Time-place learning in the zebrafish (Danio rerio)

(1)

ContentslistsavailableatScienceDirect

Behavioural

Processes

j ou rn a l h o m epa ge : w w w . e l s e v i e r . c o m / l o c a t e / b e h a v p r o c

Time-place

learning

in

the

zebraﬁsh

(Danio

rerio)

Clarissa

de

Almeida

Moura,

Ana

Carolina

Luchiari

∗

DepartamentodeFisiologia,CentrodeBiociências,UniversidadeFederaldoRioGrandedoNorte,Natal,RN,Brazil

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received3December2015 Receivedinrevisedform4April2016 Accepted11April2016

Availableonline14April2016 Keywords: Learning Socialstimulus Conspeciﬁcs Time-placelearning

a

b

s

t

r

a

c

t

Animalsexhibitactivitycyclesthatrepeatoverdays.Themostnoteworthycyclicalbehaviorsarerelated toforraging,whichgenerallyoccuratthesametimesandlocations.Thesynchronizationofanimal activ-itiesviatheassociationofdifferentplacesatdifferenttimesfortheoccurrenceofrelevantbiological eventsisknownastime-placelearning(TPL).Inthepresentstudy,weusedzebraﬁsh(Daniorerio)totest time-placelearningbasedonadifferentstimulus:socialreinforcement.Fishwerenotonlyableto asso-ciatetimeandplaceofthesocialstimulus,butalsodisplayedanticipatoryactivitypriortothearrivalof thestimulus.Furthermore,weshowthatthegroupofconspeciﬁcsisanrelevantstimulusfortime-place learningtasks,whileotherstudieshaveonlyusedfood.

1. Introduction

Animalsexhibitactivitycyclesthatrepeatoverdays,seasons,

andyears.Amongthemostnoticeablecyclicalbehaviorsare

for-ragingand resting,which generallyappear atestablishedtimes

andplaces(CarrandWilkie,1997;Dunlap,1999).Theprocesses

bywhichtheanimalssynchronizeitsactivitiestooccurat

differ-entlocationsanddifferenttimesisknownastime-placelearning

(TPL)(Wilkie,1995;Widmanetal.,2000).Forthistypeof

learn-ingtooccur,thetimedomainandspatialawarenessarethemost

importantaspects.Moreover,theabilityofTPLcanbeconsidered

adaptive,sinceitprovidesadvantagesforexploitingresourcesthat

arenotcontinuouslyavailable(Enright,1970).

GroundbreakingTPLstudieswereconducted withbees,

sug-gestingtheirabilitytoflexibilizeforagingtofindfoodatspecific

timesandlocations(Wahl,1932;Finke,1958;Koltermann,1974).

Afterbees,TPLwasalsoobservedinotherinsects (Breedetal.,

2002),birds(Wilkieetal.,1996),androdents(Meansetal.,2000;

ThorpeandWilkie,2002).Inﬁsh,thiscognitiveabilityhasreceived

littleattention,andtheavailablestudiesfocusonthetime-place

learningofsocialanimalsasaresponsetofoodresourcesthatvary

overtimeandinspace(Reebs,1993,1996,1999;Gómez-Laplaza

andMorgan,2005,Barretoetal.,2006;Brannas,2014).

Consider-ingthatfoodavailabilitychangesdaily,seasonallyandannually,

itisbelievedthatTPLisfunctionalinoptimizingthelocationand

exploitationoftheresource,aswellasinavoidingpredatorsinan

∗ Correspondingauthor.

E-mailaddress:[email protected](A.C.Luchiari).

environmentofpredictablecyclicalchange,allowingadecreasein

energycostsandconsequentincreaseinsurvival (Mulderetal.,

2013).

Althoughstudiesonﬁshrevealtheabilityoffood-relatedTPL,

learningingroupsofanimalsmayhavebeenmaskedbythecopy

phenomenon, whereby leaders of a group make the decisions

whileother individualsfollow theirbehavioral patterns(Reebs,

2000).Indeed,forsocialanimals,groupinghasadvantagesinterms

oflocatingfoodresources,foragingtime andprotectionagainst

predators(Pitcheretal.,1982;KrauseandRuxton,2002;Brown

andLaland,2003;McRobert,2004;LuchiariandFreire,2009).In

addition,thepossibilityofgroupingisunderstoodasareward(

Al-ImariandGerlai,2008),promotingasensationofwell-beingand

increasingdopaminesecretioninthebrain(Saifetal.,2013).In

thisrespect,conspeciﬁcgroupsmayactasastimulusformarkedly

socialanimals,favoringtime-placelearning.

