Welcome all to Konstanz for theASAB Summer 2019 Conference “New Frontiers in the Study of Animal Behaviour”! It is an extremely important time to be studying animal behavior. Impacting biological processes across vast scales, from the neural, genetic and endocrine interactions within organisms, to the interactions among organisms, and the emergent functional complexity of groups, populations and ecosystems,

behavior is a crucial component of almost all ecological and evolutionary processes.

With ongoing dramatic innovations in technologies, such as computer vision, global positioning systems, genetic manipulation, physiological recording, machine learning and the ability to immerse organisms into synthetic worlds, we are poised to transform the study of animal behavior and to develop approaches that do not ignore the complexities and dynamical relationships in the natural world, but rather embrace them. In addition to providing new fundamental biological and bio-medical insights, the integrative and quantitative study of behavior will increasingly be essential for the development of evidence-based policies and strategies in to conserve biodiversity in the face of unprecedented anthropogenic disturbance.

The Centre for the Advanced Study of Collective Behaviour at the University of Konstanz and the newly founded Max Planck Institute of Animal Behavior in Konstanz and Radolfzell are delighted and honored to have the opportunity to invite you all to beautiful Lake Constance at such an exciting time. We very much look forward to a dynamic, fun and productive meeting!

WEL

CO

ME

Iain D. Couzin

INSTRUCTIONS TO TALK PRESENTERS

Regular oral presentations will be 10 min long with 3 additional minutes for discussions and then 2 minutes for changing rooms. All rooms will have both Mac and PC connected to the presentation system with a current version of Powerpoint and Keynote. You will not be able to connect your own device for a contributed talk. It is highly recommended that presenters bring their talk on a USB as backup to their presentation. Don’t forget to embed media or have it in a folder with your presentation. Presentations should be formatted for 16:9 aspect ratio and labelled as: LASTNAME-ASAB-2019

Presentations must be uploaded at least 4 hours before your talk. Upload and testing can take place at the following times:

Monday, 26 August 8:00-18:00 (in A703) and 13:00-17:00 (in the Audimax)Tuesday, 27 August at 7:30-8:30 and 12:30-13:10 and 16:40-18:40 in your allocated presentation room Wednesday, 27 August at 7:30-8:30 and 12:50-14:00 in your allocated presentation room

INSTRUCTIONS TO POSTER PRESENTERS

All posters will be displayed in the foyer of the main “A” building. Assigned poster numbers are in the Index available online. Mounting materials and no Twitter signs will be provided onsite. Presenters should stand by their poster at the session on Tuesday 27 August, according to their number: evens from 16:40-17:25, and odds from 17:25-18:15.

REGISTRATION

The conference starts on Monday 26 August at the University of Konstanz with registration from 8am. The registration desk is in the main “A” building and there will be plenty of signs to direct you from the entrance. Transportation to the university is detailed below. The desk will be open all day Monday (until 18:00) and every morning (8-9am) on Tuesday and Wednesday. Last minute registration is possible via cash.

PROGRAM

You can currently access an Overview Schedule and Talk Schedule (subject to change) as well as Talk Abstracts, Poster Abstracts and Index of Presenters at uni-konstanz.de/asab-summer-2019/schedule

WIFI Username: asab2019 Password: WIVNRoC8

INSTR

UC

TION

S

https://www.uni-konstanz.de/asab-summer-2019/schedule/

LIVE STREAMING

In order to encourage open science, the presentations of plenary sessions who have given their permission will be recorded and broadcast by live stream at uni-konstanz.de/asab-summer-2019

STAFF & VOLUNTEERS

Besides registration and info desk staff, there are conference volunteers who can be identified by blue “STAFF” badges at the venue. Feel free to ask them for any assistance you may need.

BUS PASS

Delegates will receive complimentary bus passes from the City of Konstanz (tickets provided at registration). Tickets allows you to use any local bus services in Konstanz without limit from Monday 26th until Wednesday 28 August.

LUNCHES AND COFFEE

A buffet lunch is included on Tuesday 27 and Wednesday 28 as a part of the conference registration fee. Lunch is served in the “Mensa” cafeteria and seating is in a reserved space “K7” or anywhere in the cafeteria. Coffee/tea and snacks are also served during breaks (see schedule).

SOCIAL EVENTS

Welcome Reception: Monday, 26 August

Following the official opening plenary, a welcome reception will be held in an event space in the main "A" building. Complimentary finger food and drinks will be served until 21:00, after which the party will continue informally in beautiful Konstanz town.

Mainau: Tuesday, 27 August

We are hosting an excursion to the Mainau “flower” Island on day two of the conference. Entry is free of charge but requires showing your name badge to the reception desk. We will be walking together from the university (20 minutes) so please let an orgainser know if you require special assistance. Also, pack wet weather gear to ensure that you don’t miss out on experiencing this top attraction. Please note that restaurants are available on the island and are at one’s own expense.

Conference Dinner: Wednesday, 28 August

Last but by no means the least, the official conclusion of ASAB Summer 2019 will be a banquet dinner held in arguably the most important building in Konstanz: the “Konzil”. Dating back to the 1400s, the structure was the site of the Council of Konstanz (1414-1418), a convening for the election of a pope that became the largest congress of the Middle Ages .

600 years later, the venue remains an ideal convention address. A pre-dinner drink will be served from 18:30 on the Konzil’s generous terrace overlooking Lake Constance (one complimentary sparkling wine or soft drink and further drinks available for purchase). Dinner starts at 19:30 and includes a buffet with complimentary drinks until 23:00 (further drinks can be purchased from the bar). The ASAB Medal will be presented during the dinner and a DJ will take over from 22:00 to 01:00 when the event concludes. Tickets are pre-purchased and your name badge must be presented to enter. Dinner tickets can be purchased at the registration desk.

INSTR

UC

TION

S

https://www.uni-konstanz.de/asab-summer-2019/

ACK

NO

WLED

GM

ENTS

ASAB Summer 2019 Overview Schedule 26 – 28 August, University of Konstanz, Germany

– uni.kn/asab-summer-2019Workshops Plenaries Sessions

Monday, 26 August

08:00 – 09:00 Registration and welcome coffee/tea

09:00 –12:00 Workshop: Machine learning in the behavioural sciences

12:00 –13:00 Lunch

13:00  Workshop: Quantifying behavior with machine-learning13:00 –17:00

Workshop: Low-cost automa-tion of behavioral experiments with rasp-berry pi’s13:00 –16:00

Workshop: Recording and analy-zing acoustic data13:00 –17:00

17:00 –18:00 Coffee and tea break

18:00 –18.50 Welcome & Opening Plenary LectureNachum Ulanovsky: “Neural codes for natural behaviors in flying bats”

18:50 –21:00 Welcome Reception

Tuesday, 27 August

08:00 – 08:50 Registration and coffee/tea

08:50 – 09:00 Morning Update

09:00 – 09:40 Plenary Lecture 2Dora Biro: “Collective action at the cognition-culture interface”

09:40 – 10:00 Coffee and tea break

10:00 – 11:15 Session 1

11:15 – 11:45 Coffee and tea break

11:45 – 12:25 Plenary Lecture 3Jason Kerr: “Imaging brain and behaviour in freely moving mammals”Introduction by: Iain Couzin

12:25 –13:40 Lunch

13:40 –14:55 Session 2

14:55 –15:25 Coffee and tea break

15:25 –16:40 Session 3

16:40 –18:15 Poster Session

from 18:15 Guided groups depart to Mainau flower island

Wednesday, 28 August

08:00 – 08:50 Registration and coffee/tea

08:50 – 09:00 Morning Update

09:00 – 09:40 Plenary Lecture 4Mary Stoddard: “Bird’s-eye View: How Colour, Court-ship and Deception Shape the Avian Visual World”

9:40 – 10:00 Coffee and tea break

10:00 – 11:00 Session 4

11:00 – 11:30 Coffee and tea break

11.30 – 12:45 Session 5

12:45 –14:20 Lunch

13:20 –14:20 ASAB AGM

14:20 –15:00 Plenary Lecture 5Meg Crofoot: “Science of the Sociome: Tracking how interactions scale to complex societies”

15:00 –15:20 Coffee and tea break

15:20 –16:20 Session 6

16:20 –16:40 Coffee and tea break

16:40 –17:40 Session 7

17:40 – 18:00 Prize Presentation and Closing Words

18:00 – 18:30 Free

18:30 – 19:30 Apéro at Konzil (dinner ticket)

19:30 – 01:00 Conference dinner and ASAB medal presentation

08

:00

 – 1

8:0

0

Reg

istr

atio

n &

Tal

k U

ploa

d

Oral presentation abstracts 1SocialdynamicsinabasalprimatewithfacultativesocialityLluisSocias-Martínez1,2,PeterM.Kappeler1,21BehaviouralEcologyandSociobiologyUnit,DeutschesPrimatenzentrumGmbH,Germany,2DepartmentofSociobiology/Anthropology,Johann-Friedrich-BlumenbachInstituteofZoology,Georg-AugustUniversity,GermanyTheoriginsofsocialityremainwidelydebatedinbothvertebratesandinvertebrates.Amongsuggestionsforadvancingourunderstandinginthismatter,thestudyoffacultativelysocialspeciesiskey.Weinvestigatesocialdynamicsinawildpopulationofafacultativelysocialprimatespecies,thegreymouselemur(Microcebusmurinus).Inthisnocturnalsolitaryforager,nestsharingcorrespondswithcommunalbreeding,allocareandpredatormobbing.Wemonitoredsleepingassociationsof250individualsfortwoconsecutiveyearsin120nests.Preliminaryanalysesindicatethepresenceofatleastthreedifferentsocialstrategiesinbothmalesandfemales.Someindividualswereneverobservedsleepinginthecompanyofothers,asecondtypeappearstoshowstabilityinassociationsandalownumberofpartners,whileathirdtypeshowshighconnectivitywithotherthirdtypeindividuals.Infurtheranalyses,weimplementsocialnetworkmetricstoexploreassociationsdynamicsforthatsubsetofthepopulationshowinghighconnectivityandflexibilityinpartnerchoice.Ourstudymayofferaglimpseintothecharacteristicsofinitialstagesofprimatesocialityandcontributetounravellingthemechanismsandecologicalcontextsthatcontributedtoitsspreadandmaintenance.

2ParallelevolutionoftermitetunnelingwithdifferentiatedbehavioralrulesNobuakiMizumoto1,PaulM.Bardunias2,StephenC.Pratt11ArizonaStateUniversity,UnitedStates,2SUNYCollegeofEnvironmentalScienceandForestry,UnitedStatesCollectivebehaviorsbygroupinganimalsformvariousspatialpatterns,includingcoordinatedmotionandneststructures.Thesepatternsoftenvarywithinoramongspecies,whereakeyquestionishowdifferentgroup-levelpatternsemergefromindividualresponsesandtheirinteractions.Intermites,tunnelingthroughthesoilbycollectiveexcavationhasevolvedseveraltimes;howeverlittleisknownaboutthebehavioralmechanismsunderlyingthepatternformation.Hereweshowconvergentevolutionoftunnelingbehaviorsintermitesaccompaniedbydifferentiationofbehavioralrulesforcollectiveexcavation.WefoundnoveldiggingbehaviorinParaneotermessimplicicornis(Kalotermitidae),whousetheirlegstokickbacksandparticles,contrastingwithwell-knownbehaviorofReticulitermestibialisorHeterotermesaureus(Rhinotermitidae)whocarrysandwiththeirjaws.Inspiteofthisvariation,P.simplicicornisandR.tibialisshowedlessbranchingoftunnels,whileH.aureusbuiltmorebranchedtunnels.WeattributethistothehigherfrequencyofbehaviorbyH.aureusinwhichtheyexcavatesidewallsinsidecloggedtunnels,whichweconfirmedbydata-basedsimulations.Theseresultssuggestthatdifferentbehavioralrepertoirescanproducesimilartunnelingstructures,wheresmallmodificationofthebehavioralparameterisimportanttodeterminethepatterns.Basedonthesedataandongoingstudiesofbasalspecies,wediscusstheevolutionaryprocessoftermitecollectivebuilding.

