- Email: [email protected]
Robust autumn social attributes predict spring courtship skills in juvenile female brown-headed cowbirds, Molothrus ater
Animal Behaviour 85 (2013) 727e732
Contents lists available at SciVerse ScienceDirect
Animal Behaviour journal homepage: www.elsevier.com/locate/anbehav
Robust autumn social attributes predict spring courtship skills in juvenile female brown-headed cowbirds, Molothrus ater ˇ
Gregory M. Kohn*, Andrew P. King, Rebekka Dohme, Gwendwr R. Meredith, Meredith J. West Department of Psychological and Brain Sciences, Indiana University, Bloomington, IN, U.S.A.
a r t i c l e i n f o Article history: Received 2 September 2012 Initial acceptance 16 November 2012 Final acceptance 4 January 2013 Available online 28 February 2013 MS. number: A12-00663R Keywords: behavioural ontogeny brown-headed cowbird courtship behaviour juvenile sociability Molothrus ater ontogeny of courtship reproductive competence social temperament
Individuals often show predictable variation in social tendencies over time and across contexts. In species that rely on social interaction for the development of courtship behaviours, predictable variation in juvenile social behaviour may influence the ontogeny of courtship skills by exposing young individuals to different levels of contact with conspecifics. During the autumn and winter, brown-headed cowbirds utilize a social display known as the head-down to facilitate close social interaction with others. In this study we investigated whether head-downs rates are correlated across social contexts, and thus constitute a social temperament trait, and whether variation in head-down rates predicts subsequent courtship behaviour in juvenile cowbirds. During the autumn, we used a fissionefusion perturbation on a flock of juvenile females and males, and recorded the number of approaches and head-downs across social contexts. All individuals were then followed into the breeding season, where we recorded male singing behaviour, female chatter vocalization and consort status. We found that female head-down rates were correlated across contexts during the autumn, and that females that engaged in more head-down displays received more songs from males, responded to more songs with chatter vocalizations and were more likely to be in a consortship during the breeding season. As far as we know, this study presents the first demonstration that a social temperament trait can predict competent adult-like courtship behaviour in juvenile birds, and suggests that juvenile sociability may be an important developmental mechanism in the emergence of reproductive competence. The Association for the Study of Animal Behaviour. Published by Elsevier Ltd.
Little is known about the ontogeny of courtship skills in juvenile female birds and whether variation in social interaction tendencies reflects the development of these skills. Successful courtship requires sustained social attentiveness and willingness to engage others in close proximity (King & West 1987; Taborsky et al. 2012). Within many species, juveniles require close social interaction with conspecifics for the development of social skills (Suomi 1991; King & West 2002). Therefore, selection may favour predictable social tendencies early in ontogeny that facilitate close interaction, as a way of securing the experiences needed for later successful reproduction. The purpose of this study was to identify whether such predictable social tendencies exist in immature birds and whether these behaviours predict the early courtship skills needed for successful reproduction as adults. The brown-headed cowbird is an obligate brood parasite. Upon fledging from host nests, cowbirds join large flocks of conspecifics. In many locations, adults migrate in early summer, leaving
* Correspondence: G. M. Kohn, Department of Psychological and Brain Sciences, Indiana University, 1101 E. 10th Street, Bloomington, IN 47405, U.S.A. E-mail address: [email protected] (G. M. Kohn).
fledglings to form exclusively juvenile groups (Friedmann 1929). Previous studies have found that interaction with other cowbirds is essential for the development of species-typical mate preferences. For instance, juvenile cowbirds with access to experienced adults exhibit more competent courtship behaviour than juveniles without adult access (White et al. 2002). Freeberg (1998) demonstrated that juvenile females adopted song and mating preferences of adults from different subspecies with which they spent the winter. Previous work in developmental and comparative psychology has demonstrated that reliable early differences in how children and primates cope with social challenges and interact with conspecifics has long-lasting effects on the ontogeny of social and reproductive behaviour (Suomi 1991; Berman et al. 1997). Therefore, early differences in how juveniles interact with others may reflect the emergence of later behavioural skills. Within primates and birds, individuals possess ritualized social behaviours that facilitate close interaction with others while mitigating aggressive tendencies. These ritualized behaviours reflect, and help to define, an individual’s sociability. Cowbirds engage in a social display known as the head-down. During a head-down, an individual orients towards a neighbouring individual and adopts a low crouched position with the head pointed downwards in close
0003-3472/$38.00 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.anbehav.2013.01.009
728
G. M. Kohn et al. / Animal Behaviour 85 (2013) 727e732
proximity, if not touching their neighbour. The head-down is the only behaviour that brings cowbirds into close proximity for extended periods of time, and studies have shown that repeated use of the display has an appeasing influence (Selander & LaRue 1961; Robertson & Norman 1976; Scott & Grumstrup-Scott 1983). Like other affiliative displays, such as grooming in primates and allopreening in other birds, head-downs are commonly reciprocated, with cowbirds responding to another’s display with a head-down. The head-down is also seasonal and occurs mostly during the autumn and winter when cowbirds converge into large flocks (Ortega 1998). Consequently, individuals that initiate more head-downs during the autumn will have more experience in close social interactions than individuals that initiate fewer head-downs. Few studies outside primates have investigated whether robust variation in social interaction tendencies exists early in life, and whether such variation predicts the emergence of later social skills. Robustness indicates that behavioural traits are predictable across contexts, or that measures of a behaviour in one context are correlated with measures of behaviour in another context (Stamps & Groothuis 2010). The opposite of robustness is contextual specificity, where measures of behaviour in one context are not correlated with measures of the same behaviour in another context. Adult female and male cowbirds show correlated variation in autumn head-down rates across contexts (G. M. Kohn, A. P. King, U. Pott & M. J. West, unpublished data). Breeding season flocks composed of adults that initiated more head-downs during the autumn showed higher rates of egg production and courtship behaviour the following spring (G. M. Kohn, A. P. King, U. Pott & M. J. West, unpublished data). While the head-down display has been observed in juveniles, it has not been systematically studied, and little is currently known about robustness of social behaviour and its relationship to courtship skills in immature juveniles. In the current study, we manipulated group size and composition to see whether juveniles show robust patterns of head-down displays. Cowbird courtship revolves around female response to male song. Males sing directed songs by orienting towards a conspecific and giving a song within inches while performing a bowing display. As female cowbirds can withdraw during the onset of the male song display (Gros-Louis et al. 2003), males modulate the expression of their display when directing it towards females in order to reduce withdrawal (O’Loghlen & Rothstein 2010). Adult females assess song quality and communicate preferences using both visual (West & King 1988) and acoustic signals (Freed-Brown & White 2009) in response to male song displays (West et al. 2002, 2006; Miller et al. 2008). Previous studies have shown that adult females that receive more directed songs also produce more eggs (White et al. 2010). Males also exchange directed songs with other males in countersinging matches, and females may use countersinging matches as opportunities to assess male song quality and competitive ability (White et al. 2010). While female behaviour is an essential component of successful adult courtship and reproduction (West & King 1988; King et al. 2003), little is known about the early variation in juvenile female courtship behaviour. Females, which do not sing, react to directed song from a preferred male with a response chatter vocalization. Response chatters are used to indicate a preference for specific males and allow for female mate copying (Freed-Brown & White 2009). Chatter vocalizations are individually distinct and are also performed outside of singing contexts throughout the breeding season (Burnell & Rothstein 1994). These undirected chatters might advertise female identity and attract male attention (Dufty 1982), but they do not communicate mate preferences as they are often performed away from males. Females with a higher proportion of response to undirected chatters will use their chatter more selectively to communicate mate preferences. Response chatters may
aid females in formation of pair bonds through advertising mate preference and selectively reinforcing specific males. Cowbirds are a primarily monogamous species, and maintaining consorts is often essential to successful reproduction (Yokel 1986). Although juveniles often do not successfully breed during their first breeding season, they often engage in courtship behaviour (White et al. 2002). Courtship skills acquired by a juvenile may increase its reproductive success when it becomes reproductively active as an adult (White et al. 2007). Thus, in this study we considered a female’s consort status, the number of songs she received and her response chatters, as well as the proportion of the female’s response chatters to undirected chatters as measures of early courtship skills. We hypothesized that robust variation in headdown displays would emerge during the birds’ first autumn, and that higher head-down rates during the autumn would predict better courtship behaviour during the first breeding season. In particular, we expected female head-down rates to correlate with consort status, songs received and number of response chatters for females, and the number of directed songs and countersinging matches in males. METHODS Subjects All birds were originally captured in Philadelphia County, Pennsylvania, U.S.A., and housed in aviaries in Monroe County, Indiana, U.S.A. All subjects were Molothrus ater ater. For this study we used 12 juvenile females and 12 juvenile males. All birds were juveniles, and at the beginning of the study, ranged in age from 55 to 75 days old with an average age of 73 days, and were housed together for 1 month prior to the present study. Cowbirds without access to adults during the first year are not sexually competent in their first breeding season (White et al. 2002). Each bird was marked with uniquely coloured leg bands to allow for individual recognition. All birds were provided daily with a diet of vitamintreated water (Aquavite Nutritional Research), red and white millet, canary seed and a modified Bronx Zoo diet for blackbirds. Aviaries We used three aviaries that were visually isolated and substantially acoustically attenuated from each other, each with identical dimensions (9.1 21.4 3.4 m). Environmental conditions were similar throughout all aviaries with shrubs, trees, grass, covered feeding stations and access to indoor enclosures. All aviaries allowed birds to be exposed to ambient climatic conditions, wild cowbirds and the sight of predators. Data Collection During the autumn 2010, we recorded both male and female social approach and head-down displays. An approach was scored when one individual approached another individual with any part of its body within a radius of 30 cm. A head-down was scored when an individual maintained a low crouched position with its head pointed downwards so that the back of the head and neck were positioned in close proximity, if not touching, a neighbouring individual for at least 1 s. Head-downs were either initiated, when one individual initiated a head-down to a conspecific that was not in the head-down posture, or reciprocated, when an individual responded to another’s head-down posture with a reciprocal display. The identity of the individuals initiating and receiving an approach or head-down was recorded for each interaction.
