Interspecific information on predation risk affects nest site choice in a passerine bird

RESEARCH ARTICLE Open Access In e speci ic in o ma ion on p eda ion isk a ec s nes si e choice in a passe ine bi d Je e Tol anen 1,8* , Janne-Tuomas Seppänen 2,3 , Mikko Mönkkönen 4 , Robe L. Thomson 5,6 , Hannu Ylönen 7 and Jukka T. Fo sman 1,8 Abs ac Backg ound: B eeding si e choice cons i u es an impo an pa o he species niche. Nes p eda ion a ec s b eeding si e choice, and has been sugges ed o d i e niche seg ega ion and local coexis ence o species. In e speci ic social in o ma ion use may, in u n, esul in copying o ejec ion o he e ospeci ic niche cha ac e is ics and hus a ec ealized niche o e lap be ween species. We es ed expe imen ally whe he a mig a o y bi d, he pied lyca che Ficedula hypoleuca, collec s in o ma ion abou nes p eda ion isk om indi ec cues o p eda o s isi ing nes s o he e ospeci ic bi ds. Fu he mo e, we in es iga ed whe he he mig a o y bi ds can associa e such in o ma ion wi h a speci ic nes si e cha ac e is ic and gene alize he in o ma ion o hei own nes si e choice. Resul s: Ou esul s demons a e ha lyca che s can use he a e o he e ospeci ic nes ing a emp s in hei own nes si e choice, bu do so selec i ely. Young lyca che emales, when making he decision quickly, associa ed he a e o an a i icial nes wi h nes -si e cha ac e is ics and a oided he cha ac e is ic associa ed wi h highe nes p eda ion isk. Conclusions: Copying nes si e choices o success ul he e ospeci ics, and a oiding choices which led o ailed a emp s, may ampli y o coun e e ec s o nes p eda ion on niche o e lap, wi h impo an consequences o be ween-species niche di e gence-con e gence dynamics, species coexis ence and p eda o -p ey in e ac ions. Keywo ds: Social in o ma ion, Nes si e choice, P eda ion isk, Realized niche, Species coexis ence, In aspeci ic a ia ion Backg ound The niche concep is a cen al ene in he heo y o spe- cies coexis ence and communi y ecology, s a ing ha wo species canno coexis wi hou adequa e niche di e ences [1–4]. One impo an axis o he species niche is he cha - ac e is ics o o sp ing p oduc ion si e (nes , den, e c). Choice o he o sp ing p oduc ion (b eeding) si e de e - mines he a ailable esou ces and h ea s ha he animal and i s o sp ing encoun e , making i an impo an i ness- ela ed decision. Resou ces and p eda ion isk a e also a ec ed by he decisions o o he indi iduals in he communi y, including hose o o he species. B eeding si e choice d i en by a ying nes p eda ion p essu e has been shown o be an impo an mechanism a ec ing species coexis ence [5–7]. This idea is based on unc ional esponses o nes p eda o s o highe o e all nes densi y in a speci ic mic ohabi a when wo o mo e species p e e he same mic ohabi a [5,7,8]. Inc eased nes p eda ion a es in each species would hen selec o niche di e gence in mic ohabi a o nes si e choice, and he eby acili a e local coexis ence o he species [5–7]. Besides a ec ing species coexis ence, nes p eda ion is an impo an gene al selec i e o ce in animals [9,10]. I usu- ally esul s in comple e b ood loss and o sho -li ed species, ailing e en a single b eeding a emp may esul in ze o li e- ime ep oduc i e success. Howe e , nes p e- da ion isk a ies in space and ime (e.g. [11–13]). Conse- quen ly, he abili y o indi iduals o espond o cues on he ela i e isk o nes p eda ion in di e en habi a s and imes should be highly adap i e. Di ec obse a ions o o en s eal hy, widely anging and quickly mo ing nes p eda o s a e ela i ely a e e en s o an obse e , and he e o e o e li le in o ma ion o decision-making. Encoun e s wi h nes p eda o s may also h ea en he obse e i sel (e.g. [14,15]). Ye , o example, bi ds can clea ly espond o he densi y o nes p eda o s * Co espondence: [email p o ec ed] 1 Depa men o Ecology and Gene ics, Uni e si y o Oulu, 90014 Oulu, Finland 8 Cu en Add ess: Na u al Resou ces Ins i u e Finland (Luke), Uni e si y o Oulu, Paa o Ha aksen ie 3, 90014 Oulu, Finland Full lis o au ho in o ma ion is a ailable a he end o he a icle © The Au ho (s). 2018 Open Access This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0 In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons license, and indica e i changes we e made. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecommons.o g/publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed. Tol anen e al. BMC E olu iona y Biology (2018) 18:181 h ps://doi.o g/10.1186/s12862-018-1301-3 when making habi a choices [16] and o sp ing in es - men decisions [17]. A po en ial mechanism o es ima e p eda o incidence wi hou di ec isual o acous ic de ec- ion is cueing on indi ec signs o p eda o s such as odou , exc emen s o o he aces [18–22]. Beha iou , b eeding si e choices and success o o he conspeci ic o he e ospe- ci ic indi iduals may also p o ide in o ma ion abou p e- da ion isk [23–25]. A po en ially p o i able s a egy is o ollow he decisions and he esul ing success (e.g. nes dep eda ed o no ) o hose he e ospeci ics which b eed a li le ea lie han he obse e [24,26]. B eeding a emp s o such he e ospeci ics could p o ide he mos up- o-da e in o ma ion abou nes p eda ion isk. He e ospeci ic social in o ma ion use is expec ed o be mos use ul be ween ecologically simila species [24], in his case be ween species ha a e h ea ened by a simila se o p eda o s. Ins ances o social in o ma ion use, such as he e ospeci ic a ac ion and copying o beha iou s (e.g. [26–30]) can esul in posi i e i ness e ec s o he in o - ma ion use [27]. Thus social in o ma ion use may a o main aining o inc easing ecological simila i y be ween species in a communi y. Howe e , ecological simila i y also inc eases compe i ion o sha ed esou ces (e.g. ood, nes si es), a o ing dec easing ecological simila i y be ween compe ing species: niche di e gence ia cha ac e displacemen is a cen al pa adigm o species coexis ence heo y [1–4]. Nes p eda ion may he e o e igge bo h di e gence and con e gence o ealized niches, and bo h could concei ably be ampli ied by social in o ma ion use. Resul ing dynamics can be complex, scale-sensi i e, and highly dependen on local condi ions and communi y composi ion. He e, we es ed expe imen ally whe he a mig a o y, ca i y-nes ing bi d, he pied lyca che (Ficedula hypo- leuca), collec s in o ma ion abou nes p eda ion isk om indi ec cues o p eda o s isi ing nes s o a he - e ospeci ic esiden bi d, he g ea i (Pa us majo ). We hen in es iga ed whe he he mig a o y bi d can associa e such in o ma ion wi h a speci ic nes si e cha ac e is ic, a e able o gene alize he pe cei ed in o - ma ion and use i in guiding hei own nes -si e choice. G ea i s and lyca che s a e pu a i e compe i o s [31, 32], ye lyca che s use i s as a sou ce o in o ma ion in many c ucial decisions, such as b eeding si e choice ando sp ingin es men [27,30,33,34]. I pied lyca che s can associa e p eda ion isk wi h a pa icu- la nes -si e cha ac e is ic o hei pu a i e compe i o s, we expec hem o p e e acan nes si es exhibi ing he same cha ac e is ic as he ‘low- isk’nes si es o he e ospeci ics. Me hods We conduc ed he ield expe imen in no he n Finland (N 65°, E 25°) du ing b eeding seasons in 2013–2016. The expe imen al design consis ed o sepa a e nes box si es se up in habi a s sui able o b eeding pied ly- ca che s, a leas one kilome e apa o imp o e inde- pendence. Each si e included wo pai s o nes boxes (see Fig. 1 o a schema ic illus a ion o he expe imen al design). In one pai ( he ‘in o ma ion box pai ’)wecon- s uc ed inside bo h nes boxes a i icial g ea i nes s o Fig. 1 Schema ic illus a ion o he expe imen al se up Tol anen e al. BMC E olu iona y Biology (2018) 18:181 Page 2 o 8 moss and sheep hai . We assume lyca che s pe cei ed hese a i icial nes s as genuine i nes s, o a leas as he - e ospeci ic nes s, because hei own nes s consis solely o plan ma e ial (wood ba k, hay) and a e clea ly dis in- guishable om i nes s. The a i icial nes s we e andomly assigned o ea - men s wi h o wi hou simula ed nes p eda o isi s. One o he boxes ep esen ed he ‘high- isk’nes si e, whe e we sp ayed aeces and u ine o mus elids (Mus ela ni alis ni alis,Mus ela e minea and Ma es ma es; abundan nes p eda o s in he s udy a ea) dissol ed in wa e inside and ou side he nes box and on he base o he ee he nes box was a ached o. The wa e -exc e- men mix was made by mixing sawdus con aining mus- elid u ine and aeces wi h wa e and le ing he u ine and aeces dissol e in he wa e o e wo days (see [19] o mo e de ailed in o ma ion). Sawdus con aining mus- elid aeces and u ine was ob ained om cages o cap i e indi iduals a Konne esi Resea ch S a ion and Ranua Zoo. In addi ion, we pu wo pieces o mus elid aeces and ew agmen s o a common hen (Gallus Gallus domes icus) egg on op o he nes . The o he nes box wi hin he ‘in o ma ion box pai ’was ea ed using only wa e in sp aying and pu ing wo pieces o black s ing (con ol o mus elid aeces) on op o he nes , wi hou any egg agmen s. We hen a ached a neu al symbol, made o whi e plas- ic, a ound he en ance hole (3.2 cm diame e ) o each box wi hin he box pai ; one box go a iangle and he o he one a ci cle, diame e o side dimension 7.5 cm. Ac oss he expe imen al si es we sys ema ically andom- ized which symbol was associa ed wi h he ‘high- isk’nes si e ( he nes wi h simula ed p eda o isi ) so ha in hal o he si es a iangle was associa ed wi h he ‘high- isk’ si e and ice e sa. Ins ead o using some na u ally occu - ing nes -si e di e ence, such as ca i y en ance heigh o diame e o ee species, we pu pose ully used a no el, neu al cha ac e is ic pai wi h jus one e y salien bu simple con as (shape), o which bi ds should no ha e any in insic o lea ned esponse. Such designs isola e beha io al ai s unde in es iga ion om con ounding ac o s [33,35] and ha e been success ully applied in ea l- ie nes -si e choice s udies (e.g. [30,33]). The boxes wi hin he ‘in o ma ion box pai ’we e se