3PhenotypicbehaviouralvariationatdifferenthierarchicallevelsduringrapidadaptationfromfluctuatingtostableenvironmentsŽigaFišer1,SimonaKralj-Fišer2,3,Gergely,Balázs4,GergelyHorváth4,CeneFišer1*,GáborHerzeg4*1DepartmentofBiology,BiotechnicalFaculty,UniversityofLjubljana,Slovenia,2InstituteofBiology,ResearchCentreoftheSlovenianAcademyofSciencesandArts,Slovenia,3FacultyofMathematics,NaturalSciencesandInformationTechnologies,UniversityofPrimorska,Slovenia,4BehaviouralEcologyGroup,DepartmentofSystematicZoologyandEcology,EötvösLorándUniversity,Hungary,*co-seniorauthorUnderstandingphenotypicvariationinthewildhasbeenacentraltopicofbiologysinceDarwinrecognizeditasafundamentalprerequisiteforevolutionbynaturalselection.However,explainingmechanismsresponsibleforwithin-populationmaintenanceofphenotypicvariationinfitness-relatedtraitsisstillamajorchallenge.Muchofpreviousworkexaminedwhethertemporallyandspatiallyfluctuatingselectionarisingfromheterogeneousenvironmentsincreasesphenotypicvariation.Thecomplementaryperspectivepostulatingthatstabilizingselectiondecreasesphenotypicvariationwhenorganismsadapttohomogeneousenvironmentshasreceivedsurprisinglylittleattention.Weproposecaveandsurfaceenvironmentsasanidealcomparativemodelsystemtotestthishypothesis,asvariationofmostenvironmentalfactorsisstronglydiminishedincavescomparedtothesurface.Additionally,labilequantitativebehavioursseempromisingtraitstoinvestigatetheevolutionofphenotypicvariationastheycanbeinspectedacrossdifferenthierarchicallevelsincludingthewithin-individuallevel.Weappliedthislogicandrepeatedlymeasuredindividual’sbehaviourinfamiliar(6x)andunfamiliar(2x)environmentsinfourcave-surfacepopulationpairsrepresentingindependentcavecolonisationsofthefreshwatercrustaceanAsellusaquaticus,anemergingmodelorganismforrapidevolutionaryadaptation.Tocontrolforsexandlightregimeinducedvariation,wecollected30malesandfemalesperpopulationandrandomlyassignedthemintotwogroupsthatwerealternatinglyacclimatizedtodarknessanddiurnallightcyclepriorvideo-recordingtheirbehaviourinthesamelightconditions.Preliminaryanalysessuggestchangesatdifferentlevelsofphenotypicbehaviouralvariationassociatedwithcaveadaptation.Weexpectourstudywillprovidenovelinsightsonanoldquestion.

4Bet-hedgingacrossgenerationscanaffecttheevolutionofvariance-sensitivestrategieswithingenerationsThomasRayHaaland1,2,JonathanWright1,IrjaIdaRatikainen11CentreforBiodiversityDynamics,DepartmentofBiology,NorwegianUniversityofScienceandTechnology,Norway,2DepartmentofEvolutionaryBiologyandEnvironmentalStudies,UniversityofZürich,SwitzerlandInordertounderstandhoworganismscopewithongoingchangesinenvironmentalvariabilityitisimportanttoconsideralltypesofadaptationstoenvironmentaluncertaintyondifferenttime-scales.Conservativebet-hedgingrepresentsalong-termgenotype-levelstrategythatmaximizeslineagegeometricmeanfitnessinstochasticenvironmentsbydecreasingindividualfitnessvariance,despitealsoloweringarithmeticmeanfitness.Meanwhile,variance-prone(akarisk-prone)strategiesproducegreatervarianceinshort-termpayoffsbecausethisincreasesexpectedarithmeticmeanfitnessiftherelationshipbetweenpayoffsandfitnessisaccelerating.Usingtwoevolutionarysimulationmodels,weinvestigatewhetherselectionforsuchvariance-pronestrategiesarecounteractedbyselectionforbet-hedgingthatworkstoadaptivelyreducefitnessvariance.Wepredictthatvariance-pronestrategieswillbefavoredinscenarioswithmoredecisioneventsperlifetimeandwhenfitnessaccumulatesadditivelyratherthanmultiplicatively.Inourmodelvariance-pronenessevolvedinfine-grainedenvironments(withlowercorrelationsamongindividualsinenergeticstateand/orinpayoffswhenchoosingthevariabledecision),andwithlargernumbersofindependentdecisioneventsoverwhichresourcesaccumulatepriortoselection.Incontrast,geometricfitnessaccumulationcausedbycoarserenvironmentalgrainandfewerdecisioneventspriortoselectionfavorsconservativebet-hedgingviagreatervariance-aversion.Wediscussexamplesofvariance-sensitivestrategiesinoptimalforaging,migration,lifehistoriesandcooperativebreedinginlightoftheseresultsconcerningbet-hedging.Bylinkingdisparatefieldsofresearchstudyingadaptationstovariableenvironmentsweshouldbemoreabletounderstandtheeffectsinnatureofhuman-inducedrapidenvironmentalchange.

5TheroleoftheepigenomeinplasticresponsestorapidlychangingsocialenvironmentsJamesRouse1,ThomasLeech2,ElizabethDuncan1,AmandaBretman11BiologicalSciences,UniversityofLeeds,UnitedKingdom,2DepartmentPartridge,MaxPlanckInstituteforBiologyofAgeing,GermanySocialenvironmentscanbeverydynamicandhaveimportantconsequencesforfitness.Forexample,malesusesocialinformationtopredicttheamountofmatingcompetitiontheywillfaceandadjustinvestmentinparticularmatingopportunitiesaccordingly.Toaccuratelymatchtheirreproductivestrategytothisfluctuatingenvironmentrequiresplasticitythatisfastactingandreversible.Whilsttheroleoftheepigenomeinresponsestoenvironmentalchangeisbecomingmoreunderstood,ithasbeensuggestedthatitisnotinvokedinsuchflexibleplasticity.WesoughttotestthisusingaDrosophilamelanogasterfruitflymodel.Malefliesadjusttheirmatingdurationandejaculatecompositiondependingonthelevelofspermcompetitionsignalledbyexposuretorivalmalesbeforemating.Thebehaviouralcomponentofthischangeoccursquickly(under24h)andisentirelyreversible.Aprevioustranscriptomestudyshowedthattheexpressionofsomeepigeneticmodifiersissensitivetoexposuretorivalmales.Here,weusedchemicalinhibitorsandRNAitotestwhetherepigeneticremodellingisrequiredtoachievethisplasticresponse.Wefoundhistonedeacetylationiscrucialtoamale'sabilitytoplasticallyrespondtoincreasedspermcompetition,butinatissuespecificmanner.WearenowinvestigatingwhichlociaretargetedbyepigeneticremodellinginresponsetoplasticityusingChIP-seq.Overall,thissuggestsepigeneticremodellingisanimportantmechanisminshort-term,reversibleplasticphenotypes,andmustbeconsideredwhenexploringadaptationstofluctuatingenvironments.

6Istherereciprocalcooperationinnon-humanprimates?ManonK.Schweinfurth1,JosepCall11SchoolofPsychologyNeuroscience,UniversityofStAndrews,UKReciprocalcooperation,i.e.helpingthosethatwerehelpfulbefore,isaubiquitousandimportanttraitofhumansociality.Still,theevolutionaryoriginofthisbehaviourislargelyunclear,mainlybecauseitisbelievedthatourclosestlivingrelatives,otherprimates,donotexchangehelpreciprocally.Consequently,reciprocityissuggestedtohaveevolvedinthehumanlineageonly.However,recentfindingschallengethisviewbydemonstratingreciprocityinanimalsdistantlyrelatedtous,suchasratsandbats.Therefore,wesystematicallyreviewedstudiesinvestigatingreciprocityinnon-humanprimates.Contrarytocommonbelief,therearesignificantlymorepositivethannegativefindingsinbothexperimentalandobservationalstudies.Athoroughanalysisofthefindingsshowedthatreciprocityis,forexample,notconfinedtounrelatedindividuals,butthatthechoiceofcommoditiescanimpactthelikelihoodofreciprocation.Weconcludethatreciprocalcooperationinnon-humanprimatesispresentbutlargelyneglectedandnotrestrictedtohumans.Inordertodeepenourunderstandingoftheevolutionaryoriginsofreciprocity,futurestudiesshouldinvestigatewhenandhowreciprocityinnon-humananimalsemergedandhowitismaintained.

7DirectandindirectbenefitsasdriversofcomplexgroupstructureJoachimG.Frommen1,DarioJosi1,FrankGroenewoud1,2,JanaFlury1,3,FabianHeussler1,MichaelTaborsky11UniversityofBern,Switzerland,2UniversityofCambridge,UnitedKingdom,3ForschungsmuseumA.Koenig,Bonn,GermanyCooperativebreeding,i.e.individualshelpingothersintheirbroodcare,isamongthemostderivedsocialbehaviors.Itcanbeexplainedbyhelpersgainingindirectfitnessbenefitsthroughincreasingthesurvivalofrelatedindividuals.However,indirectfitnessbenefitscannotexplainwhyunrelatedindividualshelpothersinraisingoffspring.Here,directbenefitsareofimportance.Protectionfrompredatorsissuchdirectbenefitandhasbeenacknowledgedasamajordrivingforceofsociality.Still,howsuchriskrelateddirectbenefitsinteractwithindirectfitnessbenefitsindrivingtheevolutionofcomplexcooperativesocietiesislimited.WeinvestigatedthisinterplayinthecooperativelybreedingcichlidNeolamprologuspulcher.Wemeasuredgroupstructure,helpingbehavior,relatednessandreproductivesuccessineightpopulations,differinginpredationrisk.Groupstructurerelatedtopredationrisk,withgroupsinhighriskpopulationscontainingmorelargehelpersthatengagemostinpredatordefense.Inthesepopulationsthenumberoflargehelpershadastrongeffectonthebreeder'schancetoreproduce.Microsatelliteanalysesrevealedthatthedegreeofwithin-grouprelatednesswasgenerallylow.Notably,smallerhelpers,whichinvestleastindefense,weremorerelatedtobreedersinhighriskpopulations.TheseresultsindicatethatdirectfitnessbenefitsplayacrucialroleinthecooperativesystemofN.pulcher,whichcanbefurthermodifiedbyindirectfitnessgainsdependingonthepredatoryenvironment.Ourworkhighlightstheimportancetounderstandtheinterplayofdirectandindirectbenefitswhenaimingtocomprehendtheevolutionofcomplexanimalsocieties.

8Relatedness,socialstructureandhelpingbehaviourinthecooperativelybreedingcichlidNeolamprologussavoryiDarioJosi1,DikHeg1,TomohiroTakeyama2,DanielleBonfils1,DmitryA.Konovalov3,JoachimG.Frommen1,MasanoriKohda4,MichaelTaborsky11UniversityofBern,Switzerland,2OkayamaUniversity,Japan,3JamesCookUniversity,Austrailia,4OsakaCityUniversity,JapanTheevolutionarymechanismsunderlyingcooperativesocieties,wherebreedersandotherindividualscollaborateinraisingoffspring,poseanintriguingchallenge.Kinselectionisthoughttoplayanimportantroleinthedevelopmentandstabilityofsuchcooperativegroups,butmanyspeciesexhibitacomplexwithin-grouprelatednessstructure,whereindirectfitnessbenefitsalonecannotexplainhelpingbehaviour.TheLakeTanganyikacichlidNeolamprologussavoryihasbeenshowntobreedincooperativeharems,whereinuptofourbreedingfemalesformsubgroupsandmaybeassistedbyhelpers.Therelatednessstructureofthesegroupsishithertounknown,whichprecludesunderstandingoftheselectionmechanismsunderlyingapparentlyaltruisticalloparentalcare.Herewepresentthegeneticrelatednesspatternsof43groupsfromtwopopulationscontaining578individuals,using10to13microsatelliteDNAmarkers.Helpersweresignificantlymorerelatedtothebreedingmalethantothebreedingfemale.Withinsubgroups,breedertohelperandhelper-to-helperrelatednessdeclinedwithincreasinghelperage.Comparisonbetweensub-groupswithinamaleharemrevealedthathelpersweremorerelatedtothebreedingfemaletheywereassistingthantoneighbouringfemales.Immigrantsaccountedfor16.5%ofthegroupmembers,whilepatrilinealandmatrilinealinheritancewerecommonduetorapidbreederturnover.Interestingly,femaleinheritancewasmorelikelyinlargethaninsmallharems.Breederandhelperworkloadsignificantlyincreasedwiththenumberofyoungproducedwithintheharem.Ourresultshighlighttheimportanceofbothdirectandindirectfitnessbenefitsinacooperativesocietycomprisingcomplexsocialandrelatednessstructures.

9IsotocinreceptorexpressionandsocialexperiencesintheTrinidadianguppy(Poeciliareticulata)SylviaDimitriadou1,EduardaSantos1,DarrenP.Croft1,SafiK.Darden11CentreforResearchinAnimalBehaviour,UniversityofExeter,UnitedKingdom,2Biosciences,UniversityofExeterAcrosstaxaindividualscooperate,payingcostssootherscanbenefit.Whileresearchhasfocusedonthebehaviouralstrategiesunderlyingtheemergenceandmaintenanceofcooperation,littleisknownabouttheproximatemechanismsunderpinningit.Nonapeptides,suchasoxytocin,andtheirassociatedreceptorsarethoughttobeimportantregulatorsofprosocialbehaviours,includingcooperation.Here,weusetheTrinidadianguppy(Poeciliareticulata),toexploretheneuroregulatoryresponseofexperiencingcooperationordefectionfromthesocialenvironment,byexamininggeneexpressionfortheisotocinreceptor(itr=homologoustothemammalianoxytocinreceptor).WetestedfemalesfromaHighPredation(HP)andaLowPredation(LP)siteoftheGuanaporiverontheislandofTrinidadinapredatorinspectionparadigm,manipulatingwhethersocialpartnersostensiblycooperatedordefectedduringinspection.HPfishoriginatefromahabitatabundantinpredatorsandarethereforeexperiencedinpredatorexposure,comparedtofishoriginatingfromLPhabitats.Therelativeexpressionoftheitrgeneinbrainmidsectionwasthenquantified.WefoundthatinHPfishexposedtoapredator,ostensiblyexperiencingdefectionfromthesocialenvironmentledto1.5*10-1-foldhigheritrmidsectionrelativeexpressionthanexperiencingcooperation.Intheabsenceofapredator,ostensiblyexperiencingdefectionledto1.3*10-1-foldloweritrrelativeexpressioncomparedtocooperation.Wefoundnodifferenceinitrrelativeexpressionintheinexperiencedpopulation(LP).Ourfindingsdemonstratetheeffectsofcooperativeexperiencesonbrainnonapeptidereceptorgeneexpressionpatterns,andprovideinsightintotheneuromodulatorymechanismsunderlyingcooperativebehaviour.