Therefore,thehighlysocialzebraﬁsh(Engeszeretal.,2007a;

Pritchardet al.,2001; Spenceet al.,2008)canbeconsidered a

studymodelforTPLbasedonsocialreinforcement.Social

behav-iorin zebraﬁshisinnateandappersimmediatelyafterhatching

(Engeszer et al.,2007b).Ifheld isolated,zebraﬁsh shows afast

regroupingwhenplacedtogether(Kerr,1963),indicatingthesocial

grouphasapositiveeffectonmotivationtoshoal.Thus,thisstudy

aimedatassessingwhetherthezebraﬁsh(Daniorerio),whentested

individually,iscapableofassociatingplacewithspeciﬁctimesof

thedayinwhich agroupofconspeciﬁcsispresented.First,the

resultsofthisstudycontributetoincreasingknowledgeregarding

thecognitiveabilityofﬁsh,primarilyTPL,increasingthevarietyof

ﬁshtestedforthisphenomenon,aswellasaddingnewtaskstothe

behavioralinvestigationofthisspecies.Furthermore,sinceitisa

(2)

Fig.1.Schematicdiagramofthetankusedtotesttime-placelearning.Ontheright, themorningcompartment,ontheleft,theafternooncompartment.Thecentral compartmentallowspassagethroughthetwo8cm-diameterroundwindows,one locatedontheright,theotherontheleft,preventingtheanimalfromseeingallthe compartmentsatthesametime.Thevaluesrefertotheareasofthetankincm.

strongstimulusforthespecies,thesocialstimulususedindicates

thatothereventsrelevantforthespeciescanbeconsideredduring

learning,inadditiontothosecloselyassociatedwithsurvival(for

example,food).

2. Materialandmethods

2.1. Animalsandprocedures

ZebraﬁshD.rerio(Hamilton,1822)wereobtainedfromaﬁsh

farm(Natal,RioGrandedoNortestate)andkeptinstockingtanks

(2ﬁsh/L)withairedandﬁlteredwater.Four50Ltanksmakeupone

stockingunitintheclosedwatercirculationsystem,with

mechani-cal,biologicalandchemicalﬁltration,inadditiontoUVdisinfection.

Waterwasmaintainedat28±1◦C,withpH7.2andlowlevelsof

ammoniumandnitrite.Thelight cycle(ﬂuorescentlight,150lx)

wasﬁxedatlight-dark(LD),withthestartofthelightphaseat7

am.Theﬁshwerefedcommercialpelletstwiceaday(38%protein,

4%lipids,NutricomPet)andArtemiasalina.

Tenzebraﬁsh(adultsofbothsexes)fromtheaforementioned

stockwereusedtotesttime-placelearning.Alltheprocedureswith

theanimalswereauthorizedbytheAnimalEthicsCommitteeof

UniversidadeFederaldoRioGrandedoNorte(CEUA039/2015).

2.2. Experimentaldesign

Theexperimentalanimalswereindividuallytransferredtotest

tanks (100×25×25cm; length×width×height), divided

hori-zontally into three same-size compartments (33cm long): one

centralandtwolateral(Fig.1).Thecompartmentswereseparated

byopaquedividers,eachwithan8cm-diametercircularpassage

thatallowedtheﬁshtoswimbetweenthecompartments.The

pas-sagewaslocatedontherightoftherightsidedividerandonthe

leftoftheleftsidedivider,suchthattheﬁshcouldnotvisualize

morethantwocompartmentsatthesametime(Fig.1),thereby

pre-ventingthestimulusplacedinoneofthesidecompartmentsfrom

beingseenwhentheanimalwasintheoppositesidecompartment.

Acylindricalopen-frontreceptacle(10cmindiameterand10cm

high)wasﬁxedtotheupperpartofthewall,andusedtoofferthe

stimulus(conspeciﬁcgroup)atspeciﬁctimes.Theside

compart-mentsweredenominatedmorning compartment andafternoon

compartment.Thetankswerepositionedontwomultiple-layers

shelf,placed oneinfront oftheother,in awaythatthe

morn-ingcompartmentofsometankswereequivalenttotheafternoon

compartmentofsomeothertanks,inawaythatthepositionofthe

tanksdidnotallowforsidebiastotheﬁsh.Eachtankwasconstantly

aeratedthroughanexternalﬁlter(JEBO50,250L/h)locatedinthe

centralcompartmentandporousrocksinthesidecompartments.

Animalswerekeptfor5daysinexperimentaltanksfor

acclima-tizationandforafurther30daysoftheexperimentalphase.A12h

light-dark(LD)cycle(7am–7pm)wasused.Agroupof5zebraﬁsh

(samesizeandage)wereintroducedeverydayintothe

recepta-clelocatedinthemorningcompartmentat8amandremovedat

9am,andintothereceptacleoftheafternooncompartmentat5

pmandremovedat6pm,actingasastimulusfortheexperimental

ﬁshtooccupythecompartmentwherethegroupwasplaced.The

groupwasintroducedthroughareceptacle(500mL)connectedto

ahandle(2m)sothattheexperimentercouldnotbeseenbythe

animals,whichwereseparatedbyanopaquecurtain.The

stimu-luswasintroduced1haftertheonsetofthemorninglightand2h

beforetheendofthelightphase.Food(artemia)wasoffereddaily

(onceaday)atrandomtimes,alwaysinthecentralcompartment

sotheywouldnotbeassociatedwithanyotherstimulus.Theﬁsh

wasalsocheckedfrombehindthecurtainmanytimesaday,thus

providingcuesoftheexperimenterpresencetotheﬁshthatcould

notbeassociatedtoanyothercues.

Ondays15and30oftheexperimentalperiod,thebehaviorof

theanimalswasrecordedonvideofor1hand15min,startingat

7:45amand4:45pm,inordertoobserveanimalsfor15minbefore

thearrivalofthegroup(stimulus)andduringtheirentirepresence.

