Recognition memory in tree shrew (Tupaia belangeri) after repeated familiarization sessions

Recognition memory in tree shrew (Tupaia belangeri) after repeated familiarization sessions

Abbas Khani

, Gregor Rainer

aVisualCognitionLaboratory,DepartmentofMedicine,CH1700Fribourg,Switzerland

bCenterforCognition,UniversityofFribourg,CH1700Fribourg,Switzerland

Recognitionmemoriesareformedduringperceptualexperienceandallowsubsequentrecognitionof previouslyencounteredobjectsaswellastheirdistinctionfromnovelobjects.Asaconsequence,novel objectsaregenerallyexploredlongerthanfamiliarobjectsbymanyspecies.Thisnoveltypreference hasbeendocumentedinrodentsusingthenovelobjectrecognition(NOR)test,aswellisinprimates includinghumansusingpreferentiallookingtimeparadigms.Here,weexaminenoveltypreferenceusing theNORtaskintreeshrew,asmallanimalspeciesthatisconsideredtobeanintermediarybetween rodentsandprimates.Ourparadigmconsistedofthreephases:arenafamiliarization,objectfamiliariza- tionsessionswithtwoidenticalobjectsinthearenaandfinallyatestsessionfollowinga24-hretention periodwithafamiliarandanovelobjectinthearena.Weemployedtwodifferentobjectfamiliariza- tiondurations:oneandthreesessionsonconsecutivedays.Afterthreeobjectfamiliarizationsessions, treeshrewsexhibitedrobustpreferencefornovelobjectsonthetestday.Thiswasaccompaniedby significantreductioninfamiliarobjectexplorationtime,occurringlargelybetweenthefirstandsec- onddayofobjectfamiliarization.Bycontrast,treeshrewsdidnotshowasignificantpreferenceforthe novelobjectafteraone-sessionobjectfamiliarization.Nonetheless,theyspentsignificantlylesstime exploringthefamiliarobjectonthetestdaycomparedtotheobjectfamiliarizationday,indicatingthat theydidmaintainamemorytraceforthefamiliarobject.Ourstudyrevealeddifferenttimecoursesfor familiarobjecthabituationandemergenceofnoveltypreference,suggestingthatnoveltypreferenceis dependentonwell-consolidatedmemoryofthecompetingfamiliarobject.Takentogether,ourresults demonstraterobustnoveltypreferenceoftreeshrews,ingeneralsimilaritytopreviousfindingsinrodents andprimates.

1. Introduction

Treeshrews(Tupaiabelangeri)aredayactivemammalsfrom Southeast Asia and are considered as intermediates between rodentsandprimates(FuchsandFlügge,2002;Martin,1968;Petry etal.,1984).Geneticinformationisavailable,sincethetreeshrew wasselectedasoneof25mammalsforlowcoveragesequencing bytheBroadInstitute(http://www.broadinstitute.org/mammals).

Theyaresmallslendermammalsroughlythesizeofaratwitha longbushytail,andexhibitagileandquickmovementsinarbo- real environments (Luand Petry, 2003). The small size of the animal combinedwithits closeproximity toprimates froman evolutionary point of view, uniquely places the tree shrew as aninvaluablecandidatefor manybehavioral tests.Nonetheless, thebehavioralstudies donewithtree shrewshavebeen rather

∗Correspondingauthorat:VisualCognitionLaboratory,DepartmentofMedicine, CheminduMusee5,1700Fribourg,Switzerland.Tel.:+41263008689.

E-mailaddress:gregor.rainer@unifr.ch(G.Rainer).

limitedwhencomparedtomorestandardanimalmodelssuchas theratornon-humanprimate.Onelineofworkhasestablished treeshrewsasamodelanimalforsocialstress(Fuchs,2005;Fuchs andFlügge,2002,2003;Koziczetal.,2008;Zambelloetal.,2010).

Otheraspectsoftreeshrewbehaviorsuchaslearningsetforma- tion(Ohtaet al.,1985), visuo-spatiallearning(Ohlet al.,1998;

Ohland Fuchs,1999)andtransitiveinference(Takahashietal., 2008)havealsobeeninvestigated.Ingeneral,thesestudiesshow thattreeshrewshavecompetentcognitiveabilities.Forinstance, allparticipatingtreeshrewswereabletolearnaninferencetask, whereasonlyasubsetofratsachievedcriterionperformanceon thesametask(Takahashietal.,2008).Inaddition,therehasbeen someworkonvisuallybasedbehaviors(CallahanandPetry,2000;

LuandPetry,2003;Petryetal.,1984;Raczkowskietal.,1976),since treeshrews–unlikerodents–haveawelldevelopedvisualsystem andcanbereadilytrainedtoperformvariousvisuallybasedtasks.

Giventhiswell-developedandhighlydifferentiatedvisualsystem, ouraimwastoextendtherepertoireoftreeshrewvisualbehavioral paradigmsbyexaminingtheirbehaviorinanobjectrecognition memorytest.

Published in "%HKDYLRXUDO3URFHVVHV±"

which should be cited to refer to this work.

http://doc.rero.ch

analreadyfamiliarobject.Thisnoveltypreferencenecessitatesan intactrecognitionmemoryofthepreviouslyseenfamiliarobject.

The NOR test hasseveral advantages over othermemory tests includingtheabsenceofanyexternallyappliedreinforcementor rule,makingitparticularlysuitableforassessingrecognitionmem- ory.Accordingtotheintervalbetweenobjectfamiliarizationand testphases,typicallyspanningfromoneminutetoover24h,the taskissuitedforexaminationofbothshorttermmemory(STM, 1–90mininterval)andlongtermmemory(LTM,usually24hinter- val).Generally,rodentsdonotshowsignificantnoveltypreference inthelongterm(24h)memorytestintheNORtask(Ennaceurand Delacour,1988;Okudaetal.,2004;Roncaratietal.,2009;Tellez etal.,2010),withsomenotableexceptions(Botton etal.,2010;

Regeretal.,2009).ThepossiblereasonsforlargelyabsentLTMin rodentsintheNORparadigmhavenotbeendiscussedextensively inpreviousliterature.

