Different or alike? Female rainbow kribs choose males of similar consistency and dissimilar level of boldness

Article (Accepted Version)

http://sro.sussex.ac.uk

Scherer, Ulrike, Kuhnhardt, Mira and Schuett, Wiebke (2017) Different or alike? Female rainbow kribs choose males of similar consistency and dissimilar level of boldness. Animal Behaviour, 128. pp. 117-124. ISSN 0003-3472

This version is available from Sussex Research Online: http://sro.sussex.ac.uk/id/eprint/78355/

This document is made available in accordance with publisher policies and may differ from the published version or from the version of record. If you wish to cite this item you are advised to consult the publisher’s version. Please see the URL above for details on accessing the published version.

Copyright and reuse: Sussex Research Online is a digital repository of the research output of the University.

Copyright and all moral rights to the version of the paper presented here belong to the individual author(s) and/or other copyright owners. To the extent reasonable and practicable, the material made available in SRO has been checked for eligibility before being made available.

Copies of full text items generally can be reproduced, displayed or performed and given to third parties in any format or medium for personal research or study, educational, or not-for-profit purposes without prior permission or charge, provided that the authors, title and full bibliographic details are credited, a hyperlink and/or URL is given for the original metadata page and the content is not changed in any way.

1

Differentoralike?Femalerainbowkribschoosemalesofsimilarconsistency1

anddis‐similarlevelofboldness2

3

U.Scherer1,M.Kuhnhardt1andW.Schuett14

5

6

1ZoologicalInstitute,BiocentreGrindel,UniversityofHamburg,Martin‐Luther‐KingPlatz3,7

20146Hamburg,Germany8

9

10

Correspondence:11

UlrikeScherer,ZoologicalInstitute,BiocentreGrindel,UniversityofHamburg,Martin‐12

Luther‐KingPlatz3,20146Hamburg,Germany.13

E‐Mail:u.k.scherer@gmail.com14

Phone:+494042838–789415

16

2

17

Althoughtheexistenceofconsistentbetween‐individualdifferencesinbehaviour18

("personalitydifferences")hasbeenwelldocumentedduringthelastdecade,theadaptive19

valueofsuchbehaviourallimitationsstillremainsanopenfieldforresearchersofanimal20

behaviour.Personalitiesclearlyrestrictindividualsintheirabilitytoadjusttheirbehaviour21

todifferentconditions.However,sheercostsofflexibilitycannotexplainthepolymorphism22

createdbypersonalityvariation.Inacorrelativeapproach,weheretestedwhethermate23

choicemightactasamajordrivingforcemaintainingpersonalityvariationinthe24

monogamous,biparentalrainbowkrib,Pelvicachromispulcher.Wepersonality‐typedall25

malesandfemalesfortheirboldness(activityundersimulatedpredationrisk)andallowed26

femalestochoosebetweentwomalesthatdifferedintheirboldness(behaviouralleveland27

consistency).Priortothechoice,femaleswereallowedtoobservebothmales,expressing28

theirnaturalboldnesstowardsavideoanimatednaturalpredator.Bothsexesshowed29

personalitydifferencesinboldnessovertheshort‐andlong‐term.Furthermore,when30

removingside‐biasedfemales,wefoundadis‐assortativematingpreferenceforthe31

behaviourallevelandanassortativepreferenceforbehaviouralconsistencyinboldness.32

Suchpreferencepatternsmightfacilitateeffectiveparentalroleallocationduringoffspring33

careand/orprovidegeneticbenefits.Ourresultssuggestthatsexualselectionplaysan34

importantroleintheevolutionofpersonalitydifferences.35

36

Keywords:anti‐predatorbehaviour,assortative,behaviouralcompatibility,cichlid,mate37

choice,Pelvicachromispulcher,personality,risk‐taking,sexualselection,sidebias38

39

3

Individualshavetocopewithawidearrayofenvironmentalchallenges.Therefore,40

flexibilityintheexpressionofbehaviouralresponsestowardsdifferentandchanging41

conditionsshouldbefavouredbyselection(Sihetal.,2004).Yet,individualsoftenshow42

considerableconsistentbetween‐individualdifferencesinbehaviourovertimeand/or43

contexts(Boissy,1995).Suchpersonalitydifferencesarecommonthroughouttheanimal44

kingdom(reviewedinGosling,2001;Kralj‐Fišeretal.,2014)andhavebeenshownfor45

variousbehaviouraltraits,suchasactivitypattern,aggressiveness,exploratorytendencies,46

boldnessandfearfulness(reviewedinDalletal.,2004;Gosling,2001;Sihetal.,2004).47

Personalitytraitsaremoderatelyheritable(Ariyomo,Carter,etal.,2013;Patricketal.,48

2013;Reifetal.,2003;vanOersetal.,2005)andhavefitnessconsequences(e.g.Ariyomoet49

al.,2012;Dingemanseetal.,2005;Smithetal.,2008),suggestingtheyarenotmerelynon‐50

adaptivenoisethatsurroundsanadaptiveoptimum(Wilson,1998).Nevertheless,51

underlyingmechanismsthatgenerateandmaintainbehaviouralpolymorphismarelargely52

unclearandmanyaspectsofthegrowingbodyoftheoreticalframeworksstillremaintobe53

empiricallytested(reviewedine.g.Schuettetal.,2010;Wolfetal.,2010).54

55

Recently,Schuettetal.(2010)pointedoutthatsexualselectionmaybeimportantin56

generatingandmaintainingpersonalityvariationthoughthispossibilityhasrarelybeen57

tested(butseee.g.Montiglioetal.,2016;Schuettetal.,2011).Accordingtotheproposed58

framework(Schuettetal.,2010),personalitiesareexpectedtoplayanimportantrolein59

matechoicewhenapotentialmate'sbehaviouralphenotypeiseitherassociatedwith60

good/compatiblegenesthatincreaseoffspringfitness(Dingemanseetal.,2004;Ihleetal.,61

2015;Maysetal.,2004)orprovidesnon‐geneticbenefitsincreasingthereproductive62

4

successthroughparentalabilityand/orbehaviouralcompatibilitybetweenmates.While63

matechoiceforgeneticqualityandparentalabilityshouldfavourinter‐individual64

agreementinthepreferenceforabehaviouraltrait,matechoiceforgeneticorbehavioural65

compatibilityshoulddependonaninteractionbetweenmaleandfemale(geno‐or)66

phenotype(Schuettetal.,2010).Thus,matechoiceforcompatibilitywouldleadtointer‐67

individualdifferencesinmatingpreferences,creatingeitheranassortativeordis‐68

assortativematingpattern(Schuettetal.,2010).69

70

Notmanystudiestodatehaveinvestigatedtheeffectofpersonalitytraitsonmatechoice71

(reviewedinSchuettetal.,2010)andsomehaveonlyassessedthebehaviourofthechosen72

butnotthechoosingsex(Godinetal.,1996;Ophiretal.,2003).Thefewstudiesconsidering73

apotentialinterplaybetweenmaleandfemalepersonalityduringmatechoicehaveoften74

foundassortativematechoiceforvariousbehaviouraltraits,incorrelative(Gonzagaetal.,75

2010;Kralj‐Fiseretal.,2013;Mascie‐Tayloretal.,1988;Montiglioetal.,2016)or76

experimentalsettings(Schuettetal.,2011)andanincreasedreproductivesuccessof77

assortativepairs(e.g.Ariyomo&Watt,2013;Schuettetal.,2011).However,instudiesthat78

foundincreasedsuccessofassortativepairs,personalitydataareoftenobtainedpost79

pairing(Bothetal.,2005;Harrisetal.,2014;Laubuetal.,2016)notallowingtoteaseapart80

whethermatechoicewasaffectedbyindividualpersonalitiesorwhetherbehavioural81

similaritywasachievedpost‐pairinginhighlysuccessfulpairs(Laubuetal.,2016).Indirect82

evidencethatdis‐assortmentforpersonalitycansometimesbebeneficialisprovidedby83

vanOersetal.(2008),whofoundassortativepairsofgreattits,Parusmajor,toshowhigher84

ratesofextra‐pairpaternity.Generally,positiveassortmentforgenotypicorphenotypic85