Behavioronday15wasrecordedinthepresenceofthegroupto

estimatethestrength ofthisstimulus.However,onday30, the

animalswereﬁlmedwithoutthepresenceofthegroup,inorderto

assesstime-placelearninginthezebraﬁsh.

VideorecordsweremadewithSonyDCR-SX45 DigitalVideo

CameraRecordersplacedinfrontofthetanks.Analysisof

behav-ioralrecordswasconductedusingtheZebTracktrackingprogram,

developedatMatLab.Thefollowingparameterswereassessed:

res-idencetime andentryfrequency inthemorning andafternoon

compartments,averageswimmingspeed,andtotaldistance

trav-elled.Thedistanceofthecentrepointoftheﬁshfromtheprevious

sampleframetothefollowingsampleframeindicatedthedistance

travelledbetweenframes.TheZebTrackcalculatesthetotal

dis-tancetravelledbysummingthedistancesfromthepreviousand

current framesalong thetotalperiod ofrecording. Theaverage

speedtakesintoaccountthedistancetravelledbythetimespent

tocoverthatspeciﬁcdistance,alsocalculatedbytheZebTrack

soft-ware.

2.3. Statisticalanalysis

Behavioraldatafortime spentinthecompartmentsand

fre-quencyof entryinto each compartment werecompared in the

morningandafternoon,ondays15and30,usingthepaired

Stu-dent’st-test.Wealsoassessedspeedandtotaldistancetravelled

inthecompartmentswithrespecttothepresenceofthe

stimu-lus.Thesedatawerecomparedfortheperiodpriortogrouparrival

(15minbefore)andthatrelatedtothepresenceofthisstimulus,on

days15and30.Thesameparameterswerealsocomparedbetween

themorningandafternooninbothwindowsoftime(15minbefore

thegroupand60minwiththegroup).ThepairedStudent’st-test

wasusedforbothcomparisons,inordertodeterminebehavioral

changescausedbythepresence/absenceofthestimulus,andat

dif-ferenttimesoftheday.Wedisregardedalldatafromthecentral

compartment,sinceitwasapassageareaandfeedingsiteat

ran-domtimes.Thus,theanimalcouldvisitthisareatopassfromthe

rightcompartmenttotheleftortosearchforfood.Moreover,we

didnotcompareday15versusday30becausetheexperimental

conditionsweredifferent(presenceoftheshoalvs.absenceofthe

shoal),andtheshoalpresenceisabiasedcondition.

3. Results

3.1. Responsivenessonday15

Duringthe15minbeforethegroupofconspeciﬁcsarrived,there

(3)

com-0 700 1400 2100 2800 3500 Morning Afternoon R esi denc e ti me ( s) Morning compartiment Afternoon compartiment a) Anticipation Morning Afternoon

b) Presence of the group

* * 0 20 40 60 80 100 Morning Afternoon Fr equ enc y o f ent ry

Time of the day c) Anticipation

*

Morning Afternoon

Time of the day d) Presence of the group

*

Fig.2. Residencetime(aandb)andfrequencyofentry(candd)inthemorningandafternooncompartmentsonday15oftheexperimentwithzebrafish(n=10)inaTPL test.Observationsweremadebetween7:45and9:00am,and4:45and6:00pmThefirst15minofobservationindicatetheirabilityofanticipatingthearrivalofthesocial stimulus(aandc).Duringthefollowing60min,thesocialstimulus(groupwith5conspecifics)wasmaintainedintheexperimentaltank(bandd).*indicatesstatistical significance(Student’st-test,p<0.05)betweenthecompartmentscorrespondingtoeachexperimentalperiod.

partmentsinthemorning(Student’st-test,df=9t=1.29p=0.23)or

afternoon(Student’st-test,df=9t=0.72p=0.49)(Fig.2a).The

fre-quencyofentryintothemorningcompartmentwashigherinthe

morning(Student’st-test,df=9t=2.46p<0.001;Fig.2c),butdid

notdifferintheafternoon(Student’st-test,df=9t=0.15p=0.88).

Duringthepresenceofthegroup,residencetimeinthe

morn-ingcompartmentwashigherinthemorning(Student’st-test,df=9

t=5.19p<0.001),andintheafternooncompartmentinthe

after-noon(Student’st-test,df=9t=−3.18p=0.01)(Fig.2b).Weobtained

asimilarresultforfrequencyofentryintothecompartmentsinthe

morning(Student’st-test,df=9t=5.22p<0.001),andafternoon

(Student’st-test,df=9t=−2.93p=0.02)(Fig.2d).

Theaveragespeedrecordedwaslowerinthe60mininwhich

thegroupwaspresent,whencomparedtothe15minbeforetheir

arrival,inboththemorning(Student’st-test,df=9t=8.18p<0.001)

andtheafternoon(Student’st-test,df=9t=2.33p=0.04;Fig.3a).