G. M. Kohn et al. / Animal Behaviour 85 (2013) 727e732
During the breeding season, we recorded both male and female courtship behaviour, focusing on the vocal behaviour of both males and females. For male courtship behaviour, we recorded the number of female- and male-directed songs, and the number of songs within countersinging matches. A countersinging match occurs when a male reciprocates a directed song within 15 s after receiving a song from a neighbouring male. For females, we recorded the number of songs each female received from males and the number of female chatter vocalizations. Female chatter vocalizations were either response chatters or undirected chatters based on their temporal proximity to male song. Response chatters occur when a female responds to a directed male song with chatter vocalization within 1 s. Undirected chatter vocalizations occur when the females perform chatter vocalizations outside of singing contexts. Throughout the study, we used a scan-sampling procedure: behaviours were recorded, as they were observed, using voice recognition technology described in detail by White et al. (2002). When used in combination with voice recognition technology, scan sampling can accurately acquire a more comprehensive data set than focal sampling (White & Smith 2007). Autumn behaviour was recorded in blocks of 7 min scan-sampling periods and breeding season behaviours were recorded in blocks of 15 min scan-sampling periods. All observations were conducted between 0700 and 1030 hours, when cowbirds are most active, and were counterbalanced, so each observer took the same number of scan-sampling blocks in each aviary at the same time of day. All work was conducted under ASAB/ABS guidelines and approved by the Institutional Care and Use Committee of Indiana University (08-018).
729
Stage 4 Throughout the breeding season, from 1 May to 9 June, three observers conducted daily observations recording male song and female chatter vocalizations, and recorded a total of 252 15-minute observation blocks.
sex differences in the number of head-downs and whether males or females showed significant sex preferences in the head-downs they initiated or reciprocated, we used Wilcoxon signed-ranks tests. During the fusion period, we tested whether individuals preferentially interacted with their flock-mates from the fission period. Unseparated individuals were conspecifics that shared the same aviary during the fission period, whereas separated individuals were housed in the alternative aviary during the fission period. We used Wilcoxon signedranks tests to see whether individuals preferentially approached and displayed head-downs towards unseparated or separated individuals. To examine the contextual generality of head-down behaviour across periods, we used Spearman correlations to see whether individuals maintained significantly comparable rank-ordered rates of initiated and reciprocated head-down displays across the periods during the autumn. To see whether the head-down rates reflected social approach behaviour, we used Spearman correlations to examine the relationship between the number of approaches received or initiated and the number of head-downs within each period. To examine which autumn social behaviours were predictors of courtship behaviour during the spring, we used generalized linear models (GLM) with a quasi-Poisson distribution and a log link function. We used a model simplification procedure where a maximal model was simplified through stepwise elimination of nonsignificant variables. Log-likelihood tests were used for model comparisons during simplification, with a minimal adequate model defined when model comparisons exhibited a significant loss in the residual deviance with the removal of any remaining explanatory factors (Crawley 2005). We conducted four different GLMs for the females using the number of songs received, the total number of chatter vocalizations and the number of undirected and response chatter vocalizations as the dependent variables. For males we ran four different GLMs looking at the number of total, female- and male-directed song and the number of songs within countersinging matches as the dependent variables. In all models, the explanatory variables were the number of head-downs performed (both initiated and reciprocated) and approaches initiated and received across the autumn data collection period. We conducted an individual generalized linear regression with a binomial error distribution to investigate the relationship between autumn head-down rates and the proportion of response chatters. All post hoc tests included Spearman correlations to look at the strength of the relationship between the explanatory variables and the dependent variables for each model. As the explanatory variables were intercorrelated, we used variance inflation factors (VIF) to assess multicollinearity. A variance inflation factor greater than 10 is used to indicate potential multicollinearity, which makes model interpretation difficult (Chatterjee et al. 2000). We calculated VIFs for both the saturated models and for each resulting model during model simplification. In none of our presented models did the VIFs for any factor exceed 10. We considered a female to be in a consortship with a male when the songs of that particular male accounted for 75% of the songs she received and when she was one of the top two females a male sang to. To look at the influence of autumn social behaviour on consort status during the breeding season, we used ManneWhitney U tests. Specifically, ManneWhitney U tests were used to see whether consorted females showed higher rates of social approach or headdown displays during the autumn than unconsorted females.
Analysis
RESULTS
Because of small sample sizes and non-normality of data, we primarily used nonparametric statistics. We used Friedman tests and Wilcoxon signed-ranks tests to see whether the rates of head-down displays differed across periods during the autumn. To investigate
Autumn Head-down Displays
Juvenile Autumn Social Behaviour Stage 1 Baseline period. From 7 September to 11 October 2010, three observers conducted daily observations in Aviary 1, each recording approaches and head-downs, collecting a total of 165 7-minute observation blocks. Stage 2 Fission period. On 13 October, we separated birds into Aviary 1 and Aviary 2 based on their frequency of approaches during the baseline period, so that both flocks contained individuals with a similar range of approaches. Both aviaries contained 12 cowbirds including six males and six females. From 13 October to 24 October, three observers conducted daily counterbalanced observations recording approaches and head-downs, and collected a total of 195 7-minute observation blocks (98 blocks in Aviary 1, 97 blocks in Aviary 2). Stage 3 Fusion period. On 27 October, all birds were moved into Aviary 1. From 27 October to 6 November, three observers conducted daily observations recording approaches and head-downs, and collected a total of 166 7-minute observation blocks.
We observed a total of 1289 head-down displays during the autumn (median ¼ 39.0, interquartile range, IQR ¼ 4875), including
730
G. M. Kohn et al. / Animal Behaviour 85 (2013) 727e732
640 femaleefemale, 106 femaleemale, 194 maleefemale and 349 maleemale head-down displays. Females were more likely to initiate head-down displays with other females than they were with males (Wilcoxon signed-ranks: T ¼ 78, P < 0.001), whereas males showed no significant sex bias (T ¼ 19, P ¼ 0.13). Overall, the rates of head-downs were significantly higher during the fusion period than during the baseline period (T ¼ 25, N ¼ 24, P ¼ 0.007). A Friedman test showed a significant effect of period on both male (c22 ¼ 18:2, N ¼ 12, P ¼ 0.0002) and female (c22 ¼ 17:7, N ¼ 12, P ¼ 0.0001) head-down rates. Post hoc analysis demonstrated that females displayed significantly higher rates of head-downs during the fusion period than during the baseline period (Wilcoxon signed-ranks: T ¼ 25, N ¼ 12, P ¼ 0.007). For males, the rates of head-downs were significantly higher during the fusion period than during the baseline (T ¼ 6, N ¼ 12, P ¼ 0.0001) and fission (T ¼ 24, N ¼ 12, P ¼ 0.006) periods. During the fusion period, individuals initiated a higher proportion of head-downs and approaches towards unseparated individuals from the same fission flock than they did towards individuals from the alternative flock. Both females and males almost exclusively approached unseparated individuals (Wilcoxon signed-ranks test: females: T ¼ 144, N ¼ 12, P < 0.0001; males: T ¼ 127, N ¼ 12, P ¼ 0.002) and preferentially directed head-downs towards unseparated conspecifics (females: T ¼ 113.5, N ¼ 13, P < 0.0005; males: T ¼ 133, N ¼ 12, P ¼ 0.0005). Females that approached other individuals more often also initiated more headdowns, while males that received more approaches consistently initiated more head-downs (Table 1). We found no significant sex difference between the numbers of head-down displays initiated (ManneWhitney U test: U ¼ 86.5, N1 ¼ 509 females, N2 ¼ 381 males, P ¼ 0.41) or reciprocated (U ¼ 85, N1 ¼ 240 females, N2 ¼ 159 males, P ¼ 0.47) across the autumn. Across all periods during the autumn, females maintained the same rank-ordered distributions in the rates of head-down displays initiated, whereas male head-downs initiated remained correlated across the baseline to fission periods (Table 2). We observed no consistent relationships for rates of head-downs reciprocated across social periods for males or females.