up ca. 10 m apa in ees o same size and species. Abou 20 m away om he ‘in o ma ion box pai ’we se up ano he box pai , he ‘choice box pai ’. This box pai included wo acan nes boxes ca. wo me e s apa , again se up in simila ees. We pu 1.5 l o clean sawdus in o each o he ‘choice box pai ’boxes o make hese acan boxes mo e a ac i e nes ing si es han he wo in he ‘in- o ma ion box pai ’. Flyca che s s ongly p e e building hei nes on op o exis ing nes ma e ials, such as dummy [36] o dese ed (pe s. obs.) i nes s, and se ling in he ‘in o ma ion box pai ’would yield no da a on ly- ca che ’s abili y o associa e neu al symbol wi h nes p e- da ion isk. Howe e , clean sawdus in he nes box is p e e ed e en mo e s ongly han i nes s [37]. Finally, we andomly assigned he wo symbols, i- angle and ci cle, o he ‘choice box pai ’boxes. This c ea ed a se ing whe e wo equally a ac i e acan nes ing si es we e a ailable immedia ely adjacen o each o he , o lyca che o choose be ween. These nes si es di e ed only in he ype o symbol a ached a ound he ca i y en ance, while he nea by ‘in o ma ion box pai ’ ea u ed associa ion be ween one o he symbols and nes p eda ion isk. The dis ance be ween he box pai s had o be ela i ely sho o ensu e ha he bi ds se ling in he ‘choice box pai ’mos likely encoun e he simu- la ed in o ma ion in he ‘in o ma ion box pai ’be o e cons uc ing hei own nes . Due o he small spa ial scale his expe imen al design is conse a i e: i is con- cei able ha bi ds may pe cei e p eda ion isk o be uni- o m in he gene al a ea, i.e. equal in all nes boxes wi hin a si e, and consequen ly do no espond o he symbol- isk associa ion. Finding a signi ican esponse would hus gi e s ong suppo o he exis ence o abil- i ies in bi ds o i) de ec indi ec cues o p eda ion isk om obse ing he e ospeci ic nes ing a emp s and ii) associa e nes ing si e cha ac e is ics wi h ha p eda ion isk in o ma ion and iii) de elop p e e ence o “sa e ” cha ac e is ics in hei own nes -si e choice. We moni o ed he se lemen and b eeding o pied ly- ca che s by isi ing he si es usually e e y second day (occasionally e e y hi d day). Du ing each isi we e eshed he ea men s wi hin he ‘in o ma ion box pai ’( epea ed he sp aying and ensu ed ha aeces/ s ings and egg agmen s con inued o be obse able) and checked he lyca che nes s a us in he ‘choice box pai ’. We eco ded he eme ging lyca che nes s acco d- ing o a ou -le el classi ica ion: some nes ma e ial in he box, bu box loo s ill isible (le el 1), hal nes (box loo no isible) bu no cup-shape (le el 2), cup-shape o ming, bu cup no ye comple ed (le el 3) and eady nes o laying (cup comple ed; le el 4). Usually he bi ds ini ially b ough some nes ma e ial o bo h o he ‘choice’boxes, bu e en ually comple ed he nes in only one o hem. We de ined he choice o ha e happened once we obse ed a leas a wo-le el di e ence be ween he adjacen acan nes boxes. Pied lyca che males usually a i e o b eeding si es be o e emales and de end a e i o y ha may include se e al po en ial nes ca i ies (e.g. bo h nes boxes in ou ‘choice box pai ’), while he emale builds he nes and hus p esumably has mo e in luence han he male on he nes -si e choice ([14]; bu see [38]). We he e o e concen- a ed on emale beha io . We app oxima ed emale a - i al ime as he nes ini ia ion day, de ined as he day Tol anen e al. BMC E olu iona y Biology (2018) 18:181 Page 3 o 8 when a le el 1 nes was i s obse ed, o as he p e ious day in case he nes was i s obse ed as a u he le el nes . We cap u ed emales om he s ill ac i e nes s du - ing incuba ion ( wo nes we e abandoned ea lie ), de e - mined hei age as young (1-yea -old) o old (a leas 2 yea s old) based on plumage cha ac e is ics [39]and measu ed hei a sus leng h as a p oxy o body size. The alue o social in o ma ion is expec ed o show empo al deg ada ion as he ime lag be ween he eme - gence o in o ma ion (beha io o success o he in o - ma ion sou ce) and i s applica ion (by he in o ma ion use ) inc eases [24]. I is hus expec ed ha indi iduals using social in o ma ion would apply he in o ma ion quickly a e ob aining i . In addi ion, in o ma ion based on social cues is o en mo e easily a ailable compa ed o in o ma ion collec ed di ec ly om he en i onmen (e.g. by obse ing p eda o s). Thus, indi iduals ha base hei decisions on social in o ma ion a e expec ed o make as e decisions han hose who use di ec pe sonal obse a ions o ob ain in o ma ion [40–42]. To ake he ime used o make he nes si e (symbol) choice in o ac- coun , we calcula ed a a iable ‘decision ime’as he di - e ence be ween he choice da e (day when a leas a wo-le el di e ence in he nes s age was obse ed) and he app oxima e a i al da e. I social in o ma ion use esul s in as e decision-making, we expec bi ds using he simula ed in o ma ion o ha e sho e decision ime han hose no using he in o ma ion (i.e., making andom symbol choice). S a is ical analyses we e pe o med using gene alized linea models wi h binomial