5

traitsisbyfarmoreprominentintheanimalkingdomthanevidencefordis‐assortment86

(reviewedinJiangetal.,2013).87

88

Personalitytraitsconsistoftwomeasures:thebehaviourallevelandthedegreeof89

behaviouralconsistency.Althoughthereisconsiderablevariationinwithin‐individual90

behaviouralconsistency(Dingemanseetal.,2009)theeffectofsuchindividualdifferences91

inconsistencyonmatechoicehasrarelybeenconsidered(butseeSchuettetal.,2011).92

Behaviouralconsistencymightbesexuallyselectedforifitreflectsindividualquality(i.e.93

consistencyiscostlyunderchangingconditions)orifchoosingapredictable(i.e.consistent)94

mateprovidesreliableinformationaboutfutureparentalcarebehaviourpriortomating95

(Dalletal.,2004;Royleetal.,2010;Schuettetal.,2010).Forexample,afemalemightbe96

abletopredictamale'sabilitytoprotectprospectiveoffspringfromtheconsistencyin97

boldnessexpressedpriortomatechoice.98

99

Inthepresentstudy,weinvestigatedtheinfluenceofmaleandfemaleboldness(propensity100

toengageinriskybehaviour;Wilsonetal.,1994)onfemalematepreferenceinasocially101

monogamous,biparentalcichlidfromWestAfrica,therainbowkrib,Pelvicachromispulcher.102

Inthisspecies,pairsarehighlyterritorial:theydefendterritoriesandoffspringaggressively103

againstcon‐andheterospecifics.Therefore,weassumedindividualboldnesstobeatrait104

thatislikelyconsideredduringmatechoice.Furthermore,boldnesshasbeenshownto105

affectforagingsuccess(Dyeretal.,2008),eggfertilizationrates(Ariyomoetal.,2012),106

dominance(Dahlbometal.,2011),survivorship(Smithetal.,2010),andparentalcareeffort107

(Budaevetal.,1999)inotherfishspecies.Wemeasuredmaleandfemaleboldness(activity108

6

undersimulatedpredationrisk)repeatedlytotestforpersonalitydifferences.Duringmate109

choiceexperiments,femaleswerefirstallowedtoobserveabolderandashyermale110

expressingtheirnaturalboldnesstowardsapredatoranimation.Subsequentfemalemating111

preferenceforthetwomaleswasassessedinastandardmatechoicescenario.We112

consideredbothaspectsofmaleandfemalepersonality:thebehaviouralleveland113

behaviouralconsistencyofeachindividual. 114

115

Weexpectedfemalepreferencestodependonboth,thebehaviourallevelandbehavioural116

consistency,withourpredictionsbeingguidedbySchuettetal.(2010).Forthebehavioural117

level,weexpected,thatifmatechoiceisbasedonmale(parentalorgenetic)quality,118

femalesshouldshowageneralpreferenceforeitherboldorshymales(e.g.Godinetal.,119

1996;Kortetetal.,2012).Alternatively,ifmatecompatibilityismoreimportantduring120

matechoice,femalesshouldnotshowanoverallagreementbutalsoconsidertheirown121

personalityduringtheirchoice.Becausebothrainbowkribparentsprovideoffspringcare122

weconsideredthesecondpossibility,i.e.matecompatibility,tobemoreimportantformate123

choicebasedonboldness.Inspecieswithbiparentalcare,anassortativematingpreference124

forcertainbehaviouraltraitscouldreducesexualconflictoverparentalinvestment(Royle125

etal.,2010)andfacilitateoffspringcarecoordinationthroughabettersynchronisationof126

parentalactivities(Schuettetal.,2011).Dependingontheenvironmentalconditionsorthe127

biologyofthespecies,alsodis‐assortativematingmightsometimeshaveadvantages128

(Schuettetal.,2010).Forinstance,speciesthatperformseveralparentalactivitiesmight129

alsobenefitfromexpressingadis‐assortativematingpreference,facilitatingroleallocation130

andspecialisationduringoffspringcare.Often,asexualdimorphisminrolespecialisation131

7

canbeobservedwiththefemaleprovidingmoredirectoffspringcareandthemale132

defendingtheterritory(e.g.Guerraetal.,1995;Itzkowitz,1984;Neil,1984;Richteretal.,133

2010;Solomon,1993).Nevertheless,inmanyspeciesbothpartnerscanordoperformthe134

samebehaviours(seeRoyleetal.,2014forareviewontheflexibilityofparentalcare135

behaviour),andatleastpartlycompensatefortheirmates’tasksifneeded(Itzkowitz,1984;136

Laveryetal.,2010;Sasvari,1986;Storeyetal.,1994)indicatingthatsexrolesmightbeless137

fixed.Forthebehaviouralconsistency,wefolloweduptwopossiblematechoicescenarios:138

ageneralpreferenceforconsistentoverinconsistentmales,whichmightindicate139

predictabilityoflaterparentalperformance,and/orindividualquality(Royleetal.,2010;140

Schuettetal.,2010)ormatechoiceforcompatibilityleadingtoapositiveassortative141

preference(Schuettetal.,2011;Schuettetal.,2010).142

143

144

METHODS145

146

EthicalNote147

Inconsiderationofanimalwelfare,wefollowedthe"3R"framework(Russelletal.,1959).148

Todecreasethenumberofstudyanimalsneededweusedpredatoranimationsinsteadof149

livepredatorsandtestmalesformatechoicetrialswereusedtwice.Duringexperiments,150

noanimalswereharmedorexposedtoactualpredationrisk.Preyfishandpredatorswere151

keptseparatelyanddidnothavevisualcontactduringfishmaintenance.Thestudywas152

permittedbytheGerman"BehördefürGesundheitundVerbraucherschutzHamburg".153

154

8

StudyAnimalsandHoldingConditions155

StudyindividualswereobtainedfromacaptivebreedingstockattheUniversityof156

Hamburgandlocalsuppliers.Malesandfemalesusedinthisstudywere1‐2yearsoldand157

sexuallyinexperienced.Individualsweremaintainedinsame‐sexsiblinggroupsunder158

standardisedholdingconditions(100x50x25cmand200x50x25cmtanks,26±1°C159

watertemperature,aeratedandfilteredwater,weeklywaterchanges,12:12hours160

light:dark)andwerefedonceadayon5daysaweekwithArtemiaspec.On161

experimentationdays,fishwerefedafterobservations.Onedaybeforethefirstpersonality162

test,individualsweremeasuredfortheirstandardlength(males:3.8‐6.2cm,females:3.5‐163

5.1cm)usingImageJ(Schneideretal.,2012)andtransferredintoindividualtanks(25cmx164

25cmx50cm)forthedurationofexperimentaltrials(5daysperindividual).Tankswere165

endowedwithsand,halfaclaypotasshelterandaninternalfilter.Foridentification,all166

individualsweremarkedwithVIEs(visibleimplantelastomers;VIE‐NorthwestMarine167

Technology,ShawIsland,Washington,USA).Suchartificialcolourmarkshavenoinfluence168

onmatechoiceinourpopulation(Schuettetal.,2017).169

170

ExperimentalOutline171

Duringpersonalitytestingandmatechoicetrialsboldnesswasmeasuredasactivityunder172

simulatedpredationriskusingcomputeranimationsofanaturallysympatricoccurring173

predator,theAfricanobscuresnakehead,Parachannaobscura.Allmales(N=48)and174

females(N=45)usedduringmatechoiceexperimentsweretestedfortheirboldnessthree175

times(day0,day4,day33)inordertoassessthebehaviourallevelandconsistencyforall176

individuals,andshort‐andlong‐termrepeatabilityinthepopulation.Thefirstandsecond177