Table 2 Spearman rank correlations for head-down rates of brown-headed cowbirds that were initiated or reciprocated across the autumn social perturbations
Females Males
Initiated Reciprocated Initiated Reciprocated
Baselineefission
Fissionefusion
Fusionebaseline
0.87** 0.77** 0.72* 0.01
0.84* 0.45 0.23 0.69*
0.71* 0.34 0.38 0.08
*P < 0.05; **P < 0.01.
Chatter Vocalizations and Autumn Head-down Displays From 1 May to 9 June we recorded a total of 1239 (median ¼ 23.0, IQR ¼ 190.3) chatter vocalizations, including 539 (median ¼ 11.50, IQR ¼ 56.5) response chatters and 700 (median ¼ 11.34, IQR ¼ 102.5) undirected chatters. Our models predicting the total amount of chatter vocalizations and undirected chatter vocalizations found no significant effects. The model predicting the number of response chatters uncovered a significant main effect of head-downs (Poisson regression: t ¼ 2.310, N ¼ 12, P ¼ 0.04). Our post hoc correlations uncovered a significant positive relationship between the number of response chatters and the number of head-downs initiated (Spearman rank correlation: rS ¼ 0.59, N ¼ 12, P ¼ 0.04), but no significant correlation between the number of response chatters and the number of approaches initiated (rS ¼ 0.37, N ¼ 12, P ¼ 0.23) or received (rS ¼ 0.20, N ¼ 12, P ¼ 0.51). The number of female response chatters was correlated with female head-downs to other females (rS ¼ 0.57, N ¼ 12, P ¼ 0.05) but not with female head-downs towards males (rS ¼ 0.34, N ¼ 12, P ¼ 0.28). Further analysis examining the proportion of response chatters to undirected chatters revealed a significant relationship between all head-down rates in the autumn and the proportion of response chatters. The proportion of response chatters was significantly correlated with head-downs (binomial regression: t ¼ 7.79, P < 0.001, rS ¼ 0.92, N ¼ 12, P < 0.0001). Therefore, cowbirds that engaged in more head-down displays during the autumn had a higher proportion of response chatters to undirected chatters (Fig. 2).
Female Songs Received and Autumn Head-down Displays
Table 1 Spearman rank correlations between the number of approaches initiated (AI) or received (AR) by brown-headed cowbirds and the number of head-down displays (HD) within the baseline, fission and fusion periods
Females Males
AIeHD AReHD AIeHD AReHD
*P < 0.05; **P < 0.01; ***P < 0.001.
Baseline
Fission
Fusion
0.73** 0.25 0.47 0.76**
0.62* 0.5 0.16 0.57*
0.91*** 0.59* 0.77** 0.67*
150
Head-down displays
We recorded a total of 1770 songs to females (median ¼ 38, IQR ¼ 246). Our model showed a significant effect of head-downs (Poisson regression: t ¼ 2.701, P ¼ 0.02) on the number of songs a female received during spring. Post hoc correlations found a significant correlation between number of songs received and headdowns (Spearman rank correlation: rS ¼ 0.68, N ¼ 12, P ¼ 0.02), but not between number of songs received and approaches initiated (rS ¼ 0.38, N ¼ 12, P ¼ 0.22) or received (rS ¼ 0.40, N ¼ 12, P ¼ 0.20; Fig. 1). The number of songs females received was correlated with female head-downs to other females (rS ¼ 0.71, N ¼ 12, P ¼ 0.01) but not with female head-downs towards males (rS ¼ 0.21, N ¼ 12, P ¼ 0.49).
100
50
0 0
0.1
0.2 0.3 0.4 0.5 Proportion response chatter
0.6
Figure 1. Relationship between head-down rates of brown-headed cowbirds during the autumn and the proportion of response chatters in the spring. Regression lines are plotted to illustrate the strength and direction of the relationship.
G. M. Kohn et al. / Animal Behaviour 85 (2013) 727e732
600
100
50
500 400 300 200
(c)
500 400 300
100
0
600
(d)
100
50
50 100 150 200 250 300 Songs received
0
600
(e)
500 400 300 200
50 100 150 200 250 300 Songs received
0
Approaches received
150
50 100 150 200 250 300 Songs received
Approaches initated
0
Head-down displays
600
(b) Approaches recieved
(a) Approaches initated
Head-down displays
150
731
(f)
500 400 300
100 0 Not consorted Consorted Consort status
Not consorted Consorted Consort status
Not consorted Consorted Consort status
Figure 2. Relationship between the number of songs received by brown-headed cowbirds during the breeding season and the (a) number of head-down displays, (b) number of approaches initiated and (c) number of approaches received during the autumn. Lines represent nonparametric lowess regressions to illustrate the relationship between the two variables. Relationship between a female cowbird’s consort status (i.e. consorted or not consorted) and the (d) number of head-down displays, (e) number of approaches initiated and (f) number of approaches received during the autumn. Boxes represent interquartile ranges, with the median in the middle represented by a bold line; whiskers represent the range of the highest and lowest values within a range of 1.5 times the interquartile range; dots indicate data points outside this range.
Consortships Females that formed a consortship initiated more head-downs during the autumn than females that did not form a consortship (ManneWhitney U test: U ¼ 4, N1 ¼ 8, N2 ¼ 4, P ¼ 0.05), but they did not differ in the number of approaches initiated (U ¼ 6, P ¼ 0.10) or received (U ¼ 14.5, P ¼ 0.86; Fig. 1). During the spring, consorting females also received more songs (U ¼ 1, P ¼ 0.008) and produced more response chatters (U ¼ 0.5, P ¼ 0.01) than nonconsorting females. Male Singing Behaviour and Autumn Behaviour We recorded a total of 5140 songs (median ¼ 223.5, IQR ¼ 628.5) during the breeding season, including 1770 female-directed songs (median ¼ 38, IQR ¼ 246), 2407 male-directed songs (median ¼ 50.5, IQR ¼ 289.3) and 963 undirected songs (median ¼ 70.5, IQR ¼ 98.25). In total, males sang 1122 songs (median ¼ 19.5, IQR ¼ 172.25) within countersinging matches. Our GLMs showed no significant effects of approaches initiated, approaches received or head-downs on the proportion of songs to males or females, or the proportion of songs within countersinging matches. Further analysis revealed no significant correlations between any autumn social behaviours and either the frequency or proportion of singing behaviour in the spring. DISCUSSION We investigated the early behavioural correlates of courtship skills in naïve juvenile brown-headed cowbirds. In particular, we aimed to uncover whether juvenile cowbirds, without access to adults, show robust patterns of social behaviour in the autumn, and whether these patterns predict courtship skills during their first
breeding season. The rate and selectivity of head-down displays changed across the autumn, with individuals engaging in more head-down displays with more unseparated individuals when they were reunited into a large flock. Nevertheless, we discovered that juvenile female head-down rates were correlated across changes in social contexts. The ability to maintain comparable levels of social interaction across social contexts and over time may shape the expression and ontogeny of later social behaviour. We found that variation in autumn head-downs also predicted the courtship performance of juvenile females during the following breeding season. Females that initiated more head-downs were more likely to receive songs from males, to respond to songs with a response chatter and to maintain consortships with preferred mates. In contrast, while males also initiated more head-down displays when reunited, the lack of significant correlations in juvenile male headdown rates suggests that the ability to initiate close interactions depends more on the current context than on consistent individual differences. Furthermore, male head-down rates did not predict courtship performance during the breeding season. The head-down is unique, as it is the only behaviour that allows two cowbirds to remain in close proximity for extended periods of time. While the exact functional benefits of the display are debated (Scott & Grumstrup-Scott 1983; Ortega 1998), numerous studies have shown that repeated use of head-down displays decreases interindividual distances (Stevenson 1969), facilitates group cohesion and appeases agonistic tendencies (Selander & LaRue 1961; Robertson & Norman 1976; Scott & Grumstrup-Scott 1983). While previous studies have demonstrated the presence of head-downs in wild juveniles (Lowther & Rothstein 1980), here we demonstrate that juveniles’ use of the head-down display is extensive, reciprocal and selective, and shows substantial individual variability.