10Antactivity-restrhythmsvarywithageandinteractionfrequenciesofworkersHarunaFujioka1,MasatoS.Abe2,YasukazuOkada31GraduateSchoolofArtsandSciences,TheUniversityofTokyo,Japan,2RIKENCenterforAdvancedIntelligenceProject,Japan,3DepartmentofBiologicalSciences,TokyoMetropolitanUniversity,JapanSocialinsectcoloniesarehighlyorganizedsystems,wherecommunicationamongnestmates(i.e.socialinteractions)hasapivotalfunctionforcolonialorganization.Inordertofurthertheunderstandingofsocialorganization,thechronobiologicalsystemofsocialinsectspecies,particularlytheircircadianrhythm,hasrecentlyattractedmuchattention.However,gapsstillremaininourunderstandingofhowindividualactive/restrhythmsaregovernedinvarioussocialcontexts.Inthisstudy,weinvestigatetheeffectsofworker-workerinteractionsoncircadianactivityrhythms,usingthemonomorphicant,Diacammasp.Continuoustrackingofsolitaryantselucidatedcircadianactivityrhythms,bothinyoungandoldworkers.Thecolor-tagbasedautomatictrackingofmultipleworkersrevealedthatyoungoldinteractionsreducedcircadianrhythmicactivitiesinbothyoungandoldworkers,whereasyoungworkersretainedactive/restrhythmsunderyoung-youngworkerinteractions.Togetherwiththeanalysesofworker-workerinteractionfrequencies,weconcludethatinteractionsbetweenworkersindifferentage-groups(i.e.workerswithdifferenttasks)functionasdifferentcuestoalterworkeractive/restpatterns.Wediscussthepotentialrolesofworker-workerinteractionsonthechronobiologicalorganizationoftheantsociety.

11Flappingformigrating:energyexpenditurequantificationGennaroVitucci1,RenaudRonsse1,PhilippeChatelain11UniversitécatholiquedeLouvainModernhigh-resolutiontechnologyallowstrackingmigratingbirdsoverlongdistances.Dependingonspecies,durationofdisplacement,environmentalconditions,socialorganization,etc.,differentflocktopologyandflightdynamicsarise.Twomainflyingstylesaretypicallyused:soaring/glidingandflappingflight.Theformer'sperformancehasbeenstudiedindetailsforsinglebirdsasitisachievedwithfixedwings,whosemodellingreliesontheaeronautictradition.Soaring/glidingischeap,asitharvestsalltheenergyfromupwards'thermals'.Socialimplicationsfortheflockhaverecentlybeenpointedout.Nevertheless,arangeofcircumstancesmakesthechoiceofflappingflightapreferableorunavoidableoption:absenceofnearthermals,sizeoftheanimal,andstricttimerequirements,e.g.seasonalavailabilityoffoodatdestination.Ourcurrentworkfocusesonflappingflight.Inparticular,weintendtounveilobservedflockspatio-temporalsynchronizationthroughthequantificationofthemechanicalpowerrequiredforflyingingroup-andrelateddisparityinenergyconsumptionbetweenleaderandfollower.Forthispurpose,wedevelopedanovelaerodynamicmodelthatachievesatrade-offbetweencomputationalaffordabilityforalargenumberofbirdsandessentialfeaturesofcollectiveflappingflight:3Dmotion,bird-wakeinteractionandunsteadiness.Stableconfigurations,representativeofformationflight,andtransientonespredictingthecostofmaneuverareanalysed.Whereavailable,comparisonwithfield-dataarereported.Finally,applicationofourcheapcomputationaltechniqueinlightofdecision-makingstrategiesimplementationareprospected.Acknowledgements:ThisworkhasreceivedfundingfromtheBelgianJointResearchActivityRevealFlight.

12IndividualtrackingrevealsconsistentheadingsinmigratinghawkmothsMylesH.M.Menz1,2,MartinWikelski1,21MaxPlanckInstituteofAnimalBehavior,Germany,2UniversityofKonstanz,GermanyInsectsarethemostdiverseandabundantgroupofterrestrialmigrants.However,incomparisontovertebrates,themigratorypathwaysofinsectsandbehaviourenroutearerelativelypoorlyunderstood.Whileindividualtrackingcanallowforinsectstobemonitoredduringmigration,thistechniquehasrarelybeenapplied,primarilyduetomostinsectspeciesbeingtoosmalltocarryradio-transmitters.Herewepresentthefirststudytoindividuallytracknocturnalmigratinginsects,toidentifymigratoryroutesandunderstandfactorsaffectingflightbehaviour.UsingaCessna172aeroplane,wetrackedindividualhawkmoths(Acherontiaatropos,Sphingidae),fittedwith0.25gradio-transmitters.Duringtracking,allmothsmaintainedaconsistentheading,whichpersistedacrossdaysinsomeindividuals.However,therewasvariationinheadingsamongindividuals.Furthermore,weanalysedwinddatatodetermineifmothscompensatedforwinddriftinordertomaintainconsistentheadings.

13Closer-to-home'strategybenefitsjuvenilesurvivalinalong-distancemigratorybirdYachangCheng1,2,WolfgangFiedler1,2,MartinWikelski1,2,3,AndreaFlack1,2,31DepartmentofMigrationandImmuno-Ecology,MaxPlanckInstituteforOrnithology,Germany,2DepartmentofBiology,UniversityofKonstanz,Germany,3CentrefortheAdvancedStudyofCollectiveBehaviour,UniversityofKonstanz,GermanyHuman-inducedchangesinclimateandenvironmentarechallengingtheexistenceofmigratoryspecies.Specieswithdiverseandflexiblemigratorybehavioursufferlessfrompopulationdecline,astheyaremorecapabletoresponsebyalteringmigratorybehaviour.Attheindividual-level,variationsinmigratorybehaviourmayleadtodifferencesinfitnessandsubsequentlyinfluencedemographicdynamics.UsinglifetimeGPSbio-loggingdatafrom169whitestorks(Ciconiaciconia),weanswerwhethertheirrecentlyshortenedmigrationhassurvivalbenefitduringthejuvenilestage,theriskiestlifeperiodformanymigrants.Wealsoexplorehowothervariationsinmigratorydecisions(i.e.time,destination),movementactivity(measuredbytheoverallbodydynamicacceleration),andearlylifeconditionsinfluencejuveniles'survival.Weobservedthatfirstautumnmigrationwastheriskiestperiodforjuvenilewhitestorks.Individualsthatmigratedshorterdistancesandfledgedearlierexperiencedlowermortalityrisk.Inaddition,highermovementactivityandoverwintering'closer-to-home'inEuropeandNorthAfrica(84.21%oftrackedindividualsadoptedthisnewstrategy)wereassociatedwithhighersurvival.Ourstudyshowshowquicklyavianmigrantscanchangelifehistorydecisionslinkedtofitnessandthushelpsustounderstandandpredicthowmigrantsrespondtothechangingworld.

14Seasonalcontrastsinindividualconsistencyoforientalhoneybuzzards'migrationShokoSugasawa1,HiroyoshiHiguchi21CentreforBiologicalDiversity,SchoolofBiology,UniversityofStAndrews,UnitedKingdom,2ResearchandEducationCentreforNaturalSciences,KeioUniversity,JapanIndividualconsistency,orhowrepeatableindividualsare,inmigrationcanhelpusunderstandthemechanismsofmigration.Moststudiesreportedthatbirdsaremoreconsistentinthetimingthanintheroutesorstopoversitesduringmigration,butsomespecialistspeciesshowedtheoppositepatterns,beingmoreconsistentinspatialthantemporalaspectsofmigration.Onepossibleexplanationforthiscontrastisthatspecialistsrelyonparticularfoodorhabitatresources,whichrestrictthemigratoryroutestheycantake,leadingtohighspatialconsistency.Ifthisisthecase,theeffectofspecialistforagingshouldbecomeapparentonlywhenbirdsforage,insteadoffastingandflyingcontinuously.Totestthiseffect,weanalysedindividualconsistencyinmigrationoftheorientalhoneybuzzard(Pernisptilorhynchus),aspecialistraptorthatfeedsonhoneybeesandwasps,usingalong-termtrackingdataset.Ashoneybuzzardsmakeextendedstopoversduringwhichtheyforageinspringbutnotinautumn,thespatialconsistencyshouldbehigherinspringthaninautumn.HoneybuzzardswerehighlyconsistentinboththeirmigratoryroutesandstopoversitesinSoutheastAsia,butonlyduringspringmigration.Whilebirdsshowedsignificantrepeatabilityintimingofmigrationbothinautumnandspring,theseasonaldifferencewaslessconspicuouscomparedtothatinroutes.Ourresultshighlightanimportantlinkbetweenspecies'migratoryconsistencyandforagingecology.

15DifferentialuseofenergyavailableinthelandscapebytwosoaringbirdspeciesMartinaScacco1,2,AndreaFlack1,2,OlivierDuriez3,MartinWikelski1,2,KamranSafi1,21DepartmentofMigrationandImmuno-ecology,MaxPlanckInstituteforOrnithology,Germany,2DepartmentofBiology,UniversityofKonstanz,Germany,3Centred’EcologieFonctionnelleetEvolutive,UniversitédeMontpellier,FranceSoaringbirdsusetheenergyavailableintheenvironmentintheformofatmosphericuplifts,tosubsidizetheirflightandmoveacrossthelandscape.Theirmovementpatternisthereforeshapedbythespatialandtemporalavailabilityofuplifts,resultingfromaninteractionoflocalatmosphericconditionswiththeunderlyinglandscapestructure.Sofar,theenergyavailableinthelandscapeandthecostoftransportofsoaringbirdshavebeenrelatedtoatmosphericinformationonly.Herewecomparedtheaccuracyofstaticlandscapefeatures(topography,landcover)andcommonlyusedupliftestimators(basedonatmosphericinformation)inpredictingtheflightbehaviouroftwoobligatesoaringspecies,thewhitestorkCiconiaciconiaandthegriffonvultureGypsfulvus.Weusedsoaringandflappingflightlocationsof67individualsasindicativeofthepresenceandabsenceofuplifts.WefoundthatstaticlandscapefeaturesalonecanpredictandmaptheupliftsavailabletothetwospeciesacrossEurope.Bothspeciesstronglyrelyontheavailabilityofuplifts.However,theupliftavailabilitymapssuggestedspecies-specificdifferencesintheuseofthelandscapeandtheavailableenergy.Theseupliftavailabilitymapshighlighttheimportanceofconsideringinter-specificdifferences,eveninspecieswithsimilarflightbehaviour,whengeneralizingthecomplexrelationshipbetweenenvironmentandmovementpatterns.Thesemapsprovideabasetoexploretheeffectsthatchangesinthelandscapestructurehaveontheenergyexpenditurefordifferentsoaringspecies,allowidentifyinglow-costmovementcorridorsandcanultimatelyinformtheplanningofanthropogenicdevelopments.

16Life-longrepeatabilityandheritabilityofsexualdisplaybehaviorincaptive-bredAfricanhoubarabustardsEnricoSorato1,LoïcLesobre1,YvesHingrat11RenecoInternationalWildlifeconsultants,UnitedArabEmiratesBetween-individualvariationinbehaviour(i.e.personality)hasbeendescribedinawiderangeofanimaltaxa.However,personalitieshavebeenmostlyinvestigatedalongafewbehaviouralaxes,typicallyinvolvingactivity,exploration,boldnessandsociability,whileotherbehaviouraldomainshavebeenlargelyneglected.Notably,despitethedocumentedexistenceofdiscretealternativematingstrategiesacrossseveralspecies,typicallyeithercondition-dependentorundersimpleMendeliangeneticcontrol,fewstudieshaveinvestigatedcontinuousindividualdifferencesinsexualbehaviouranditsdegreeofquantitativepolygeniccontrol.Wethereforeassessedbetween-individualvariationinsexualbehaviour,anditsunderlayinggeneticversusenvironmentalcomponents,inacaptive-bredpopulationoftheNorthAfricanhoubarabustard(Chlamydotisundulataundulata),apromiscuouslong-livedspecies,whereinmalesdisplayinexplodedlekstoattractfemales.Usingalargesampleofmorethan1000000behaviouralobservationsacrossmorethan3000captivemales,andspanningaperiodof15years,wefoundlong-termindividualrepeatabilityintheextentofsexualdisplayinginisolationandintheoccurrenceofprecopulatorydisplaybehaviourtowardsadummyfemale,togetherwithindividualdifferencesintheseasonalexpressionofthesebehaviours,andage-dependentindividualtrajectories,includingvariationinsenescencepatterns.Wefurtherhighlightassociationsbetweenbehaviour,bodysizeandcondition.Among-individualvariationinthesetraitswasunderlaidbysignificantgeneticcomponents.Wediscussfactorsthatmayhavegeneratedandmaintainedheritablevariationinsexualbehaviour,andtherelevanceofthesefindingsforthefieldsofanimalpersonalityandsexualselection.

