontrastsshowedthatper-formanceforYoung(0.03)andOld-High(wassignicantlyhigherthanforOld-Low,whereasthedifferencebetweenYoungandOld-High(0.4)wasnotsignicant.Therefore,consistentwithsubjectse-lection,Old-HighparticipantsperformedaswellasYoungparticipantswhereasOld-Lowparticipantsper-formedreliablybelowYoungparticipants.Givendif-ferencesingroupsize,theresultswerecheckedusingnonparametrictests.AKruskalWallistestconasignicantgroupeffect(0.02),aKolmogorovSmirnovtestconrmedthecriticaldifferencebetweenOld-HighandOld-Lowgroups(0.03),andaWil-coxontestconrmedthataccuracywaslowerforsourcethanforrecallmemory(PETDataPFCregionsshowingsignicanteffectsoftaskandgroupinteractionsarelistedinTable3.Alltheregionsthatshowedasignicanttaskeffectsinoneormoreofthegroups(Young,Old-Low,andOld-Highcolumns)alsoshowedasignicanttaskgroupinter-action(rightmostcolumn).Intherecall-minus-sourcecontrast,Youngadultsshowedsignicantactivationsinleftdorsolateral(Brodmannarea(BA)9)andven-trolateral(BA47/44)regions.Old-LowandOld-Highadultsshowedonlytheleftventrolateralactivation,whichwasweakerinOld-Highparticipantsthanintheothertwogroups.Conrmingtheseimpressions,taskinteractionsindicatedthattheleftdor-solateralactivationwasgreaterinYoungthaninOld-LowandOld-Highparticipantsandthattheleftven-trolateralactivationwasgreaterinYoungandOld-LowparticipantsthaninOld-Highparticipants.Inthesource-minus-recallcontrast,YoungadultsshowedsignicantPFCactivationsinrightdorsolat-eral(9/45)andrightanterior(BA10)regions.Old-LowparticipantsshowedactivationsinthesametworightPFCregions,althoughthedorsolateralactivationwasweakerthaninYoungadultsandtheanterioractiva-tionwasstrongerthaninYoungadults.Old-Highpar-ticipantsalsoshowedstrongeractivitythanYoungadultsinrightanteriorPFC,butthisgroupdidnotshowtherightdorsolateralactivation.Moreimpor-tantly,Old-HighparticipantsshowedaleftanteriorPFCactivationnotshownbyeitherYoungorOld-Lowparticipants.GrouptaskinteractionsconrmedthattherightdorsolateralPFCactivationwasgreaterinYoungthaninOld-LowandOld-Highparticipants,whereastherightanteriorPFCactivationwasgreaterinOld-HighandOld-LowparticipantsthaninYoungparticipants.GrouptaskinteractionsalsoconthecriticalndingthattheleftanterioractivationwascantlygreaterinOld-HighthaninYoungandOld-Lowparticipants.Thus,consistentwiththecompensationhypothesis,onlyOld-Highparticipantsshowedareductioninhemisphericasymmetry,andthisreductionwasfoundduringthemostdemandingmemorytask,thatis,theTABLE3PFCRegionsShowingSignicantTaskEffectsandTaskGroupInteractions PFCregionBATaskeffectsgroupinteractionsYoungOld-LowOld-HighxyzZxyzZxyzZxyzZLeftdorsolateral93044363.9Y2644363.3Leftventrolateral/insula47/4428443.634083.9282883.1Y/OL282244.7Rightdorsolateral9/454420285.54218404.2YOL/OH4214203.5Rightanterior10345083.62456124.3265644.6OL/OHY225644.1Leftanterior10385044.1OH385042.7Y,young;OL,old-low;OH,old-high.CABEZAETAL. sourcememorytask.AsdepictedinFig.2,PFCactivityduringsourcememorywasrightlateralizedinYoungandOld-LowparticipantsbutbilateralinOld-Highparticipants.Old-LowparticipantsshowedstrongeractivitythanYoungparticipantsinrightanteriorPFC,butthelocationandlateralizationofPFCactivitydur-ingsourcememorywerethesameasinYoungadults.Incontrast,Old-HighparticipantsdidnotshowtherightdorsolateralPFCactivationshownbyYoungandOld-Lowparticipantsandshowedinsteadaleftante-riorPFCactivationthatwasnotdisplayedbytheothertwogroups.Thepresentresultsprovidestrongsupportforthecompensationhypothesisofage-relatedhemisphericasymmetryreductions.AsshowninFig.2,PFCactiv-ityduringsourcememorywasrightlateralizedinYoungandOld-LowparticipantsbutbilateralinOld-Highparticipants.Thus,consistentwiththecompen-sationhypothesis,anage-relatedasymmetryreductionwasfoundforthebestperforminggroupduringthemostdemandingtask.ThisndingsuggeststhatOld-HighparticipantsrespondedtotheretrievaldemandsofthesourcememorytaskbyrecruitingbilateralPFCregions.Incontrast,theresultsarenotconsistentwiththededifferentiationhypothesisofage-relatedasym-metryreductions.Ifreducedlateralizationisjustan-otherexampleofthedeleteriouseffectsofagingonthebrain(e.g.,atrophy),thenitshouldhaveoccurredinthegroupofelderlyadultsdisplayingmorepronouncedage-relatedcognitivedecits(Old-Lowparticipants),butitdidnot.Onthecontrary,reducedlateralizationwasfoundinthegroupofelderlythatperformedaswellasyoungadults,suggestingthatitisabeneratherthanadetrimentalchange.Thepresentresultssuggestthatintermsoffunc-tionalcompensationadditionalactivitywithinthesamehemisphereislesseffectivethantherecruitmentofhomologousregionsinthecontralateralhemisphere.Duringsourcememory,bothgroupsofolderadultsshowedgreateractivityinthesamerightanteriorPFCregionrecruitedbyyoungadults.Ifoneassumesthatthisage-relatedincreaseinwithin-hemisphereactiva-tionreectedanattemptoffunctionalcompensation,onewouldhavetoconcludethattheattemptwasun-successfulbecauseOld-LowparticipantsperformedcantlymorepoorlythanYoungadultsinthesourcememorytask.Incontrast,therecruitmentofahomologousregioninthecontralateralhemispherewasonlyshownbyOld-Highparticipants,whoper-formedaswellasYoungadultsinthesourcememorytask.Thus,thepresentresultssuggestthatwhetherage-relatedincreasesinactivityaresuccessfulintermsofcompensationdependsonwhethertheyinvolvethesameregionsengagedbyyoungadultsoradditional FIG.2.PFCactivityduringsourcememorywasrightlateralizedinYoungandOld-LowparticipantsbutbilateralinOld-HighCOMPENSATORYACTIVITYINHIGH-PERFORMINGELDERLY regionsinthecontralateralhemisphere.Onepossibleexplanationisthatadditionalwithin-hemisphereac-tivitydoesnotinvolveanetworkmodication,whereasadditionalcontralateralactivityinvolvestherecruit-mentofanalternativenetwork.Inthoseterms,thepresentresultssuggestthatOld-Lowengagedasimi-larmemorynetworkasYoungparticipantsbutuseditlessefciently,whereasOld-Highparticipantscom-pensatedage-relatedmemorydeclinebyreorganizingmemorynetworks.ThecognitivefunctionsoftheparticularPFCregionsshowingage-relatedchangesinactivationcanbein-ferredfrompreviousfunctionalneuroimagingre-search.Itisgenerallyheldthatepisodicretrievalin-volvestwoprocesses:generateprocessesinwhichcandidateinformationisretrievedandrecognizepro-cessesthatselectfromamongtheretrievedinforma-tion.WehaverecentlyproposedthattheleftPFCismoreinvolvedinproductionprocesses,whereastherightPFCismoreinvolvedinmonitoringprocessesetal.,2001).Accordingly,thepresentresultssuggestthatduringsourcememory,Old-Lowpartici-pantstriedtocompensatememorydecitsbyrecruit-ingadditionalmonitoringprocessesmediatedbyrightPFC,whileOld-Highparticipantsalsotappedaddi-tionalproductionprocessesmediatedbyleftPFC.Theseideasareconsistentwiththepresentbehavioralresultsbecauseadditionalmonitoringprocessesarelesslikelytocounteractinsufcientinformationrecov-erythanareadditionalproductionprocesses.Asfortheage-relateddecreasesindorsolateralPFCactivity(leftBA9duringrecallandrightBA9/45duringrecall)shownbybothgroupsofolderparticipants,theyareconsistentwithevidencethatdorsolateralPFCactivityisparticularlysensitivetoaging(RypmaandEsposito,2000;Rypmaetal.,2001).Age-relatedre-ductionsindorsolateralPFCmayreectworkingmemorydecitsassociatedwithcognitiveaging(RypmaandDEsposito,2000;Zacksetal.,etal.,AlthoughthepresentstudyfocusedonPFCactivity,thereissomeevidencethatage-relatedasymmetryreductionscanoccurbeyondPFC.Forexample,inaPETstudyonfaceencoding(Gradyetal.,2002),age-relatedasymmetryreductionswerefoundnotonlyinPFCbutalsointemporalandparietalregions.Also,inaPETstudyonfaceperception(Gradyetal.,positivecorrelationsbetweentemporoparietalactivityandmemoryperformancewerefoundinthelefthemi-sphereforyoungadultsbutbilaterallyforoldadults.Moreover,inanfMRIstudyoninhibitorycontrol(Niel-etal.,2002),youngadultsshowedamorebilateralactivationpatterninPFCaswellasparietalregions.Thus,severalfunctionalneuroimagingstudiessuggestthatage-relatedasymmetryreductionsmayalsoapplytotemporalandparietalregions.Ifreducedhemisphericasymmetryinoldadultscanhaveacompensatoryfunction,howcouldwetakead-vantageofthismechanisminordertoattenuateage-relatedcognitivedecline?Theanswertothisquestiondependsonwhetherage-relatedasymmetryreductionshaveacognitiveoraneuralorigin(foradiscussion,seeCabeza,2002).Ifage-relatedasymmetryreductionsectachangeincognitivestrategies,thenstrategiesthatleadtobilateralPFCrecruitmentcouldbeidenti-edandtaughttolow-performingolderadults.How-ever,itisunlikelythatage-relatedasymmetryreduc-tionsareprimarilycognitiveinorigin,becausetheyhavebeenobservedforsimpleperceptual(Gradyetal.,1994,2000)andmotor(Calauttietal.,2001)tasksinwhichcognitivestrategiesplaylittleornorole.More-over,thereisevidencethatmanipulationsofcognitivestrategiesmayaffecttheoveralllevelofPFCactivityinolderadultswithoutaffectingage-relatedasymmetryreductions(Loganetal.,2002).Ifage-relatedasymme-tryreductionshaveaneuralorigin,thenagoodunder-standingoftheneuralmechanismsunderlyingthesereductionsmayeventuallyleadtothedevelopmentofdrugsandothertherapies.Regardlessofwhethertheinterventionsaremadeatthecognitiveorneurallevel,functionalneuroimagingtechniquescouldprovidenon-invasivemeasurestoguidetheinterventionprocess.Insummary,duringasourcememorytask,youngadultsandlow-performingolderadultsrecruitedsim-ilarrightPFCregions,whereashigh-performingolderadultsengagedPFCregionsbilaterally.Theseresultssuggestthatlow-performingolderadultsrecruitasim-ilarnetworkofbrainregionsasyoungadultsbutusetheminefciently,whereashigh-performingolderadultscounteractage-relatedneuraldeclinebyreorga-nizingbrainfunctions.ThisresearchwassupportedbyAHFMR(Alberta,Canada),NSERC(Canada),andDukeUniversity.WethankLarsNybergforcommentsonearlierversionsofthisarticle.Babcock,R.L.,Laguna,K.D.,andRoesch,S.C.1997.Acomparisonofthefactorstructureofprocessingspeedforyoungerandolderadults:Testingtheassumptionofmeasurementequivalenceacrossagegroups.Psychol.Agingckman,L.,Almkvist,O.,Andersson,J.,Nordberg,A.,Windblad,B.,Rineck,R.,andLm,B.1997.Brainactivationinyoungandolderadultsduringimplicitandexplicitretrieval.J.Cogn.Baltes,P.B.,andLindenberger,U.1997.Emergenceofapowerfulconnectionbetweensensoryandcognitivefunctionsacrosstheadultlifespan:Anewwindowtothestudyofcognitiveaging?Psychol.AgingBellis,T.J.,Nicol,T.,andKraus,N.2000.Agingaffectshemisphericasymmetryintheneuralrepresentationofspeechsounds.J.Neu-CABEZAETAL. Brion,J.P.,Demeurisse,G.,andCapon,A.1989.Evidenceofcorticalreorganizationinhemipareticpatients.Buckner,R.L.,Corbetta,M.,Schatz,J.,Raichle,M.E.,andPe-tersen,S.E.1996.Preservedspeechabilitiesandcompensationfollowingprefrontaldamage.Proc.Natl.Acad.Sci.USACabeza,R.2001.Functionalneuroimagingofcognitiveaging.InHandbookofFunctionalNeuroimagingofCognition(R.CabezaandA.Kingstone,Eds.),pp.331377.MITPress,Cambridge,MA.Cabeza,R.2002.Hemisphericasymmetryreductioninoldadults:TheHAROLDModel.Psychol.AgingCabeza,R.,Grady,C.L.,Nyberg,L.,McIntosh,A.R.,Tulving,E.,Kapur,S.,Jennings,J.M.,Houle,S.,andCraik,F.I.M.1997.Age-relateddifferencesinneuralactivityduringmemoryencodingandretrieval:Apositronemissiontomographystudy.J.Neurosci.Cabeza,R.,Locantore,J.K.,andAnderson,N.D.2001.Lateraliza-tionofprefrontalcortexactivityduringepisodicmemoryretrieval:Evidencefortheproduction-monitoringhypothesis.J.Cognit.inpress.Cabeza,R.,andNyberg,L.2000.ImagingCognitionII:Anempiricalreviewof275PETandfMRIstudies.J.Cogn.Neurosci.Calautti,C.,Serrati,C.,andBaron,J.-C.2001.Effectsofageonbrainactivationduringauditory-cuedthumb-to-indexopposition:Apositronemissiontomographystudy.Cao,Y.,Vikingstad,E.M.,PaigeGeorge,K.,Johnson,A.F.,andWelch,K.M.A.1999.Corticallanguageactivationinstrokepa-tientsrecoveringfromaphasiawithfunctionalMRI.Christensen,H.,Mackinnon,A.J.,Korten,A.E.,Jorm,A.F.,Hen-derson,A.S.,Jacomb,P.,andRodgers,B.1999.Ananalysisofdiversityinthecognitiveperformanceofelderlycommunitydwell-ers:Individualdifferencesinchangescoresasafunctionofage.Psychol.AgingCicinelli,P.,Traversa,R.,andRossini,P.M.1997.Post-strokereor-ganizationofbrainmotoroutputtothehand:A24monthfol-low-upwithfocalmagnetictranscranialstimulation.cephalogr.Clin.Neurophysiol.Craik,F.I.M.,andSalthouse,T.A.2000.HandbookofAgingandCognitionII.Erlbaum,Mahwah,NJ.Delis,D.C.,Kramer,J.H.,Kaplan,E.,andOber,B.A.1987.CaliforniaVerbalLearningTest:AdultVersionManual.ThePsy-chologicalCorporation,SanAntonio,TX.Demeurisse,G.,andCapon,A.1991.Brainactivationduringalinguistictaskinconductionaphasia.DiPiero,V.,Chollet,F.M.,MacCarthy,P.,Lenzi,G.L.,andFrack-owiak,R.S.1992.Motorrecoveryafteracuteischaemicstroke:Ametabolicstudy.J.Neurol.Neurosurg.PsychiatryDixit,N.K.,Gerton,B.K.,Dohn,P.,Meyer-Lindenberg,A.,andBerman,K.F.2000.Age-relatedchangesinrCBFactivationdur-inganN-Backworkingmemoryparadigmoccurpriortoage50.(Part2):S94.Engelien,A.,Silbersweig,D.,Stern,E.,Huber,W.,Doring,W.,Frith,C.,andFrackowiak,R.S.1995.Thefunctionalanatomyofrecov-eryfromauditoryagnosia.APETstudyofsoundcategorizationinaneurologicalpatientandnormalcontrols.Fletcher,P.C.,andHenson,R.N.A.2001.Frontallobesandhumanmemory:Insightsfromfunctionalneuroimaging.Glisky,E.L.,Polster,M.R.,andRouthieaux,B.C.1995.Doubledissociationbetweenitemandsourcememory.Grady,C.L.,Bernstein,L.J.,Beig,S.,andSiegenthaler,A.L.2002.Theeffectsofencodingstrategyonage-relatedchangesinthefunctionalneuroanatomyoffacememory.Psychol.AgingGrady,C.L.,Maisog,J.M.,Horwitz,B.,Ungerleider,L.G.,Mentis,M.J.,Salerno,J.A.,Pietrini,P.,Wagner,E.,andHaxby,J.V.1994.Age-relatedchangesincorticalbloodowactivationduringvisualprocessingoffacesandlocation.J.Neurosci.Grady,C.L.,McIntosh,A.R.,Horwitz,B.,andRapoport,S.I.2000.Age-relatedchangesintheneuralcorrelatesofdegradedandnon-degradedfaceprocessing.Cogn.Neuropsychol.Honda,M.,Nagamine,T.,Fukuyama,H.,Yonekura,Y.,Kimura,J.,andShibasaki,H.1997.Movement-relatedcorticalpotentialsandregionalcerebralbllodowinpatientswithstrokeaftermotorJ.Neurol.Sci.Kelley,W.M.,Miezin,F.M.,McDermott,K.B.,Buckner,R.L.,Raichle,M.E.,Cohen,N.J.,Ollinger,J.M.,Akbudak,E.,Conturo,T.E.,Snyder,A.Z.,andPetersen,S.E.1998.Hemisphericspecial-izationinhumandorsalfrontalcortexandmedialtemporallobeforverbalandnonverbalmemoryencoding.Li,S.,Lindenberger,U.,andFrensch,P.A.2000.Unifyingcognitiveaging:Fromneuromodulationtorepresentationtocognition.Li,S.-C.,andLindenberger,U.1999.Cross-levelunication:Acom-putationalexplorationofthelinkbetweendeteriorationofneuro-transmittersystemsdedifferentiationofcognitiveabilitiesinoldage.InCognitiveNeuroscienceofMemory(L.-G.NilssonandH.J.Markowitsch,Eds.),pp.103146.Hogrefe&Huber,Seattle.Logan,J.M.,Sanders,A.L.,Snyder,A.Z.,Morris,J.C.,andBuckner,R.L.2002.Under-recruitmentandnonselectiverecruit-ment:Dissociableneuralmechanismsassociatedwithaging.Madden,D.J.,Turkington,T.G.,Provenzale,J.M.,Denny,L.L.,Hawk,T.C.,Gottlob,L.R.,andColeman,R.E.1999.Adultagedifferencesinfunctionalneuroanatomyofverbalrecognitionmem-Hum.BrainMapp.McDermott,K.B.,Buckner,R.L.,Petersen,S.E.,Kelley,W.M.,andSanders,A.L.1999.Set-andcode-specicactivationinfrontalcortex:AnfMRIstudyofencodingandretrievaloffacesandwords