Thetotaldistancetravelledbytheindividuals wasgreater

dur-ingexposuretothegroup,onlyintheafternoon(Student’st-test,

morningdf=9t=−0.72p=0.49;afternoondf=9t=−2.36p=0.04;

Fig.3b).Whentheparameterswerecomparedbetweenthe

morn-ingandafternoonforeachtimeperiod(15minbeforethegroupand

60minwiththegroup),therewasnosigniﬁcantdifferencein

aver-agespeed(Student’st-test,df=9t=0.17p=0.87;Fig.3a)ortotal

distancetravelled(Student’st-test,df=9t=0.59p=0.57;Fig.3b)

inthe15minbeforethegrouparrived,or duringthe60minof

exposuretothem(Student’st-test,averagespeeddf=9t=−1.65

p=0.14;totaldistancetravelleddf=9t=0.48p=0.64;Fig.3).

3.2. Determinationoflearningonday30

Inthe15minbeforethegroupwasintroducedintothetank,

the ﬁsh remained mostly in the morning compartment in the

morning(Student’st-test,df=9t=6.68p<0.001;Fig.4a),butin

theafternoontherewasnodifferencebetweenthetimespentin

eachcompartment(Student’st-test,df=9t=−1.95p=0.08).The

frequencyofentrywasalsohigherinthemorningcompartment

in themorning (Student’st-test, df=9 t=2.55p=0.03)and the

afternooncompartmentin theafternoon(Student’st-test,df=9

t=−2.18p=0.05)(Fig.4c).

Duringthe60minthatthegroupwasexpectedtoarrive,even

intheabsenceofastimulus,theﬁshremainedforalongertimein

themorningcompartmentinthemorning(Student’st-test,df=9

t=3.41p=0.01)andtheafternooncompartmentintheafternoon

(Student’st-test,df=9t=−3.28p=0.01)(Fig.4b).Thefrequencyof

entryintothecompartmentswashigherinthemorning(Student’s

t-test,df=9t=5.83p<0.001)andafternoon(Student’st-test,df=9

t=−3.73p=0.01)(Fig.4d).

Therewasnosigniﬁcantdifferenceinaveragespeedbetween

the15minprecedingthegroup’sarrivalandthe60minwhereit

waspresent,inboththemorningandafternoon(Student’st-test,

morningdf=9t=−0.31p=0.77,afternoondf=9t=−0.59p=0.57;

Fig.5a).Thetotaldistancetravelledalsodidnotdifferbetweenthe

15minpriortothegroup’sarrivalandthetimeitwasusuallyin

thetank,inthemorning(Student’st-test,df=9t=−0.97p=0.36),

ortheafternoon(Student’st-test,df=9t=0.67p=0.52)(Fig.5b).

Theparametersalsodidnotdifferwhenthemorningandafternoon

werecomparedforeachtimeperiod(15minbeforethearrivalof

thegroupand60minwithoutit),bothduringthe15previous

min-utes(Student’st-test,averagespeeddf=9t=−1.31p=0.22;total

distancetravelleddf=9t=0.09p=0.93),andinthe60minwithout

thepresenceofthegroup(Student’st-test,averagespeedt=−1.64

(4)

Fig.3.Averagespeed(a)andtotaldistancetravelled(b)bythezebraﬁsh(n=10)onday15oftheexperimentaltestforTPL.Theanimalswereobservedfor15minbefore presentationofthesocialstimulus(from7:45to8:00am,andfrom4:45and6:00pm)andforthe60minthatthestimuluswaspresent(from8:00to9:00am,andfrom 5:00to6:00pm).*indicatesstatisticalsigniﬁcance(Student’st-test,p<0.05)betweentheperiodbeforepresentationofthestimulusandduringthepresenceofthesocial stimulus.

Fig.4.Residencetime(aandb)andfrequencyofentry(candd)inthemorningandafternooncompartmentsonday30oftheexperimentwithzebrafish(n=10)inthe TPLtest.Observationsweremadefrom7:45to9:00am,andfrom4:45to6:00pmThefirst15minofobservationindicatetheabilitytoanticipatethearrivalofthesocial stimulus(aandc).Duringthenext60min,thesocialstimuluswasnotintroducedintotheexperimentaltank(bandd).*indicatesstatisticalsignificance(Student’st-test, p<0.05)betweenthecompartmentscorrespondstoeachexperimentalperiod.

4. Discussion

Ourresultsshowthatzebraﬁshnotonlystayedlongerwhere

conspeciﬁcswerepresentatpredeterminedtimes,butalso

exhib-ited anticipatory ability, remaining longer and visiting more

frequentlythecompartmentofthetankwherethegroupof

con-speciﬁcs waspresent,for 15min beforethestimulusitselfwas

offered.Theseresultsarerelevant,sincetheyindicatethatzebraﬁsh

arenotonlyabletoassociateenvironmentalcues(timeandspace)

withimportantevents(socialstimulus),butalsosuggeststhe

abil-itytoanticipaterelevantsigns.

Fewstudieshavediscussedtheabilityoftime-placeassociation

inﬁsh,andtothebestofourknowledge,thereareonly10studiesin

theliteratureontheissue(Reebs,1993,1996,1999;Gómez-Laplaza

andMorgan,2005;Delicioetal.,2006;Barretoetal.,2006;Delicio

andBarreto,2008;HeydarnejadandPurser,2008;Ebrahimietal., 2013;Brannas,2014),mostofwhichusegroupsofﬁshandallof

themusefoodasthelearningstimulus.