e o dis ibu ion in P o- g am R ( e sion 3.3.1; [43]). The ull model included he a iables emale age, a sus leng h, a i al ime (bo h linea and quad a ic e ms), decision ime, yea (2013–2016) and chosen symbol ( iangle o ci cle). In addi ion, we i ed wo-way in e ac ions be ween emale age and a sus leng h, a i al ime and deci- sion ime, as well as wo-way in e ac ions be ween a sus leng h and a i al ime and decision ime. Con- inuous explana o y a iables we e mean-cen e ed (a - i al ime o yea -speci ic means) p io o analyses. Since he analysis o he ull da a indica ed an in e - ac ion ela ed o emale age, we epea ed he analyses o old and young emales sepa a ely. In hese analyses he ull model was he same as o he ull da a, bu wi hou he emale age a iable and i s in e ac ions. A e de ining he ull model, all biologically eason- able models unde he ull model we e i ed o he da a and Akaike’s in o ma ion c i e ion [44] co ec ed o small sample size, AICc, was used o ank he models. To ake model selec ion unce ain y in o accoun we de i ed op model se s ha included all models wi h ΔAICc < 6, bu wi h he cons ain ha models ha we e mo e complex e sions o a model wi hlowe AICcwe eomi ed[45]. I mo e han one model was included in he inal op model se , we e alua ed he ela i e suppo be ween he models using e idence a ios ( a ios o model Akaike weigh s; [46]). We also p esen e ec sizes wi h 95% con i- dence in e als o a iables in he bes suppo ed models. Collinea i ies o con inuous explana o y a i- ables we e es ima ed using Pea son co ela ion coe i- cien s, bu all pai -wise co ela ions we e low ( < 0.29). Bo h ull and inal models wi hin he op model se s (i included a leas wo explana o y a iables) we e also es ed o o e all mul icollinea i y using a iance in la ion ac o s (VIF), bu all VIFs we e accep able (< 2.8). Also o e dispe sion le els o ull and inal models we e accep able (sum o squa ed Pea son esiduals / esidual d < 1.14). Resul s We eco ded 113 pied lyca che nes si e choices wi hin he ‘choice box pai s’, bu wo nes s we e abandoned be o e we could cap u e he emales. The e o e he da a se used in he analysis included 111 nes si e choices, o which 62 we e made by old emales and 49 by young emales. O e all, lyca che s chose he symbol depic ing ‘low- isk’nes si e in 62 o 111 (56%) cases; old emales in 32 o 62 (52%) and young emales in 30 o 49 (61%) cases. Only 15 lyca che pai s se led on op o he sim- ula ed i nes s (se en and eigh pai s on op o nes s wi h and wi hou simula ed nes p eda o isi , espec - i ely) showing ha adding sawdus in he ‘choice box pai s’was an e icien way o ge he lyca che s o se le wi hin he desi ed box pai . The analysis o he ull da a se (n=111 choices) esul ed in a op model se o wo models. The bes sup- po ed model included he in e ac ion be ween emale age and decision ime (Table 1), and was conside ably be e suppo ed han he second, in e cep -only model (ΔAICc = 4.29; e idence a io 0.9 / 0.1 = 9.0). Da a was he e o e analysed sepa a ely o old and young lyca che emales. Table 1 Pa ame e es ima es and hei 95% con idence in e als in he bes -suppo ed models o ull da a, old and young emale da a. S a is ics o s a is ically signi ican a iables (95% CI excluding ze o) in explaining pied lyca che nes si e choices in bold Da a se Va iable Es ima e 95% CI Full da a In e cep 0.058 −0.443 –0.561 Age_young 0.363 −0.443 –1.185 Decision ime 0.037 −0.128 –0.214 Age_young:Decision ime −0.474 −0.845 –−0.142 Old emales In e cep 0.065 −0.435 –0.567 Young emales In e cep 0.523 −0.105 –1.194 Decision ime −0.437 −0.770 –−0.151 Tol anen e al. BMC E olu iona y Biology (2018) 18:181 Page 4 o 8 The op model se o old emales only included he in e cep -only model. The in e cep did no di e om ze o indica ing andom nes si e choice in espec o he symbol a ound he en ance hole (Table 1). Fo young emales he op model se included only one model indica ing s ong nega i e e ec o decision ime (Table 1). We illus a e he e ec o decision ime on he nes si e choice o young emales in Fig. 2a, and o com- pa ison illus a e he same e ec on choices o old emales in Fig. 2b. Bo h igu es a e based on he models including only he e ec o decision ime ( he bes model o young and he hi d bes model o old emale da a). Those young emales ha made he nes si e (symbol) choice ela i ely quickly, decision ime < 3 days, p e e ed he ‘low- isk’symbol: 18 o 23 (78%) emales chose he ‘low- isk’symbol. The p e e ence o ‘low- isk’symbol de- c eased wi h inc easing decision ime and disappea ed when decision ime eached 5 days o mo e (Fig. 2a). The e a e e y ew da a poin s a longes decision imes (n= 4 o decision ime > 6 days), hus e e sal o p e e - ence a la ge alues should no be in e ed, despi e he g aphical appea ance. Decision ime was no ela ed o nes si e choice in old emales (Fig. 2b). Discussion Ou esul s demons a e ha lyca che s a e able o use he a e o p eceding he e ospeci ic nes ing a emp s in hei own nes -si e choice –bu