17Experimentallyflight-impairedfemalesshowhigherlevelsofextra-pairpaternityinapasserinebirdMireiaPlaza1,AlejandroCantarero2,DiegoGil1,JuanMoreno11EvolutiorayEcologyDepartment,NationalMuseumofNaturalSciences,Spain,2SectionofEcology,UniversityofTurku,FinlandThereisyetnoconsensusonthereasonswhyfemalesengageinextra-paircopulations(EPC)andalthoughinsomespeciestheymayaccrueindirectbenefits,theseeffectsarebynomeanscommon.Thesexualconflicthypothesispositsthatextra-pairpaternity(EPP)istheresultofstrongselectionformalepursuitofEPCwithoutrealbenefitsforfemales.Inordertotestthishypothesis,weexperimentallyreducedwingarea(reversiblytyingtogethersomeprimaryfeathers),therebyincreasingwingloading(bodymass/wingarea),whichisnegativelyassociatedwithflyingabilityandthus,withcapacitytoescapefromunwantedcopulations,inagroupofpiedflycatcherfemales(Ficedulahypoleuca).WecomparedEPPintheirbroodswiththoseinagroupofcontrolfemales.OurresultsshowedasignificantincreaseinEPPinnestsoftheexperimentaltreatmentcomparedwithcontrolnests.Theseresultssuggestthatinourstudypopulation,EPPcouldbepartlyaproductoffemalecapacitytoavoidEPCs.

18OntheevolutionofmonandrousmatingstrategyinmothsAllyHarari1,YardenDodi1,HadassSteinitz11DepartmentofEntomology,TheVolcaniCenter,IsraelInmostsexualbreedingspeciesfemalesarepolyandrousandthusmostfemalesmatemorethanonce,whereasintheminorityofspeciesfemalearemonandrous,wherebymostfemalesinthespeciesmateonceonly.Malesinbothmatingsystemstypicallymatemorethanonce.Duringmating,malemothstransfertofemalesaspermatophorefullwithnutrientsandtwotypesofsperm,eupyrene,thefertilizingspermthatisproducedinlimitedamounts,andapyrene,anonefertilizingsperm,thatisproducedinhighnumbersandisassumedtohavearoleinspermcompetition.TestingthecostsandbenefitsofrematinginthemonandrousmothLobesiabotrana,weappliedstrongselectinforpolyandry,producing70%ofrematingfemales,vs90%ofmonandrousfemalesinthewildtype.Afterafewgenerationswemeasuredthespermcontentintestesofvirginandmatedmalesandfoundthatsonsofpolyandrousfemaleshavehadanincreasedamountoftheapyrenespermandtheytransferredtofemalesmoreoftheapyrenespermthansonsofmonandrousfemales.Searchingforacostimposedonmalesandfemalesinthepolyandrouslinewecomparedfitnessparametersoffemalesandmalesinthetwoselectedlinesandmalepreferencetomonadrousandpolyandrousfemales.

19Theroleofmale-maleinteractionsinfemalereproductivedecisionsNinaBircher1,2,KeesvanOers1,2,CamillaHinde1,MarcNaguib11BehaviouralEcologyGroup,WageningenUniversityResearch,Netherlands,2DepartmentofAnimalEcology,NetherlandsInstituteofEcology,NetherlandsConspicuousmaleadvertisementssignalsplayakeyroleinfemalechoiceofsocialandcopulationpartners.Inspecieswheremalesengageinmale-maleinteractions,femalesmayeavesdroponsuchinteractionsandbeabletodirectlycomparecompetingmaleswithlittlecosts.However,littleisknownaboutwhatkindofinformationfemalesmaygatheraboutmaleswheneavesdropping,whetherfemaleschangetheirbehaviourinthelong-terminresponseandwhetherreproductivedecisionsbasedonmaleperformanceinsuchinteractionsleadtobenefitsforfemales.Hereweexploretheroleofeavesdroppingonmalesonginteractionsinapopulationofwildgreattits(Parusmajor)combiningdatafromarecorderarray,microsatellitegenotyping,cross-fosterexperiments,interactiveplaybacksandautomatedradiotracking.Wepresentresultsonthelong-termspatialresponsesoffemalestomaleinteractionsandmaletraitsthatmaybehonestlyindicatedbymaleperformanceduringsuchcontests,providingnewinsightsintotheroleofmalesonginteractionsforfemalereproductivedecisions.

20FitnessconsequencesoffemalealternativereproductivetactitsinhousemiceBarbaraKönig1,ManuelaFerrari1,AnnaK.Lindholm11UniversityofZurich,SwitzerlandAlternativereproductivetactics(ARTs)aredefinedasdiscretedifferencesinmorphological,physiologicaland/orbehaviouraltraitsassociatedwithreproduction,whichoccurwithinthesamesexandpopulation.Housemice(Musmusculusdomesticus)providearareexampleforARTsinfemales,whichcaneitherreartheiryoungsolitarily,ortogetherwithoneorseveralotherfemalesinacommunalnest.Weassessedthefitnessconsequencesofcommunalandsolitarybreedinginawildpopulationtounderstandhowthetwotacticscanbeevolutionarilystable.Femalesswitchedbetweenthetwotactics,pointingtowardsthembeingtwotacticswithinasinglestrategy.Communalbreedingresultedinreducedpupsurvivalandnegativelyimpactedfemalereproductivesuccess.Olderandlikelyheavierfemalesmoreoftenrearedtheirlitterssolitarily,indicatingthatfemalesuseaconditiondependentstrategy.Solitarybreedingseemsthemoresuccessfultacticandonlyyoungerandlikelylesscompetitivefemalesmightoptforcommunalnursing,evenatthecostofincreasedpupmortality.ThisstudyemphasizestheimportanceofanalysingphenotypicplasticityanditsroleincooperationinthecontextoffemaleARTs.

21Causesandconsequencesofbaboon(Papioursinus)urban-foraginginCapeTown,SouthAfricaA.Bracken1,J.M.O’Riain1,C.Christensen1,I.Fürtbauer1,A.J.King11InstituteforCommunitiesandWildlifeinAfrica,DepartmentofBiologicalSciences,UniversityofCapeTown,SouthAfrica,1DepartmentofBiosciences,CollegeofScience,SwanseaUniversity,UnitedKingdomHuman-inducedenvironmentalchangespresentbothopportunitiesandchallengesforwildlife.Whilemanyindividuals,populationsandspeciesarenegativelyaffectedbychange,somemodifytheirbehaviours(withintheirmorphologicalandphysiologicalconstraints)andtakeadvantageofnewopportunitiespresentedinhuman-changedlandscapes.'Urban-foraging'isonesuchbehaviouralmodification,enablingindividualstoexploithighenergy,human-derivedfoods(e.g.cropsandfood-waste),butthisbehaviourcanalsoresultinconservationconflictsandchallengesforthemanagementofwildanimalpopulations.IntheCapePeninsulaofSouthAfrica,chacmababoons(Papioursinus)liveinspatialoverlapwithhumansandcommonlyurban-forage:'raiding'fruitingtrees,residentialandcommercialproperties,eventakingfooddirectlyfrompeople.TheCityofCapeTownhiresamanagementprogrammethatemploysfield-rangerstoeffectively'herd'thebaboonsoutoftheurbanspace.Here,weinvestigatethecausesandconsequencesofurban-foragingforababoontrooplivingattheurbanedge.Wecombinebehaviouraldatafromcustom-builttrackingcollars(forn=16adults,recordinghighresolutionGPSandaccelerationdata)withnon-invasivelyassessedphysiologicalstresslevelsandnutritionalstate,andinformationonenvironmentalrisksandrewards.Wepresentpreliminaryfindingsbasedonindividualtrajectorydatacollectedat1Hzresolutionoverseveralmonths,andexplorethefactorsthatpredictthefrequencyanddurationofindividualurban-foraging.Finally,wediscussourfindingsinthecontextofcurrentmanagementpracticesforreducinghuman-baboonconflictonthePeninsula,andconsiderourresultsinunderstandinghowurban-foragingspeciesmanagethecostsandbenefitsoflivingalongsidehumans.

22Movementtrackingrevealsthatparrotsexploitcyclic,human-derivedfoodopportunitiesinurbanhabitatsGaelleFehlmann1,2,LucyAplin1,21CognitiveandCulturalecology,MaxPlanckInstituteforAnimalBehavior,Germany,2CentrefortheAdvancedStudyofCollectiveBehaviour,UniversityofKonstanz,GermanyForanimalsthatpartlyexploithuman-derivedfood,urbanenvironmentspresentacomplexfoodlandscape.Thisincludesbothgeographicalandtemporalaspects,suchastheflowofpeoplehavinglunchincityparksovertheweekendorprovisioningfromresidentialbalconiesbeforework.Predictingtheavailabilityofsuchresourcesintimeandspacerepresentscognitiveandnavigationalchallenges.Here,wepresentpioneeringworkthataimstouncovercognitiveprocessesunderlyingthenavigationalandforagingdecisionmakingofsulphur-crestedcockatoos(Cacatuagalerita),alargesocialparrotthathassuccessfullyinvadedorpersistedinhuman-modifiedhabitatsacrossAustralia.Usingover30,000recorddatafromthelongtermcitizenscienceproject'Wing-tags'wecreateadynamicrepresentationofcyclicfoodavailabilitiesarisingfromrecreationalbirdfeedingincentralSydney,andmatchthesewith2yearsofhighfrequencyGPSdatacollectedon8sulphurcrestedcockatoos.Birds'activitydailypatternscoincidewithcollectivepatternsemergingfromhumanactivities.Inaddition,birdsshowsitefidelityandregulardirectionalflightstospecificlocationsindenseurbanareas,suggestingknowledgeandmemoryofregular,geographicallyspecificforagingopportunities.Buildingonthesemaps,wediscusstheroleofcognitiveprocessessuchasmemoryandmentalmapswhennavigatingacrossurbanlandscapestofindfood.Ourworkrepresentsanovelrepresentationofthecitypulsefromahumanandabirdpointofview,andadvancesourunderstandingofhowcognitioncanfacilitatesomespeciestocoexistandinteractwithhumansinlargecities.

23ThegeometryofdecisionmakingVivekHariSridhar1,2,3,RenaudBastien1,2,3,PaulSzyszka4,NirS.Gov5,IainD.Couzin1,2,31DepartmentofCollectiveBehaviour,MaxPlanckInstituteofAnimalBehavior,Germany,2DepartmentofBiology,UniversityofKonstanz,Germany,3CentrefortheadvancedStudyofCollectiveBehaviour,UniversityofKonstanz,Germany,4DepartmentofZoology,UniversityofOtago,NewZealand,5DepartmentofChemicalPhysics,TheWeizmannInstituteofScience,IsraelFromgeneticandneuralnetworkstoantcoloniesandwildebeestherds,collectivebehaviourhasevolvedacrosslevelsofbiologicalcomplexity.Incontrasttomostartificialsystems,decentralisedcontrolisoftenasignatureofsuchsystems.Foranimalsinthenaturalworld,survivaloftendependsonthebehaviouralrulesthatindividualsadoptandthedecisionstheymakeinresponsetothelocationofconspecifics,resourcesandthreatsintheirenvironment.Mostofthedecision-makingliteratureinbothanimalbehaviourandneuroscienceislimitedtoindividualsexposedtoatwo-choicecontext.Butthisisnotnecessarilytrueoffreely-movinganimalsinthewild.Individualsmayencountermultipleoptionsandrewardsrelyontheanimalchoosingtherightoptionwithinreasonabletime.Myworkelucidateshowthebrainbreakssymmetrywhenfacedwithann-choicedecisionscenarioandhowthislinkstoconsensusdecision-making,previouslydescribedinfishschoolsandbaboontroops.Bylinkingdecision-makingatthesetwoscalesi.e.individualandgroup,Irevealunifyingprinciplesofdecision-makinginEuclideanspace.