Inourstudy,weuseda groupof5zebraﬁsh asa time-place

learning stimulus. In the 15min beforethe group arrived, the

zebraﬁshenteredthecorrectcompartmentmorefrequentlyonly

inthemorning.Althoughthetimespentinthecompartmentsdid

notdifferbetweenthemorningandafternoon(Fig.2a),thegreater

frequencyobservedinthemorning(Fig.2c)indicatesanticipation

ofthestimulus.Anticipatoryactivityhaspreviouslybeen

demon-stratedingroupsofArcticchar(Salvelinusalpinus)whenfacedwith

afoodstimulus(Brannasetal.,2005).Anticipatingatemporalevent

maybeadvantageous,providingbeneﬁtsintermsoffood,

part-ners,orrefugeacquisition(Davisand Bardach,1965;Reebsand

(5)

pos-0 5 10 15 20 Anticipation Test A v er ag e sp ee d ( cm /s)

a)

0 1000 2000 3000 4000 Anticipation Test T o ta l di st anc e tr av ell ed ( cm ) Morning Afternoon

b)

Fig.5.Averagespeed(a)andtotaldistancetravelled(b)bythezebraﬁsh(n=10)onday30oftheTPLexperimentaltest.Theanimalswereobservedfor15minbefore presentationofthesocialstimulus(from7:45to8:00am,andfrom4:45to6:00pm)andduringtheperiodinwhichthestimuluswaspresent,butwithoutthepresenceof socialstimulus(from8:00to9:00am,andfrom5:00to6:00pm).*indicatesstatisticalsigniﬁcance(Student’st-test,p<0.05)betweentheperiodspriortopresentationof thestimulationandthepresenceofthestimulus.

siblybecausea15-day periodmaynot havebeensufﬁcient for

individualstomemorizetheexacttimeofstimuluspresentation.

Moreover,lightmayhavebeenusedtopredictthegroup’sarrival,

favoringtheperceptionofthetimeinthemorningthatthistook

place,sincethestimulusoccurred1haftertheonsetofthelight

phase(7am).Intheafternoon,thestimulusoccurred1hbefore

lightsoff(7pm),suchthattheanimalsmusthavehadmore

dif-ﬁcultyinassociatingthegroup’sarrivalwiththesubsequentlight

cue.

On day 15 of behavioral records, the experimental animals

remainedsigniﬁcantlylongernearthegroupofconspeciﬁcsduring

theexposuretimeinthemorningandafternoon(Fig.2aandb).This

reinforcedtheimportanceofconspecificsforzebrafish,confirming

thesocialbehaviorreportedinotherstudies(Wrightetal.,2003;

McRobert,2004;RuhlandMcRobert,2005;Engeszeretal.,2007a; PaciorekandMcRobert,2012).Withrespecttotheaveragespeed,

itwashigherinthe15minbeforethegroup’sarrival,bothduring

themorningandtheafternoon(Fig.3a),indicatingﬁshdecreased

swimmingspeedandstayedlongerwiththegroupduringboth

presentationtimeperiods.

Studiesontime-placelearningusuallyusedlongtestingtimes

(Reebs,1996;Widmanetal.,2000;Delicioetal.,2006;Ebrahimi

etal.,2013),sincetheyreportonlearningthatinvolvesnotonly

entrainmentoftheinternaloscillator,butalsothedailyroutineof

timeandplaceassociation(Biebach,1989).Thus,after30daysof

theexperimentweobservedthatinthe15minbeforethepresence

ofthestimulus,theanimalsenteredmorefrequently(Fig.4c)and

spentmoretimeinthecorrectcompartments(Fig.4a),inboththe

morningandafternoon.Theseresultscorroboratethehypothesisof

acircadianstrategyinthetime-placelearningprocess.According

toCarrandWilkie(1999),animalsthatusecircadianstrategylearn

thatthetimeoftheeventhasﬁxedperiodicity,whichisassociated

withdifferentstagesofendogenouscircadianphases.Thus,animals

canusethisinformationtopredicttheexacttime thatanevent

mustoccur.Indeed,studieswithratsshowanticipationof

time-placelearninginaT-mazewithfoodreward(DeibelandThorpe,

2013).

Howsomeanimalslearn toassociatetime-placestimuli and

othersdonot,remainslargelyunknown.Manyanimalslivein

envi-ronmentsinwhichfoodresources,sexualpartnersandpredators

varyperiodicallyandpredictablyovertime(Rijnsdorpetal.,1981;

SilverandBittman,1984;Beckeretal.,1993;Wilkieetal.,1996).

Whenindividualspredicteventsandlearntoassociatethemwith

theplaceand time of occurrence,theymay obtainadvantages,

whicharelikelylinkedtotheevolutionofcognitiveandcircadian

systems(Enright,1970;Daan,1981;Aschoff,1989;Reebs,1996;

CarrandWilkie,1999).Evidencethattime-placelearningis

associ-atedwithendogenouscircadianrhythmwasproposedbyVander

Zeeetal.(2008).TheseauthorsshowedthatCry1andCry2

knock-outmicewereunabletolearntime-placeassociation,whilewild

micelearnedinbothcircadiancycleandunderconstantlight.The

biologicalclockincludesanauto-regulatedandoscillating

tempo-rizingsystemwithdeﬁnedanatomicalstructuresandDNAcoded

oscillators(MarquesandMenna-Barreto,2003).Eventhoughonly

fewstudiesapproachthesystemstructureinzebraﬁsh,itisknown

thatthisspeciespresentsabiologicalclock(Cahilletal.,1998).