Theweaknoveltypreferenceobservedafter24hinrodentsis thoughttobeduetoafailuretomaintainamemoryofthefamiliar objectoverthisperiod.ThispoorLTMprovidesanopportunityfor testingmemoryenhancers(Okudaetal.,2004;Wongetal.,2010;

Roncaratietal.,2009).Bycontrast,rodentsdonotprovideagood modelforassessingdetrimentaleffectsonLTM.Nevertheless,to date,almostallNORtestshavebeendoneinrodentsandthistest isincreasinglyusedinrodentanimalmodelsofvariousdiseases includingAlzheimer’sDisease(Donkinetal.,2010;Grecoet al., 2010;Hillenetal.,2010;Luetal.,2009;Mourietal.,2007;Nishida etal.,2006;Taglialatelaetal.,2009),schizophrenia(Hauseretal., 2009;McLeanetal.,2009;Neilletal.,2010;Ozawaetal.,2006;

Powellet al.,2007; Viganoetal., 2009),and Down’ssyndrome (Lockrowetal.,2010).Theabsenceofanyexternalreinforcement intheNORtestisparticularlyimportantasitmakesitmorecom- parabletomemorytestsinhuman,whicharenormallyconducted withouttheuseofpositiveornegativereinforcerssuchasfoodand electricshock(EnnaceurandDelacour,1988).Additionally,validat- ingtheNORtaskinananimalmodelthatisphylogeneticallycloser tohumanscouldstrengthentheextrapolationofanimalstudiesto ourunderstandingofhumancognitivefunctions,aswellasfacili- tatetranslationalmedicineinmemory-relatedcomplications.

Here,weusedtheNORtaskintreeshrewandshowedrobust noveltypreferencefollowingthree,butnotonedayofobjectfamil- iarization.Inaddition,wefoundthatexplorationtimeforfamiliar objectswasalreadyreducedfollowingasingleobjectfamiliariza- tionsession.Ourstudyprovidesnovelinsightsintothedynamics ofrecognitionmemoryformationinthetreeshrew.

2. Materialsandmethods

2.1. Animals

Adultmaleandfemaletreeshrewswerehousedinlargecages thatwereconnectedthroughatubetoanestingbox.Atotalof13 animals(4males)aged1–5yearsoldwereusedintheexperiments.

Theywerekeptinatemperaturecontrolledroomandmaintained onastandard12-hlight/darkcycle(lightonat06:00).Foodand water wereavailable adlibitum. Allanimalswerenaïve tothe

fieldbox(lengthandwidth;60cm×60cm)coveredbyapyramidal ceilingatthetop.Thecamerawasmountedataheightofabout 70cm.Arenawallsweremadeofdark-grey-paintedwoodandthe arenawasilluminatedwithLEDlightsfixedontheceiling.The arenawasplacedinaquiettestingroomclosetotheanimalroom.

TheobjectstoberememberedwereobjectsmadeofLego(approx.

5cm×10cm×10cm)inexperiments1and2andmetalsoftdrink cans(redwithwhitewritingandlogoonit;approx.6cmdiam- eter×12cmheight)andawhiteplasticcontainerwitharedlid (approx.7cm×7cm×10cm)inexperiment2.

2.3. Procedure

Experiment1.Theexperimentconsistedofthreephases:Arena familiarization;Objectfamiliarization;andTestphase.Inallphases, everyanimalwascarriedinsidethenestingboxtothetestroom whichwasconnectedtothearenausingaflexibletube.Thegate wasopenedandtheanimalsweregiventheopportunitytoenter andexplorethearena.Inallphasesandinallexperiments,the animalswereabletofreelycommutebetweenthearenaandthe nestingboxthroughtheconnecting tube.Thisallowedthemto returntoafamiliarandsafeplaceafterexploration.Attheendof theexperiment,thegatewasclosedwhiletheanimalswereinthe nestingbox,andthenanimalswerecarriedbacktotheircage.Pilot experimentsshowedthattreeshrewsrequirealotoftimetoget familiartothearena.Thearenafamiliarizationtrialswereconsid- eredcompletewhen10minelapsedsincethefirsttimetheanimal hadenteredthearena.

Afterthreedaysoffamiliarizationtotheemptyarena,twoiden- ticalobjectswereincludedinthearenaintheobjectfamiliarization phase.Againanimalscouldcommutefreelybetweenthearenaand thenestingboxandnowexploreobjectsinsidethearena.Every animalwasallowedfiveminutes(startingwhentheanimalfirst enteredthearena)toexploreobjectsinasingleobjectfamiliariza- tionsessionandvideowasrecordedforeveryanimal.Afteralltrials, thearenaandobjectswerecleanedby70%ethanoltoavoidany olfactorytrails.Theanimalsweresubsequentlyreturnedtotheir homecageandaftera24-hretentionperiod,oneoftheobjectswas exchangedwithanovelobject,andtreeshrewswereagainbrought tothearenaandwereallowedtoexplorethefamiliarandnovel objectsduringthetestphase.Familiarandnovelobjectsusedin thisexperimentweremadeofLegoandwerecounterbalancedfor differentanimals.Everyanimalwasallowedfiveminutestonav- igateinthearenaandfreelyexploreobjectswhiletheirbehavior wasrecordedbythecamera.Fig.1showsthenavigationmapofa treeshrewsuperimposedonthearenawithrespectiveobjectsdur- ingobjectfamiliarizationandtestphases.Itshouldbenotedthatall phasesofallexperimentswereconductedinaclosedsystemand theexperimenterdidnothandleanimalsorotherwiseintervene duringtheexperiment.Theexperimenterwasmonitoringanimal behavioronlineonacomputerscreenthroughtheconnectedcam- era.