do so selec i ely. Young lyca che emales, when making he decision quickly, as- socia ed he simula ed a e o an a i icial he e ospeci ic nes wi h a neu al nes -si e cha ac e is ic and p e e ed he cha ac e is ic associa ed wi h lowe nes p eda ion isk (o a oided he cha ac e is ic associa ed wi h highe isk). The choices o old emales we e andom i espec i e o he ime used in decision-making. P e ious s udies ha e shown ha bi ds may ob ain in- o ma ion abou p eda o s indi ec ly ia ol ac ion [18, 21] o by obse ing exc emen s o mammalian p eda o s and a oid nes si es ha p eda o s appa en ly had isi ed [19,20,22]. Ou expe imen demons a es a mo e com- plica ed esponse whe eby bi ds de i e in o ma ion on nes p eda ion isk om he e ospeci ic nes ing a emp s (ano he dimension o indi ec in o ma ion) and associ- a e he in o ma ion o a speci ic nes si e cha ac e is ic. This de i ed in o ma ion may hen be applied o o he nes si e loca ions wi h simila cha ac e is ics. Fo ex- ample, ege a ion cha ac e is ics a e o en consis en ly associa ed wi h high and low p eda ion isk [5–7] and by linking he obse ed success o o he s and ege a ion cha ac e is ics, indi iduals could choose sa e nes si es (o a oid isky si es) acco dingly. I is likely ha nes ing a emp s o he e ospeci ics and he associa ed mic o- habi a ea u es a e mo e easily obse ed han p eda o s hemsel es, making hei associa ion an a ailable in o - ma ion sou ce o indi iduals. Howe e , he alue o such in o ma ion depends on he spa io- empo al a i- a ion in p eda o ype and beha io and i is expec ed o dec ease wi h inc easing dis ance and delay om he e en ha gene a ed he in o ma ion [24]. Gi en ha nes p eda ion poses a signi ican h ea o b eeding bi ds [9,10], i is expec ed ha indi iduals would collec in o ma ion abou p eda o incidence in speci ic nes si es o a eas. Why hen did no all he ly- ca che emales p e e he ‘low- isk’nes si es in ou expe imen ? Social in o ma ion use s a egies ha e been obse ed o exhibi age-speci ic a ia ion [34,47–50], p obably due o di e ences in pe sonal expe ience o abili y and oppo uni y o collec in o ma ion pe sonally. Old (a leas 2-yea -old) emales p obably ha e p io b eeding expe ience and may be be e in collec ing in o ma ion pe sonally (i.e. di ec ly, no indi ec ly ia o he indi iduals). They may mo e eadily ely on Fig. 2 The p obabili y o (a) young and (b) old pied lyca che emales o choose he nes si e wi h he ‘low- isk’symbol a ached a ound he en ance hole in ela ion o he decision ime. Dashed lines along he solid ed line ( he p edic ed p obabili y) depic he 95% con idence in e als. Ve ical dashed line depic s he expec ed p obabili y le el (0.50) unde andom nes si e choice. Do s ep esen da a poin s (each do may include se e al da a poin s). No e he di e en scale o he x-axis be ween (a) and (b) Tol anen e al. BMC E olu iona y Biology (2018) 18:181 Page 5 o 8 pe sonal in o ma ion, despi e also ha ing ob ained he social in o ma ion. Fo example, old emales may ha e pe cei ed he simula ed social in o ma ion bu hen may ha e obse ed ha he e a e no p eda o s p esen in he a ea (any mo e) and hus selec ed he nes si e andomly in espec o he symbol. Young (1-yea -old) emales do no ha e p io b eeding expe ience and may hus p e e o use social in o ma ion. Due o he small spa ial dis- pe sion o he expe imen al nes boxes, mo e expe i- enced old emales may also ha e pe cei ed he p eda ion isk o be equal among all he boxes esul ing in andom nes si e choice. We obse ed a ia ion also among he young emales: only hose ha made hei nes si e choice decisions quickly used he simula ed social in o ma ion. This is consis en wi h he hypo hesis ha he alue o social in o ma ion de- c eases wi h inc easing delay be ween he e en gene a ing he in o ma ion and he applica ion o ha in o ma ion by he obse e [24]. On he o he hand, social in o ma ion use also enables as e in o ma ion collec ion and hus also as e decision-making compa ed o using (only) pe sonal in o ma ion [40–42]. Ou di ec obse a ions o lyca che beha iou a he expe imen al si es showed ha emales may isi and he e o e de i e in o ma ion om all he nes boxes wi hin a si e in jus a ew minu es. By elying on he in o ma ion a ailable in he o he nes s hey could he e o e make he nes si e choice immedia ely a e a i ing o he si e. Nes si e choice by emales ha ook longe o make a decision was indi e en o symbols and hei associa ed p eda ion ea men , and is consis en wi h hem ob aining nes si e quali y in o ma ion pe sonally. Females making slowe decisions may also ha e explo ed he su oundings and encoun e ed o he expe imen al si es wi h con as ing symbol-in o ma ion associa ions ( hese we e andomized ac oss si es) inc easing he likelihood o andom choices. These esul s p o ide empi ical e idence o he hypo heses o as e decision-making when elying on social in o ma- ion [40–42] as a esponse o deg ading in o ma ion alue wi h ime ([24]; see also [51]), mainly obse ed in g oup o