9

testseriesofmaleboldnesstestswereintegratedintomatechoicetrials(N=45),allowing178

femalestoobservetwomalesexpressingtheirnaturalboldness.Aftertheobservation,179

femaleswereallowedtochoosebetweenthetwomalestheyhadjustobservedina180

standardmatechoicetest(seeMateChoiceTrials).Fortheremainingboldnesstrials(third181

seriesofmaleboldnesstestsandallfemaleboldnesstests)thetestprocedurewasidentical182

tothoseintegratedintomatechoicetrialstoensureequaltestconditionsthroughout.183

184

BoldnessTest185

Boldnesstestswereconductedinatesttank(waterlevel10cm,watertemperature26±186

1°C;Figure1),whichwasdividedintothreecompartments:twoparalleltestcompartments187

inwhichtwoindividualscouldbetestedfortheirboldnessatthesametimeandanadjacent188

observercompartment.Aone‐waymirrorbetweentheobserverandthetestcompartments189

allowedtheobservertoseethetestindividualsbutinhibitedtestindividualstoseethe190

observer.Ontheothershortside,testcompartmentsfacedacomputermonitor(Dell,191

UltraSharpU2412M61cm,24”)forthepresentationofpredatoranimations.Removable192

opaquedividersbetweenthetestandtheobservercompartmentsaswellasbetweenthe193

testcompartmentsandthemonitorallowedvisualseparationduringacclimationbefore194

trials.195

196

Priortoaboldnesstest,weintroducedtwosame‐sexindividuals(fordetailsseealsoMate197

ChoiceTrials)intoaclearcylinder(diameter=11cm)each,onepertestcompartment(test198

compartmentswerepermanentlyvisuallyseparatedfromeachother).Anobserverofthe199

oppositesexwasintroducedintotheobservercompartmentbeingallowedtofreelyswim200

10

around.Anobserverwasalwaysintroduced(eveninmaleandfemalepersonalityteststhat201

werenotintegratedintomatechoicetrials)becauseitmaybepossiblethatchemicalcues202

weretransmittedfromtheobservertothetestcompartmentsdespitephysicalseparation.203

Aftera15minacclimation,theopaquedividerswereremovedallowingfreeviewofthe204

animation(testindividualsandobserver)andtestindividuals(observer).Afteranother1205

minthecylinderswereremovedandthetestperiodof11minstarted.Trialswerevideo‐206

recordedfromabovewithnohumanbeingpresentduringtrialsandthetesttankwas207

surroundedwithwhitePlexiglastoavoiddisturbances.Individualswerealwaysboldness‐208

typedatthesametimeofday±30mintoaccountforpotentialeffectsoftimeofdayand209

hungerlevelonindividualactivitypattern(Ariyomoetal.,2015;MacPhailetal.,2009).In210

eachboldnesstest,individualswereexposedtoarandomlychosenanimationshowinga211

predatorspecimentheyhadnotseenbefore.212

213

Predatoranimations(N=4,eachusinganotherspecimen)werepreparedusing214

PowerPoint©followingFischeretal.(2014).Animationsdisplayedastillphotographofthe215

predatorswimmingbackandforthinfrontofawhitebackground.Wehavevalidatedthis216

method:P.pulcherdecreasedtheiractivityinresponsetopredatoranimationscomparedto217

acontrolwhilenodifferenceinresponsetowardsalivepredatorandtheanimationwas218

found(Schereretal.,2017).219

220

Boldnesswasmeasuredasindividualactivity(totaldistancemoved;cm)fromthevideo221

recordingsusingthetrackingsoftwareEthovisionXT11(Noldus,Wageningen,The222

Netherlands).Theactivitywasassessedforatestperiodof10min,beginning1minafter223

11

thestartofthevideo.Forallindividualsthebehaviourallevelwasdefinedasthemean224

activityofthefirstandsecondtestseries.Behaviouralconsistencywascalculatedfollowing225

Ioannouetal.(2016)astheabsolutevalueofthedifferenceinactivitybetweenthefirstand226

secondboldnesstest.WefurtherdividedthemeasureofIoannouetal.(2016)bythetotal227

variationinthepopulation(rangeofactivitywithinfirstandsecondboldnesstest).As228

suggestedbyDingemanseetal.(2009),suchanindexwouldprovideameasurethatis229

standardisedinrelationtothepopulation.Wecalculatedbehaviouralconsistencyformales230

andfemalesseparately.Valuesforconsistencycanrangefrom0(highconsistency)to1231

(lowconsistency).232

233

MateChoiceTrials234

Matechoicetrialsconsistedoftwoparts:theabovedescribedobservationanda235

subsequentchoice.Duringobservation,thefemalecouldobservetwomalesshowingtheir236

naturalboldness(seeBoldnessTest).Subsequentmatechoicewasconductedimmediately237

aftertheobservationinastandarddichotomouschoicetest,suitabletopredictmate238

preferencefromtheamountoftimespentwithamaleincichlids(Dechaume‐Moncharmont239

etal.,2011;Thünkenetal.,2007).Thechoicechamber(35x100x25cm,waterlevel=10240

cm)wasseparatedintothreecompartmentswiththefemalecompartmentbeinginthe241

middle(60x35x25cm)andamalecompartmentateachside(20x35x25cm).242

243

Tobeginthechoicetest,wetransferredthefemaleandthetwomalesshehadjustobserved244

fromtheboldnesstesttanktothechoicechamber.Maleswererandomlyassignedtothe245

twomalecompartments.Allindividualswereallowedtoacclimatefor10minwhilebeing246

12

visuallyseparatedfromeachother.Then,opaquedividerswereremovedandthefirsttest247

periodof12minbegan.Thereafter,theprocedurewasrepeatedwiththemalesswitching248

sidestotakeaccountforapotentialsidebias(again10minacclimationfollowing12min249

testperiod).Toavoiddisturbancesthechoicechamberwassurroundedwithwhite250

Plexiglasandnohumanwaspresentduringtrials.Trialswerevideo‐recordedfromabove.251

252

Eachfemalewasusedonceduringmatechoicetrials.Thetwomalesusedinamatechoice253

trialwerematchedforsize(standardlengthdifference≤5%,i.e.≤3mm)andfamilybut254

otherwiserandomlychosen.Thefemaleobserveroriginatedfromadifferentfamilythan255

themales.256

257

Theassociationtimeforthetwomaleswasdeterminedfrombothtestperiods(i.e.20min)258

usingEthovisionXT11.Testperiodswereanalysedfor10min,starting2minafterthestart259

ofthevideo.Theassociationtimewasdefinedasthetimethefemalespentwithin5cm260

distancetoeachmalecompartment(whichcorrespondstoca.onefishlength;hereafter261

“preferencezone”).Femalestrengthofpreferencewasthenquantifiedastherelative262

amountoftimeshespentinthepreferencezoneoftheboldmale(associationtimeforthe263

boldmalewasdividedbytheassociationtimeforbothmales;e.g.Dugatkin,1996;264

Makowiczetal.,2010).Foreachmatechoicetest,theboldmalewasdefinedasthemale265

beingmoreactiveduringtheboldnesstestandtheshymalewasdefinedasbeingtheless266

activemale(mean±SEforabsolutesimilaritybetweenshyandboldmales:behavioural267

level=975.95±147.81;behaviouralconsistency=0.11±0.02;pleaseseeStatistical268

Analysesforcalculationofsimilarityindices).Also,wecalculatedthesidebiasforall269

13

femalesandconsideredafemalebeingside‐biasedwhenshespentmorethan80%ofthe270

totaltimespentinpreferencezones(bothtestperiods)injustonezone,regardlesswhich271

malewasthere(Poschadeletal.,2009;Schlüteretal.,1998).272

273

StatisticalAnalyses274

AlldataanalyseswereconductedinR3.2.3(RCoreTeam,2015).Totestforpersonality275

differencesrepeatabilityofourmeasureforboldness(activityundersimulatedpredation276