24PlasticpollutionimpairsshellselectionbehaviourinhermitcrabsAndrewCrump1,CharlotteMullens1,EmilyJ.Bethel2,GarethArnott11InstituteforGlobalFoodSecurity,SchoolofBiologicalSciences,Queen’sUniversityBelfast,UnitedKingdom,2CentreforResearchinBrainandBehaviour,SchoolofNaturalSciencesandPsychology,LiverpoolJohnMooresUniversity,UnitedKingdomMicroplastics(plastics

25SensorybiologyinballooningspidersEricaL.Morley1UniversityofBristol,UnitedKingdomSomespidersandotherwinglessarthropods,suchascaterpillarsandspidermites,disperseaeriallyoverhundredsofkilometresbyballooning.Technicallyamisnomer,ballooninginvolvesthearthropodreleasingstrandsofsilkonwhichsufficientforcesacttoproviderapidliftandtakeoff.Airmovementfromwind,thermalsandtheirassociateddragforcescangeneratethelifttomaketheseanimalsair-borne,howeveranalternativehypothesisisthatelectrostaticforcescouldgeneratelift.Atmosphericelectricity,ortheatmosphericpotentialgradient(APG),ispresentatalltimesandvarieswithweatherpatterns;fromaround+120V/mondayswithclearskiestoupto±10kV/munderstormcloudsandmist.UnderecologicalconditionsspidersandotherarthropodswillbesubjecttobothairmovementsandelectricfieldsprovidedbytheAPG.Here,theabilityofspiderstodetectandrespondbehaviourallytoelectricfieldsistested.IshowthatspidersattempttoballooninresponsetoelectricfieldsaloneandIexplorehowwindande-fieldscaninteracttotriggerballooningbehaviour.Thediscoverythatterrestrialorganismscandetectelectricfieldsatatmosphericlevelsopensupanentirelynewfieldinsensorybiology,demandingtheuseoftechnologyfromthephysicalsciencestomakemeasurementspossible.Understandinghowballooningbehaviouristriggered,andinwhichecologicalconditions,willprovideanewtoolforunderstandingthedistributionandabundanceofarthropodspeciesthatuseballooningtodisperse.

26Altruisticbet-hedginginanaridzonecooperativebreederPabloCapilla-Lasheras1,AlastairJ.Wilson1,XavierHarrison2,EmmaWood1,AndrewJ.Young11CentreforEcologyandConservation,UniversityofExeter,UnitedKingdom,2GeoffreyPopeBuilding,UniversityofExeter,UnitedKingdom.Altruisticspeciesaregloballyassociatedwitharidandunpredictableenvironments,highlightingtheimportanceofenvironmentalconditionsasadriverfortheevolutionofaltruism.Recenttheoreticalworksuggeststhatwhentheenvironmentisunpredictable,altruisticbehaviourscanbeselectedtoreducevarianceinfitnessratherthanincreasesinmeanfitness,astraditionallythought.Thisnewhypothesis,so-calledaltruisticbet-hedging,providesanovelframeworktotestthemechanismsbywhichenvironmentalconditionspromotetheappearanceofaltruism.However,thepredictionsofaltruisticbet-hedgingtheoryhavenotbeenexplicitlytestedyet.Inthistalkweoutlineourresearchshowingthatenvironmentalandcooperativefactorsinwhite-browedsparrowweaversocietieshaveeffectsonreproductivesuccessthatcloselymatchthosepredictedbyaltruisticbet-hedgingtheory.Wefindthatthepresenceof(female)helpersreducesvarianceinreproductivesuccessbutnotitsmean.Furthermore,weexplicitlyshowthatthesedifferencesinreproductivesuccessareinpartexplainedbyhowthepresenceoffemalehelpersinteractswithenvironmentalconditions.Ourresultsproviderareevidencesuggestingthataltruisticbet-hedgingmayexplaincooperationinanaridzonebird.Furthermore,ourfindingsrevealhiddencostsofcooperationandhighlightthattheexpectedoveralleffectofcooperationonreproductivesuccesswilldependontherelativefrequencyofdifferentenvironmentalconditions.Variationintherelativefrequencyofenvironmentalconditionsduetogeographicalfactorsorclimatechangeisexpectedtomodifytheextenttowhichselectionforcooperationarisesfromeffectsonthemeanorvariance.

27BehaviouraltraitsthatdefinesocialdominancearethesamethatreducesocialinfluenceMRodriguez-Santiago1,PNührenberg2,3,JDerry1,ODeussen3,FFrancisco2,3,LKGarrison1,SGarza2,3,HAHofmann1,4,5,MWildner3,ALJordan1-31DepartmentofIntegrativeBiology,TheUniversityofTexasatAustin,UnitedStates,2DepartmentofCollectiveBehaviour,MaxPlanckInstituteofAnimalBehavior,Germany,3CentrefortheAdvancedStudyofCollectiveBehaviour,UniversityofKonstanz,Germany,4InstituteforCellularandMolecularBiology,TheUniversityofTexasatAustin,UnitedStates,5InstituteforNeuroscience,TheUniversityofTexasatAustin,UnitedStatesThetraitsthatdefinesocialdominanceinmanysocieties–aggression,coercion,andinfluenceovergroupdecisions–maybetheverytraitsthatreduceinfluenceinothercontexts.Hereweexaminethebehaviorandinfluenceofdominantandsubordinatemalecichlidfishindifferentcontexts.Wefindthatunderstandardsocialconditions,dominantmalesarethemostaggressivegroupmembersandconcomitantlyhavethegreatestinfluenceovergroupbehavior.However,inamorecomplexgroupscenario–respondingtoasimpleassociationtask–dominantmaleshavetheweakestinfluenceovergroupbehavior.Insteadsubordinatemales,withlowaggressionandlittleinfluenceovertypicalgroupbehavior,havethegreatestinfluenceovertheirgroupsinthecontextoftheassociationtask.Althoughsubordinatemaleshadlittlebehavioralconnectionwithothergroupmembers,theywerehighlyconnectedbothspatiallyandvisually–likelyimportanttypesofconnectionwhenrespondingtoavisualtask.Moreover,thebehavioralstimulithatcausedgroupmovementshaddifferentsignal-to-noiseratiosfordominantandsubordinatemales.Responsetotheassociationtaskwascharacterizedbyarapid,directionalswimtowardthelightcue,whichwaskinematicallydifferentfromnormalswimming,butsimilartotheaversivechasingfrequentlydisplayedbydominantmales.Incontrast,subordinatemalesonlydisplayedthisbehaviorintheassociationtask.Weconcludethatitwastheverybehaviorsthatmadedominantmalesinfluentialinonecontextthatcausedthemtolackinfluenceinanother.

28Epigeneticsandenvironmentalchange:adaptationversusconstraintDustinRubenstein11ColumbiaUniversity,DepartmentofEcology,EvolutionandEnvironmentalBiology,UnitedStatesIforganismsaretopersistinthefaceofclimatechange,theymustbeabletodealnotonlywithincreasingtemperatures,butalsogreaterclimaticvariation.Oneoftheprimarywaysanimalscopewithenvironmentalchangeisthroughphenotypicplasticity,theabilitytorespondtoenvironmentalcuesthroughphenotypicadjustment.Formanyanimals,plasticityduringdevelopmentcaninfluencebehaviourandfitnesslaterinlife,bothpositively(adaptation)andnegatively(constraint).Usingsuperbstarlings(Lamprotornissuperbus),whichinhabitarangeofEastAfricanenvironmentswhererainfallvarieswithinandamongyears,Iwillexplorethemolecularmechanismsthatunderlieplasticityinthegenomeintheformofepigeneticchange.Specifically,IwillexaminehowpatternsofDNAmethylationacrossthestarlinggenomevarywithrainfallduringdevelopment,discussinggenesthatshowsignaturesofadaptivecapacityversusthosethatinsteadshowsignaturesofbeingconstrainedbyearlylifeconditions.IwillthencomparepatternsofDNAmethylationacrossthegenomefrombirdscollectedalonganecologicalgradientspanninghundredsofkilometersthatvariesinthedegreeofrainfallvariabilityandpredictability.Together,thesetwostudieswillnotonlyillustratethedifferentwaysthatenvironmentalconditionsshapepatternsofDNAmethylationacrossthegenome,buttheywillenablemetodevelopanevolutionaryframeworkforintegratingideasofadaptationandconstraintinthecontextofclimatechangeandbehaviouralepigenetics.

29TransgenerationaleffectsofsimpleandcomplexenvironmentsonbehaviourinzebrafishMichaelGreen1,WillSwaney11LiverpoolJohnMooresUniversity,UnitedKingdomPhenotypicplasticityisafundamentalmechanismbywhichorganismscanadapttocopewithenvironmentalchange,suchplasticityrepresentsashort-termmechanismofadaptationwithinthelifetimeofanindividual.However,recentevidencesuggeststhattheinheritanceofacquiredtraitsviaepigeneticmechanismsmayalsorepresentamechanismforadaptationtocertainenvironmentsovermultiplegenerations.Therearemanyoutstandingquestionsregardingtheadaptiverelevanceofthesemechanisms,theenvironmentalcontextsunderwhichthistypeofinheritancemayoccur,thepersistenceofsucheffectsandhowcommontheyare.WeexaminedtheinheritanceofacquiredbehaviouraltraitsusingacombinationofenvironmentalmanipulationsandautomatedanimaltrackingwithDaniorerio.Adultmalezebrafishwereexposedtosimpleorcomplexphysicalenvironmentsandtheirbehaviourwasquantifiedusingautomatedvideotrackingandcustomanalysisscripts.Malesofbothgroupswerethenbredwithseparateunmanipulatedfemales,andbehaviouroftheF1offspringwasquantifiedbothaslarvaeandasadults.Socialbehaviour,activityandanxiety-likebehaviourwereallinfluencedbytheenvironmentintheparents,butwealsofoundevidencefortransmissionofbehaviourtoF1offspring.WethenrepeatedthemanipulationagainusingmalesfromthisF1offspringcohortandexaminedtheinheritanceoftraitsinF2offspringwithvaryingparentaland/orgrand-parentalenvironmentalexperience.Theseresultssuggestthatpaternalenvironmentalexperiencecanbeinheritednon-geneticallyinzebrafish,affectingbehaviourinsubsequentgenerations.

31Adaptivemotorcontrol:Slope-dependentmodulationofmuscleco-contractioninfreelywalkingstickinsectsYannickGünzel1,ThierryHoinville1,2,JosefSchmitz1,2,VolkerDürr1,21DepartmentofBiologicalCybernetics,BielefeldUniversity,GermanyTerrestriallocomotioninavariableandunpredictableenvironmentrequiresanimalstoadjusttheirlimbmovementsappropriately.Forexample,theabilitytocopewithdisturbanceslikechangesinsubstrateslopeiscrucialforlocomotionatintendedspeedanddirection.Negotiatinginclinesisparticularlyinterestingasitisassociatedwithchangesinloaddistributionacrossthebody.Flexibilityinmotorcontrolisthoughttobeachievedbyeitherutilizingdistinctmotorpatterns,orbyclosed-loopcontrolthroughsensoryfeedback.Mammals,forexample,adjusttheirmotoroutputinanticipationofaninclineandusefunctionallydistincttransitionstridesbetweenlevelandslopewalking(GottschallandNichols,2011).Sofar,slope-dependentchangesininsectshavebeenstudiedmainlyundersteady-stateconditions,thusneglectingthedynamicsofadaptationinresponsetochangesinloaddistribution.Here,weinvestigatedhowfreelywalkingstickinsects(Carausiusmorosus)masteredastepchangeinwalkwayslope(±45°).Forthis,wesimultaneouslyrecordedtheirwhole-bodykinematicsandhindlegmuscleactivity.Wemoreoverusedasimplifiedmechanicalmodeltoestimatethevaryingmechanicaldemandduetoslopetransitions.Ourresultssuggestthatthesetransitionsinvolveonlylittlekinematicadjustments,butaconsiderablereductioninmuscleco-contractionassociatedwithshiftsinhowgroundreactionforcesdistribute.Sincetheobservedchangesoccurredgraduallywitheachsubsequentlegsteppingontotheincline,weconcludethatstickinsectmotoradaptationtoinclinesisachievedreflexivelyandcontinuously,dependingontheongoingloadsensoryfeedback,ratherthanthroughswitchingbetweenpre-wireddistinctmotorprograms.