Onecouldarguethatinsteadoftime-placelearningbasedonthe

circadiantiming,ﬁshmayhaveshownwin-stay,lose-shiftstrategy.

Thispossibility,however,maybediscardedinthepresentstudy

becausewin-stay,lose-shiftstrategyusetooccurwithinashorter

than8hperiodoftimeandtheabsenceoftheshoalinthe

morn-ingcompartmentinthemorningperiodwouldleadtoanincrease

in thetime spent in theafternoon compartment, or,at least a

higherfrequencyofentriesintheafternooncompartment,

indicat-ingtheﬁshwerecheckingtheafternooncompartment.However,

weobservedﬁshstayedlongerinthemorningcompartmentduring

themorningandintheafternooncompartmentduringthe

after-noon(Fig.4bandd).

Therefore,ourresultsregardingthetimezebraﬁshspentonday

30ineachcompartmentasafunctionofthestimulus(Fig.4b),in

whichlearningwastestedwithoutthepresenceofasocial

stimu-lus,showsthetime-placelearningabilityofthisanimal.Thus,we

corroboratedearlier studieswithﬁsh(Reebs, 1993,1996,1999;

Gómez-Laplaza and Morgan, 2005;Barreto et al., 2006; Delicio etal.,2006;DelicioandBarreto,2008;Heydarnejadand Purser, 2008;Ebrahimietal.,2013;Brannas,2014)andfoundthat

time-placelearningabilitymaybeinvolvedwithrelevantstimuliforthe

speciesunderstudy,inthiscase,thesocialsignal.Wedidnotfound

signiﬁcantdifferencesin both averagespeed and totaldistance

travelledbeforeandduringthe60minoftestonthe30◦day,

possi-blyduetheabsenceofthestimulusﬁshdidnotchangeswimming

behaviorpatternonthisprobeday.

Finally,itisimportanttoemphasizethatinthepresentstudy,

thesightofagroupofconspeciﬁcswaseffectiveasastimulusfor

thelearningtaskproposed.Asocialstimuluscanbeeffectivelyused

duetothehighlysocialcharacteristicofzebraﬁsh(Wrightetal.,

2003;Engeszeretal.,2004;Luchiarietal.,2015).Otherstudies

withsocialsignalsasalearningstimulusindicatethevalidityof

thissignal,suchasthestudybyKarnikandGerlai(2012)usinga

conspeciﬁcimageforconditioningofplace,Silveiraetal.(2015)

usingconspeciﬁcsinassociativeconditioningtasks,andLuchiari

etal.(2015)usingagroupofconspeciﬁcsinalatentlearningtask.

Duetothehighsociabilityofzebraﬁsh,futurestudieson

(6)

orfood,maycontributetoabetterunderstandingoftheanimal’s

behaviorandlearningcopystrategies.Furthermore,sinceourstudy

doesnotdiscusstheday-to-day developmentof thetime-place

association,additionalstudiesareneededtofollowthebehavior

patternsthroughouttheentireexperimentalperiod.Otherpoints

forfutureinvestigationthatcouldnotbecoveredinthepresent

studyaretheofferoffoodinadifferentpartofthetank,which

forcedustoeliminatethecentralcompartmentfromTPLanalyses,

andthebaselinebehavioralanalysisinordertocoverthe

distribu-tionoftheﬁshbeforetraining.Despitenotusingthiscompartment

or following theanimal for the 30days of thetest, our results

arerobustinthattheyshowthetime-placeassociationabilityof

zebraﬁsh,asdemonstratedbydataonday30ofthetest(Fig.4).

Inourstudy,D.rerioexhibitedtime-placelearning,aswellas

anticipatedthearrivalofthestimulus.Moreover,weshowedthat

thegroupofconspeciﬁcsisanrevevantstimulusfor time-place

learningtasks,inwhichtherearenorecordsofstudiesusingthis

typeofstimulus.Here,weusedacircadianphotoperiod(LD),but

time-placelearningstudiesintheabsenceoflightzeitgebersare

neededtoinvestigateTPLwithoutlightsignalstoindicatetime.

Studieswithmolecularanalysesmustalsobeconductedtoindicate

theassociationbetweenthebiologicalclockandthistypeoftask,

essentialforunderstandingtheendogenousmechanismsinvolved

inTPL.

Acknowledgements

WethankMsSilveira,V.A.M.andSantos,T.S.forhelpin

collect-ingdataforthisarticle.

References

Al-Imari,L.,Gerlai,R.,2008.Sightofconspeciﬁcsasrewardinassociativelearning inzebraﬁsh(Daniorerio).Behav.BrainRes.189,216–219.

Aschoff,J.,1989.Temporalorientation:circadianclocksinanimalsandhumans. Anim.Behav.37,881–896.

Barreto,R.E.,Rodrigues,P.,Luchiari,A.C.,Delicio,H.C.,2006.Time-placelearningin individuallyrearedangelﬁsh,butnotinpearlcichlid.Behav.Process.73, 367–372.