aging o da e [52,53]. Cueing on nes p eda o p esence ia indi ec sou ces h ough he e ospeci ic nes ing a emp s conside ably inc eases he amoun o a ailable in o ma ion ha can be acqui ed sa ely and quickly by b eeding bi ds. This beha - io may ha e impo an implica ions o ealized niche o e lap be ween species and he e o e o species coexis - ence, communi y ecology, and ul ima ely e olu ion. Whe eas nes p eda ion pe se selec s o niche di e gence in nes si e choice o coexis ing species [5–7], he e ospeci- ic in o ma ion use in ela ion o nes p eda ion isk, and mo e speci ically copying o nes si e cha ac e is ics o sa e nes si es, may esul in main enance o simila i y and e en enhanced con e gence in o he ‘low- isk / p eda o ee niche space’. On he o he hand, a oidance o cha ac e is ics o isky nes si es esul s in di e gence in he ‘high- isk niche space’. Since nes p eda ion exe s a s ong selec ion p essu e in bi ds [9,10], niche e olu ion should p oceed owa ds sa e nes si es in bo h in e ac ing species (in o ma ion sou ce and in o ma ion use ). I simi- la cha ac e is ics de ine sa e nes si es in bo h species, hey should show niche con e gence. He e ospeci ic in o - ma ion use could hen add on he independen species -speci ic e ec s and accele a e con e gence o he ealized niches be ween he species. In o ma ion use and conse- quen he e ospeci ic a ac ion also p o ide a po en ial explana ion o he obse a ions whe e compe i o s sha e simila mic ohabi a s despi e hey su e highe nes p e- da ion han when b eeding alone (c . [7]). On he o he hand, con e gence in nes si e cha ac e is- ics ia he e ospeci ic in o ma ion use would esul in highe nes densi y in speci ic nes si es. I nes p eda o s espond unc ionally and begin o p e e such mic ohabi a s in sea ching p ey, inc eased p eda ion isk in his mic o- habi a would selec o di e gence in nes si e niche be ween he wo species (c . [5–7]). As a esul , he in e ac - ing species (in o ma ion sou ce and in o ma ion use ) could end up in luc ua ing ealized-niche di e gence con e gence dynamics. Pace, ampli ude and spa ial scale o hose dynamics would depend on he unc ional espon- si eness and o he cha ac e is ics o he p eda o commu- ni y and he p e alence o he e ospeci ic in o ma ion use among he p ey species. S ong unc ional esponsi eness o p eda o s coupled wi h equen he e ospeci ic in o ma- ion use among he p ey species would esul in ela i ely apid luc ua ions occu ing wi hin ew gene a ions, while weake esponsi eness and less equen in o ma ion use could yield modes luc ua ions o e longe empo al scales. Spa ial a ia ion in p eda o and p ey communi ies and hei esponsi eness and eliance on in e speci ic in o ma- ion use would c ea e localized dynamics, and consequen ly di e si y ecological in e ac ions (p eda o -p ey and in o - ma ion use dynamics) a la ge spa ial scales. Conclusions This s udy demons a es ha bi ds i) can de ec indi ec cues o p eda ion isk om obse ing he e ospeci ics nes - ing a emp s and ii) can associa e nes ing si e cha ac e is ics wi h ha p eda ion isk in oma ion and iii) can –bu only condi ionally do –de elop p e e ence o “sa e ”cha ac e - is ics in hei own nes si e choice. Such in e speci ic social in o ma ion use in ela ion o nes p eda ion isk may a ec ealized niche dynamics among coexis ing species wi h im- po an implica ions o species coexis ence and commu- ni y dynamics. These indings also add o he accumula ing e idence o be ween-indi idual a ia ion in social in o ma- ion use pa e ns; bo h due o age- ela ed di e ences and also due o wi hin-age-g oup a ia ion. Gi en he subs an- ial po en ial o indi idual le el a ia ion in beha iou o Tol anen e al. BMC E olu iona y Biology (2018) 18:181 Page 6 o 8 a ec key ecological and demog aphic p ocesses, also in- cluding species coexis ence and communi y ecology and e olu ion [54–57], we should aim o ho oughly unde s and he a ia ion in in o ma ion use. Addi ional ile Addi ional ile 1: The da a se suppo ing he a icle. (XLSX 16 kb) Abb e ia ions AICc: Akaike’s in o ma ion c i e ion co ec ed o small sample size; VIF: Va iance in la ion ac o ; ΔAICc: Di e ence in AICc uni s o he ocal model compa ed o he model wi h lowes AICc Acknowledgemen s We hank Mikko Ka jalainen o help in he ield and he Konne esi Resea ch S a ion and he Ranua Zoo o p o iding ma e ials. Commen s om wo e iewe s in Pee age o Science imp o ed he manusc ip . Funding This s udy was unded by Socie as p o Fauna e Flo a Fennica, Oska Ö lunds S i else (g an s o JT), Kone Founda ion (JTF) and Academy o Finland (g an s #122665, #125720 o JTF and #138049 o RLT). A ailabili y o da a and ma e ials All da a analysed du ing his s udy a e included in his published a icle as a supplemen a y ile (Addi ional ile 1). Au ho s’con ibu ions All au ho s concei ed he ideas and designed me hodology. JT collec ed and analysed he da a and led he w i ing o he manusc ip . All au ho s con ibu ed c i ically o he d a s and ga e inal app o al o publica ion. E hics app o al and consen o pa icipa e Handling o bi ds was done acco ding o he guidelines o he Finnish Ringing Cen e. O he e hics app o al no applicable. Consen o publica ion No applicable. Compe ing in e es s The au ho s decla e ha hey ha e no compe ing in e es s. Publishe ’sNo e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps and ins i u ional a ilia ions. Au ho de ails 1 Depa men o Ecology and Gene ics, Uni e si y o Oulu, 90014 Oulu, Finland. 