32DynamicalnetworkformationofC.elegansTakumaSugi1,HiroshiIto2,KenH.Nagai31MolecularNeuroscienceResearchCenter,ShigaUniversityofMedicalScience,Japan,2FacultyofDesign,KyushuUniversity,Japan,3SchoolofMaterialsScience,JapanAdvancedInstituteofScienceandTechnology,Japan.Understandingphysicalrulesunderlyingcollectivemotionsrequiresperturbationofcontrollableparametersinself-propelledparticles.However,controllingparametersinanimalsisgenerallynoteasy,whichmakesphysicalrulesunderlyinganimals'collectivebehaviorselusive.Here,wefindthataconventionalmodelanimal,C.elegans,collectivelyformsdynamicalnetworksofbundle-shapedaggregateswithsimplephysicalrules.WeinvestigatethedependenceoftheC.elegansnetworkformationonvariousextrinsicparameters(materialofsubstrate,ambienthumidityanddensityofworms).Takingadvantageofwell-establishedC.elegansgenetics,wealsocontrolintrinsicparameters(geneticallydeterminedmotility)bymutationsandbyforcedneuralactivationviaoptogenetics.Furthermore,wedevelopaminimalagent-basedmodelthatreproducesthedynamicalnetworkformationanditsdependenceontheparameters,suggestingthatthekeyfactorsarealignmentofwormsaftercollisionandsmoothturning.Ourfindingsimplythattheconceptsofactivematterphysicsmayhelpustounderstandbiologicalfunctionsofanimalgroups.Ref,Sugi*etal.NatureCommun,2019;Sugi*etal.PNAS,2014;Sugietal.NatureNeurosci,2011

33Batsadjusttheirmouth-gapetorapidlynarrowtheiracoustic'fieldofview'OfriEitan1,MorTaub1,AmirZviran1,2,ArjanBoonman1,VladimiarTourbabin3,AnthonyJ.Weiss2,YossiYovel1,41DepartmentofZoology,TelAvivUniversity,Israel,2SchoolofElectricalEngineering,TelAvivUniversity,Israel,3DepartmentofElectricalandComputerEngineering,Ben-GurionUniversityoftheNegev,Israel,4SagolSchoolofNeuroscience,TelAvivUniversity,IsraelEcholocatingbatsarerenownedfortheirabilitytocontrolthesensoryinformationthattheyacquire.Byrapidlychangingtheirecholocationsignaldesignandtiming,batscanadjusttherateandaccuracywhichtheyacquireinformationabouttheenvironment.Muchlessisunderstoodaboutbats'abilitytocontrolthespatialaspectsoftheiremission,thatistheirabilitytocontrolthesectorofspacewhichtheyscanusingasingleemission(beamforming).Weusedalarge(wideband)microphonearrayandahigh-endtrackingsystemtoreconstructhighresolutionbeamsandmouthmovementsofPipistrelluskuhliibatsastheyweresearchingforandlandingonatargetinalargeflightroom.Weshowthatbatsrapidlynarrowtheirbio-sonarfieldofviewwhenscanningatarget.On-targethorizontalsonar-beamswere~23degreesnarrowerthanoff-targetbeams.Bydirectlymeasuringbats'mouth-gape,wefoundthatthisbeamadjustmentwasmediatedbychangesinthemouthgape:batsopenedtheirmouthtonarrowthebeamandviceversa.Acousticsimulationsconfirmedtheseresultsandalsosuggestedthatthebatsnarrowedtheverticalbeam-widthby~50degrees.Thebatsusedthesamesensorystrategyinthepresenceofloudmaskingnoiseimplyingthatbeamadjustmentsarenotusedfordealingwithnoise.Thefunctionofthenarrowingofthebeamwasprobablytoimprovesignal-to-noise-ratiowhichincreasedbyatleast60%.Thenarrowingofthebeam-widthcouldalsoplayaroleinimprovingspatiallocalization.

34Neuralmechanismsofsimplegroupingbehaviour:nonapeptideregulationofshoalinginfishWilliamT.Swaney11LiverpoolJohnMooresUniversity,UnitedKingdomThenonapeptidesareahighlyconservedfamilyofneuropeptidehormonesthathavebeenrepeatedlyshowntoregulatesocialbehavioursacrossvertebratetaxa.Inmammalsthenonapeptidesoxytocinandvasopressinhavebeenshowntoplaykeyrolesincomplexbehaviourssuchaspairbonding,socialrecognition,parentalbehaviouretc.Howeversimplegroupingbehaviourisoftendifficulttostudyinmanymammals,andrecentresearchinothertaxahasstartedtoexploretheroleofnonapeptidesingroupingbehaviour.Wehaveexaminedtheroleofnonapeptidesinshoalingbehaviourindifferentfishspeciesandpopulations,usingbehavioural,pharmacologicalandneurobiologicalapproachestoinvestigatehowtheteleostnonapeptidesisotocinandvasotocininfluenceshoalingbehaviour.Inguppies,wehaveshownthatadministrationofisotocinincreasesshoalingtendency,whilevasotocinreducesit.Wehaveshownthatincreasesinshoalingtendencyseeninpredator-exposedguppiesareassociatedwithchangesinvasotocinsignallingandsensitivity.Inzebrafish,wehavefoundthatincreasedenvironmentalcomplexityleadstoincreasedshoalingtendencieswithinafewweeks,andthatthisbehaviouralshiftisassociatedwithincreasedisotocinneuronnumbers,specificallyinthemagnocellularportionofthepre-opticarea.Unlikeinmammals,behaviouralphenotypesinfishseemtobeassociatedwithchangesinnonapeptidesignalling,ratherthannonapeptidereceptordistribution.Thisworkindicatesthattheroleofnonapeptidesinsocialbehaviourisconservedacrossvertebrates,butwithimportantmechanisticdifferencesbetweentaxa,andthatnonapeptidesignallingplaysacriticalroleinsimplegroupingbehaviourinvertebrates,aswellascomplexsocialbehaviour.

35EscapefromthreatinthepresenceofobstaclesPhilipShamash1RubenVale1,TiagoBranco11SainsburyWellcomeCentre,UnitedKingdomHavingafunctionalunderstandingoftheworldisfundamentalforchoosingandexecutingadaptiveactions.Itisnotknownhowthemammalianbrainbuildsandrepresentsthisunderstanding.Hereweaddressthisproblembystudyingasetofethologicalactionsthatrelyonknowledgethatanimalshaveabouttheirspatialenvironment.Usingnewbehaviouralparadigms,weinvestigatethestrategiesthatmiceusetoescapefromimminentthreatswhenthereareobstaclesonthewaytotheirshelter.Wefindahierarchyofstrategies,inwhichmicefirstfollowinnatenavigationalrulesthatleadtosuboptimalescapepaths.Next,theylearnfromthespatialstatisticsoftheseinefficientescapes:insubsequenttrials,theyrundirectlytoanintermediategoalthatprovidesaccesstotheshelter,suchasanedgeoftheobstacle.Atthisstage,previouslytakenescapepathsarealsoconvertedtomemorizedtrajectories;whentheenvironmentchangesacutely,thereisabalancebetweenexploitingtheseknownroutesandfindingnew,efficientpathstotheshelter.Theseresultsidentifystrategiesthatmiceusetoselectescaperoutesbasedonanunderstandingoftheirsurroundings,andtheyprovideastartingpointforinvestigatinghowthebrainusesinternalmodelsoftheworldduringnaturalbehaviours.

36Inter-groupdispersalinVulturineguineafowlJamesA.Klarevas-Irby1,2,3,DanaiPapageorgiou1,2,3,DamienR.Farine1,2,31MaxPlanckInstituteofAnimalBehavior,Germany,2DepartmentofBiology,UniversityofKonstanz,Germany,3CenterfortheAdvancedStudyofCollectiveBehaviour,UniversityofKonstanz,GermanyDispersalisoneoftheprimarydriversofpopulationdynamics,connectivity,andgeneflow.Forsocialspecies,dispersalbehaviorsarenotonlydependentonthephysicalpropertiesoftheirenvironment,butalsoonthesociallandscape.Manyspeciesusesocialinformation(e.g.conspecificdensities)todecidewhentodepartandwheretosettle,possiblyasanindicatorofunderlyinghabitatqualityortoavoidcompetition.However,forgroup-livingspecies,theunderlyinghabitatislikelylessimportanttodispersersthanthedistributionandsocialstructureofgroups.Howdosubadultsdispersethroughthesociallandscape?Forexample,individualswhoarepartofmorecohesivesocialgroups,orwhoreceivemoreaffiliativeinteractionsarelesslikelytodisperse.Similarly,morecohesivesocialgroupsmaybemoreresistanttotheimmigrationofnewindividuals,thusaffectingsettlement.TheVulturineguineafowl(Acrylliumvulturinum)isahighly-gregariousbirdspecieswhichlivesinhighly-cohesivegroups.Usinga2-yeardatasetcomprisingover800individualsspanning18groupsthatarepermanentlytrackedusingGPStags,Iquantifyhowinter-individualinteractions,andthehistoryanddynamicsofinteractionsamonggroups,predictwhichindividualsdisperseandwheretheydisperseto.Theseuniquedataprovidesomeofthefirstinsightsintoboththelarge-scalemovementsofindividualsacrosssociallandscapesandthefine-scalesocialthatdictatesthedecisionsthatindividualsmake.

37Individualinformationaccessresultingfromfine-scalemovementdecisionsBlairR.Costelloe1,2,BenjaminKoger1,2,JacobM.Graving1,2,IainD.Couzin1,21MaxPlanckInstituteofAnimalBehavior,Germany,2CenterfortheAdvancedStudyofCollectiveBehaviour,UniversityofKonstanz,GermanyAnimalsmustconstantlymakedecisionsthataffecttheirfitness(e.g.wheretogo,whattoeat,whentobevigilant).Inordertomakebeneficialdecisions,theymustcollectandintegrateinformationandfromtheirphysicalsurroundingsandsocialpartners.Incomplexnaturalenvironmentsananimal'sfine-scalebehaviorcanstronglyinfluenceitsaccesstosuchinformation.Tostudythis,weuseddronestofilmfree-rangingzebraherdsinKenyaandcreate3Dmodelsofthesurroundingenvironment.Usingnovelanalyticalmethods,weextractedmovementandbehavioraldatafromthesevideos,andreconstructedindividuals'visualfields.Herewepresenttheresultsofouranalyses,inwhichweexplorehowindividuals'fine-scalemovementandbehavioraldecisionsaffecttheirvisualaccesstosocialandenvironmentalinformation.Bycomparingobservedpathstorandomizedandoptimizedpaths,weexploretheextenttowhichindividualsprioritizevisualinformationaccessastheymovethroughcomplexhabitats.Wealsoconsidertheimplicationsofindividualstrategiesforgroup-levelphenomenasuchascollectivedetectionandinformationtransfer.

38Theinfluenceofsocialrelationshipsonleader-followerdecisionsinhighlydynamicbirdflocksYohanSassi1,2,3,A.Maldonado-Chaparro2,3,4,DamienR.Farine2,3,4,51UniversitédeBourgogneFrancheCompté,France,2DepartmentofCollectiveBehavior,MaxPlanckInstituteofAnimalBehavior,Germany,3CentrefortheAdvancedStudyofCollectiveBehaviour,UniversityofKonstanz,Germany,4DepartmentofBiology,UniversityofKonstanz,Germany,5EdwardGreyInstitute,DepartmentofZoology,UniversityofOxford,UnitedKingdomIngroup-livinganimals,pre-existingsocialrelationshipscanunderlieleader-followerdecisions.Primatesthatliveinstablegroups,forinstance,tendtofollowtheirpreferredaffiliatesduringcollectivemovements.However,suchhighlystablegroupsarerelativelyrareinnature;itremainsthereforeunclearwhethersuchleader-followerdecisionrulescouldalsotakeplaceinmorefluidsocialsystems.Here,weinvestigatetheextenttowhichsocialrelationshipspredictleader-followereventsincollectivemovementsofcolonialbirdsthatformfission-fusionflocks.Weusedahigh-resolutionsystemtotrack6flocksof28captivezebrafinches(Taeniopygiaguttata),usingthepreciserecordoftheirlocationsovertimetocalculatetheirsocialassociationstrengthswithothers(i.e.theirsocialnetworks)andtheirfine-scalemovementsovertime(i.e.leader-followerevents).Thenweusedgeneralizedlinearmodelstorelateprevioussocialassociationswithcurrentleader-followereventsduringcollectivemovements.Ourresultsindicatethattheprobabilitytofollowanindividualispositivelycorrelatedwithsocialassociationstrengthandnegativelycorrelatedwiththedistancebetweenleaderandfollowerpriortothemovementtakingplace.Thesefindingscorroboratepredictionsofself-organizingcollectivemovementinwhichindividualstendtoreacttotheirnearbyconspecifics;buttheyalsohighlightthateveninfluidsocialsystemsindividualscanhaveahigherprobabilitytofollowthosethattheyaremorestronglyassociatedwith.Ourstudysuggeststhatthe'follow-a-friend'rulecanbedetectednotonlyinspeciesthatliveinstablegroupsbutalsoinmoredynamicsocialsystems.