Becker,P.H.,Frank,D.,Sudmann,S.R.,1993.Temporalandspatialpatternof commontern(Sternahirundo)foragingintheWaddenSea.Oecologia93, 389–393.

Biebach,H.,1989.Time-and-placelearningbygardenwarblers,sylviaborin.Anim. Behav.37,353–360.

Brannas,E.,Berglund,U.,Eriksson,L.O.,2005.Timelearningandanticipatory activityingroupsofArcticcharr.Ethology111,681–692.

Brannas,E.,2014.Time–placelearningandleader–followerrelationshipsinArctic charrSalvelinusalpinus.J.FishBiol.84,133–144.

Breed,M.D.,Stocker,E.M.,Baumgartner,L.K.,Vargas,S.A.,2002.Time–place learningandtheecologyofrecruitmentinastinglessbee,Trigonaamalthea (Hymenoptera,Apidae).Apidologie33,251–258.

Brown,C.,Laland,K.N.,2003.Sociallearninginﬁshes.ARev.Fish.4,280–288. Cahill,G.M.,Hurd,M.W.,Batchelor,M.M.,1998.Circadianrhythmicityinthe

locomotoractivityoflarvalzebraﬁsh.Neuroreport9,3445–3449.

Carr,J.A.R.,Wilkie,D.M.,1997.Ordinal,phase,andintervaltiming.In:Bradshaw, C.M.,Szabadi,E.(Eds.),TimeandBehavior:PsychologicalandNeurobiological Analyses.Elsevier,Amsterdam,pp.265–327.

Carr,J.A.R.,Wilkie,D.M.,1999.Ratsarereluctanttousecircadiantiminginadaily time-placetask.Behav.Process.44,287–299.

Daan,S.,1981.Adaptivedailystrategiesinbehavior.In:Aschoff,J.(Ed.),Biological Rhythms.Verlag,USA,pp.275–298.

Davis,R.E.,Bardach,J.E.,1965.Time-coordinatedprefeedingactivityinﬁsh.Anim. Behav.13,154–162.

Deibel,S.H.,Thorpe,C.M.,2013.Theeffectsofresponsecostandspecies-typical behaviorsonadailytime-placelearningtask.Learn.Behav.41,42–53. Delicio,H.C.,Barreto,R.E.,2008.Time-placelearninginfood-restrictedniletilapia.

Behav.Proc.77,126–130.

Delicio,H.C.,Luchiari,A.C.,Barreto,R.E.,Marcondes,A.L.,2006.Testingtime–place learninginthecichlidﬁshNiletilapia.J.Ethol.24,195–200.

Dunlap,J.C.,1999.Molecularbasesforcircadianclocks.Cell96,271–290. Ebrahimi,E.,Heydarnejad,M.S.,Sattari,M.,Bani,A.,Kasheﬁ,P.,2013.Atestof

time–placelearninginsingleandgroupedgreatsturgeonHusohuso.Acta Ethol.16,77–82.

Engeszer,R.E.,Ryan,M.J.,Parichy,D.M.,2004.Learnedsocialpreferencein zebraﬁsh.Curr.Biol.14,881–884.

Engeszer,R.E.,Alberici,D.A.,Barbiano,L.,Ryan,M.J.,Parichy,D.M.,2007a.Timing andplasticityofshoalingbehaviourinthezebraﬁsh.Daniorerio.Anim.Behav. 74,1269–1275.

Engeszer,R.E.,Patterson,L.B.,Rao,A.A.,Parichy,D.M.,2007b.Zebrafishinthewild: areviewofnaturalhistoryandnewnotesfromthefield.Zebrafish4,21–40. Enright,J.T.,1970.Ecologicalaspectsofendogenousrhythmicity.Rev.Ecol.Syst.1,

221–238.

Finke,I.,1958.ZeitgedächtnisundSonnenorientierungderBienenLehramtsarbeit Naturw.LudwigMaximiliansUniversityMunich,Munich.

Gómez-Laplaza,L.M.,Morgan,E.,2005.Time–placelearninginthecichlid angelﬁsh,Pterophyllumscalare.Behav.Process.70,177–181. Heydarnejad,M.S.,Purser,J.,2008.Speciﬁcindividualsofrainbowtrout

(Oncorhynchusmykiss)areabletoshowtime-placelearning.Turk.J.Biol.32, 209–229.

Karnik,I.,Gerlai,R.,2012.Canzebrafishlearnspatialtasks?Anempiricalanalysis ofplaceandsingleCS-USassociativelearning.Behav.Brain.Res.233,415–421. Kerr,J.P.,1963.Groupingbehaviourofthezebrafishasinfluencedbysocial

isolation.Am.Zool.2,532–533.

Koltermann,R.,1974.Periodicityintheactivityandlearningperformanceofthe honeybeeinexperimentalanalysisofinsectbehavior.In:Browne,B.L.(Ed.), ExperimentalAnalysisofInsectBehaviour.Springer-Verlag,Berlin,pp. 218–227.

Krause,J.,Ruxton,G.D.,2002.LivinginGroups.OxfordUniversityPressOxford. Luchiari,A.C.,Freire,F.A.M.,2009.Effectsofenvironmentalcolourongrowthof

Niletilapia,Oreochromisniloticus(Linnaeus,1758),maintainedindividuallyor ingroups.J.Appl.Ichthyol.2,162–167.