risk)wasassessedwithlinearmixedeffectmodels(LMMs)usingtherptR‐package277

(Schielzethetal.,2013).Weassessedshort‐termrepeatability(boldnesstest:day0,day4)278

aswellaslong‐termrepeatability(boldnesstest:day4,day33)forsexesseparatelywith279

1000bootstrappingrunsand1000permutations.Significancewasinferredwhenthe95%280

CIdidnotincludezero.Activitywassquareroot‐transformedfornormalityandmodels281

werefitforGaussianerrorstructure.282

283

Totestforageneralpreferenceforboldorshymales,weranaLMMwithfemalestrengthof284

preferenceforboldmalesastheresponseandmaleIDasrandomeffect.Wedidnotinclude285

anyfixedeffects.Tocheckforadeviationfromrandomchoice(i.e.strengthofpreference=286

50%)weobtainedthe95%CIoftheestimatedmean.Apreferenceforeitherboldorshy287

maleswouldbeindicatediftheCIdoesnotinclude0.50.Similarly,wetestedforageneral288

preferenceforbehaviouralconsistencybyrunninganullmodelwithfemalestrengthof289

preferenceforthemaleshowingthehigherconsistencyduringtheobservationasthe290

responseandmaleIDasrandomeffect.Apreferenceforeitherconsistencyorinconsistency291

wouldberevealedifthe95%CIofthemeandoesnotinclude0.50.292

14

293

Totestfor(dis)‐assortativefemalematechoicewefittedaLMMwithfemalestrengthof294

preferenceforboldmalesastheresponsevariableandmaleIDasrandomterm.Asfixed295

effectsweincludedrelativesimilarityforthebehaviourallevelandrelativesimilarityfor296

thebehaviouralconsistencybetweenthefemaleandthemalesshesawduringthe297

observationphaseandmatechoicetest.Tocalculaterelativesimilarity(forleveland298

consistency,respectively),wefirstcomputeddifference‐scorebasedsimilaritybetweenthe299

femaleandeachofthetwomales(boldandshy)astheabsolutevalueofthedifferencein300

therespectivebehaviour(e.g.Gaunt,2006;Luoetal.,2005;Montiglioetal.,2016)between301

thefemaleandtheboldmale,andthefemaleandtheshymale.Thus,similarity(inleveland302

consistency,respectively)ishighestatzeroanddis‐similarityincreaseswithincreasing303

values.RelativesimilaritywasthencalculatedfollowingGasparinietal.(2015):the304

similaritybetweenthefemaleandtheboldmalewassubtractedfromthesimilarity305

betweenthefemaleandtheshymale.Positivevaluesforrelativesimilarity(inleveland306

consistency,respectively)indicatehighersimilaritybetweenthefemaleandtheboldmale307

whilenegativevaluesindicatetheshymaleismoresimilartothefemalethanthebold308

male.Priortotheanalysis,wez‐transformedbothrelativesimilarityforthebehavioural309

levelandforthebehaviouralconsistencyforstandardisation.310

311

Weusedthelme4‐package(Batesetal.,2015)forLMMs.Weusedstepwisebackward312

modelsimplificationtofittheminimumadequatemodel.PartialR2withCL(confidence313

level)werecalculatedforexplanatoryvariablesusingtheapproachsuggestedbyNakagawa314

etal.(2013),implementedinther2glmm‐package(Jaeger,2016).Fornon‐significant315

15

explanatoryvariableswereportedregressionestimatesandpartialR2ofthemodelbefore316

thetermwasdropped.Modelassumptionswerevisuallyensuredthroughmodeldiagnosis317

plots.Forallanalyses,femalestrengthofpreferencewasarcsine‐squareroot‐transformed318

fornormality.Wehadaprioridecidedtoexcludeside‐biasedfemales(N=6)from319

preferenceanalyses(Dosenetal.,2004;Hoysaketal.,2007;Knieletal.,2015;Schluppetal.,320

1999;Schlüteretal.,1998;Williamsetal.,2010).Bydefinition,aside‐biasedfemaleshows321

contradictorypreferencesduringthetwotestperiodsofachoicetest.Theremovalofsuch322

inconsistentbehaviourthatappearsrandominregardtothepresentedmalesiscrucialas323

toremovefemalesthatwouldnotexpressamatingpreferenceforthepresentedmalesbut324

ratherapreferencefor(oragainst)aspecificsideofthechoicechamber(e.g.becauseofa325

lackofmotivation).Leavingsuchbiasedpreferencedatainthedatasetwouldartificially326

increasethesamplesizeanddistorttheactualpreferencepattern.Ontheotherhand,327

removingside‐biasedfemalesfromthedatasetcanlowerthebehaviouralrange328

representedinthisstudy.Astherearedifferentapproachesbutnocommonagreementin329

howtohandlesidebiasesinmatechoicetrials,weperformedallpreferenceanalysestwice,330

oncewithandoncewithoutremovingside‐biasedfemales(N=45).Thoughwehere331

considerbothapproaches,weadvocatetheremovalofclearlybiasedpreferencedatafrom332

analysesandwillthereforemainlyfocusonthepresentationofpreferenceanalyses333

performedwithoutobvioussidebiasesinthedata.334

335

RESULTS336

337

16

Malesandfemalesweresignificantlyrepeatableintheirboldnessovertheshort‐term338

(LMMmales:R=0.507,SE=0.110,CI=[0.246,0.686],N=48;LMMfemales:R=0.604,SE339

=0.097,CI=[0.380,0.763],N=45)andlong‐term(LMMmales:R=0.463,SE=0.113,CI=340

[0.233,0.657],N=48;LMMfemales:R=0.557,SE=0.111,CI=[0.311,0.732],N=42).341

342

Wefoundnogeneralpreferenceforeitherboldorshymales(meanpreferenceforbold343

males:46.5%;95%CI=[40.8,52.1%]).Also,wedidnotdetectageneralpreferencefor344

maleconsistency(meanpreferenceforconsistentmales:53.5%,95%CI=[47.8,58.9%]).345

346

Femalestrengthofpreferencefortheboldmalesignificantlydecreasedwithincreasing347

relativesimilarityinthebehaviourallevel(LMM:χ21=10.572,N=39,P=0.001,coefficient348

±SE(standardised)=‐0.091±0.026;R2=0.242,CL=[0.056,0.475];Figure2a).Further,349

femalestrengthofpreferenceincreasedwithincreasingrelativesimilarityinbehavioural350

consistency(LMM:χ21=4.528,N=39,P=0.033,coefficient±SE(standardised)=0.058±351

0.026;R2=0.114,CL=[0.003,0.341];Figure2b).352

353

Whenperformingpreferenceanalysiswithouttheremovalofside‐biasedfemales,we354

receivedsimilarresultswithregardtofemalestrengthofpreferenceforboldmales(mean355

preference:46.5%;95%CI=[41.5,51.6%])andforconsistentmales(meanpreference:356

53.9%;95%CI=[49.1,59.1%])notshowingadeviationfromrandomchoice.However,357

differenttotheanalysiswithremovedsidebiases,relativesimilarityinthebehavioural358

leveltendedtonegativelyinfluencefemalepreferenceforboldmales(LMM:χ21=2.885,N=359

45,P=0.089,coefficient±SE(standardised)=‐0.043±0.034;R2=0.066,CL=[0.001,360

17

0.258])andrelativesimilarityinbehaviouralconsistencydidnotaffectfemalepreference361