Experiment 2. This experiment examined how retention is boostedbyrepeatedexposuretofamiliarobjects.Itwasconducted onemonthaftertheendofexperiment1,andwassimilartothe first experiment except that every animal had just one day of

http://doc.rero.ch

Fig.1. Tracksofmovementsofatreeshrewduringobjectfamiliarizationandtestphasessuperimposedontheimageofthearenawithrespectiveobjects.(A)Thetracing oftheanimal’smovementsduringanobjectfamiliarizationphaseshowsasimilaramountofexplorationofbothidenticalobjects.(B)Theanimalspentmoretimenearthe novelobject(top)comparedtothefamiliarobject(bottom).

arenafamiliarizationfollowedbythreesubsequentdaysofobject familiarizationwithonesessionperday.Thetimespentexploring eachobjectandthetotalobjectexplorationtimeoneveryobject familiarizationdayandonthetestdaywerecalculatedforevery animal.Thisexperimentwasrepeatedtwicewithdifferentsetsof objects.Todistinguishbetweenthesetworepetitions,hereinafter wewillrefertothefirstone(softdrinkcanandplasticbottleas objects)asexperiment2aandthesecondoneasexperiment2b (objectsmadeofLego).Therewasathreeweekintervalbetween theexperiments2aand2b.Inadditiontofiveanimalsthattook partinexperiment1,twomoreanimalswereincludedinexperi- ment2a(butoneinexperiment2b).Asinexperiment1,familiar andnovelobjectswerecounterbalancedamonganimalsineach experiment.

Controlexperiment.Wecounterbalancedobjects(asfamiliarand novel)acrossanimalstoprecludea possiblebiastowardsacer- tainobject thatcouldcontaminatetheresults. Asanadditional control,weperformedanotherseparatepreferencecontrolexperi- ment.Fivenewanimalstookpartinthisexperiment.Theprocedure consistedofarenafamiliarizationandtestphasesandtherewasno objectfamiliarizationphase.Afterthreedaysofarenafamiliariza- tion,theanimalsweregivenfiveminutestoenterintothearenaand exploretwodifferentobjectsmadeofLego.Afterthistestphase,the controlexperimentwasconcludedandtheanimalswerereturned totheirhomecages.

2.4. Dataanalysesandstatistics

Videos recorded during the object familiarization and test phaseswereanalyzedofflinetodeterminethetimeeveryanimal exploredeachobject.Pointingofthenoseatanobjectatadis- tanceoflessthanonecmortouchinganobjectwereconsidered criteriaforobjectexploration.Wedeterminedboththetimespent exploringeachobjectandthetotalobjectexplorationtime.Toana- lyzeanimals’noveltypreference,wecomparednovelandfamiliar objectexplorationtimes.Inaddition,adiscriminationindex(DI) wascalculatedasthedifferenceintimeexploringnovelandfamil- iarobjectsdividedbytotalexplorationtime.

DI[%]= (NE−FE) (NE+FE)×100

whereNEand FE refertonovelobjectexploration andfamiliar objectexplorationtimesrespectively.Tohaveanindexforcom- parison,wealsocalculatedahypotheticaldiscriminationindexfor theobjectfamiliarization phaseusingthesameformula.Inthis case,thefamiliar objectthat waslater tobereplacedwiththe novelobjectservedasthehypotheticalnovelobject.Alldataare expressedasmean±SEM.Statisticalanalysisusedone-samplet- tests,student’spairedt-testsandrepeatedmeasuresanalysisof variance(ANOVA)followedbyTukey’sandNewman–Keulspost

hoctestsformultiplecomparisons.Aprobabilitylevellessthan 0.05wasconsideredasstatisticalsignificance.

3. Results

Beforedescribing theresultsof thenovelobject recognition (NOR)tests,wepresenttheresultsofacontrolexperimentcon- ductedtoensurethatanimalshadnointrinsicpreferenceamong pairs of objectsused in theNORtask. Thiscontrol experiment wasconductedinfiveanimalsthatallwerenaïvetoanybehav- ioralexperiment includingtheNORtest.Apairedt-testshowed that animals spent similar amount of time (t_obj1=18.8±5.4s, t_obj2=19.1±5.8s)toexploreeachofthetwodifferentobjectswhen theyhadnofamiliarizationtoeitherofthesetwoobjects(t₄=−0.29, p=0.78).ThissuggeststhattheresultsobservedintheNORtests belowareinfactduetorecognitionmemoryfollowingfamiliariza- tionandnotartifactsresultingfromourchoiceofobjectstimuli.

3.1. Experiment1

Thisexperimentexaminedrecognitionmemorybymeansof novelobjectpreferenceintreeshrewusinga24-hretentionperiod and a single session of exposuretofamiliar objects.We found that animalsspenta similaramountof time (t_obj1=20.2±5.4s, t_obj2=17.1±2.9s)exploringeachofthetwoidenticalobjectsdur- ingtheobjectfamiliarizationphase,demonstratingnopreference foroneoftheobjectsorlocations(pairedt-test;t₄=0.54,p=0.62).

Consideringthetotalexplorationtime asameasureofanimals’

motivationforobjectexploration,animals’totalexplorationdur- ingtheobjectfamiliarizationandtestphaseswascomparedusing a paired t-test. The comparison showed there was no signifi- cant difference in total exploration time (t_total-fam=37.3±6.7s, t_total-test=31.5±6.5s) between object familiarization and test phase(t₄=1.43;p=0.23;Fig.2A).Toexaminewhethertherewas neverthelessanymemorytraceofhavingpreviouslyexploredthe familiarobject,wetested iftherewasasignificantreductionin explorationtimeofthefamiliarobjectbetweentheobjectfamil- iarizationandtestphases.Thus,theaveragetimespentexploring eachofthefamiliarobjectsduringtheobjectfamiliarizationphase (t=18.7±3.3s)wascomparedwiththetimespentexploringthe familiar object during the test phase (t=11.8±2.8s). A paired t-testshowedthattherewasindeedsignificantreductioninfamil- iarobjectexploration(t₄=3.53,p=0.02;Fig.2B).Duringthetest phase, a reduction in motivation toexplore thefamiliar object (i.e.partialhabituation) wasthus accompaniedbyanincreased motivationtoexplore anewlyincludednovelobject,leadingto overall unchanged exploration time as compared tothe object familiarizationphase.Toexaminethedifferencesbetweenexplo- rationofnovel(t_novel=19.7±4.2s)andfamiliar(t_fam=11.8±2.8s) objectsonthetestday,apairedt-testonfamiliarandnovelobjects’

http://doc.rero.ch

Fig.2.Resultsofexperiment1.(A)Totalexplorationonthetestandobjectfamiliarizationdays.Therewasnosignificantdifferencebetweentotalexplorationontheobject familiarizationandtestdays.(B)Familiarobjectexplorationontheobjectfamiliarizationandontestdays.(C)Novelandfamiliarobjectexplorationonthetestday.(D)Novel objectexplorationshowedfairlystronglinearcorrelationtothetotalexplorationonthetestday.EachsymbolinAandBandeachcircleinCcorrespondstooneanimal.Obj.

fam.=objectfamiliarization.

explorationtimesrevealedanon-significanttrendtowardsexplo- ration preference for the novel object during the test phase (t4=−2.71,p=0.053;Fig.2C).