Luchiari,A.C.,Salajan,D.C.,Gerlai,R.,2015.Acuteandchronicalcohol administration:effectsonperformanceofzebraﬁshinalatentlearningtask. Behav.BrainRes.282,76–83.

Marques,N.,Menna-Barreto,L.,2003.In:Marques,N.andMenna-Barreto,L.(Ed). Cronobiologia:PrincípioseAplicac¸ões.SãoPaulo,pp.224.

McRobert,S.P.,2004.Shoalingbehaviorinﬁsh.In:Bekoff,M.(Ed.),The EncyclopediaofAnimalBehavior.PhoenixGreenwood,pp.1012–1016. Means,L.W.,Ginn,S.R.,Arolfo,M.P.,Pence,J.D.,2000.Breakfastinthenookand

dinnerinthediningroom:time-of-daydiscriminationinrats.Behav.Process 49,21–33.

Mulder,C.K.,Gerkema,M.P.,VanderZee,E.A.,2013.Circadianclocksandmemory: time-placelearning.Front.Mol.Neurosci.6,8.

Paciorek,T.,McRobert,S.,2012.Dailyvariationintheshoalingbehaviorof zebraﬁshDaniorerio.Curr.Zool.58,129–137.

Pitcher,T.J.,Magurran,A.E.,Winﬁeld,I.,1982.Fishinalargershoalsﬁndfood faster.Behav.Ecol.Sociobiol.10,149–151.

Pritchard,V.L.,Lawrence,J.,Butlin,R.K.,Krause,J.,2001.Shoalchoiceinzebraﬁsh, Daniorerio:theinﬂuenceofshoalsizeandactivity.Anim.Behav.62, 1085–1088.

Reebs,S.G.,Gallant,B.,1997.Foodanticipatoryactivityasacueforlocal enhancementingoldenshiners(Pisces,Cyprinidae,Notemigonuscrysoleucas). Ethology103,1060–1069.

Reebs,S.G.,1993.Atestoftime–placelearninginacichlidﬁsh.Behav.Process.30, 273–282.

Reebs,S.G.,1996.Time-placelearningingoldenshiners(Pisces:cyprinidae). Behav.Process.36,253–262.

Reebs,S.G.,1999.Time-placelearningbasedonfoodbutnotonpredationriskina ﬁsh,theInangaGalaxiasmaculatus”.Ethology105,361–371.

Reebs,S.G.,2000.Canaminorityofinformedleadersdeterminetheforaging movementsofaﬁshshoal?Anim.Behav.59,403–409.

Rijnsdorp,A.,Daan,S.,Dijkstra,C.,1981.Huntinginthekestrel,Falcotinnunculus, andtheadaptivesigniﬁcanceofdailyhabits.Oecologia50,391–406. Ruhl,N.,McRobert,S.P.,2005.Theeffectofsexandshoalsizeonshoaling

behaviourinDaniorerio.J.FishBiol.67,1318–1326.

Saif,M.,Chatterjee,D.,Buske,C.,Gerlai,R.,2013.Sightofconspeciﬁcimages induceschangesinneurochemistryinzebraﬁsh.Behav.BrainRes.243, 294–299.

Silveira,M.M.,Oliveira,J.J.,Luchiari,A.C.,2015.DuskydamselﬁshStegastesfuscus relationallearning:evidencesfromassociativeandspatialtasks.J.FishBiol.86, 1109–1120.

Silver,R.,Bittman,E.L.,1984.Reproductivemechanisms:interactionofcircadian andintervaltiming.Ann.NYAcad.Sci.423,488–514.

Spence,R.,Gerlach,G.,Lawrence,C.,Smith,C.,2008.Thebehaviourandecologyof thezebraﬁsh:Daniorerio.Biol.Rev.Camb.Philos.Soc.83,13–34.

Thorpe,C.M.,Wilkie,D.M.,2002.Unequalintervaltime–placelearning.Behav. Process.58,157–166.

VanderZee,E.A.,Havekes,R.,Barf,R.P.,Hut,R.A.,Nijholt,I.M.,Jacobs,E.H., Gerkema,M.P.,2008.Circadiantime-placelearninginmicedependsonCry genes.Curr.Biol.18,844–848.

Wahl,O.,1932.Neueuntersuchungenüberdaszeitgedächtnisderbienen.J.Comp. Physiol.16,529–589.

Widman,D.R.,Gordon,D.,Timberlake,W.,2000.Responsecostandtime–place discriminationbyratsinmazetasks.Anim.Learn.Behav.28,298–309. Wilkie,D.M.,Carr,J.A.R.,Siegenthaler,A.,1996.Fieldobservationsoftime–place

behaviourinscavengingbirds.Behav.Process.38,77–88. Wilkie,D.M.,1995.Time-placelearning.NewDir.Psychol.Sci.4,85–89. Wright,D.,Rimmer,L.B.,Pritchard,V.L.,Krause,J.,Butlin,R.K.,2003.Interand

intra-populationvariationinshoalingandboldnessinthezebraﬁsh(Danio rerio).Naturwissenschaften90,374–377.