732
G. M. Kohn et al. / Animal Behaviour 85 (2013) 727e732
Robust variation in social tendencies has been observed in a wide diversity of species (Mills & Faure 2000; Sibbald et al. 2005; Kralj-Fiser et al. 2007; Cote et al. 2008; Gibbons et al. 2010; Kohn et al. 2011; Koski 2011; Wey & Blumstein 2012), but its influence on ontogeny is just being recognized (Stamps 2003; Taborsky & Oliveira 2012). The predictability of head-downs across changes in group size and composition demonstrates that juvenile females vary in how they interact. By initiating more head-downs with more individuals, some females acquire more experience in close social interaction than other females. In the autumn, cowbirds converge into large roosting and feeding aggregations and must manage close interactions with a diversity of conspecifics. Social interaction during the autumn and winter has been shown to shape later female reproductive behaviour and mate preferences (Freeberg 1997, 1998; Freeberg et al. 1999). In particular, sex assortment in cowbird flocks is high (Kohn et al. 2011), and close interaction with other females during autumn may expose individuals to cues regarding the quality of male song displays (Gros-Louis et al. 2003) and shape later mate preferences (West et al. 2006). During the spring, cowbirds focus on courtship and sustaining consortships with preferred mates (Friedmann 1929; Ortega 1998). Juvenile females that initiated more head-downs with females in the autumn received more song during the spring and used more response chatters. The attentiveness to social cues required for cowbird courtship may select for individuals that can initiate and sustain close interactions. Females that are more tolerant of another’s proximity will have more opportunities to assess song and respond with a response chatter, and this may ultimately result in the formation of strong consortships. In adult cowbirds, more sociable females use response chatters more often (G. M. Kohn, unpublished data) and show increased reproductive output (Kohn et al. 2012). While more work needs to be done exploring the developmental mechanisms linking early social attributes and reproduction, our findings demonstrate that robust variation in autumn affiliative displays predicts the ontogenesis of courtship skills in birds in their first year. Acknowledgments We thank Robert Seyfarth and an anonymous referee for their comments on the manuscript. All work was conducted under ASAB/ ABS guidelines and approved by the Institutional Care and Use Committee of Indiana University (08-018). References Berman, C. M., Rasmussen, K. L. R. & Suomi, S. J. 1997. Group size, infant development and social networks in free-ranging rhesus monkeys. Animal Behaviour, 53, 405e421. Burnell, K. & Rothstein, S. I. 1994. Variation in the structure of female brownheaded cowbird vocalizations and its relation to vocal function and development. Condor, 96, 703e715. Chatterjee, S., Hadi, A. S. & Price, B. 2000. Regression Analysis by Example. 3rd edn. New York: J. Wiley. Cote, J., Dreiss, A. & Clobert, J. 2008. Social personality trait and fitness. Proceedings of the Royal Society B, 275, 2851e2858. Crawley, M. J. 2005. Statistics: an Introduction Using R. Chichester: J. Wiley. Dufty, A. M., Jr. 1982. Response of brown-headed cowbirds to simulated conspecific intruders. Animal Behaviour, 30, 1043e1052. Freeberg, T. M. 1997. Cultural transmission of behaviors facilitating assortative courtship and mating in cowbirds (Molothrus ater) Ph.D. thesis, Indiana University. Freeberg, T. M. 1998. The cultural transmission of courtship patterns in cowbirds, Molothrus ater. Animal Behaviour, 56, 1063e1073. Freeberg, T. M., Duncan, S. D., Kast, T. L. & Enstrom, D. A. 1999. Cultural influences on female mate choice: an experimental test in cowbirds, Molothrus ater. Animal Behaviour, 57, 421e426. Freed-Brown, G. & White, D. J. 2009. Acoustic mate copying: female cowbirds attend to other females’ vocalizations to modify their song preferences. Proceedings of the Royal Society B, 276, 3319e3325. Friedmann, H. 1929. The Cowbirds: a Study in the Biology of Social Parasitism. Springfield, Illinois: C. C. Thomas. Gibbons, J. M., Lawrence, A. B. & Haskell, M. J. 2010. Measuring sociability in dairy cows. Applied Animal Behaviour Science, 122, 84e91.
Gros-Louis, J., White, D. J., King, A. P. & West, M. J. 2003. Female brown-headed cowbirds’ (Molothrus ater) social assortment changes in response to male song: a potential source of public information. Behavioral Ecology and Sociobiology, 53, 163e173. King, A. P. & West, M. J. 1987. The experience of experience: an exogenetic program for social competence. In: Perspectives in Ethology (Ed. by P. P. G. Bateson & P. H. Klopfer), pp. 153e182. New York: Plenum. King, A. P. & West, M. J. 2002. Ontogeny of competence. In: Conceptions of Development (Ed. by D. J. Lewkowicz & R. Lickliter), pp. 77e104. New York: Psychology Press. King, A. P., White, D. J. & West, M. J. 2003. Female proximity stimulates development of male competition in juvenile brown-headed cowbirds, Molothrus ater. Animal Behaviour, 66, 817e828. Kohn, G. M., King, A. P., Scherschel, L. L. & West, M. J. 2011. Social niches and sex assortment: uncovering the developmental ecology of brown-headed cowbirds, Molothrus ater. Animal Behaviour, 82, 1015e1022. Kohn, G. M., King, A. P., Dohme, R., Meredith, G. R. & West, M. J. 2012. In the company of cowbirds (Molothrus ater): robust patterns of sociability predict reproductive performance. Journal of Comparative Psychology, published online 1 October 2012, http://dx.doi.org/10.1037/a0029681. Koski, S. E. 2011. Social personality traits in chimpanzees: temporal stability and structure of behaviourally assessed personality traits in three captive populations. Behavioral Ecology and Sociobiology, 65, 2161e2174. Kralj-Fiser, S., Scheiber, I. B. R., Blejec, A., Moestl, E. & Kotrschal, K. 2007. Individualities in a flock of free-roaming greylag geese: behavioral and physiological consistency over time and across situations. Hormones and Behavior, 51, 239e248. Lowther, P. E. & Rothstein, S. I. 1980. Head-down or ‘preening-invitation’ displays involving juvenile brown-headed cowbirds. Condor, 82, 459e460. Miller, J. L., King, A. P. & West, M. J. 2008. Female social networks influence male vocal development in brown-headed cowbirds (Molothrus ater). Animal Behaviour, 76, 931e941. Mills, A. D. & Faure, J. M. 2000. Ease of capture in lines of Japanese quail (Coturnix japonica) subjected to contrasting selection for fear or sociability. Applied Animal Behaviour Science, 69, 125e134. O’Loghlen, A. L. & Rothstein, S. I. 2010. Multimodal signalling in a songbird: male audiovisual displays vary significantly by social context in brown-headed cowbird. Animal Behaviour, 79, 1285e1292. Ortega, C. P.1998. Cowbirds and Other Brood Parasites. Tucson: University of Arizona Press. Robertson, R. J. & Norman, R. F. 1976. Behavioral defenses to brood parasitism by potential hosts of the brown-headed cowbirds. Condor, 78, 166e173. Scott, T. W. & Grumstrup-Scott, J. M. 1983. Why do brown-headed cowbirds perform the head-down display? Auk, 100, 139e148. Selander, R. K. & LaRue, C. J. 1961. Interspecific preening invitation display of parasitic cowbirds. Auk, 78, 473e504. Sibbald, A. M., Elston, D. A., Smith, D. J. F. & Erhard, H. W. 2005. A method for assessing the relative sociability of individuals within groups: an example with grazing sheep. Applied Animal Behaviour Science, 91, 57e73. Stamps, J. 2003. Behavioural processes affecting development: Tinbergen’s fourth question comes of age. Animal Behaviour, 66, 1e13. Stamps, J. & Groothuis, T. G. G. 2010. The development of animal personality: relevance, concepts and perspectives. Biological Reviews, 85, 301e325. Stevenson, M. 1969. Agonistic behavior in the cowbird Molothrus ater Ph.D. thesis, Kansas State University. Suomi, S. J. 1991. Uptight and laid-back monkeys: individual differences in the response to social challenges. In: Plasticity of Development (Ed. by S. E. Brauth), pp. 27e56. Cambridge, Massachusetts: MIT Press. Taborsky, B. & Oliveira, R. F. 2012. Social competence: an evolutionary approach. Trends in Ecology & Evolution, 27, 679e688. Taborsky, B., Arnold, C., Junker, J. & Tschopp, A. 2012. The early social environment affects social competence in a cooperative breeder. Animal Behaviour, 83, 1067e1074. West, M. J. & King, A. P. 1988. Female visual displays affect the development of male song in the cowbird. Nature, 334, 244e246. West, M. J., White, D. J. & King, A. P. 2002. Female brown-headed cowbirds’ (Molothrus ater) organization and behaviour reflects male social dynamics. Animal Behaviour, 63, 377e385. West, M. J., King, A. P., White, D. J., Gros-Louis, J. & Freed-Brown, G. 2006. The development of local song preferences in female cowbirds (Molothrus ater): flock living stimulates learning. Ethology, 112, 1095e1107. Wey, T. W. & Blumstein, D. T. 2012. Social attributes and associated performance measures in marmots: bigger male bullies and weakly affiliating females have higher annual reproductive success. Behavioral Ecology and Sociobiology, 66, 1075e1085. White, D. J. & Smith, V. A. 2007. Testing measures of animal social association by computer simulation. Behaviour, 144, 1447e1468. White, D. J., King, A. P. & West, M. J. 2002. Facultative development of courtship and communication in juvenile male cowbirds (Molothrus ater). Behavioral Ecology, 13, 487e496. White, D. J., Gros-Louis, J., King, A. P., Papakhian, M. & West, M. J. 2007. Constructing culture in cowbirds (Molothrus ater). Journal of Comparative Psychology, 121, 113e123. White, D. J., King, A. P., West, M. J., Gros-Louis, J. & Tuttle, E. M. 2010. Effects of singing on copulation success and egg production in brown-headed cowbirds, Molothrus ater. Behavioral Ecology, 21, 211e218. Yokel, D. A. 1986. Monogamy and brood parasitism: an unlikely pair. Animal Behaviour, 34, 1348e1358.