Finally, given the remarkablevariability in total exploration amonganimals,weexaminedhowmuchtimeindividualanimals spentexploringnovelandfamiliarobjectsasafunctionoftotal explorationtime(Fig.2D).Wefoundthatbothnovelandfamiliar object exploration times increased withincreasing total explo- rationtime.Linearregressionshowedthatnovelobjectexploration amountedtoroughly62%oftotalexplorationaftera24hreten- tionperiodfollowingtheobjectfamiliarizationsession(c=0.618, r²=0.915, p=0.0108). Thus, despite the large variation of total explorationtimeamonganimals,aconstantfractionofthistime wasspentexploringthenovelobject.

3.2. Experiment2

Giventherelativelyweak24-hmemoryretentioninthefirst experiment,weproceededtoexaminewhetherrepeatedexposure tofamiliarobjectsduringmultipleobjectfamiliarizationsessions onconsecutivedayscanenhancenoveltypreferenceduringthetest phase.Thisexperimentwasconductedtwice(Experiments2aand 2b)usingtwodifferentsetsofobjectstoruleoutstimulusspecific confounds.Similartoexperiment1,animalsspentasimilaramount of time (Experiment2a: t_obj1=10.2±2.2s, t_obj2=9.7±2.3s;and Experiment 2b: t_obj1=17.4±2.8s, t_obj2=17.1±2.5s) exploring eachofthefamiliarobjectsduringtheobjectfamiliarizationphase (t6=0.18,p=0.86;andt5=0.16,p=0.88forexperiments2aand2b respectively).Animals’totalexplorationduringtheobjectfamil- iarizationandtestphasewascomparedusingrepeatedmeasures

one-wayANOVAsfollowed byNewman–Keuls posthoctest for multiple comparisons. This revealed a significant difference between the total exploration time (t_total-fam1=31.7±7.3s and 52.6±8.0s, t_total-fam2=15.9±3.6s and 25.7±4.6s, t_total-fam3=12±2.9s and 25.2±5.2s, t_total-test=27.0±5.0s and 33.4±5.8sforthreedaysofobjectfamiliarizationandthetestday, F(3,27)=4.88, p=0.012and F(3,23)=9.97, p<0.001,for experi- ments2aand2brespectively;seeFig.3A).Posthoctestsrevealed thattherewasasignificantdifferencebetweenthetotalexploration timeonthefirstdayofobjectfamiliarizationascomparedwiththe total explorationonthe second(p<0.05; p<0.001,respectively for2aand2b)andonthethird(p<0.05;p<0.01,respectivelyfor 2aand2b)daysofobjectfamiliarization.Thisshowsthatoncon- secutivedaysofobjectfamiliarization,animalsshowadiminished motivationforfamiliarobjectexploration.Despiteanincreasefor someanimalsonthetestday,thetotalexplorationtimedidnot significantlydifferbetweenthethirdobjectfamiliarizationdayand thetestday.Thus,theincreasedmotivationforobjectexploration duetotheinclusionofanovelobjectcouldnotcompensateforthe prolongedhabituationdevelopedduringsuccessivedaysofobject familiarization(Fig.3A)Additionally,toexaminethetimecourse ofreductioninexplorationtimeoffamiliarobjectsthroughoutthe objectfamiliarizationphase,theaverageoftimespentexploring a familiar object during the object familiarization phase was comparedtothefamiliarobjectexplorationtimeduringthetest phase.Wefoundthatfamiliarobjectexplorationtime generally decreasedwithrepeatedexposuretothefamiliarobjectonsub- sequentdays.Repeatedmeasuresone-wayANOVAsfollowedby Tukey’sposthoctestformultiplecomparisonsrevealedthatthere wasasignificantreductioninexplorationtimeoffamiliarobjects

http://doc.rero.ch

Fig.3.Resultsofexperiment2(2aand2b).(A)Totalexplorationduringtheobjectfamiliarizationphase(threedays)andtestphase(oneday).(B)Familiarobjectexploration onthreeconsecutivedaysofobjectfamiliarizationandonthetestday.(C)Novelandfamiliarobjectexplorationonthetestday.(D)Novelobjectexplorationshowedastrong linearcorrelationtothetotalexplorationonthetestday.EachsymbolinAandBandeachcircleinCcorrespondstooneanimal.InD,solidcirclesandtrianglescorrespond toexperiment2b.‘first’,‘second’and‘third’refertodaysofobjectfamiliarization.

(F(3, 27)=8.55, p=0.001; F(3, 23)=19.1, p<0.0001, respectively for 2a and 2b; Fig.3B). Post hoc Tukeytests showed that the significantreduction inexplorationtime occurredjustafterthe first dayof object familiarization and there was nosignificant differencebetweendays2and3ofobjectfamiliarizationandthe testdayinboth experiments(orrepetitions;Fig.3B).Thus,the strongestreductioninfamiliarobjectexplorationtimeoccurred betweenthefirstandsecondobjectfamiliarizationdays.

Toexaminenoveltypreferenceonthetestday,pairedt-tests wereemployedonnoveland familiarobjectexploration times.

Wefoundthat animalsspentsignificantly moretime exploring the novel object compared to the familiar object (Experiment 2a:t_fam=4.3±1.1s,t_novel=22.6±4.8s,t₆=−3.718,p<0.01;Exper- iment2b:t_fam=5.7±1.3s,t_novel=27.7±5.1s, t₅=−4.64, p<0.01, respectively for 2a and 2b; Fig. 3C). Thus, after three days of repeatedexposuretofamiliarobjects,treeshrew demonstrated robustrecognitionmemoryrevealedbyrobustpreferencefornovel objectexploration.