(LMM:χ21=2.279,N=45,P=0.131,coefficient±SE(standardised)=0.040±0.025;R2=362

0.052,CL=[0.000,0.235]).363

364

365

DISCUSSION366

367

BothsexesofP.pulchershowedconsistentshort‐andlong‐termpersonalitydifferencesfor368

boldness.Wedidnotdetectanoverallagreementinfemalematingpreferenceforeither369

malelevelorconsistencyofboldness.However,wefounddis‐assortativefemalechoicefor370

thelevelofboldness.Also,femalepreferenceincreasedwithsimilarityinbehavioural371

consistency,suggestingassortativechoiceforconsistencyinboldness(whenside‐biased372

femaleswereremoved).373

374

Thedis‐assortativepreferenceforthebehaviouralleveliscontradictorytotheresultsof375

mostothermatechoicestudiestestingforbehavioural(dis‐)assortmentthatmainly376

reportedassortativematingpreferences(e.g.Montiglioetal.,2016;Schuettetal.,2011).At377

thispoint,wecanonlyspeculateaboutpossibleadaptivebenefitsofadis‐assortative378

preference.Behaviouraldis‐similaritycouldpossiblyincreasewithin‐pairbehavioural379

and/orgeneticcompatibility(Schuettetal.,2010).Behaviouralcompatibilityhasprimarily380

beendiscussedforbiparentalspecieswhenbothparentsperformmoreorlessthesame381

parentalactivity,forinstanceoffspringprovisioninginsomebirds(Royleetal.,2010).In382

zebrafinches,Taeniopygiaguttata,forinstance,similarityinthebehaviourallevelhasbeen383

18

showntoincreasepaircompatibility(e.g.Schuettetal.,2011).However,whenspecies384

performvariousparentalactivitiestheymightsometimesbenefitfromexpressingadis‐385

assortativematingpreference,facilitatingroleallocationduringoffspringcare.InP.pulcher,386

parentstypicallydividethelabourwithoneindividualstayingmorewiththeoffspringand387

theotheronedefendingtheterritory.Thoughsexualdimorphisminrolespecialisationhas388

beendescribedformanycichlids(McKayeetal.,2008;Neil,1984;Richteretal.,2010),sex389

rolesmightnotbeentirelystrictinthespeciesandmayratherdependontheinterplay390

betweenmaleandfemalepersonality.Itzkowitzetal.(2005)haveshownthatmaleand391

femaleparentconvictcichlids,Archocentrusnigrofasciatum,changedtheirdefense392

behaviourinresponsetothemate'sbodysize,regardlessofthesex.Thisresultindicates393

thatparentalroleallocationmayinsomespeciesratherdependonthemate'sbehaviour394

andphysiologythanonthesexitself.Behaviouraldis‐similarityinboldnessmayfacilitate395

labourdivisionwiththebolderindividualdefendingtheterritoryandtheshyerindividual396

stayingwiththeyoung,regardlessofthesex.Hence,dis‐assortativematingforpersonality397

couldsometimesleadtoinvertedparentalcarerolesthoughthishasnotbeeninvestigated398

yet.Also,anincreasedgeneticcompatibilitythroughdis‐similaritycouldbepossibleifdis‐399

assortativematingleadstoheterozygoteoffspringthataremoreviable(Charlesworthetal.,400

1987;Dingemanseetal.,2004).Forexample,Marshalletal.(2003)showedastrong401

correlationbetweenindividualgeneticdiversityandabehaviouraltrait,songcomplexity,in402

sedgewarblers,Acrocephalusschoenobaenus.Femaleschosetomatewithmalesthat403

increasedoffspringgeneticdiversity(Marshalletal.,2003).Seddonetal.(2004)foundmale404

heterozygositytobecorrelatedwithterritorysizeandsongstructureinmale(butnot405

female)subdesertmesite,Moniasbenschi.406

19

407

Further,wefoundassortativematechoicefortheconsistencyofboldness.Thefewstudies408

thathaveassessedthelinkbetweenbehaviouralconsistencyandsexualselectionfounda409

positiverelationshipbetweenconsistencyandreproductivesuccess(Boteroetal.,2009;410

Byers,2006)andahigherreproductivesuccessofpairsmatchedforbehavioural411

consistency(Schuettetal.,2011).Schuettetal.(2011)haveshownthatpairsmatchedfor412

consistencyraisedfosterfledglingsofbetterbodycondition,indicatingthepossible413

mechanismdrivingassortmentforbehaviouralconsistencymightbeahigherefficiencyin414

theprovisionofparentalcare.415

416

Clearly,ourstudyislimitedbythecorrelativedesign,notallowingtospecificallyaddress417

thecausalityunderlyingthepreferencepattern.Furtherexaminationsusingbehavioural418

manipulationsarenowneededtodecoupleboldnessfrompotentiallycorrelatedtraitsthat419

mightinfluencematechoice,toensurethepreferencepatternwefoundisunequivocally420

relatedtoindividualbehaviour.Moreover,itshouldbementionedthatourmeasurefor421

behaviouralconsistencyderivedfromonlytwomeasurements.Weareherefacingacritical422

trade‐off.Whilemultiplemeasurementscanleadtoachangeinbehaviourcausedbythe423

numberoftimestested,e.g.throughhabituationorsensitization(Belletal.,2009;Stampset424

al.,2012),themeasurementerrorishigherwhenonlytestedtwice.Inthisparticularstudy,425

wetestedindividualresponsestowardsunfamiliarpredatoranimations,presentedina426

novelsituation.Ourmeasurementforboldnesswouldlikelybeaffectedbypriorexperience427

andfamiliaritywithtestconditions,makingitdifficulttoreceivethesamenatureof428

measureforboldnesswhentestedmultipletimes.However,thestrengthofourstudyis429

20

thatfemalescouldobservemaleboldnessdirectlybeforematechoicetrialswhiletheywere430

hiddenbehindone‐wayglassandpartitions.Thisway,malescouldexpresstheirnatural431

behaviourwithoutbeingaffectedbythefemale'spresence.Adecouplingofobservationand432

choiceensuredfemalepreferencenotbeingconfoundedbythepresenceofapredator.433

434

Conclusions435

Insummary,weprovidesuggestiveevidencethatsexualselectionmayrepresentakeyrole436

intheevolutionofpersonalitydifferences.Femalesshowedadis‐assortativemating437

preferenceforthelevelofboldnessandanassortativepreferenceforthedegreeof438

behaviouralconsistency.Ourresultsindicatematechoiceforbehaviouraland/orgenetic439

compatibilitythoughonlyassessedinacorrelativeapproach.Suchamatingpreference440

mightimproveparentalcareefficiencythroughfacilitationofparentalroleallocation441

and/ortoincreaseoffspringfitnessthroughgeneticbenefits.Noticeable,thehandlingof442