Contents lists available at SciVerse ScienceDirect
Animal Behaviour journal homepage: www.elsevier.com/locate/anbehav
Robust autumn social attributes predict spring courtship skills in juvenile female brown-headed cowbirds, Molothrus ater ˇ
Gregory M. Kohn*, Andrew P. King, Rebekka Dohme, Gwendwr R. Meredith, Meredith J. West Department of Psychological and Brain Sciences, Indiana University, Bloomington, IN, U.S.A.
a r t i c l e i n f o Article history: Received 2 September 2012 Initial acceptance 16 November 2012 Final acceptance 4 January 2013 Available online 28 February 2013 MS. number: A12-00663R Keywords: behavioural ontogeny brown-headed cowbird courtship behaviour juvenile sociability Molothrus ater ontogeny of courtship reproductive competence social temperament
Individuals often show predictable variation in social tendencies over time and across contexts. In species that rely on social interaction for the development of courtship behaviours, predictable variation in juvenile social behaviour may influence the ontogeny of courtship skills by exposing young individuals to different levels of contact with conspecifics. During the autumn and winter, brown-headed cowbirds utilize a social display known as the head-down to facilitate close social interaction with others. In this study we investigated whether head-downs rates are correlated across social contexts, and thus constitute a social temperament trait, and whether variation in head-down rates predicts subsequent courtship behaviour in juvenile cowbirds. During the autumn, we used a fissionefusion perturbation on a flock of juvenile females and males, and recorded the number of approaches and head-downs across social contexts. All individuals were then followed into the breeding season, where we recorded male singing behaviour, female chatter vocalization and consort status. We found that female head-down rates were correlated across contexts during the autumn, and that females that engaged in more head-down displays received more songs from males, responded to more songs with chatter vocalizations and were more likely to be in a consortship during the breeding season. As far as we know, this study presents the first demonstration that a social temperament trait can predict competent adult-like courtship behaviour in juvenile birds, and suggests that juvenile sociability may be an important developmental mechanism in the emergence of reproductive competence. The Association for the Study of Animal Behaviour. Published by Elsevier Ltd.
Little is known about the ontogeny of courtship skills in juvenile female birds and whether variation in social interaction tendencies reflects the development of these skills. Successful courtship requires sustained social attentiveness and willingness to engage others in close proximity (King & West 1987; Taborsky et al. 2012). Within many species, juveniles require close social interaction with conspecifics for the development of social skills (Suomi 1991; King & West 2002). Therefore, selection may favour predictable social tendencies early in ontogeny that facilitate close interaction, as a way of securing the experiences needed for later successful reproduction. The purpose of this study was to identify whether such predictable social tendencies exist in immature birds and whether these behaviours predict the early courtship skills needed for successful reproduction as adults. The brown-headed cowbird is an obligate brood parasite. Upon fledging from host nests, cowbirds join large flocks of conspecifics. In many locations, adults migrate in early summer, leaving
* Correspondence: G. M. Kohn, Department of Psychological and Brain Sciences, Indiana University, 1101 E. 10th Street, Bloomington, IN 47405, U.S.A. E-mail address: [email protected] (G. M. Kohn).
fledglings to form exclusively juvenile groups (Friedmann 1929). Previous studies have found that interaction with other cowbirds is essential for the development of species-typical mate preferences. For instance, juvenile cowbirds with access to experienced adults exhibit more competent courtship behaviour than juveniles without adult access (White et al. 2002). Freeberg (1998) demonstrated that juvenile females adopted song and mating preferences of adults from different subspecies with which they spent the winter. Previous work in developmental and comparative psychology has demonstrated that reliable early differences in how children and primates cope with social challenges and interact with conspecifics has long-lasting effects on the ontogeny of social and reproductive behaviour (Suomi 1991; Berman et al. 1997). Therefore, early differences in how juveniles interact with others may reflect the emergence of later behavioural skills. Within primates and birds, individuals possess ritualized social behaviours that facilitate close interaction with others while mitigating aggressive tendencies. These ritualized behaviours reflect, and help to define, an individual’s sociability. Cowbirds engage in a social display known as the head-down. During a head-down, an individual orients towards a neighbouring individual and adopts a low crouched position with the head pointed downwards in close
0003-3472/$38.00 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.anbehav.2013.01.009
728
G. M. Kohn et al. / Animal Behaviour 85 (2013) 727e732
proximity, if not touching their neighbour. The head-down is the only behaviour that brings cowbirds into close proximity for extended periods of time, and studies have shown that repeated use of the display has an appeasing influence (Selander & LaRue 1961; Robertson & Norman 1976; Scott & Grumstrup-Scott 1983). Like other affiliative displays, such as grooming in primates and allopreening in other birds, head-downs are commonly reciprocated, with cowbirds responding to another’s display with a head-down. The head-down is also seasonal and occurs mostly during the autumn and winter when cowbirds converge into large flocks (Ortega 1998). Consequently, individuals that initiate more head-downs during the autumn will have more experience in close social interactions than individuals that initiate fewer head-downs. Few studies outside primates have investigated whether robust variation in social interaction tendencies exists early in life, and whether such variation predicts the emergence of later social skills. Robustness indicates that behavioural traits are predictable across contexts, or that measures of a behaviour in one context are correlated with measures of behaviour in another context (Stamps & Groothuis 2010). The opposite of robustness is contextual specificity, where measures of behaviour in one context are not correlated with measures of the same behaviour in another context. Adult female and male cowbirds show correlated variation in autumn head-down rates across contexts (G. M. Kohn, A. P. King, U. Pott & M. J. West, unpublished data). Breeding season flocks composed of adults that initiated more head-downs during the autumn showed higher rates of egg production and courtship behaviour the following spring (G. M. Kohn, A. P. King, U. Pott & M. J. West, unpublished data). While the head-down display has been observed in juveniles, it has not been systematically studied, and little is currently known about robustness of social behaviour and its relationship to courtship skills in immature juveniles. In the current study, we manipulated group size and composition to see whether juveniles show robust patterns of head-down displays. Cowbird courtship revolves around female response to male song. Males sing directed songs by orienting towards a conspecific and giving a song within inches while performing a bowing display. As female cowbirds can withdraw during the onset of the male song display (Gros-Louis et al. 2003), males modulate the expression of their display when directing it towards females in order to reduce withdrawal (O’Loghlen & Rothstein 2010). Adult females assess song quality and communicate preferences using both visual (West & King 1988) and acoustic signals (Freed-Brown & White 2009) in response to male song displays (West et al. 2002, 2006; Miller et al. 2008). Previous studies have shown that adult females that receive more directed songs also produce more eggs (White et al. 2010). Males also exchange directed songs with other males in countersinging matches, and females may use countersinging matches as opportunities to assess male song quality and competitive ability (White et al. 2010). While female behaviour is an essential component of successful adult courtship and reproduction (West & King 1988; King et al. 2003), little is known about the early variation in juvenile female courtship behaviour. Females, which do not sing, react to directed song from a preferred male with a response chatter vocalization. Response chatters are used to indicate a preference for specific males and allow for female mate copying (Freed-Brown & White 2009). Chatter vocalizations are individually distinct and are also performed outside of singing contexts throughout the breeding season (Burnell & Rothstein 1994). These undirected chatters might advertise female identity and attract male attention (Dufty 1982), but they do not communicate mate preferences as they are often performed away from males. Females with a higher proportion of response to undirected chatters will use their chatter more selectively to communicate mate preferences. Response chatters may
aid females in formation of pair bonds through advertising mate preference and selectively reinforcing specific males. Cowbirds are a primarily monogamous species, and maintaining consorts is often essential to successful reproduction (Yokel 1986). Although juveniles often do not successfully breed during their first breeding season, they often engage in courtship behaviour (White et al. 2002). Courtship skills acquired by a juvenile may increase its reproductive success when it becomes reproductively active as an adult (White et al. 2007). Thus, in this study we considered a female’s consort status, the number of songs she received and her response chatters, as well as the proportion of the female’s response chatters to undirected chatters as measures of early courtship skills. We hypothesized that robust variation in headdown displays would emerge during the birds’ first autumn, and that higher head-down rates during the autumn would predict better courtship behaviour during the first breeding season. In particular, we expected female head-down rates to correlate with consort status, songs received and number of response chatters for females, and the number of directed songs and countersinging matches in males. METHODS Subjects All birds were originally captured in Philadelphia County, Pennsylvania, U.S.A., and housed in aviaries in Monroe County, Indiana, U.S.A. All subjects were Molothrus ater ater. For this study we used 12 juvenile females and 12 juvenile males. All birds were juveniles, and at the beginning of the study, ranged in age from 55 to 75 days old with an average age of 73 days, and were housed together for 1 month prior to the present study. Cowbirds without access to adults during the first year are not sexually competent in their first breeding season (White et al. 2002). Each bird was marked with uniquely coloured leg bands to allow for individual recognition. All birds were provided daily with a diet of vitamintreated water (Aquavite Nutritional Research), red and white millet, canary seed and a modified Bronx Zoo diet for blackbirds. Aviaries We used three aviaries that were visually isolated and substantially acoustically attenuated from each other, each with identical dimensions (9.1 21.4 3.4 m). Environmental conditions were similar throughout all aviaries with shrubs, trees, grass, covered feeding stations and access to indoor enclosures. All aviaries allowed birds to be exposed to ambient climatic conditions, wild cowbirds and the sight of predators. Data Collection During the autumn 2010, we recorded both male and female social approach and head-down displays. An approach was scored when one individual approached another individual with any part of its body within a radius of 30 cm. A head-down was scored when an individual maintained a low crouched position with its head pointed downwards so that the back of the head and neck were positioned in close proximity, if not touching, a neighbouring individual for at least 1 s. Head-downs were either initiated, when one individual initiated a head-down to a conspecific that was not in the head-down posture, or reciprocated, when an individual responded to another’s head-down posture with a reciprocal display. The identity of the individuals initiating and receiving an approach or head-down was recorded for each interaction.
