Optimallearningmethodsoftendifferbasedontaskgoals,whetheritbeenhancinggeneralizedknowledgethat canbeflexiblyappliedtonewcontextsorrememberingspecificdetailsofasingleevent.Researchhas identifieddifferentneuralcodesthatsupportthesetwodifferenttypesoflearning:integratedrepresentations storedinanteriorhippocampus(aHPC)andmedialprefrontalcortex(mPFC)promoteknowledgethatis generalizable,whereaspatternseparatedrepresentationsstoredinposteriorhippocampus(pHPC)andlateral prefrontalcortex(lPFC)promotedescriminationandspecificity.However,thereisagenerallackof understandingofhowintegratedandseparatedneuralrepresentationsmaycompeteorinteract,andhow differentlearningexperiencescanshapetheformationandretentionofdifferentneuralrepresentationstodrive subsequentbehaviors.Thus,theoverarchinggoalofthisproposalistoidentifythemechanismsandtraining conditionsthatpromotethesuccessfulacquisitionandretentionofdifferenttypesofknowledge.Weproposea neuro-computationalapproachtoexaminecategorylearning,whichallowsustosimulteneouslymeasureand assesstrade-offsassociatedwiththeacquisitionandretentionofgeneralized(category-level)anddetailed (exemplar-level)knowledgeasafunctionofdifferentlearningexperiences.
In Aim1, wewillusemodel-based fMRIanalysestocharacterizehowattentionbiasesformationofneuralrepresentationsinaHPCvia interactionswithmPFCtoimprovegeneralizationatthecostofspecificity.
In Aim2, wewillmanipulatestimulus presentationordertodeterminehowdifferentlearningexperiencescanbiasindividualstoformgeneral (integrated)vs.specific(pattern-separated)knowledgerepresentationsinHPCandPFCusingrepresentational similarityanalysis(RSA)techniques.
In Aim3, wewilluseconnectivitymeasuresandRSAtoidentifyneural patternswithintheHPC-PFCcircuitthatpredictsuccessfulconsolidationofgeneralvs.specificinformationand determinetheconditionsassociatedwithsuccessfullong-termretentionofknowledge.Findingsfromthese proposedstudiescanhelpuncoverwaystoimprovetrainingparadigmsandtargetlearninginterventionsto individualizeandoptimizeknowledgeretention.ThisresearchisparticularlyrelevanttopatientswithAlzeimer?s disease,epilepsy,anddementia,whoexperienceprofounddeficitsinlearningandmemoryretention.
Theproposedresearchexamineshowneuralrepresentationsareshapedduringdifferentlearningexperiences topromotelearningandmemoryretentionthroughdistinctmechamisms.Indoingso,thisresearchwillprovide importantinsightsintotargetinglearninginterventionsthatcanenhancetheacquisitionandretentionof knowledgeinvariousclinicalpopulations.Findingsfromthisresearchwillprovideimportantinsightsinto helpingindividualssufferingfromneurologicaldisorderssuchasAlzheimer?sdisease,epilepsy,andstroke- induceddementiascompensateforassociatedlearningandmemorydeficitsbyoptimzingtrainingconditions thatpromotetheformationofneuralrepresentationsassociatedwithlong-termretention.