sidebiasessignificantlyaffectedourresults.Whilewefoundaneffectofbehavioural443

similarityinlevelandconsistencywhenremovingsidebiases,wecouldnotdetectsuch444

effectswithoutremovingside‐biasedfemalesfromthedata.Thisdiscrepancyinresults445

underlinestheimportanceoftakingtheapproachusedintoconsiderationwhencomparing446

theresultsofdifferentmatechoicestudies.Thehandlingofsidebiasesinmatechoice447

studiesisnottrivialandcanlargelyaffectexperimentaloutcomes.448

449

450

ACKNOWLEDGEMENTS451

452

21

ThisresearchwasfoundedbyDeutscheForschungsgemeinschaft(SCHU‐2927/2‐1,grantto453

W.S.).WethankF.X.Dechaume‐Moncharmontandtwoanonymousreviewersfortheir454

constructivecomments.455

456

457

REFERENCES458

459

Ariyomo,T.O.,Carter,M.,&Watt,P.J.(2013).Heritabilityofboldnessandaggressivenessin460

thezebrafish.BehaviorGenetics,43,161‐167.461

462

Ariyomo,T.O.,&Watt,P.J.(2012).Theeffectofvariationinboldnessandaggressivenesson463

the reproductive success of zebrafish. Animal Behaviour, 83(1), 41‐46.464

doi:10.1016/j.anbehav.2011.10.004465

466

Ariyomo, T. O., & Watt, P. J. (2013). Disassortative mating for boldness decreases467

reproductive success in the guppy. Behavioral Ecology, 24(6), 1320‐1326.468

doi:10.1093/beheco/art070469

470

Ariyomo,T.O.,&Watt,P.J.(2015).Effectofhungerlevelandtimeofdayonboldnessand471

aggression in the zebrafishDanio rerio. Journal of FishBiology, 86(6), 1852‐1859.472

doi:10.1111/jfb.12674473

474

22

Bates,D.,Mächler,M.,Bolker,B.,&Walker,S. (2015).FittingLinearMixed‐EffectsModels475

Usinglme4.JournalofStatisticalSoftware,67(1),1‐48.doi:10.18637/jss.v067.i01476

477

Bell,A.M.,Hankison,S. J.,&Laskowski,L.(2009).Therepeatabilityofbehaviour:ameta‐478

analysis.AnimalBehaviour,77,771‐783.479

480

Boissy,A. (1995).Fearand fearfulness inanimals.TheQuarterlyReviewofBiology,70(2),481

165‐191.482

483

Botero, C. A., Rossman, R. J., Caro, L. M., Stenzler, L. M., Lovette, I. J., de Kort, S. R., &484

Vehrencamp,S.L.(2009).Syllabletypeconsistencyisrelatedtoage,socialstatusand485

reproductive success in the tropical mockingbird. Animal Behaviour, 77, 701‐706.486

doi:10.1016/j.anbehav.20487

488

Both, C., Dingemanse,N. J., Drent, P. J., & Tinbergen, J.M. (2005). Pairs of extreme avian489

personalities have highest reproductive success. Journal of Animal Ecology, 74(4),490

667‐674.doi:10.1111/j.1365‐2656.2005.00962.x491

492

Budaev,S.V.,Zworykin,D.D.,&Mochek,A.D.(1999).Individualdifferencesinparentalcare493

andbehaviourprofileintheconvictcichlid:acorrelationstudy.AnimalBehaviour,58,494

195‐202.495

496

23

Byers, B. E. (2006). Extrapair paternity in chestnut‐sided warblers is correlated with497

consistent vocal performance. Behavioral Ecology, 18(1), 130‐136.498

doi:10.1093/beheco/arl058499

500

Charlesworth, D., & Charlesworth, B. (1987). Inbreeding depression and its evolutionary501

consequences.AnnualReviewofEcologyandSystematics,18,237‐268.502

503

Dahlbom, S. J., Lagman, D., Lundstedt‐Enkel, K., Sundstrom, L. F., & Winberg, S. (2011).504

Boldness predicts social status in zebrafish (Danio rerio).PLoSOne,6(8), e23565.505

doi:10.1371/journal.pone.0023565506

507

Dall,S.R.X.,Houston,A.I.,&McNamara,J.M.(2004).Thebehaviouralecologyofpersonality:508

consistentindividualdifferencesfromanadaptiveperspective.EcologyLetters,7(8),509

734‐739.doi:10.1111/j.1461‐0248.2004.00618.x510

511

Dechaume‐Moncharmont,F.X.,Cornuau, J.H.,Keddar, I., Ihle,M.,Motreuil,S.,&Cezilly,F.512

(2011). Rapid assessment of female preference formale size predicts subsequent513

choiceof spawningpartner ina sociallymonogamouscichlid fish.ComptesRendus514

Biologies,334(12),906‐910.doi:10.1016/j.crvi.2011.08.004515

516

Dingemanse,N. J.,Both,C.,Drent,P. J.,&Tinbergen, J.M. (2004).Fitnessconsequencesof517

avianpersonalitiesinafluctuatingenvironment.ProceedingsofTheRoyalSocietyB,518

271(1541),847‐852.doi:10.1098/rspb.2004.2680519

24

520

Dingemanse,N. J.,Kazem,A. J.,Réale,D.,&Wright, J. (2009).Behaviouralreactionnorms:521

animalpersonalitymeetsindividualplasticity.TrendsinEcologyandEvolution,25(2),522

81‐89.doi:10.1016/j.tree.2009.07.013523

524

Dingemanse,N.J.,&Reale,D.(2005).Naturalselectionandanimalpersonality.Behaviour,525

142,1165‐1190.526

527

Dosen, L. D., &Montomerie, R. (2004). Female size influencesmate preferences ofmale528

guppies.Ethology,110,245‐255.529

530

Dugatkin, L. A. (1996). Interface between culturally based preferences and genetic531

preferences: Femalemate choice inPoecilia reticulata.Proceedingsof theNational532

AcademyofSciencesUSA,93,2770‐2773.533

534

Dyer, J.R.G.,Croft,D.P.,Morrell,L. J.,&Krause, J. (2008). Shoal compositiondetermines535

foraging success in the guppy. Behavioral Ecology, 20(1), 165‐171.536

doi:10.1093/beheco/arn129537

538

Fischer,S.,Hess,S.,Oberhummer,E.,Burlaud,R.,Fernandez,A.A.,Frommen,J.G.,&Taborsky,539

B.(2014).Animatedimagesasatooltostudyvisualcommunication:acasestudyina540

cooperatively breeding cichlid. Behaviour, 151(12‐13), 1921‐1942.541

doi:10.1163/1568539x‐00003223542

25

543

Gasparini,C.,Congiu,L.,&Pilastro,A. (2015).Majorhistocompatibilitycomplexsimilarity544

andsexual selection:differentdoesnotalwaysmeanattractive.MolecularEcology,545

24(16),4286‐4295.doi:10.1111/mec.13222546

547

Gaunt, R. (2006). Couple similarity andmarital satisfaction: are similar spouses happier?548

JournalofPersonality,74(5),1401‐1420.doi:10.1111/j.1467‐6494.2006.00414.x549

550

Godin,J.‐G.J.,&Dugatkin,L.A.(1996).Femalematingpreferenceforboldmalesintheguppy,551

Poeciliareticulata.ProceedingsoftheNationalAcademyofSciencesUSA,93,10262‐552

10267.553

554

Gonzaga,G.C.,Carter,S.,&Buckwalter, J.G. (2010).Assortativemating,convergence,and555

satisfaction in married couples. Personal Relationships, 17(4), 634‐644.556

doi:10.1111/j.1475‐6811.2010.01309.x557

558

Gosling,S.D.(2001).Frommicetomen:Whatcanwelearnaboutpersonalityfromanimal559

research?PsychologicalBulletin,127,45‐86.560

561

Guerra,M.,&Drummond,H. (1995).Reversedsexualsizedimorphismandparental care:562

minimaldivisionoflabourintheblue‐footedbooby.Behaviour,132,479‐496.563

564

26

Harris,M.R.,&Siefferman,L.(2014).Interspecificcompetitioninfluencesfitnessbenefitsof565

assortativematingforterritorialaggressionineasternbluebirds(Sialiasialis).PLoS566