G. M. Kohn et al. / Animal Behaviour 85 (2013) 727e732
During the breeding season, we recorded both male and female courtship behaviour, focusing on the vocal behaviour of both males and females. For male courtship behaviour, we recorded the number of female- and male-directed songs, and the number of songs within countersinging matches. A countersinging match occurs when a male reciprocates a directed song within 15 s after receiving a song from a neighbouring male. For females, we recorded the number of songs each female received from males and the number of female chatter vocalizations. Female chatter vocalizations were either response chatters or undirected chatters based on their temporal proximity to male song. Response chatters occur when a female responds to a directed male song with chatter vocalization within 1 s. Undirected chatter vocalizations occur when the females perform chatter vocalizations outside of singing contexts. Throughout the study, we used a scan-sampling procedure: behaviours were recorded, as they were observed, using voice recognition technology described in detail by White et al. (2002). When used in combination with voice recognition technology, scan sampling can accurately acquire a more comprehensive data set than focal sampling (White & Smith 2007). Autumn behaviour was recorded in blocks of 7 min scan-sampling periods and breeding season behaviours were recorded in blocks of 15 min scan-sampling periods. All observations were conducted between 0700 and 1030 hours, when cowbirds are most active, and were counterbalanced, so each observer took the same number of scan-sampling blocks in each aviary at the same time of day. All work was conducted under ASAB/ABS guidelines and approved by the Institutional Care and Use Committee of Indiana University (08-018).
729
Stage 4 Throughout the breeding season, from 1 May to 9 June, three observers conducted daily observations recording male song and female chatter vocalizations, and recorded a total of 252 15-minute observation blocks.
sex differences in the number of head-downs and whether males or females showed significant sex preferences in the head-downs they initiated or reciprocated, we used Wilcoxon signed-ranks tests. During the fusion period, we tested whether individuals preferentially interacted with their flock-mates from the fission period. Unseparated individuals were conspecifics that shared the same aviary during the fission period, whereas separated individuals were housed in the alternative aviary during the fission period. We used Wilcoxon signedranks tests to see whether individuals preferentially approached and displayed head-downs towards unseparated or separated individuals. To examine the contextual generality of head-down behaviour across periods, we used Spearman correlations to see whether individuals maintained significantly comparable rank-ordered rates of initiated and reciprocated head-down displays across the periods during the autumn. To see whether the head-down rates reflected social approach behaviour, we used Spearman correlations to examine the relationship between the number of approaches received or initiated and the number of head-downs within each period. To examine which autumn social behaviours were predictors of courtship behaviour during the spring, we used generalized linear models (GLM) with a quasi-Poisson distribution and a log link function. We used a model simplification procedure where a maximal model was simplified through stepwise elimination of nonsignificant variables. Log-likelihood tests were used for model comparisons during simplification, with a minimal adequate model defined when model comparisons exhibited a significant loss in the residual deviance with the removal of any remaining explanatory factors (Crawley 2005). We conducted four different GLMs for the females using the number of songs received, the total number of chatter vocalizations and the number of undirected and response chatter vocalizations as the dependent variables. For males we ran four different GLMs looking at the number of total, female- and male-directed song and the number of songs within countersinging matches as the dependent variables. In all models, the explanatory variables were the number of head-downs performed (both initiated and reciprocated) and approaches initiated and received across the autumn data collection period. We conducted an individual generalized linear regression with a binomial error distribution to investigate the relationship between autumn head-down rates and the proportion of response chatters. All post hoc tests included Spearman correlations to look at the strength of the relationship between the explanatory variables and the dependent variables for each model. As the explanatory variables were intercorrelated, we used variance inflation factors (VIF) to assess multicollinearity. A variance inflation factor greater than 10 is used to indicate potential multicollinearity, which makes model interpretation difficult (Chatterjee et al. 2000). We calculated VIFs for both the saturated models and for each resulting model during model simplification. In none of our presented models did the VIFs for any factor exceed 10. We considered a female to be in a consortship with a male when the songs of that particular male accounted for 75% of the songs she received and when she was one of the top two females a male sang to. To look at the influence of autumn social behaviour on consort status during the breeding season, we used ManneWhitney U tests. Specifically, ManneWhitney U tests were used to see whether consorted females showed higher rates of social approach or headdown displays during the autumn than unconsorted females.
Analysis
RESULTS
Because of small sample sizes and non-normality of data, we primarily used nonparametric statistics. We used Friedman tests and Wilcoxon signed-ranks tests to see whether the rates of head-down displays differed across periods during the autumn. To investigate
Autumn Head-down Displays
Juvenile Autumn Social Behaviour Stage 1 Baseline period. From 7 September to 11 October 2010, three observers conducted daily observations in Aviary 1, each recording approaches and head-downs, collecting a total of 165 7-minute observation blocks. Stage 2 Fission period. On 13 October, we separated birds into Aviary 1 and Aviary 2 based on their frequency of approaches during the baseline period, so that both flocks contained individuals with a similar range of approaches. Both aviaries contained 12 cowbirds including six males and six females. From 13 October to 24 October, three observers conducted daily counterbalanced observations recording approaches and head-downs, and collected a total of 195 7-minute observation blocks (98 blocks in Aviary 1, 97 blocks in Aviary 2). Stage 3 Fusion period. On 27 October, all birds were moved into Aviary 1. From 27 October to 6 November, three observers conducted daily observations recording approaches and head-downs, and collected a total of 166 7-minute observation blocks.
We observed a total of 1289 head-down displays during the autumn (median ¼ 39.0, interquartile range, IQR ¼ 4875), including
730
G. M. Kohn et al. / Animal Behaviour 85 (2013) 727e732
640 femaleefemale, 106 femaleemale, 194 maleefemale and 349 maleemale head-down displays. Females were more likely to initiate head-down displays with other females than they were with males (Wilcoxon signed-ranks: T ¼ 78, P < 0.001), whereas males showed no significant sex bias (T ¼ 19, P ¼ 0.13). Overall, the rates of head-downs were significantly higher during the fusion period than during the baseline period (T ¼ 25, N ¼ 24, P ¼ 0.007). A Friedman test showed a significant effect of period on both male (c22 ¼ 18:2, N ¼ 12, P ¼ 0.0002) and female (c22 ¼ 17:7, N ¼ 12, P ¼ 0.0001) head-down rates. Post hoc analysis demonstrated that females displayed significantly higher rates of head-downs during the fusion period than during the baseline period (Wilcoxon signed-ranks: T ¼ 25, N ¼ 12, P ¼ 0.007). For males, the rates of head-downs were significantly higher during the fusion period than during the baseline (T ¼ 6, N ¼ 12, P ¼ 0.0001) and fission (T ¼ 24, N ¼ 12, P ¼ 0.006) periods. During the fusion period, individuals initiated a higher proportion of head-downs and approaches towards unseparated individuals from the same fission flock than they did towards individuals from the alternative flock. Both females and males almost exclusively approached unseparated individuals (Wilcoxon signed-ranks test: females: T ¼ 144, N ¼ 12, P < 0.0001; males: T ¼ 127, N ¼ 12, P ¼ 0.002) and preferentially directed head-downs towards unseparated conspecifics (females: T ¼ 113.5, N ¼ 13, P < 0.0005; males: T ¼ 133, N ¼ 12, P ¼ 0.0005). Females that approached other individuals more often also initiated more headdowns, while males that received more approaches consistently initiated more head-downs (Table 1). We found no significant sex difference between the numbers of head-down displays initiated (ManneWhitney U test: U ¼ 86.5, N1 ¼ 509 females, N2 ¼ 381 males, P ¼ 0.41) or reciprocated (U ¼ 85, N1 ¼ 240 females, N2 ¼ 159 males, P ¼ 0.47) across the autumn. Across all periods during the autumn, females maintained the same rank-ordered distributions in the rates of head-down displays initiated, whereas male head-downs initiated remained correlated across the baseline to fission periods (Table 2). We observed no consistent relationships for rates of head-downs reciprocated across social periods for males or females.