Finally, we examined what fractionof thetotal exploration time individual animals spent on novel object exploration. To explore this, noveland familiar object exploration times were plottedagainsttotalexplorationtime(Fig.3D).Linearregression showedthattherewasastrongcorrelationbetweennovelobject andtotalexplorationtimeduringthetestphase(c=0.84,r²=0.95, p<0.0001).Onaverage,animalsthusspentabout84%ofthetotal explorationtimeexploringthenovelobject.Bycontrast,familiar objectexplorationwasonlyweaklycorrelatedwithtotalexplo- rationtime,andanimalsspentonaverageonlyaboutfiveseconds exploringthefamiliarobject.

3.3. Discriminationindex

Sincethediscriminationindexis acommonly usedmeasure fornoveltypreference,wecalculateddiscriminationindexesfor allexperiments.Aone-sample t-test showedthatthere wasno significantdiscriminationbetween twodifferent objectsduring thecontrolexperiment(t4=0.31,p=0.77)intheabsenceofobject familiarization.Inbothexperiments1and2,duringobjectfamil- iarization,thediscriminationindex(familiarvs.hypotheticalnovel object)wasalsonotsignificantlydifferentfromzero(onesample t-tests;p=0.81,p=0.88,p=0.96,respectivelyforexperiments1, 2aand2b;Fig.4).Forexperiment1,aonesamplet-testshowed thatthediscriminationindexbetweenthenoveland thefamil- iarobjectdidnotsignificantlydifferfromzero(p=0.105;Fig.4A).

However,thediscriminationindexcalculatedforexperiments2a and 2brevealed thatanimalsexhibited a significantpreference forthenovelobject (p<0.001;p<0.0001,respectively;Fig.4B).

Thediscriminationindexthusyieldedsimilarresultstotheabove analysesbasedonexplorationtime,providingfurtherevidencefor robustnoveltypreferenceintreeshrewfollowingrepeatedobject familiarizationsessions.

4. Discussion

Thisstudy examinedlong term recognitionmemoryin tree shrewusingthenovelobjectrecognitiontestasamodeltestof spontaneousrecognition memorywithouttraining. The experi- mentsdemonstratedthattreeshrewsexhibitrecognitionmemory asshownbyarobustnoveltypreferencefollowingthree-day,but

http://doc.rero.ch

Fig.4.Discriminationindexinexperiments1and2.Hypotheticaldiscriminationbetweentwoidenticalobjectsfromfamiliarizationphasewasplottedforallexperiments asareferencefordiscriminationbetweennovelandfamiliarobjectsonthetestday.PositivevaluesonYaxisindicatenoveltypreference.(A)Discriminationindexfor experiment1.(B)Discriminationindexforexperiment2.Obj.fam.=objectfamiliarization.

notone-day,familiarizationtothefamiliarobjects.Inaone-day objectfamiliarizationparadigm,treeshrewsshowedsomemod- erateevidenceforrecognitionmemoryformation,asrevealedby significantreductioninexplorationtimeofthefamiliarobjectafter a24-hdelayfollowingobjectfamiliarization.Repeatedexposure tothesamefamiliarobjectshasbeenusedrelativelyinfrequently intheNORparadigm(Saveetal.,1992;Weibleetal.,2009),and thesepreviousexperimentsdidnotspecificallyaddresseffectsof repeatedexposureonmemoryconsolidation.Thissimplemodifica- tionintheprotocolinourexperimentledtoarobustdiscrimination betweenthefamiliarandnovelobject,reflectingaveryvividmem- oryofthefamiliarobject.Contrastingresultsofnoveltypreference inthree-dayandone-dayobjectfamiliarizationparadigmsofthe taskdemonstratesthatrepeatedexposuretofamiliarobjectssignif- icantlyconsolidateslongtermmemory.Thisfindingisincontrast tothesuggestionofGaskinetal.(2010)thatincreasedexposureto familiarobjectsdoesnotleadtoanincreasednoveltypreference.A speciesdifferenceinexploratorybehaviorbetweentreeshrewand rodentsmightaccountforthesedivergentresults.Additionally,a methodologicaldifferencebetweenthestudiesmightalsoplaya role:WhileGaskinetal.repeatedlyusedthesamefamiliarobject, coupledwithdifferentnewobjects,insubsequenttestsessionsto increasefamiliarity,weusedmultipleobjectfamiliarizationdays withthesamefamiliarobjectsforthispurpose.Thus,prolonged objectfamiliarizationinourexperimentscouldefficientlyconsoli- datememoryofthefamiliarobject,whileusingthesamefamiliar objectinsubsequenttestingsessions,asintheGaskinetal.study, mighthavefailedtoconsolidatethismemoryduetointerference fromanewobjectoneverysubsequenttesttrial.

Our experiments showed that while reduction in familiar objectsexploration(i.e.habituation)occursoversuccessivedaysof exposure,thestrongesthabituationtakesplacefollowingthefirst objectfamiliarizationsessioninbothone-dayandthree-dayobject familiarizationparadigmsoflongtermNORtask.Despitethelackof significantnoveltypreferenceintheone-dayobjectfamiliarization paradigmofthetask,significantreductioninfamiliarobjectexplo- rationtimeemergesonthetestdayalreadyafteraone-dayobject familiarizationphase.Onemayarguethatreducedexplorationof thefamiliarobjectisduetothepresenceofanovelobjectrather thanhabituationtothefamiliarobjectperse.However,asignificant habituationtothefamiliarobjectalsooccurredinathree-dayobject familiarizationparadigmfollowingthefirstdayofobjectfamiliar- ization.Sincethissignificanthabituationwasevidentonthesecond dayofobjectfamiliarizationintheabsenceofanovelobject,itfol- lowsthattheprocessofhabituationtofamiliarobjectsdevelops followingaone-sessionexposuretotheseobjectsindependently fromthepresenceofanovelobject.Takentogether,thestrongest habituationtothefamiliarobjectsoccurredfollowingthefirstday ofobjectfamiliarization,whilesubsequentobjectfamiliarization daysresultedinrelativelyweakadditionalhabituation.