.J.Cogn.Neurosci.Mitrushina,M.,andSatz,P.1991.Analysisoflongitudinalcovari-ancestructuresinassessmentofstabilityofcognitivefunctionsinBrainDysfunctionMorcom,A.M.,Good,C.D.,Frackowiak,R.S.,andRugg,M.D.2002.Ageeffectsontheneuralcorrelatesofsuccessfulencoding.inpress.Netz,J.,Lammers,T.,andHomberg,V.1997.Reorganizationofmotoroutputinthenon-affectedhemisphereafterstroke.Nielson,K.A.,Langenecker,S.A.,andGaravan,H.P.2002.Differ-encesinthefunctionalneuroanatomyofinhibitorycontrolacrosstheadultlifespan.Psychol.AgingNyberg,L.,Cabeza,R.,andTulving,E.1996.PETstudiesofencodingandretrieval:TheHERAmodel.PsychonomicBull.Rev.Ohyama,M.,Senda,M.,Kitamura,S.,Ishii,K.,Mishina,M.,andTerashi,A.1996.Roleofthenondominanthemisphereandun-damagedareaduringwordrepetitioninpoststrokeaphasics.Raz,N.2000.Agingofthebrainanditsimpactoncognitiveperfor-mance:Integrationofstructuralandfunctionalndings.InbookofAgingandCognition—II(F.I.M.CraikandT.A.Salt-house,Eds.).Erlbaum,Mahwah,NJ.Reuter-Lorenz,P.,Jonides,J.,Smith,E.S.,Hartley,A.,Miller,A.,Marshuetz,C.,andKoeppe,R.A.2000.Agedifferencesinthefrontallateralizationofverbalandspatialworkingmemoryre-vealedbyPET.J.Cogn.Neurosci.COMPENSATORYACTIVITYINHIGH-PERFORMINGELDERLY Reuter-Lorenz,P.A.,Stanczak,L.,andMiller,A.C.1999.Neuralrecruitmentandcognitiveaging:Twohemispheresarebetterthanone,especiallyasyouage.Psychol.Sci.Rypma,B.,andDEsposito,M.2000.Isolatingtheneuralmecha-nismsofage-relatedchangesinhumanworkingmemory.Rypma,B.,Prabhakaran,V.,Desmond,J.D.,andGabrieli,J.D.E.2001.AgedifferencesinprefrontalcorticalactivityinworkingPsychol.AgingSilvestrini,M.,Cupini,L.M.,Placidi,F.,Diomedi,M.,andBernardi,G.1998.Bilateralhemisphericactivationintheearlyrecoveryofmotorfunctionafterstroke.Silvestrini,M.,Troisi,E.,Matteis,M.,Razzano,C.,andCaltagirone,C.1993.Correlationsofowvelocitychangesduringmentalac-tivityandrecoveryfromaphasiainischemicstroke.Smith,E.E.,andJonides,J.1997.Workingmemory:AviewfromCogn.Psychol.Stebbins,G.T.,Carrillo,M.C.,Dorman,J.,Dirksen,C.,Desond,J.,Turner,D.A.,Bennett,D.A.,Wilson,R.S.,Glover,G.,andGabrieli,D.E.2002.Agingeffectsonmemoryencodinginthefrontallobes.Psychol.AgingTalairach,J.,andTournoux,P.1988.ACo-planarSterotacticAtlasoftheHumanBrain.Thieme,Stuttgart,Germany.Thomas,C.,Altenmuller,E.,Marckmann,G.,Kahrs,J.,andDich-gans,J.1997.Languageprocessinginaphasia:Changesinlater-alizationpatternsduringrecoveryreectcerebralplasticityinElectroencephalogr.Clin.Neurophysiol.Thulborn,K.R.,Carpenter,P.A.,andJust,M.A.1999.Plasticityoflanguage-relatedbrainfunctionduringrecoveryfromstroke.Wechsler,D.1987.WechslerMemoryScaleThePsycho-logicalCorporation,SanAntonio,TX.Weiller,C.,Isensee,C.,Rijntsjes,M.,Huber,W.,Muller,S.,Bier,D.,Dutschka,K.,Woods,R.P.,Noth,J.,andDiener,H.C.1995.RecoveryfromWernickesAphasia:Apositronemissiontomogra-phystudy.Ann.Neurol.Zacks,R.T.,Hasher,L.,andLi,K.Z.H.2000.Humanmemory.InHandbookofAgingandCognitionII.(F.I.M.CraikandT.A.Salthouse,Eds.),Vol.293357.Erlbaum,Mahwah,NJ.CABEZAETAL.