One,9(2),e88668.doi:10.1371/journal.pone.0088668567

568

Hoysak,D.J.,&Godin,J.‐G.J.(2007).Repeatabilityofmalematechoiceinthemosquitofish,569

Gambusia holbrooki. Ethology, 113(10), 1007‐1018. doi:10.1111/j.1439‐570

0310.2007.01413.x571

572

Ihle, M., Kempenaers, B., & Forstmeier, W. (2015). Fitness benefits of mate choice for573

compatibility in a socially monogamous species. PLoS biology, 13(9), e1002248.574

doi:10.1371/journal.pbio.1002248575

576

Ioannou,C.C.,&Dall,S.R.(2016).Individualsthatareconsistentinrisk‐takingbenefitduring577

collectiveforaging.ScientificReports,6,33991.doi:10.1038/srep33991578

579

Itzkowitz,M.(1984).Parentaldivisionof laborinamonogomousfish.Behaviour,89,251‐580

260.581

582

Itzkowitz, M., Santangelo, N., Cleveland, A., Bockelman, A., & Richter, M. (2005). Is the583

selectionofsex‐typicalparentalrolesbasedonanassessmentprocess?Atestinthe584

monogamous convict cichlid fish. Animal Behaviour, 69(1), 95‐105.585

doi:10.1016/j.anbehav.2003.12.027586

587

27

Jaeger,B. (2016). r2glmm:ComputesRsquared formixed(multilevel)models.Rpackage588

version0.1.1.Retrievedfromhttps://CRAN.R‐project.org/package=r2glmm589

590

Jiang,Y.,Bolnick,D.I.,&Kirkpatrick,M.(2013).Assortativematinginanimals.TheAmerican591

Naturalist,181(6),125‐138.doi:10.1086/670160592

593

Kniel,N.,Durler,C.,Hecht,I.,Heinbach,V.,Zimmermann,L.,&Witte,K.(2015).Novelmate594

preference throughmate‐choice copying in zebra finches: sexes differ. Behavioral595

Ecology,26(2),647‐655.doi:10.1093/beheco/aru241596

597

Kortet,R.,Niemelä,P. T., Vainikka,A.,&Laakso, J. (2012). Femalespreferboldmales; an598

analysisofboldness,matechoice,andbacterialresistanceinthefieldcricketGryllus599

integer.EcologicalParasitologyandImmunology,1,1‐6.doi:10.4303/epi/235580600

601

Kralj‐Fiser, S., Sanguino Mostajo, G. A., Preik, O., Pekar, S., & Schneider, J. M. (2013).602

Assortativematingbyaggressivenesstypeinorbweavingspiders.BehavioralEcology,603

24(4),824‐831.doi:10.1093/beheco/art030604

605

Kralj‐Fišer, S., & Schuett,W. (2014). Studyingpersonality variation in invertebrates:why606

bother?AnimalBehaviour,91,41‐52.doi:10.1016/j.anbehav.2014.02.016607

608

28

Laubu,C.,Dechaume‐Moncharmont,F.X.,Motreuil,S.,&Schweitzer,C.(2016).Mismatched609

partners that achievepostpairingbehavioral similarity improve their reproductive610

success.ScienceAdvances,2(3),e1501013.doi:10.1126/sciadv.1501013611

612

Lavery,R.J.,&Reebs,S.G.(2010).Effectofmateremovaloncurrentandsubsequentparental613

care in the convict cichlid (Pisces: Cichlidae). Ethology, 97(4), 265‐277.614

doi:10.1111/j.1439‐0310.1994.tb01046.x615

616

Luo, S., & Klohnen, E. C. (2005). Assortativemating andmarital quality in new‐lywed: A617

couple‐centeredapproach.JournalofPersonalityandSocialPsychology,88,304–326.618

619

MacPhail, R. C., Brooks, J., Hunter, D. L., Padnos, B., Irons, T. D., & Padilla, S. (2009).620

Locomotion in larval zebrafish: Influence of time of day, lighting and ethanol.621

Neurotoxicology,30(1),52‐58.doi:10.1016/j.neuro.2008.09.011622

623

Makowicz,A.,Plath,M.,&Schlupp,I.(2010).Maleguppies(Poeciliareticulata)adjusttheir624

mate choice behaviour to the presence of an audience.Behaviour, 147(13), 1657‐625

1674.doi:10.1163/000579510x528206626

627

Marshall,R.C.,Buchanan,K.L.,&Catchpole,C.K. (2003). Sexual selectionand individual628

genetic diversity in a songbird. Proceedings of TheRoyal Society B, 270, 248‐250.629

doi:10.1098/rsbl.2003.0081630

631

29

Mascie‐Taylor,C.G.N.,&Vandenberg,S.G.(1988).AssortativematingforIQandpersonality632

duetopropinquityandpersonalpreference.BehaviorGenetics,18,339‐345.633

634

Mays,H.L.,Jr.,&Hill,G.E.(2004).Choosingmates:goodgenesversusgenesthatareagood635

fit.TrendsinEcologyandEvolution,19(10),554‐559.doi:10.1016/j.tree.2004.07.018636

637

McKaye,K.R.,&Murry,B.A.(2008).SexroledifferentiationinbrooddefensebyNicaraguan638

cichlidfish,Amphilophusxiloanensis.CaribbeanJournalofScience,44,13‐20.639

640

Montiglio,P.O.,Wey,T.W.,Chang,A.T.,Fogarty,S.,&Sih,A.(2016).Multiplematingreveals641

complexpatternsofassortativematingbypersonalityandbodysize.JournalofAnimal642

Ecology,85(1),125‐135.doi:10.1111/1365‐2656.12436643

644

Nakagawa,S.,&Schielzeth,H.(2013).AgeneralandsimplemethodforobtainingR2from645

generalizedlinearmixed‐effectsmodels.MethodsinEcologyandEvolution,4(2),133‐646

142.doi:10.1111/j.2041‐210x.2012.00261.x647

648

Neil,S.J.(1984).FieldstudiesofthebehavioralecologyandagonisticbehaviorofCichlasoma649

meeki(Pisces:Cichlidae).EnvironmentalBiologyofFishes,10,59‐68.650

651

Ophir,A.G.,&Galef,B.G.(2003).FemaleJapanesequailthat‘eavesdrop’onfightingmales652

prefer losers to winners. Animal Behaviour, 66(2), 399‐407.653

doi:10.1006/anbe.2003.2230654

30

655

Patrick, S. C., Charmantier, A., & Weimerskirch, H. (2013). Differences in boldness are656

repeatableandheritableinalong‐livedmarinepredator.EcologyandEvolution,3(13),657

4291‐4299.doi:10.1002/ece3.748658

659

Poschadel,J.R.,Plath,M.,&Schlupp,I.(2009).Divergentfemalematingpreferenceinaclonal660

fish.actaethologica,12(1),55‐60.doi:10.1007/s10211‐009‐0055‐8661

662

R Core Team. (2015). R: A language and environment for statistical computing. Vienna,663

Austria: R Foundation for Statistical Computing. Retrieved from http://www.R‐664

project.org/665

666

Reif,A.,&Lesch,K.‐P.(2003).Towardamoleculararchitectureofpersonality.Behavioural667

BrainResearch,139,1‐20.doi:10.1016/S0166‐4328(02)00267‐X668

669

Richter,M.,Santangelo,N.,&Itzkowitz,M.(2010).Biparentaldivisionofrolesintheconvict670

cichlidfish:influenceofintrudernumbersandlocations.EthologyEcology&Evolution,671

17,1‐15.doi:10.1080/08927014.2005.9522611672

673

Royle,N.J.,Russell,A.F.,&Wilson,A.J.(2014).Theevolutionofflexibleparenting.Science,674