InagreementwiththeoriginalworkontheNORtask(Ennaceur and Delacour, 1988), our experiment showed that despite the lackofasignificantdifferenceintotalexplorationintheone-day objectfamiliarizationparadigmofthetask,significanthabituation doesoccurtothefamiliarobject.Whilehabituationeffectswere strongestafterthefirstdayofobjectfamiliarization,robustnov- eltypreferenceonlyoccurredaftermultiplefamiliarizationdays withfamiliarobjects,i.e.significanthabituationtofamiliarobjectis

http://doc.rero.ch

notsufficienttoelicitsignificantnoveltypreference,butastronger degreeofhabituation,whichlieswellabovesignificance,isrequired to produce robust novelty preference. We suggest that during thesecondandthirdobjectfamiliarizationdays,animalsfurther explorethefamiliarobjects,resultinginaconsolidatedrepresen- tationinlong-termmemory.Awell-consolidatedrepresentation ofthefamiliarobjectisaprerequisiteforobservingrobustnovelty preferenceintheNORtask.Itisthuspossiblethattheobjectsare alreadyencodedandstoredinlong-termmemoryduringthefirst visualencounter.However,inorderfortheencodedinformationto beeasilyavailableforcomparisoninthepresenceofanovelobject, theencodedinformationneedstobeconsolidatedbyfurtherfamil- iarizationsessions.Inthecontextofnovelobjectrecognition,the threephenomenaofmemoryconsolidation,habituationandnov- eltypreferencearestronglyinter-relatedandcannotbediscussed independentofeachother.Itshouldbenotedthatthenotionof noveltypreferenceinthistaskinevitablyentailstheexistenceof analreadyfamiliarobjectthathaslostitsnoveltytotheanimal,i.e.

hasbeenhabituated.So,wesuggestthathabituationandnovelty preferenceareindicationsofmemoryandarebehavioralprocesses accompanyingtheformationofrecognitionmemories.

Despiteremarkableindividualdifferencesintotalexploration amongstanimals,weobservedastronglinearcorrelationbetween totalexplorationandnovelobjectexplorationindicatingaroughly constantpercentageofnoveltypreferenceamonganimalsineach experiment. This indicates that individual differences in total explorationinfluence thedegree of novelobject exploration in aproportionatemanner,leavingtheratioofnovelobjectexplo- rationtimetototalexplorationtimeunchanged.Theseindividual differencesmayresultfromvariation ofexploratorymotivation amonganimalsoralternativelyfromdifferencesinthedynamics of object processing. However,regardless of the causeof indi- vidualdifferences,theproportionofnoveltypreferenceremains constant.The strikinglinearity of novelobjectexploration asa functionoftotalexplorationsuggeststhatthedegreeofnovelty preferenceinbothone-dayand three-dayobjectfamiliarization paradigmsoftheexperiment remainsconstantandisnotinflu- enced by the general motivation of animal for exploration or differentialprocessingdynamics.Thissuggeststhatindividualdif- ferences in exploratory motivation among animals result from changesinmotivationalbraincircuitsthatarelargelyindependent fromthecircuitsinvolvedinobjectmemory.

Most previous behavioral studies in tree shrews have used rewardslikesugaredwater,almondorfoodtomotivateanimals fortaskparticipation(CallahanandPetry,2000;Ohletal.,1998;

Petryet al.,1984).To our knowledge, ourstudy is thefirst to demonstratebehavioralpreferenceeffectsintreeshrewbasedon intrinsicmotivationintheabsenceofexternalrewards.Whether sucheffects couldbe demonstratedhad beenquestionabledue tothefast-movingandstress-vulnerablecharacteristicsofthese animalswhenexposedtonewenvironments.Weovercamethis limitationbyusinganarenacloselyresemblingtheanimals’home cage,combinedwiththeuseofnestingboxconnectedtothearena providinganopentwo-waypathwaytothenestingbox.Thissetup providesa situationtotest animals’exploratorybehavior while minimizinganimals’levelofstressatthesametime.

Previous work has shown that rodents exhibit robust nov- eltypreferencefollowingshort-termretentionintervals.However, thereareinconsistentresultsintheliteratureregardinglong-term memoryretentioneffects in rodents: While manystudies have shownalackofsignificantretentioninalongterm(24h)mem- oryNORtask(EnnaceurandDelacour,1988;Okudaetal.,2004;

Roncaratietal.,2009;Tellezetal.,2010),afewrecentstudieshave demonstratedsuchlong-termretentioneffects(Bottonetal.,2010;

Regeretal.,2009).Thisdisparityismostlikelyduetodifferencesin methodologicalnuanceslikethedegreeofobjectfamiliarization,

themannerofhandlinganimalsortheanimals’environmentand theirlevelofstress.However,mostofthestudiesofthiskinddid notdiscussthereasonsaccountingforthelackofLTMintheNOR taskandfocusedontheeffectsofcertaintreatmentsonmemory.

Notehowever,thattheabovestudiesinrodentsgenerallyused asingleobjectfamiliarizationsession.Anexaminationofrodent behaviorintheNORtaskusingmultipleobjectfamiliarizationses- sionsisrequiredtoaddresswhethertreeshrewsareindeedmore effectiveintermsofencodingmemoriesforobjectsthanrodents,as ourfindingssuggest.Nonetheless,animmediateadvantageoftree shrewoverrodentsisitsphylogeneticclosenesstohumansdespite havingalmostthesamesizeastherat.So,in additiontofilling thegapbetweenrodentsandprimatesforcomparativestudies,the treeshrewisincreasinglyputforwardasapotentialcandidateasa modelforstudiesofbothbraindisordersandbasicunderstanding ofbrainfunction(Caoetal.,2003).

The visual paired comparison or preferential looking tasks are closely related to the NOR paradigm, and have been used to test recognition memory in human infants and non-human primates.Usingthesetasks,noveltypreferencehasbeendemon- stratedfollowing short-termretention periods(Bachevalier and Nemanic,2008;PascalisanddeSchonen,1994;JutrasandBuffalo, 2010;Zeameretal.,2010)inbothhumaninfantsandmonkeys.

A number ofstudies have alsodocumented novelty preference followinglong-termretention intervals(Gundersonand Swartz, 1985;Pascalisetal.,1998).Inaddition,theeffectoffamiliarization timeandre-familiarizationonnoveltypreferencehasbeenstud- iedinhumanandnon-humanprimates(GundersonandSwartz, 1986; Monk et al., 1996; Morgan and Hayne, 2006).Available evidencesuggeststhatgenerallyincreasedfamiliarizationandre- familiarizationresultinincreasednoveltypreference.Inparticular, Monket al.(1996)showed thatonly infantmonkeys,who had are-familiarizationtrialprecedingthetesttrialinvisualpaired- comparisontestexhibitednoveltyresponses.Thissimilarityinthe effectofre-familiarizationonvisualrecognitionmemorybetween treeshrewandmonkeyinfantsisconsistentwiththeclosephyloge- neticrelationshipbetweenthesespeciesandsuggeststreeshrews asattractivemodelanimalsforfurtherrecognitionmemoryinves- tigations.