Table 2 Spearman rank correlations for head-down rates of brown-headed cowbirds that were initiated or reciprocated across the autumn social perturbations
Females Males
Initiated Reciprocated Initiated Reciprocated
Baselineefission
Fissionefusion
Fusionebaseline
0.87** 0.77** 0.72* 0.01
0.84* 0.45 0.23 0.69*
0.71* 0.34 0.38 0.08
*P < 0.05; **P < 0.01.
Chatter Vocalizations and Autumn Head-down Displays From 1 May to 9 June we recorded a total of 1239 (median ¼ 23.0, IQR ¼ 190.3) chatter vocalizations, including 539 (median ¼ 11.50, IQR ¼ 56.5) response chatters and 700 (median ¼ 11.34, IQR ¼ 102.5) undirected chatters. Our models predicting the total amount of chatter vocalizations and undirected chatter vocalizations found no significant effects. The model predicting the number of response chatters uncovered a significant main effect of head-downs (Poisson regression: t ¼ 2.310, N ¼ 12, P ¼ 0.04). Our post hoc correlations uncovered a significant positive relationship between the number of response chatters and the number of head-downs initiated (Spearman rank correlation: rS ¼ 0.59, N ¼ 12, P ¼ 0.04), but no significant correlation between the number of response chatters and the number of approaches initiated (rS ¼ 0.37, N ¼ 12, P ¼ 0.23) or received (rS ¼ 0.20, N ¼ 12, P ¼ 0.51). The number of female response chatters was correlated with female head-downs to other females (rS ¼ 0.57, N ¼ 12, P ¼ 0.05) but not with female head-downs towards males (rS ¼ 0.34, N ¼ 12, P ¼ 0.28). Further analysis examining the proportion of response chatters to undirected chatters revealed a significant relationship between all head-down rates in the autumn and the proportion of response chatters. The proportion of response chatters was significantly correlated with head-downs (binomial regression: t ¼ 7.79, P < 0.001, rS ¼ 0.92, N ¼ 12, P < 0.0001). Therefore, cowbirds that engaged in more head-down displays during the autumn had a higher proportion of response chatters to undirected chatters (Fig. 2).
Female Songs Received and Autumn Head-down Displays
Table 1 Spearman rank correlations between the number of approaches initiated (AI) or received (AR) by brown-headed cowbirds and the number of head-down displays (HD) within the baseline, fission and fusion periods
Females Males
AIeHD AReHD AIeHD AReHD
*P < 0.05; **P < 0.01; ***P < 0.001.
Baseline
Fission
Fusion
0.73** 0.25 0.47 0.76**
0.62* 0.5 0.16 0.57*
0.91*** 0.59* 0.77** 0.67*
150
Head-down displays
We recorded a total of 1770 songs to females (median ¼ 38, IQR ¼ 246). Our model showed a significant effect of head-downs (Poisson regression: t ¼ 2.701, P ¼ 0.02) on the number of songs a female received during spring. Post hoc correlations found a significant correlation between number of songs received and headdowns (Spearman rank correlation: rS ¼ 0.68, N ¼ 12, P ¼ 0.02), but not between number of songs received and approaches initiated (rS ¼ 0.38, N ¼ 12, P ¼ 0.22) or received (rS ¼ 0.40, N ¼ 12, P ¼ 0.20; Fig. 1). The number of songs females received was correlated with female head-downs to other females (rS ¼ 0.71, N ¼ 12, P ¼ 0.01) but not with female head-downs towards males (rS ¼ 0.21, N ¼ 12, P ¼ 0.49).
100
50
0 0
0.1
0.2 0.3 0.4 0.5 Proportion response chatter
0.6
Figure 1. Relationship between head-down rates of brown-headed cowbirds during the autumn and the proportion of response chatters in the spring. Regression lines are plotted to illustrate the strength and direction of the relationship.
G. M. Kohn et al. / Animal Behaviour 85 (2013) 727e732
600
100
50
500 400 300 200
(c)
500 400 300
100
0
600
(d)
100
50
50 100 150 200 250 300 Songs received
0
600
(e)
500 400 300 200
50 100 150 200 250 300 Songs received
0
Approaches received
150
50 100 150 200 250 300 Songs received
Approaches initated
0
Head-down displays
600
(b) Approaches recieved
(a) Approaches initated
Head-down displays
150
731
(f)
500 400 300
100 0 Not consorted Consorted Consort status
Not consorted Consorted Consort status
Not consorted Consorted Consort status
Figure 2. Relationship between the number of songs received by brown-headed cowbirds during the breeding season and the (a) number of head-down displays, (b) number of approaches initiated and (c) number of approaches received during the autumn. Lines represent nonparametric lowess regressions to illustrate the relationship between the two variables. Relationship between a female cowbird’s consort status (i.e. consorted or not consorted) and the (d) number of head-down displays, (e) number of approaches initiated and (f) number of approaches received during the autumn. Boxes represent interquartile ranges, with the median in the middle represented by a bold line; whiskers represent the range of the highest and lowest values within a range of 1.5 times the interquartile range; dots indicate data points outside this range.
Consortships Females that formed a consortship initiated more head-downs during the autumn than females that did not form a consortship (ManneWhitney U test: U ¼ 4, N1 ¼ 8, N2 ¼ 4, P ¼ 0.05), but they did not differ in the number of approaches initiated (U ¼ 6, P ¼ 0.10) or received (U ¼ 14.5, P ¼ 0.86; Fig. 1). During the spring, consorting females also received more songs (U ¼ 1, P ¼ 0.008) and produced more response chatters (U ¼ 0.5, P ¼ 0.01) than nonconsorting females. Male Singing Behaviour and Autumn Behaviour We recorded a total of 5140 songs (median ¼ 223.5, IQR ¼ 628.5) during the breeding season, including 1770 female-directed songs (median ¼ 38, IQR ¼ 246), 2407 male-directed songs (median ¼ 50.5, IQR ¼ 289.3) and 963 undirected songs (median ¼ 70.5, IQR ¼ 98.25). In total, males sang 1122 songs (median ¼ 19.5, IQR ¼ 172.25) within countersinging matches. Our GLMs showed no significant effects of approaches initiated, approaches received or head-downs on the proportion of songs to males or females, or the proportion of songs within countersinging matches. Further analysis revealed no significant correlations between any autumn social behaviours and either the frequency or proportion of singing behaviour in the spring. DISCUSSION We investigated the early behavioural correlates of courtship skills in naïve juvenile brown-headed cowbirds. In particular, we aimed to uncover whether juvenile cowbirds, without access to adults, show robust patterns of social behaviour in the autumn, and whether these patterns predict courtship skills during their first
breeding season. The rate and selectivity of head-down displays changed across the autumn, with individuals engaging in more head-down displays with more unseparated individuals when they were reunited into a large flock. Nevertheless, we discovered that juvenile female head-down rates were correlated across changes in social contexts. The ability to maintain comparable levels of social interaction across social contexts and over time may shape the expression and ontogeny of later social behaviour. We found that variation in autumn head-downs also predicted the courtship performance of juvenile females during the following breeding season. Females that initiated more head-downs were more likely to receive songs from males, to respond to songs with a response chatter and to maintain consortships with preferred mates. In contrast, while males also initiated more head-down displays when reunited, the lack of significant correlations in juvenile male headdown rates suggests that the ability to initiate close interactions depends more on the current context than on consistent individual differences. Furthermore, male head-down rates did not predict courtship performance during the breeding season. The head-down is unique, as it is the only behaviour that allows two cowbirds to remain in close proximity for extended periods of time. While the exact functional benefits of the display are debated (Scott & Grumstrup-Scott 1983; Ortega 1998), numerous studies have shown that repeated use of head-down displays decreases interindividual distances (Stevenson 1969), facilitates group cohesion and appeases agonistic tendencies (Selander & LaRue 1961; Robertson & Norman 1976; Scott & Grumstrup-Scott 1983). While previous studies have demonstrated the presence of head-downs in wild juveniles (Lowther & Rothstein 1980), here we demonstrate that juveniles’ use of the head-down display is extensive, reciprocal and selective, and shows substantial individual variability.