39Theevolutionofmenopauseinresidentkillerwhales(andotherinterestinganimals)Daniel.W.Franks1,S.Nattrass1,L.Brent2,D.P.Croft2,S.Ellis2,M.A.Cant3,M.Weiss2,4,B.Wright5,E.Stredulinsky5,T.Doniol-Valcroze5,J.K.B.Ford5,K.C.Balcomb41DepartmentofBiologyandDepartmentofComputerScience,UniversityofYork,UnitedKingdom,2CentreforResearchinAnimalBehaviour,CollegeofLifeandEnvironmentalSciences,UniversityofExeter,Exeter,UnitedKingdom,3CentreforEcologyandConservation,UniversityofExeterinCornwall,UnitedKingdom,4CenterforWhaleResearch,UnitedStates,5PacificBiologicalStation,FisheriesandOceansCanada,CanadaWhyfemalesofsomespeciesceaseovulationbeforetheendoftheirnaturallifespanisalongstandingpuzzleinlife-historyevolution.Inhumans,aswellassomenaturalpopulationsoftoothedwhales,reproductiveagingoccursmuchfasterthansomaticagingandfemalesexhibitprolongedpost-reproductivelifespans(PRLSs).DeterminingthemechanismsandfunctionsthatunderpinPRLSshasprovedasignificantchallenge.HereIsummariseourwork,bringtogetherbothclassicandmodernhypothesesproposedtoexplainPRLSsandlife-historyevolutionanddiscusstheirapplicationwithparticularreferencetoourstudiesofkillerwhales.IndoingsoIhighlighttheneedtoconsidermultipleinteractingexplanationsfortheevolutionofPRLSsanddiscussthekeyroleofsocialstructure.

39MovementpatternsandleadershipinamultilevelsocialgroupKatalinOzogány1,ViolaKerekes1,2,MátéNagy3,4,5,ZoltánBarta61DepartmentofEvolutionaryZoologyandHumanBiology,UniversityofDebrecen,Hungary,2HortobagyNationalParkDirectorate,Hungary,3DepartmentofCollectiveBehaviour,MaxPlanckInstituteofAnimalBehavior,Germany,4DepartmentofBiology,UniversityofKonstanz,Germany,5MTA-ELTEStatisticalandBiologicalPhysicsResearchGroup,HungarianAcademyofSciences,Hungary,6MTA-DEBehaviouralEcologyResearchGroup,DepartmentofEvolutionaryZoologyandHumanBiology,UniversityofDebrecen,HungaryPatternsinanimalmovementarisefrominteractionsbetweenmovingindividuals,andhencearemodulatedbysocialrelationswithinthegroup.Thus,thespatialdistributionandmovementpatternsofindividualsoftenreflectthestructureofthesocietytheanimalsarelivingin.Theoccurringparticularmovementpatternsmayalsoinfluencethedynamicsofleadingwithinthegroup.Toinvestigatetherelationshipbetweenmovementpatternsandleadership,weobservedthecollectivemotionoffree-rangingPrzewalskihorsesinHortobágyNationalPark,Hungary.Thesehorsesareespeciallysuitabletosuchastudybecauseinthisreservetheyliveinacomplexmultilevelsociety:haremsuniteinalargeherd.Inthisstudyweusednovelmethods,sinceweneedthecontinuousandsimultaneouspositiondataathighfrequencyforwildanimalsinnaturalenvironment.Ourobservationalmethodisbasedonaerialvideosoftwodrones,oneprovideshighresolutionmotiondataforallanimalsintheherd,whiletheotherensuresindividualrecognition.Inthisway,wecollectedtrajectorydataregardingthedailymovementsofaround250individualsatthesametime,whileknowingtheidentityofallofthem.Usingthedatageneratedfromtheaerialfootage,wearestudyingthemotionpatternsandleadershipdynamicsinsidetheharemsandinsidethewholeherd.Sinceallthehorsesarerecognisedandtheirlife-andharemhistorywasrecordedregularlyforthepast20years,motiondatacanberelatedtomanyindividualandgroupcharacteristics.

40Conflictsofinterestandcollectivedecision-makinginwhite-facedcapuchins(Cebuscapucinus)GraceH.Davis1,2,MargaretC.Crofoot1,21UniversityofCalifornia,Davis,UnitedStates,2SmithsonianTropicalResearchInstitute,PanamaHowdogroupsofindividualsdecidewheretogoandwhattodo?Toobtainthebenefitsofsociality,animalgroupsmustremaincohesive,reachingconsensusandcoordinatingimportantdailyactivitieslikeforaging.However,foragingstrategiesofindividualgroupmembersoftenvary,creatingconflictsofinterestaboutforagingchoices.Whenconflictsofinterestaboutwhenandwheretofeedexistinagroup,someindividualsmustcompromisetheirpreferredbehavior,presumablyatacosttothemselves.Thisstudyextendsthemarginalvaluetheoremtoindividualslivinginstablesocialgroupstogeneratepredictionsaboutthemechanismsunderpinninghowgroupsreachconsensus.Usingasix-monthstudyoftwowhite-facedcapuchinmonkeygroups(Cebuscapucinus)onBarroColoradoIsland,Panama,weexamineindividualdifferencesinoptimalpatchdeparturetimeandhowthesetranslateintocollectivedecisions.Wetestthehypothesisthatdominance,size,andageimpacthowlongindividualsprefertoremaininaforagingtree,creatingconflictsofinterestoverwhentoleave.UsingthefocaltreemethodongroupfeedingsinAttaleabutyraceapalms,wecalculateexactindividualfeedingratesforallgroupmembersineachpalm.Thesefeedingratesgenerateindividualforaginggaincurvesthatpredictoptimaldeparturetimes.Wefurtheranalyzeindividualattemptstoinitiategroupmovementtounderstandwhoexertsinfluenceongroupdecisions.Resultsindicateindividualsdifferintheirpreferreddeparturetimesandgroupdecisionsaresharedbetweenadults.Together,thiscapturesimportantelementsofgroupdecision-makinginsocialprimates:whentogoandwhodecides.

41DronesandDeepLearningRevealVisually-mediatedCollectiveDecisionMakingintheWildBenjaminKoger1,2,3,BlairR.Costelloe1,2,3,JacobM.Graving1,2,3,IainD.Couzin1,2,31DepartmentofCollectiveBehavior,MaxPlanckInstituteofAnimalBehavior,Germany,2CentrefortheAdvancedStudyofCollectiveBehaviour,UniversityofKonstanz,Germany,3DepartmentofBiology,UniversityofKonstanz,GermanyWeusevideorecordingdronesandcustombuiltcomputervisionalgorithmstonotonlyautomaticallyrecordthelocationofallindividualsinungulateherdsinthewildbutalsodetectninebodykey-points,thingslikefrontofthehead,shoulders,andtail,forallindividualsinthegroup.Wethenembedthisinformationinfinescalegeoreferenced3Dlandscapemapswithcentimeterlevelprecision.Westudyherdsofplainszebra,Grevy'szebra,impala,andafricanbuffaloincentralKenya.Usingarelayofdroneswerecordherdsforuptoanhourduringwhichtimeweobservetheanimalsinanundisturbedstateandthenpurposelywalktowardthemonfoottocreateadisturbancetheydetectandrespondto.Afterwardsweuseanadditionaldronetocreate3Dmapsoftheexactareatheanimalsmovedthrough(sometimesovermorethanasquarekilometer)withpixelresolutionofafewcentimeters.Sinceweknoweachindividualslocation,headpositionanddirection,andtheirexactpositionwithintheenvironment,westudytherolevisualcommunicationplaysingroupdecisionmakingincomplexenvironmentsunderheavypredationriskandobserveindividuals'strategiesforoptimizingtheacquisitionofbothsocialandenvironmentalinformationwhilealsoeffectivelyforagingandremainingsafefrompredation.

42EnergysavinginfishschoolLiangLi1,2,3,4,MateNagy1,2,5,JacobM.Graving1,2,3,JosephBak-Coleman6,GuangmingXie4,7,8,IainD.Couzin1,21DepartmentofCollectiveBehaviour,MaxPlanckInstituteforOrnithology,Germany,2DepartmentofBiology,UniversityofKonstanz,Germany,3CentrefortheAdvancedStudyofCollectiveBehaviour,UniversityofKonstanz,Germany,4StateKeyLaboratoryforTurbulenceandComplexSystems,CollegeofEngineering,PekingUniversity,China,5MTA-ELTEStatisticalandBiologicalPhysicsResearchGroup,HungarianAcademyofSciences,Hungary,6DepartmentofEcologyandEvolutionaryBiology,PrincetonUniversity,UnitedStates.,7IntelligentBiomimeticDesignLab,CollegeofEngineering,PekingUniversity,China,8InstituteofOceanResearch,PekingUniversity,ChinaMovementthroughfluidsimposessteepenergeticcostsonorganismsthatswim,glide,andfly.Ithaslongbeenproposedthatgroup-livingspeciesminimisethesecoststhroughcollectivecoordination,butwhilethisappearstobetrueforflockingbirds,thereislittleevidencethatschoolingfishcoordinatetheirmovementtosaveenergywhenswimming.Inadditiontoscarcedirectevidence,wealsolackabiologically-plausibletheorytodescribehowfishshouldbehavewhenobtaininghydrodynamicbenefitsfromothers.Mostmodelsofcollectiveenergysavingsareoverlysimplisticormakerestrictive,unrealisticassumptionsaboutspatialpositions,whichtendtobehighlydynamicinrealgroups.Consequently,despiteanabundanceofpredictionsthatenergysavingispossible,wehaveyettoreconcileexistingtheorywithnaturalbehaviour.Toaddresstheseproblems,weemployanintegrativeexperimentalandtheoreticalapproach.First,weuseaphysicalmodeloffish-likerobotstoderiveanew,foundationaltheoryofhydrodynamicinteractionsthatgeneralisestoanygroupstructure.Ourmodelrevealsthatregardlessofthespecificspatialarrangement,fishcanobtainenergeticbenefitsfromaleadingneighbourbyadjustingtheirswimmingpatternusingasimple,linearrelationship.Wefurtherconductexperimentswithpairsoffreely-swimminggoldfish(Carassiusauratus)andfindthatthisspeciesengagesinthesamedynamicalvortexphasematchingpredictedbyourtheorytosaveenergy.Ourresultsofferimportantinsightsintotheecologyofschoolingfish,andourtheoreticalmodelcouldbereadilyappliedtothedevelopmentofefficientunderwaterautonomousvehicles.

43DeepattentionnetworksrevealtherulesofcollectivemotioninzebrafishFranciscoJ.H.Heras11ChampalimaudResearch,PortugalAvarietyofsimplemodelshasbeenproposedtounderstandthecollectivemotionofanimals.Simplemodelscanbeinsightfulbutlackimportantelementsnecessarytopredictthemotionofeachindividualinthecollective.Addingmoredetailincreasespredictabilitybutcanmakemodelstoocomplextobeinsightful.Herewereporthowthemodularstructureofdeepattentionnetworkscanobtainamodelofcollectivebehaviorthatissimultaneouslypredictiveandinsightful.Ourmodeldescribeszebrafishpairwiseinteractions,Daniorerio,asapproximatelyrepulsive,attractiveorasalignment,butonlywhenmovingslowly.Athighvelocities,interactionscorrespondonlytoalignmentoralignmentmixedwithrepulsionatclosedistances.Themodelcapturesaggregationofinformationfromdifferentneighboursasaweightedaverage.Weightsarehigherforneighboursthatareclose,inacollisionpathormovingfasterinfrontalandlaterallocations.Theseweightseffectivelyselectadynamicalnumberofneighbours,fromasingleonetoupto12,oftenchanginginlessthanasecond.Byfittingthemodeltogroupsoffishofdifferentages,weexploredtheontogenyofcollectivebehaviour.Predictionaccuraciesarehigherforolderfish.Interactionmapsshowstrongerattractionandorientationinolderfish,withsmootherattentionmaps.Wealsoexploredtheeffectofgroupsizeandhabituationtotheenvironmentontherulesofcollectivebehaviour.

44idtracker.ai:trackingallindividualsinsmallorlargecollectivesofunmarkedanimalsFranciscoRomero-Ferrero11ChampalimaudResearch,PortugalUnderstandingofanimalcollectivesislimitedbytheabilitytotrackeachindividual.Determininganimaltrajectoriesfromvideorecordingsfacestheproblemofmaintainingcorrectanimalidentificationsafterindividualstouch,crossorareoccludedbyenvironmentalfeatures.Wepresentidtracker.ai,analgorithmandsoftwarethatextractsalltrajectoriesfromvideo,withhighidentificationaccuracyforcollectivesofupto100individuals.First,aspecies-agnosticpreprocessingextractsimagesfromthevideo.Then,afirstconvolutionalnetworkdetectswhenanimalstouchorcross,andasecondconvolutionalnetworkidentifiestheanimalsalongthevideo.Trainingexamplesforbothnetworksaresafelyextractedusingasetofheuristicswithouthumansintheloop.Atrainingandidentificationprotocoladaptstotheconditionsofthevideoandtrackingdifficulty.Post-processingstepsareappliedtoensurethecontinuityofthetrajectories.Thankstothetransferlearningcapacitiesoftheconvolutionalnetworks,idtracker.aicanbeusedindifferentwaystomatchanimalidentitiesacrossvideos.Themodularityofthisopen-sourceandfreesoftwaretrackingsystemallowsresearcherstouseitandadaptittotheirneeds.Inparticular,weshowhowweuseidtracker.aiandapredictivemodelbasedondeepattentionnetworkstostudyinformationtransferacrossindividualsingroupsofjuvenilezebrafish.