2 Na u e and Game Managemen T us , Dege by, Finland. 3 Open Science Cen e, Uni e si y o Jy askyla, PO Box 35, 40014 Uni e si y o Jy askyla, Jy äskylä, Finland. 4 Depa men o Biological and En i onmen al Sciences, Uni e si y o Jy askyla, PO Box 35, 40014 Uni e si y o Jy askyla, Jy äskylä, Finland. 5 Sec ion o Ecology, Depa men o Biology, Uni e si y o Tu ku, 20014 Tu ku, Finland. 6 Pe cy Fi zPa ick Ins i u e o A ican O ni hology, DST-NRF Cen e o Excellence, Uni e si y o Cape Town, 7701 Rondebosch, Sou h A ica. 7 Depa men o Biological and En i onmen al Science, Uni e si y o Jy askyla, Konne esi Resea ch S a ion, 44300 Konne esi, Finland. 8 Cu en Add ess: Na u al Resou ces Ins i u e Finland (Luke), Uni e si y o Oulu, Paa o Ha aksen ie 3, 90014 Oulu, Finland. Recei ed: 28 Feb ua y 2018 Accep ed: 20 No embe 2018 Re e ences 1. Hu chinson GE. Concluding ema ks. Cold Sp ing Ha b Symp Quan Biol. 1957;22:415–27. 2. MacA hu R, Le ins R. The limi ing simila i y, con e gence, and di e gence o coexis ing species. Am Na . 1967;101:377–85. 3. Leibold MA. The niche concep e isi ed: mechanis ic models and communi y con ex . Ecology. 1995;76:1371–82. 4. Chesson P. Mechanisms o main enance o species di e si y. Annu Re Ecol Sys. 2000;31:343–66. 5. Ma in TE. On he ad an age o being di e en : Nes p eda ion and he coexis ence o bi d species. PNAS. 1988;85:2196–9. 6. Ma in TE. Fi ness cos s o esou ce o e lap among coexis ing bi d species. Na u e. 1996;380:338–40. 7. Ma in PR, Ma in TE. Ecological and i ness consequences o species coexis ence: a emo al expe imen wi h wood wa ble s. Ecology. 2001;82:189–206. 8. Schmid KA, Whelan CJ. P eda o -media ed in e ac ions be ween and wi hin guilds o nes ing songbi ds: expe imen al and obse a ional e idence. Am Na . 1998;152:393–402. 9. Ma in TE. Nes p eda ion and nes si es. Bioscience. 1993;43:523–32. 10. Ibáñez-Alámo JD, Mag a h RD, O eyza JC, Chal oun AD, Ha TM, Schmid KA, Thomson RL, Ma in TE. Nes p eda ion esea ch: ecen indings and u u e pe spec i es. J O ni hol. 2015;156(Supplemen 1):247–62. 11. Schmid KA, Os eld RS, Smy h KN. Spa ial he e ogenei y in p eda o ac i i y, nes su i o ship, and nes -si e selec ion in wo o es h ushes. Oecologia. 2006;148:22–9. 12. Halupka L, Halupka K, Klimczuk E, Sz wie nia H. Coping wi h shi ing nes p eda ion e uges by Eu opean eed wa ble s Ac ocephalus sci paceus. PLoS One. 2014;9:e115456. 13. Ringelman KM, Eadie JM, Acke man JT, Sih A, Loughman DL, Ya is GS, Oldenbu ge SL, McLand ess MR. Spa io empo al pa e ns o duck nes densi y and p eda ion isk: a mul i-scale analysis o 18 yea s and mo e han 10 000 nes s. Oikos. 2017;126:332–8. 14. Lundbe g A, Ala alo RV. The pied lyca che . London: T & A D Poyse ; 1992. 15. S ojano ic D, Webb MH, Alde man R, Po i io LI, Heinsohn R. Disco e y o a no el p eda o e eals ex eme bu highly a iable mo ali y o an endange ed mig a o y bi d. Di e s Dis ib. 2014;20:1200–7. 16. Fon aine JJ, Ma in TE. Habi a selec ion esponses o pa en s o o sp ing p eda ion isk: an expe imen al es . Am Na . 2006;168:811–8. 17. Fon aine JJ, Ma in TE. Pa en bi ds assess nes p eda ion isk and adjus hei ep oduc i e s a egies. Ecol Le . 2006;9:428–34. 18. Amo L, Gal án I, Tomás G, Sanz JJ. P eda o odou ecogni ion and a oidance in a songbi d. Func Ecol. 2008;22:289–93. 19. Mönkkönen M, Fo sman JT, Kananoja T, Ylönen H. Indi ec cues o nes p eda ion isk and a ian ep oduc i e decisions. Biol Le . 2009;5:176–8. 20. Eichholz MW, Dassow JA, S a o d JD, Wea he head PJ. Expe imen al e idence ha nes ing ducks use mammalian u ine o assess p eda o abundance. Auk. 2012;129:638–44. 21. Zida J, Lø lie H. Scen o he enemy: beha iou al esponses o p eda o aecal odou in he owl. Anim Beha . 2012;84:547–54. 22. Fo sman JT, Mönkkönen M, Ko pimäki E, Thomson RL. Mammalian nes p eda o eces as a cue in a ian habi a selec ion decisions. Beha Ecol. 2013;24:262–6. 23. Danchin É, Gi aldeau L-A, Valone TJ, Wagne RH. Public in o ma ion: om nosy neighbo s o cul u al e olu ion. Science. 2004;305:487–91. 24. Seppänen J-T, Fo sman JT, Mönkkönen M, Thomson RL. Social in o ma ion use is a p ocess ac oss ime, space, and ecology, eaching he e ospeci ics. Ecology. 2007;88:1622–33. 25. Schmid KA, Dall SRX, an Gils JA. The ecology o in o ma ion: an o e iew on he ecological signi icance o making in o med decisions. Oikos. 2010;119:304–16. 26. Mönkkönen M, Helle P, Soppela K. Nume ical and beha iou al esponses o mig an passe ines o expe imen al manipula ion o esiden i s (Pa us spp.): He e ospeci ic a ac ion in no he n b eeding bi d communi ies? Oecologia. 