732
G. M. Kohn et al. / Animal Behaviour 85 (2013) 727e732
Robust variation in social tendencies has been observed in a wide diversity of species (Mills & Faure 2000; Sibbald et al. 2005; Kralj-Fiser et al. 2007; Cote et al. 2008; Gibbons et al. 2010; Kohn et al. 2011; Koski 2011; Wey & Blumstein 2012), but its influence on ontogeny is just being recognized (Stamps 2003; Taborsky & Oliveira 2012). The predictability of head-downs across changes in group size and composition demonstrates that juvenile females vary in how they interact. By initiating more head-downs with more individuals, some females acquire more experience in close social interaction than other females. In the autumn, cowbirds converge into large roosting and feeding aggregations and must manage close interactions with a diversity of conspecifics. Social interaction during the autumn and winter has been shown to shape later female reproductive behaviour and mate preferences (Freeberg 1997, 1998; Freeberg et al. 1999). In particular, sex assortment in cowbird flocks is high (Kohn et al. 2011), and close interaction with other females during autumn may expose individuals to cues regarding the quality of male song displays (Gros-Louis et al. 2003) and shape later mate preferences (West et al. 2006). During the spring, cowbirds focus on courtship and sustaining consortships with preferred mates (Friedmann 1929; Ortega 1998). Juvenile females that initiated more head-downs with females in the autumn received more song during the spring and used more response chatters. The attentiveness to social cues required for cowbird courtship may select for individuals that can initiate and sustain close interactions. Females that are more tolerant of another’s proximity will have more opportunities to assess song and respond with a response chatter, and this may ultimately result in the formation of strong consortships. In adult cowbirds, more sociable females use response chatters more often (G. M. Kohn, unpublished data) and show increased reproductive output (Kohn et al. 2012). While more work needs to be done exploring the developmental mechanisms linking early social attributes and reproduction, our findings demonstrate that robust variation in autumn affiliative displays predicts the ontogenesis of courtship skills in birds in their first year. Acknowledgments We thank Robert Seyfarth and an anonymous referee for their comments on the manuscript. All work was conducted under ASAB/ ABS guidelines and approved by the Institutional Care and Use Committee of Indiana University (08-018). References Berman, C. M., Rasmussen, K. L. R. & Suomi, S. J. 1997. Group size, infant development and social networks in free-ranging rhesus monkeys. Animal Behaviour, 53, 405e421. Burnell, K. & Rothstein, S. I. 1994. Variation in the structure of female brownheaded cowbird vocalizations and its relation to vocal function and development. Condor, 96, 703e715. Chatterjee, S., Hadi, A. S. & Price, B. 2000. Regression Analysis by Example. 3rd edn. New York: J. Wiley. Cote, J., Dreiss, A. & Clobert, J. 2008. Social personality trait and fitness. Proceedings of the Royal Society B, 275, 2851e2858. Crawley, M. J. 2005. Statistics: an Introduction Using R. Chichester: J. Wiley. Dufty, A. M., Jr. 1982. Response of brown-headed cowbirds to simulated conspecific intruders. Animal Behaviour, 30, 1043e1052. Freeberg, T. M. 1997. Cultural transmission of behaviors facilitating assortative courtship and mating in cowbirds (Molothrus ater) Ph.D. thesis, Indiana University. Freeberg, T. M. 1998. The cultural transmission of courtship patterns in cowbirds, Molothrus ater. Animal Behaviour, 56, 1063e1073. Freeberg, T. M., Duncan, S. D., Kast, T. L. & Enstrom, D. A. 1999. Cultural influences on female mate choice: an experimental test in cowbirds, Molothrus ater. Animal Behaviour, 57, 421e426. Freed-Brown, G. & White, D. J. 2009. Acoustic mate copying: female cowbirds attend to other females’ vocalizations to modify their song preferences. Proceedings of the Royal Society B, 276, 3319e3325. Friedmann, H. 1929. The Cowbirds: a Study in the Biology of Social Parasitism. Springfield, Illinois: C. C. Thomas. Gibbons, J. M., Lawrence, A. B. & Haskell, M. J. 2010. Measuring sociability in dairy cows. Applied Animal Behaviour Science, 122, 84e91.
Gros-Louis, J., White, D. J., King, A. P. & West, M. J. 2003. Female brown-headed cowbirds’ (Molothrus ater) social assortment changes in response to male song: a potential source of public information. Behavioral Ecology and Sociobiology, 53, 163e173. King, A. P. & West, M. J. 1987. The experience of experience: an exogenetic program for social competence. In: Perspectives in Ethology (Ed. by P. P. G. Bateson & P. H. Klopfer), pp. 153e182. New York: Plenum. King, A. P. & West, M. J. 2002. Ontogeny of competence. In: Conceptions of Development (Ed. by D. J. Lewkowicz & R. Lickliter), pp. 77e104. New York: Psychology Press. King, A. P., White, D. J. & West, M. J. 2003. Female proximity stimulates development of male competition in juvenile brown-headed cowbirds, Molothrus ater. Animal Behaviour, 66, 817e828. Kohn, G. M., King, A. P., Scherschel, L. L. & West, M. J. 2011. Social niches and sex assortment: uncovering the developmental ecology of brown-headed cowbirds, Molothrus ater. Animal Behaviour, 82, 1015e1022. Kohn, G. M., King, A. P., Dohme, R., Meredith, G. R. & West, M. J. 2012. In the company of cowbirds (Molothrus ater): robust patterns of sociability predict reproductive performance. Journal of Comparative Psychology, published online 1 October 2012, http://dx.doi.org/10.1037/a0029681. Koski, S. E. 2011. Social personality traits in chimpanzees: temporal stability and structure of behaviourally assessed personality traits in three captive populations. Behavioral Ecology and Sociobiology, 65, 2161e2174. Kralj-Fiser, S., Scheiber, I. B. R., Blejec, A., Moestl, E. & Kotrschal, K. 2007. Individualities in a flock of free-roaming greylag geese: behavioral and physiological consistency over time and across situations. Hormones and Behavior, 51, 239e248. Lowther, P. E. & Rothstein, S. I. 1980. Head-down or ‘preening-invitation’ displays involving juvenile brown-headed cowbirds. Condor, 82, 459e460. Miller, J. L., King, A. P. & West, M. J. 2008. Female social networks influence male vocal development in brown-headed cowbirds (Molothrus ater). Animal Behaviour, 76, 931e941. Mills, A. D. & Faure, J. M. 2000. Ease of capture in lines of Japanese quail (Coturnix japonica) subjected to contrasting selection for fear or sociability. Applied Animal Behaviour Science, 69, 125e134. O’Loghlen, A. L. & Rothstein, S. I. 2010. Multimodal signalling in a songbird: male audiovisual displays vary significantly by social context in brown-headed cowbird. Animal Behaviour, 79, 1285e1292. Ortega, C. P.1998. Cowbirds and Other Brood Parasites. Tucson: University of Arizona Press. Robertson, R. J. & Norman, R. F. 1976. Behavioral defenses to brood parasitism by potential hosts of the brown-headed cowbirds. Condor, 78, 166e173. Scott, T. W. & Grumstrup-Scott, J. M. 1983. Why do brown-headed cowbirds perform the head-down display? Auk, 100, 139e148. Selander, R. K. & LaRue, C. J. 1961. Interspecific preening invitation display of parasitic cowbirds. Auk, 78, 473e504. Sibbald, A. M., Elston, D. A., Smith, D. J. F. & Erhard, H. W. 2005. A method for assessing the relative sociability of individuals within groups: an example with grazing sheep. Applied Animal Behaviour Science, 91, 57e73. Stamps, J. 2003. Behavioural processes affecting development: Tinbergen’s fourth question comes of age. Animal Behaviour, 66, 1e13. Stamps, J. & Groothuis, T. G. G. 2010. The development of animal personality: relevance, concepts and perspectives. Biological Reviews, 85, 301e325. Stevenson, M. 1969. Agonistic behavior in the cowbird Molothrus ater Ph.D. thesis, Kansas State University. Suomi, S. J. 1991. Uptight and laid-back monkeys: individual differences in the response to social challenges. In: Plasticity of Development (Ed. by S. E. Brauth), pp. 27e56. Cambridge, Massachusetts: MIT Press. Taborsky, B. & Oliveira, R. F. 2012. Social competence: an evolutionary approach. Trends in Ecology & Evolution, 27, 679e688. Taborsky, B., Arnold, C., Junker, J. & Tschopp, A. 2012. The early social environment affects social competence in a cooperative breeder. Animal Behaviour, 83, 1067e1074. West, M. J. & King, A. P. 1988. Female visual displays affect the development of male song in the cowbird. Nature, 334, 244e246. West, M. J., White, D. J. & King, A. P. 2002. Female brown-headed cowbirds’ (Molothrus ater) organization and behaviour reflects male social dynamics. Animal Behaviour, 63, 377e385. West, M. J., King, A. P., White, D. J., Gros-Louis, J. & Freed-Brown, G. 2006. The development of local song preferences in female cowbirds (Molothrus ater): flock living stimulates learning. Ethology, 112, 1095e1107. Wey, T. W. & Blumstein, D. T. 2012. Social attributes and associated performance measures in marmots: bigger male bullies and weakly affiliating females have higher annual reproductive success. Behavioral Ecology and Sociobiology, 66, 1075e1085. White, D. J. & Smith, V. A. 2007. Testing measures of animal social association by computer simulation. Behaviour, 144, 1447e1468. White, D. J., King, A. P. & West, M. J. 2002. Facultative development of courtship and communication in juvenile male cowbirds (Molothrus ater). Behavioral Ecology, 13, 487e496. White, D. J., Gros-Louis, J., King, A. P., Papakhian, M. & West, M. J. 2007. Constructing culture in cowbirds (Molothrus ater). Journal of Comparative Psychology, 121, 113e123. White, D. J., King, A. P., West, M. J., Gros-Louis, J. & Tuttle, E. M. 2010. Effects of singing on copulation success and egg production in brown-headed cowbirds, Molothrus ater. Behavioral Ecology, 21, 211e218. Yokel, D. A. 1986. Monogamy and brood parasitism: an unlikely pair. Animal Behaviour, 34, 1348e1358.