Acknowledgements

ThisworkwassupportedbyEURYIgrantPE0033-117106and bytheUniversityofFribourg.TheauthorswouldliketothankP.

LavanexandE.Fuchsforhelpfulcommentsontheearlierversions ofthismanuscript,aswellasR.Kretzforhelpfuladvice.

References

Aubele,T.,Kaufman,R.,Montalmant,F.,Kritzer,M.F.,2008.Effectsofgonadectomy andhormonereplacementonaspontaneousnovelobjectrecognitiontaskin adultmalerats.Horm.Behav.54,244–252.

Bachevalier,J.,Nemanic,S.,2008.Memoryforspatiallocationandobject-place associationsaredifferentlyprocessedbythehippocampalformation,parahip- pocampalareasTH/TFandperirhinalcortex.Hippocampus18,64–80.

Botton,P.H.,Costa,M.S.,Ardais,A.P.,Mioranzza,S.,Souza,D.O.,daRocha,J.B.,Por- ciúncula,L.O.,2010.Caffeinepreventsdisruptionofmemoryconsolidationin theinhibitoryavoidanceandnovelobjectrecognitiontasksbyscopolaminein adultmice.Behav.BrainRes.214,254–259.

Callahan,T.L.,Petry,H.M.,2000.Psychophysicalmeasurementoftemporalmodula- tionsensitivityinthetreeshrew(Tupaiabelangeri).VisionRes.40,455–458.

Cao,J.,Yang,E.B.,Su,J.J.,Li,Y.,Chow,P.,2003.Thetreeshrews:adjunctsandalter- nativestoprimatesasmodelsforbiomedicalresearch.J.Med.Primatol.32, 123–130.

Donkin,J.J.,Stukas,S.,Hirsch-Reinshagen,V.,Namjoshi,D.,Wilkinson,A.,May,S., Chan,J.,Fan,J.,Collins,J.,Wellington,C.L.,2010.ATP-bindingcassettetransporter A1mediatesthebeneficialeffectsoftheliver-X-receptoragonistGW3965on objectrecognitionmemoryandamyloidburdeninAPP/PS1mice.J.Biol.Chem.

285,34144–34154.

Ennaceur,A.,Delacour,J.,1988.Anewone-trialtestforneurobiologicalstudiesof memoryinrats1:behavioraldata.Behav.BrainRes.31,47–59.

http://doc.rero.ch

Tezapsidis,N.,Casadesus,G.,2010.Leptinreducespathologyandimproves memoryinatransgenicmousemodelofAlzheimer’sdisease.J.AlzheimersDis.

19,1155–1167.

Gunderson,V.M.,Swartz,K.B.,1985.Visualrecognitionininfantpigtailedmacaques aftera24-hourdelay.Am.J.Prim.8,259–264.

Gunderson,V.M.,Swartz,K.B.,1986.Effectsoffamiliarizationtimeonvisualrecog- nitionmemoryininfantpigtailedmacaques(Macacanemestrina).Dev.Psychol 22,477–480.

Hauser,T.A.,Kucinski, A.,Jordan, K.G.,Gatto,G.J.,Wersinger,S.R., Hesse,R.A., Stachowiak,E.K.,Stachowiak,M.K.,Papke,R.L.,Lippiello,P.M.,Bencherif,M., 2009.TC-5619:analpha7neuronalnicotinicreceptor-selectiveagonistthat demonstratesefficacyinanimalmodelsofthepositiveandnegativesymp- tomsandcognitive dysfunctionofschizophrenia.Biochem.Pharmacol.78, 803–812.

Hillen,H.,Barghorn,S.,Striebinger,A.,Labkovsky,B.,Müller,R.,Nimmrich,V.,Nolte, M.W.,Perez-Cruz,C.,vanderAuwera,I.,vanLeuven,F.,vanGaalen,M.,Bespalov, A.Y.,Schoemaker,H.,Sullivan,J.P.,Ebert,U.,2010.Generationandtherapeu- ticefficacyofhighlyoligomer-specificbeta-amyloidantibodies.J.Neurosci.30, 10369–10379.

Jurdak,N.,Kanarek,R.B.,2009.Sucrose-inducedobesityimpairsnovelobjectrecog- nitionlearninginyoungrats.Physiol.Behav.96,1–5.

Jutras,M.J.,Buffalo,E.A.,2010.Recognitionmemorysignalsinthemacaquehip- pocampus.Proc.Natl.Acad.Sci.U.S.A.107,401–406.

Kozicz,T.,Bordewin,L.A.,Czéh,B.,Fuchs,E.,Roubos,E.W.,2008.Chronicpsychoso- cialstressaffectscorticotropin-releasingfactorintheparaventricularnucleus andcentralextendedamygdalaaswellasurocortin1inthenon-preganglionic Edinger-Westphalnucleusofthetreeshrew.Psychoneuroendocrinology33, 741–754.

Lockrow,J.,Boger,H.,Bimonte-Nelson,H.,Granholm,A.C.,2010.Effectsoflong- termmemantineonmemoryandneuropathologyinTs65Dnmice,amodelfor Downsyndrome.Behav.BrainRes.,http://dx.doi.org/10.1016/j.bbr.2010.03.036, 251658240.

Lu,H.D.,Petry,H.M.,2003.Temporalmodulationsensitivityoftreeshrewretinal ganglioncells.Vis.Neurosci.20,363–372.

Lu,P.,Mamiya,T.,Lu,L.L.,Mouri,A.,Zou,L.,Nagai,T.,Hiramatsu,M.,Ikejima,T., Nabeshima,T.,2009.Silibininpreventsamyloidbetapeptide-inducedmemory impairmentandoxidativestressinmice.Br.J.Pharmacol.157,1270–1277.

Martin,R.D.,1968.Towardsanewdefinitionofprimates.Man3,377–401.