45IndividualidentificationusingdeeplearninginwildbirdsAndréC.Ferreira1,2,LilianaR.Silva2,FrancescoRenna3,JulienP.Renoult1,DamienR.Farine4,5,6,7,RitaCovas2,8,ClaireDoutrelant1,81CEFE-UMR5175CNRS-UniversitédeMontpellier,France,2CIBIO-InBio,ResearchCentreinBiodiversityandGeneticResources,CampusAgráriodeVairão,Portugal,3InstitutodeTelecomunicaçõeseFaculdadedeCiênciasdaUniversidadedoPorto,Portugal,4DepartmentofCollectiveBehaviour,MaxPlanckInstituteofAnimalBehavior,Germany,5ChairofBiodiversityandCollectiveBehaviour,DepartmentofBiology,UniversityofKonstanz,Germany,6CentrefortheAdvancedStudyofCollectiveBehaviour,UniversityofKonstanz,Germany,7EdwardGreyInstituteofFieldOrnithology,DepartmentofZoology,UniversityofOxford,UnitedKingdom,8FitzPatrickInstituteofAfricanOrnithology,DST-NRFCentreofExcellence,UniversityofCapeTown,SouthAfricaIndividualidentificationisessentialtomostanimalstudiesinecology,evolutionandconservation.Todate,thishasbeenachievedmostlythroughmethodsrelyingonmarkingindividuals(e.g.colourrings)andusinghumanobserversfordataacquisition.Whilepresentingseveraladvantages,thesemethodscanbetimeconsumingandlabour-intensive,therebyhamperinghighratesofdatacollection.Recenttechnologicalandanalyticaladvances,suchasRFIDanddeeplearningcanhelptoovercometheselimitationsbyautomatizingdatacollectionandanalysis.Here,wepresentamethodbasedonphotosthatallowstheindividualidentificationandbehaviouralstudyofasmallpasserinebird,thesociableweaver,Philetairussocius.First,wedescribeanautomatedmethod(basedonRFID)forefficientcollectionoflargesamplesofindividuallylabelledimagesofbirdsinthewild,whicharerequiredfortrainingconvolutionalneuralnetworks(CNNs).Second,wegothroughtheprocessoftrainingaCNNtobuildanaccurateclassifierofthefocalindividuals.Finally,wetestthegeneralizationcapabilityofourmodelsbypredictingtheidentityofthebirdsfromimagesthatwerecollectedwithdifferentcamerasandindifferentcontextsfromtheonesoriginallyusedfortrainingtheCNNs.Weshowthatitispossibletoidentifybirdswith>90%accuracy,evenwhentheconditionsofthetrainingdatasetsdifferfromtheimagestowhichtheclassifiercanpotentiallybeapplied.TheseresultsprovideapracticalsolutionforcollectinglargetrainingdatasetsandillustratethepotentialuseofCNNforindividualidentificationinwildpopulations.

463D-trackingcollectiveescapeinwildgroupsofDamselfishSimonGingins1,2,3,RenaudBastien1,2,3,MátéNagy1,2,3,AlexJordan1,2,3,IainCouzin1,2,31DepartmentofCollectiveBehaviour,MaxPlanckInstituteforOrnithology,Germany,2DepartmentofBiology,UniversityofKonstanz,Germany,3CentrefortheAdvancedStudyofCollectiveBehaviour,UniversityofKonstanz,GermanyInanimalcollectives,itmightonlytakeoneorafewindividualstodetectathreatinordertotriggeracollectiveescapeatthelevelofthegroup.Foranimalslivinginhierarchically-structuredgroups,itremainsunclearwhether,andhow,dominancerelationshipsshapethewayinformationspreads.Iwillpresenthowweinvestigatedtheimpactofsocialhierarchyoncollectivesensinginsocially-structuredgroupsofdamselfishDascyllusmarginatus.Inthefield,weexperimentallytriggeredcollectiveescaperesponsesusingaloomstimulusprojectedonaniPad.Thisallowedustocreatesituationswherealimitednumberofindividualshaddirectaccesstotheinformation.Usingmultiplecameras,wetrackedin3Dthemovementsofallgroupmembersduringtheseexperimentally-inducedcollectiveescapes.Inordertodisentanglebetweenprivateandsocialaccesstoinformation,wethencomputedthevisualconnectionsbetweenalltheindividualsandtheloomstimulus.Iwilldiscusswhetherthenumberandthesocialstatusoftheinitiator(s)affectthespeedatwhichinformationspreadswithinthegroup,andtheintensityoftheescaperesponseintheotherindividuals.

47Allyouneedisstatisticalphysics?Group-vs.Individual-OptimizationagainstPredatorAttacksPascalP.Klamser1,PawelRomanczuk11InstuteforTheoreticalBiology,HumboldtUniversitätzuBerlin,GermanyAnimalcollectivescanbeframedasanensembleofagentsfollowingidenticalrulesmodeledassocialforces.Thisframingisalsocommoninstatisticalphysics,whichsuggeststhatgeneralprinciplesfromstatisticalphysicsalsoholdforthedescriptionofanimalcollectives.Oneofthemostprominentonesisthatasystematatransitionfromanorderedtoanunorderedstatehasthestrongestresponsetoanexternalglobalfield.Basedonthisconceptofmaximalresponseatthetransitionitwasconjecturedthatnaturalsystemsshouldevolvetothistransition('criticality'),i.e.representexamplesofself-organizedcriticalsystems.Hereweinvestigatethevalidityofthisconceptusinganindividual-basedmodelforacollectiveofagentsbeingattackedbyasinglepredatorwithdifferentattackschemes.Wefindthatundergroup-leveloptimizationtheorder-transitionisfavored.However,notimprovedreactionbutspatialstructureisthemaindriver.Ontheotherhand,underindividual-leveloptimization,i.e.naturalevolutionlikeoptimization,thecollectiveisnotshowingself-organizedcriticalitybutratherevolvesintotheorderedphaseawayfromthetransition.Themaincauseofthistrendinevolutionisthespatialself-sortingofagentsaccordingtotheirparameters.Ourworkshowshowdifficultitistotranslateconceptsfromstatisticalphysicstoheterogeneousandspatiallyexplicitbiologicalsystems.However,wealsoemphasizehowimportantandfruitfulthefeedbackbetweenthetwodisciplinescanbe.

48LaOlá'wavesofthesulphurmolly:adaptiveanti-predatorbehaviorofanextremophilefishJulianeLukas1,2,PawelRomanczuk2,JaeugChoi2,JensKrause1,2,DavidBierbach11DepartmentofBiologyandEcologyofFishes,Leibniz-InstituteofFreshwaterEcologyandInlandFisheries,Germany,2FacultyofLifeSciences,HumboldtUniversityofBerlin,GermanyTheabilitytodetectandreacttowardsapotentialpredatoriscrucialforanimalsasfailuretorespondisusuallyfatal.However,anti-predatorresponsesoftencomeatacostandreactionswhilenooronlyinactivepredatorsarepresentwillreducetheprey'soverallfitness.InMexico,sulphide-richandseverelyhypoxicspringsarecolonizedbytheendemicsulphurmolly(Poeciliasulphuraria).Tocopewiththesephysicochemicalstresses,sulphurmolliesperformaquaticsurfacerespiration,resultinginthebuild-upoflargeaggregationsdirectlyatthewatersurface,wheretheyareparticularlyvulnerabletoavianpredation.Followingabirdattack,fishschoolsproduceaseriesofsynchronizedcollectivewavesbyrepeatedlydivingdowninacascade-likemanner.Asdivingintothehypoxicwatercolumniscostly,fishmayhaveevolvedmechanismstodistinguishbetweenpotentialthreatsandharmlessenvironmentaldisturbancessuchasflyoversofnon-predatorybirds.Wefoundthatonlyspecializedpiscivoreslikekingfishersandegretsproducelarge-scale,repeatedcollectiveresponsesofthesulphurmollies.Inalaboratoryexperiment,wepresentedfishwithartificialdisturbancecuesandfoundthatbimodalstimulationsconsistingofacombinedvisualandacousticcueinducedsignificantlystrongerescapebehaviorsthaneithercuealone.Mostpiscivorousbirdsproducebothvisualandacousticcuesduringtheirhunting,whilemostnon-dangerousdisturbancesareonlyassociatedwithasinglecue.Thus,weassumethatsulphurmolliesusepredator-specificcuesetstodistinguishdangerousfromnon-dangerousdisturbancesandreactonlytothelatter.

49Anti-predatorcostsandbenefitsofleadership:experimentalevidenceusingvirtualpreyattackedbyrealpredatorsChristosC.Ioannou1,FlorenceRocque2,JamesE.Herbert-Read1,CallumDuffield3,JoshA.Firth4,51SchoolofBiologicalSciences,UniversityofBristol,UnitedKingdom,2DepartmentofPsychology,UniversityofPortsmouth,UnitedKingdom,3MammalianBehaviourandEvolutionGroup,InstituteofIntegrativeBiology,UniversityofLiverpool,UnitedKingdom,4EdwardGreyInstitute,DepartmentofZoology,UniversityofOxford,UnitedKingdom,5MertonCollege,UniversityofOxford,UnitedKingdomThemajorcostofleadershipinmovinganimalgroupshaslongbeenassumedtobepredation,withindividualsleadingfromthefrontofgroupsbeingtargetedmoreoftenbypredators.Nevertheless,empiricalevidenceforthisislimitedandexperimentaltestsareentirelylacking.Toavoidconfoundingeffectsassociatedwithobservationalstudies,wepresentedasimulationofvirtualpreytorealfishpredatorstodirectlyassessthepredationcostofleadership.Preyleadingothersareatgreaterriskthanthoseinthemiddleofgroups,confirmingthatanybenefitsofleadingmaybeoffsetbypredationcosts.Importantly,however,followersconferanetsafetybenefittoleaders,aspreyleadingotherswerelesslikelytobeattackedcomparedtosolitaryprey.Wealsofindthatthepredatorspreferentiallyattackedwhensolitaryindividualsweremorefrequent,butthiseffectwasrelativelyweakcomparedtothepreferenceforattackingsolitarypreyduringanattack.Ourresultssuggestthatgoal-orientatedindividuals,i.e.potentialleaders,areunderselectivepressuretomaintaingroupcohesion,favouringeffectiveleadershipratherthangroupfragmentation.

50Whenandhow:temporalpatterns,defensivebehavioursandadaptivebenefitsassociatedtomaleparentalcareintheglassfrogCentrolenesavageiAnaMarıáOspina-L.1,PaulaNavarro-Salcedo2,JuliánRios-Soto3,SebastiánDuarte-Marıń2,FernandoVargas-Salinas21ColeccióndeSonidosAmbientales,InstitutoAlexandervonHumboldt,VilladeLeyva,Colombia,2GrupodeEvolución,Ecologı́ayConservación,ProgramadeBiologıá,FacultaddeCienciasBásicasyTecnologı́as,UniversidaddelQuindı́o,Armenia,Colombia,3SecretariadeGestiónRuralyDesarrolloSostenible,Filandia,ColombiaRecordinginformationaboutadaptivebenefitsandvariationofparentalinvestmentinglassfrogs,isimportantforabetterunderstandingoftheevolutionarybehavioralecologyinanurans.Inthisstudy,werecordedvariationsinparentalinvestmentbymalesoftheglassfrogCentrolenesavageiattwotemporalscales;andweperformedfieldexperimentstosimulaterisksofpredationbysmallinvertebrates,totestthelevelofaggressionofmalescaringeggs.BetweenFebruary2016andNovember2018,wemonitored87malesand154eggclutchesinapopulationofC.savageiintheCentralAndesofColombia.Parentalinvestmentwashigheratnightthanatnoon,anditdecreaseasembryosdevelopandbecomemoreindependent.Malescaringeggsexhibithigherlevelsofaggressionthansolitarymales,bybitingandkickingwiththeirlegsthebrushtipwhensimulatingpredationattacks;moreover,thesemalesalsospentmoretimetoleratingthestimulusbeforeflee.OurresultsshowthatmaleparentalcarebehaviorinC.savageiisadaptive.However,parentalinvestmentisaplasticbehaviorandvariesacrossembryonicdevelopment,duringthedayandlikelyacrossseasonsorpopulationswithdifferentenvironmentalconditions.

51SwarmintelligenceandtheavoidanceofparasitesinschoolingfishAngelaAlbi1,JacobDavidson1,TristanWalter1,VivekH.Sridhar1,JakeGraving1,IainCouzin11DepartmentofCollectiveBehaviour,MaxPlanckInstituteofAnimalBehavior,GermanyDiseasetransmissionandparasitismisthoughttobeoneofthemajorcostsofgroupliving.Nonetheless,whenmanyindividualsinteract,avarietyofbehavioralpatternsemergeenhancinginformationprocessingandtheabilitytodiscriminatesubtledifferencesintheenvironment.Withthisstudy,weexplorehowgroupsoffishcopewiththepresenceofanectoparasiteandifgrouplivingfacilitatesdiseaseavoidance.Attheindividuallevel,fisharemanuallyinfectedandmonitoredforthemetaboliccostsandbehavioraleffectsoftheinfectionusingphysiologytestsinaswimtunnel.Atthegrouplevel,weuseindividualtrackingmethodsandreconstructcollect