1990;85:218–25. 27. Fo sman JT, Seppänen J-T, Mönkkönen M. Posi i e i ness consequences o in e speci ic in e ac ion wi h a po en ial compe i o . P oc R Soc Lond B. 2002;269:1619–23. 28. Fle che RJ J . Species in e ac ions and popula ion densi y media e he use o social cues o habi a selec ion. J Anim Ecol. 2007;76:598–606. 29. Sebas ián-Gonzáles E, Sánchez-Zapa a JA, Bo ella F, O askainen O. Tes ing he he e ospeci ic a ac ion hypo hesis wi h ime-se ies da a on species co- occu ence. P oc R Soc B. 2010;277:2983–90. 30. Loukola OJ, Seppänen J-T, K ams I, To inen SS, Fo sman JT. Obse ed i ness may a ec niche o e lap in compe ing species ia selec i e social in o ma ion use. Am Na . 2013;182:474–83. Tol anen e al. BMC E olu iona y Biology (2018) 18:181 Page 7 o 8 31. Gus a sson L. In e speci ic compe i ion lowe s i ness in colla ed lyca che s Ficedula albicollis: an expe imen al demons a ion. Ecology. 1987;68:291–6. 32. Fo sman JT, Thomson RL, Seppänen J-T. Mechanisms and i ness e ec s o in e speci ic in o ma ion use be ween mig an and esiden bi ds. Beha Ecol. 2007;18:888–94. 33. Seppänen J-T, Fo sman JT. In e speci ic social lea ning: no el p e e ence can be acqui ed om a compe ing species. Cu Biol. 2007;17:1–5. 34. Fo sman JT, Seppänen J-T, Nykänen IL. Obse ed he e ospeci ic clu ch size can a ec o sp ing in es men decisions. Biol Le . 2012;8:341–3. 35. Ala alo RV, Mappes J. T acking he e olu ion o wa ning signals. Na u e. 1996;382:708–10. 36. Fo sman JT, Seppänen J-T. Lea ning wha (no ) o do: es ing ejec ion and copying o simula ed he e ospeci ic beha iou al ai s. Anim Beha . 2011;81:879–83. 37. Loukola OJ, Seppänen J-T, Fo sman JT. Pied lyca che s nes o e o he nes s, bu would p e e no o. O nis Fenn. 2014;91:201–8. 38. Loukola OJ, Seppänen J-T, Fo sman JT. In aspeci ic social in o ma ion use in he selec ion o nes si e cha ac e is ics. Anim Beha . 2012;83:629–33. 39. Jenni L, Winkle R. Moul and ageing o Eu opean passe ines. London: Academic P ess; 1994. 40. Valone TJ. G oup o aging, public in o ma ion, and pa ch es ima ion. Oikos. 1989;56:357–63. 41. Valone TJ, Temple on JJ. Public in o ma ion o he assessmen o quali y: a widesp ead social phenomenon. Phil T ans R Soc Lond B. 2002;357:1549–57. 42. Gale BG J , Laland KN. Social lea ning in animals: empi ical s udies and heo e ical models. Bioscience. 2005;55:489–99. 43. R Co e Team. R: A language and en i onmen o s a is ical. compu ing. Vienna: R Founda ion o S a is ical Compu ing; 2016. h ps://www.R-p ojec .o g. 44. Akaike H. A new look a he s a is ical model iden i ica ion. IEEE T ans Au om Con ol. 1974;AC-19:716–23. 45. Richa ds SA, Whi ingham MJ, S ephens PA. Model selec ion and model a e aging in beha iou al ecology: he u ili y o he IT-AIC amewo k. Beha Ecol Sociobiol. 2011;65:77–89. 46. Bu nham KP, Ande son DR. Model selec ion and mul imodel in e ence: a p ac ical in o ma ion- heo e ic app oach. 2nd ed. New Yo k: Sp inge - Ve lag; 2002. 47. Doligez B, Pä T, Danchin E, Clobe J, Gus a sson L. A ailabili y and use o public in o ma ion and conspeci ic densi y o se lemen decisions in he colla ed lyca che . J Anim Ecol. 2004;73:75–87. 48. Hahn BA, Sil e man ED. Social cues acili a e habi a selec ion: Ame ican eds a s es ablish b eeding e i o ies in esponse o song. Biol Le . 2006;2:337–40. 49. Saleh N, Chi ka L. The impo ance o expe ience in he in e p e a ion o conspeci ic chemical signals. Beha Ecol Sociobiol. 2006;61:215–20. 50. Fa ine DR, Spence KA, Booge NJ. Ea ly-li e s ess igge s ju enile zeb a inches o swi ch social lea ning s a egies. Cu Biol. 2015;25:2184–8. 51. Thomson RL, Si kiä PM, Ville s A, Laaksonen T. Tempo al peaks in social in o ma ion: p ospec o s in es iga e conspeci ic nes s a e a simula ed p eda o isi . Beha Ecol Sociobiol. 2013;67:905–11. 52. Temple on JJ, Gi aldeau L-A. Vica ious sampling: he use o pe sonal and public in o ma ion by s a lings o aging in a simple pa chy en i onmen . Beha Ecol Sociobiol. 1996;38:105–14. 53. Smi h JW, Benkman CW, Co ey K. The use and misuse o public in o ma ion by o aging ed c ossbills. Beha Ecol. 1999;10:54–62. 54. Bolnick DI, S anbäck R, Fo dyce JA, Yang LH, Da is JM, Hulsey CD, Fo is e ML. The ecology o indi iduals: incidence and implica ions o indi idual specializa ion. Am Na . 2003;161:1–28. 55. Bolnick DI, Ama aseka e P, A aújo MS, Bü ge R, Le ine JM, No ak M, Rudol VHW, Sch eibe SJ, U ban MC, Vasseu DA. Why in aspeci ic ai a ia ion ma e s in communi y ecology. T ends Ecol E ol. 2011;26:183–92. 56. Violle C, Enquis BJ, McGill BJ, Jiang L, Albe CH, Hulsho C, Jung V, Messie J. The e u n o he a iance: in aspeci ic a iabili y in communi y ecology. T ends Ecol E ol. 2012;27:244–52. 57. Tu co e MM, Le ine JM. Pheno ypic plas ici y and species coexis ence. T ends Ecol E ol. 2016;31:803–13. Tol anen e al. BMC E olu iona y Biology (2018) 18:181 Page 8 o 8