McLean,S.L.,Idris,N.F.,Woolley,M.L.,Neill,J.C.,2009.D(1)-likereceptoractiva- tionimprovesPCP-inducedcognitivedeficitsinanimalmodels:Implicationsfor mechanismsofimprovedcognitivefunctioninschizophrenia.Eur.Neuropsy- chopharmacol.19,440–450.

Monk,C.S.,Gunderson,V.M.,Grant,K.S.,Mechling,J.L.,1996.Ademonstrationofthe memorysavingseffectininfantmonkeys.Dev.Psychol.32,1051–1055.

Morgan,K.,Hayne,H.,2006.Theeffectofencodingtimeonretentionbyinfantsand youngchildren.InfantBehav.Dev.29,599–602.

Mouri,A.,Noda,Y.,Hara,H.,Mizoguchi,H.,Tabira,T.,Nabeshima,T.,2007.Oral vaccinationwithaviralvectorcontainingAbetacDNAattenuatesage-related Abetaaccumulationandmemorydeficitswithoutcausinginflammationina mouseAlzheimermodel.FASEBJ.21,2135–2148.

Neill,J.C.,Barnes,S.,Cook,S.,Grayson,B.,Idris,N.F.,McLean,S.L.,Snigdha,S., Rajagopal,L.,Harte,M.K.,2010.Animalmodelsofcognitivedysfunctionand

nitionmemoryrequiretraining-associatedemotionalarousal.Proc.Natl.Acad.

Sci.U.S.A.101,853–858.

Ozawa,K.,Hashimoto,K.,Kishimoto,T.,Shimizu,E.,Ishikura,H.,Iyo,M.,2006.

Immuneactivationduringpregnancyinmiceleadstodopaminergichyperfunc- tionandcognitiveimpairmentintheoffspring:aneurodevelopmentalanimal modelofschizophrenia.Biol.Psychiatry59,546–554.

Pascalis,O.,deHaan,M.,Nelson,C.A.,deSchonen,S.,1998.Long-termrecognition memoryforfacesassessedbyvisualpairedcomparisonin3-and6-month-old infants.J.Exp.Psychol.Learn.Mem.Cogn.24,249–260.

Pascalis,O.,deSchonen,S.,1994.Recognitionmemoryin3-to4-day-oldhuman neonates.Neuroreport5,1721–1724.

Petry,H.M.,Fox,R.,Casagrande,V.A.,1984.Spatialcontrastsensitivityofthetree shrew.VisionRes.24,1037–1042.

Powell,K.J.,Hori,S.E.,Leslie,R.,Andrieux,A.,Schellinck,H.,Thorne,M.,Robertson, G.S.,2007.CognitiveimpairmentsintheSTOPnullmousemodelofschizophre- nia.Behav.Neurosci.121,826–835.

Raczkowski,D.,Casagrande,V.A.,Diamond,I.T.,1976.Visualneglectinthetree shrewafterinterruptionofthedescendingprojectionsofthedeepsuperior colliculus.Exp.Neurol.50,14–29.

Reger,M.L.,Hovda,D.A.,Giza,C.C.,2009.Ontogenyofratrecognitionmemorymea- suredbythenovelobjectrecognitiontask.Dev.Psychobiol.51,672–678.

Roncarati,R.,Scali,C.,Comery,T.A.,Grauer,S.M.,Aschmi,S.,Bothmann,H.,Jow,B., Kowal,D.,Gianfriddo,M.,Kelley,C.,Zanelli,U.,Ghiron,C.,Haydar,S.,Dunlop, J.,Terstappen,G.C.,2009.Procognitiveandneuroprotectiveactivityofanovel alpha7nicotinicacetylcholinereceptoragonistfortreatmentofneurodegener- ativeandcognitivedisorders.J.Pharmacol.Exp.Ther.329,459–468.

Save,E.,Poucet,B.,Foreman,N.,Buhot,M.C.,1992.Objectexplorationandreactions tospatialandnonspatialchangesinhoodedratsfollowingdamagetoparietal cortexorhippocampalformation.Behav.Neurosci.106,447–456.

Taglialatela,G.,Hogan,D.,Zhang,W.R.,Dineley,K.T.,2009.Intermediate-andlong- termrecognition memorydeficitsinTg2576micearereversedwithacute calcineurininhibition.Behav.BrainRes.200,95–99.

Takahashi,M.,Ushitani,T.,Fujita,K.,2008.Inferencebasedontransitiverelationin treeshrews(Tupaiabelangeri)andrats(Ratusnorvegicus)onaspatialdiscrim- inationtask.Psychol.Rec.58,215–227.

Tellez,R.,Rocha,L.,Castillo,C.,Meneses,A.,2010.Autoradiographicstudyofsero- tonintransporterduringmemoryformation.Behav.BrainRes.212,12–26.

Vigano,D.,Guidali,C.,Petrosino,S.,Realini,N.,Rubino,T.,DiMarzo,V.,Parolaro, D.,2009.Involvementoftheendocannabinoidsysteminphencyclidine-induced cognitivedeficitsmodellingschizophrenia.Int.J.Neuropharmacol.12,599–614.

Weible,A.P.,Rowland,D.C.,Pang,R.,Kentros,C.,2009.Neuralcorrelatesofnovel objectandnovellocationrecognitionbehaviorinthemouseanteriorcingulate cortex.J.Neurophysiol.102,2055–2068.

Wong,F.K.,Lee,S.H.,Atcha,Z.,Ong,A.B.,Pemberton,D.J.,Chen,W.S.,2010.Rasagiline improveslearningandmemoryinyounghealthyrats.Behav.Pharmacol.21, 278–282.

Zambello,E.,Fuchs,E.,Abumaria,N.,Rygula,R.,Domenici,E.,Caberlotto,L.,2010.

ChronicpsychosocialstressaltersNPYsystem:differenteffectsinratandtree shrew.Prog.Neuropsychopharmacol.Biol.Psychiatry34,122–130.

Zeamer,A.,Heuer,E., Bachevalier,J.,2010. Developmentaltrajectoryofobject recognitionmemoryininfantrhesusmacaqueswithandwithoutneonatalhip- pocampallesions.J.Neurosci.30,9157–9165.

http://doc.rero.ch