346(6198),776‐781.675

676

31

Royle,N.J.,Schuett,W.,&Dall,S.R.X.(2010).Behavioralconsistencyandtheresolutionof677

sexual conflict over parental investment. Behavioral Ecology, 21(6), 1125‐1130.678

doi:10.1093/beheco/arq156679

680

Russell,W.M. S., &Burch, R. L. (1959).Theprinciplesofhumane experimental technique.681

LondonW.C.I.:MethuenandCo.,Ltd.682

683

Sasvari,L.(1986).Reproductiveeffortofwidowedbirds.JournalofAnimalEcology,55,553‐684

564.685

686

Scherer,U.,Godin, J.G. J.,&Schuett,W. (2017).Validationof2D‐animatedpictures as an687

investigativetoolinthebehaviouralsciences–acasestudywithaWestAfricancichlid688

fish,Pelvicachromispulcher.submittedmanuscript.689

690

Schielzeth,H.,&Nakagawa, S. (2013). rptR: Repeatability forGaussian andnon‐Gaussian691

data.https://R‐Forge.R‐project.org/projects/rptr/. 692

693

Schlupp, I., Waschulewski, M., & Ryan, M. J. (1999). Female preferences for naturally‐694

occurringnovelmaletraits.Behaviour,136,519‐527.695

696

Schlüter,A.,Parzefall,J.,&Schlupp,I.(1998).Femalepreferenceforsymmetricalverticalbars697

inmalesailfinmollies.AnimalBehaviour,56,147‐153.698

699

32

Schneider,C.A.,Rasband,W.S.,&Eliceiri,K.W.(2012).NIHImagetoImageJ:25yearsof700

imageanalysis.Naturemethods,9(7),671‐675.doi:PMID22930834701

702

Schuett, W., Dall, S. R. X., & Royle, N. J. (2011). Pairs of zebra finches with similar703

‘personalities’ make better parents. Animal Behaviour, 81(3), 609‐618.704

doi:10.1016/j.anbehav.2010.12.006705

706

Schuett,W.,Godin,J.G.J.,&Dall,S.R.X.(2011).Dofemalezebrafinches,Taeniopygiaguttata,707

choose their mates based on their ‘personality’? Ethology, 117(10), 908‐917.708

doi:10.1111/j.1439‐0310.2011.01945.x709

710

Schuett,W,Nava,TF,Rahmlow,T&UScherer(2017).ArtificialVisibleImplantElastomer711

(VIE) tags of different colour and symmetry do not influence mate choice in a712

cichlid.Behaviour,inpress.713

714

Schuett, W., Tregenza, T., & Dall, S. R. (2010). Sexual selection and animal personality.715

BiologicalReviews,85(2),217‐246.doi:10.1111/j.1469‐185X.2009.00101.x716

717

Seddon, N., Amos,W.,Mulder, R. A., & Tobias, J. A. (2004). Male heterozygosity predicts718

territorysize, songstructureandreproductivesuccess inacooperativelybreeding719

bird. Proceedings of The Royal Society B, 271(1550), 1823‐1829.720

doi:10.1098/rspb.2004.2805721

722

33

Sih,A.,Bell,A.,&Johnson,J.C.(2004).Behavioralsyndromes:anecologicalandevolutionary723

overview. Trends in Ecology and Evolution, 19(7), 372‐378.724

doi:10.1016/j.tree.2004.04.009725

726

Sih,A.,Bell,A.M.,Johnson,J.C.,&Ziemba,R.E.(2004).Behavioralsyndromes:anintegrative727

overview.TheQuarterlyReviewofBiology,79(3),241‐277.728

729

Smith,B.R.,&Blumstein,D.T.(2008).Fitnessconsequencesofpersonality:ameta‐analysis.730

BehavioralEcology,19(2),448‐455.doi:10.1093/beheco/arm144731

732

Smith, B. R., & Blumstein, D. T. (2010). Behavioral types as predictors of survival in733

Trinidadian guppies (Poecilia reticulata). Behavioral Ecology, 21(5), 919‐926.734

doi:10.1093/beheco/arq084735

736

Solomon,N.G. (1993).Comparisonofparentalbehavior inmaleand femaleprairievoles737

(Microtus ochrogaster). Canadian Journal of Zoology, 71(2), 434‐437.738

doi:10.1139/z93‐061739

740

Stamps,J.A.,Briffa,M.,&Biro,P.A.(2012).Unpredictableanimals:individualdifferencesin741

intraindividual variability (IIV). Animal Behaviour, 83(6), 1325‐1334.742

doi:10.1016/j.anbehav.2012.02.017743

744

34

Storey,A.E.,Bradbury,C.G.,&Joyce,T.L.(1994).Nestattendanceinmalemeadowvoles:the745

roleofthefemaleinregulatingmaleinteractionswithpups.1994,47,1037‐1046.746

747

Thünken,T.,Bakker,T.C.M.,Baldauf,S.A.,&Kullmann,H.(2007).Active inbreeding ina748

cichlidfishanditsadaptivesignificance.CurrentBiology,17,225‐229.749

750

van Oers, K., de Jong, G., van Noordwijk, A. J., Kempenaers, B., & Drent, P. J. (2005).751

Contributionofgeneticstothestudyofanimalpersonalities:areviewofcasestudies.752

Behaviour,142,1185‐1206.753

754

vanOers,K.,Drent,P.J.,Dingemanse,N.J.,&Kempenaers,B.(2008).Personalityisassociated755

withextrapairpaternityingreattits,Parusmajor.AnimalBehaviour,76(3),555‐563.756

doi:10.1016/j.anbehav.2008.03.011757

758

Williams,T.H.,&Mendelson,T.C.(2010).Behavioralisolationbasedonvisualsignalsina759

sympatricpairofdarterspecies.Ethology,116(11),1038‐1049.doi:10.1111/j.1439‐760

0310.2010.01816.x761

762

Wilson,D.S.(1998).Adaptiveindividualdifferencewithinsinglepopulations.Philosophical763

TransactionsoftheRoyalSocietyB,353,199‐205.764

765

Wilson, D. S., Clark, A. B., Coleman, K., & Dearstyne, T. (1994). Shyness and boldness in766

humansandotheranimals.TrendsinEcology&Evolution,9,442‐446.767

35

768

Wolf, M., & Weissing, F. J. (2010). An explanatory framework for adaptive personality769

differences.PhilosophicalTransactionsoftheRoyalSocietyB,365(1560),3959‐3968.770

doi:10.1098/rstb.2010.0215771

772

773

36

FIGURES774

Figure1:Experimentalset‐upfortheboldnesstest.Twosame‐sexfocalindividuals

(visuallyseparated)wereexposedtoavideoanimationofapredator.Testindividuals

wereobservedbyafishoftheothersexbutcouldthemselvesnotseetheobserver:the

observercompartmentwasendowedwithaone‐waymirroralignedwithanangleof45°

towardsthetestcompartmentsprovidingavisualcoverfortheobserver.Fishnotto

scale.

775

37

Figure2:Femalestrengthofpreferencefortheboldmaleindependenceofrelative

similarityin(a)theleveland(b)theconsistencyofboldness.Positivesimilarityvalues

indicatetheboldmalewasmoresimilartothefemalethantheshymale,negativevalues

indicatehighersimilaritybetweenthefemaleandtheshymale.Datavisualisationon

originaldata,strengthofpreferencewasarcsine‐squareroot‐transformedforanalyses.

776