Search

Vocal behaviour offers a window into understanding the social life and evolution of animals. Colobine monkeys show great interspecific and interpopulation variation in their social organization and behaviour. Recent research has shown that Rwenzori Angolan colobus (Colobus angolensis ruwenzorii) differ substantially from other black-and-white colobus in forming a multi-level society. No previous research has been conducted on the communication of C. a. ruwenzorii, but the social complexity hypothesis for communication suggests that more complex societies should evolve more complex communication repertoires. Our objective was to catalogue the vocal repertoire of C. a. ruwenzorii at Nabugabo, Uganda, and to compare it with the data available on congeners regarding intergroup tolerance, vocal repertoire size, and acoustic and behavioural features of vocal communication. Vocalizations were subject to spectrographic and behavioural analysis, and a descriptive analysis of each vocalization type was made. The influence of a few environmental and social factors on calling rates was also examined. We describe five vocalizations (i.e., the snort, roar, squeak, scream and pok) and one non-vocal signal (i.e., the tongue click) in this subspecies and their contexts. Distinct alarm calls are made for dogs, and these are given more often near the edge of the forest where humans frequent. We did not find that C. a. ruwenzorii showed a greater vocal repertoire than C. guereza or C. polykomos, which do not live in multi-level societies. Further, preliminary data do not indicate greater calling rates in larger core units of C. a. ruwenzorii compared to smaller units. These findings support the view that these primates’ vocalizations tend to be relatively conserved despite large differences in social organization.

There are many ways for signalling systems to be complex, one of which is a large signal repertoire. However, it is frequently challenging to estimate larger repertoires accurately. We present the first description of repertoire size for a subspecies of marsh wren on the Gulf Coast of North America,Cistothorus palustris thryophilus, using four repertoire estimation techniques (simple enumeration, Wildenthal curve-fitting, Davidson & Wilkinson curve-fitting, and the Coupon Collector model). We demonstrate that the repertoire estimation methods in question perform with varying degrees of accuracy under natural conditions, and propose that rather than establishing an absolute repertoire size for a given individual it may instead be necessary to use multiple techniques to establish a repertoire range. We find thatC. p. thryophilushas a song repertoire size falling well within the range of repertoires produced by other marsh wren subspecies in eastern North America, although it may be on the larger end of this range. We find no evidence supporting a correlation between migratory distance and increased repertoire size in this species, as might be predicted under a sexual selection hypothesis.

Many factors could influence repertoire size and function. Here, we provide a review of the descriptions of vocal repertoires of New World jays and a unifying terminology for function of their vocalizations. We summarized 19 different functional types of vocal behaviour shared by at least two species, for review purposes. Comparing the 14 studied species, we also classified vocalizations into four major classes defined a priori: ‘Social Integration’, ‘Conflict Resolution’, ‘Predator-related’, and ‘Breeding’. Cooperative breeding, as well as group size, are related not only to repertoire size, but to how species allocate their vocalizations into functional classes. We discuss the tendencies of Conflict Resolution and Social Integration and the possibility of the existence and implications of a trade-off between those functional classes. We identified gaps in the current knowledge on vocalizations of the studied species, and emphasize the importance of empiric evidence of vocalization function in the New World jays.

Consistent individual differences in animal behaviour are an increasingly common focus of research across various behavioural and biological sciences. Such ‘animal personalities’ comprise a diverse repertoire of behavioural tendencies, recently expanding to incorporate the social domain. Aggression and peace, hallmarks of many social systems including that of humans, warrant integration with this literature. Specifically, animal personality research should consider the potential role of stable conflict and post-conflict behavioural tendencies. We focus our discussion primarily on examples in nonhuman primates and humans, but suggest that individual variation patterns are relevant for any social species in which these phenomena exist. In highly gregarious species, an individual’s conflict and post-conflict tendencies can affect the strength and stability of its social bonds. Because social relationships in turn impact survival and reproductive success, we also encourage future work to investigate the ultimate (i.e., fitness-relevant) consequences of individual variation in aggressive and peaceful behaviour.

The relationship between song sharing among male songbirds and geographic distance between their territories is a function of juvenile dispersal distance and length of the song learning period. I analysed the spatial pattern of syllable sharing in white-throated thrushes (Turdus assimilis) to infer their song learning strategy. Songs of 23 males from three different sites (2.5–21 km) in Puntarenas, Costa Rica, were studied using visual classification, spectrogram correlation and multidimensional scaling analysis. In total, 8505 syllables were classified into 296 syllable types. All males shared syllables but most types were unique to particular males, suggesting that males invent most of their repertoires. Repertoire dissimilarity increased with geographic distance. Within sites, syllable sharing was higher between intermediate than between adjacent neighbours. These results suggest that juvenile dispersal distance is

Cultural evolution is a powerful process shaping behavioural phenotypes of many species including our own. Killer whales are one of the species with relatively well-studied vocal culture. Pods have distinct dialects comprising a mix of unique and shared call types; calves adopt the call repertoire of their matriline through social learning. We review different aspects of killer whale acoustic communication to provide insights into the cultural transmission and gene-culture co-evolution processes that produce the extreme diversity of group and population repertoires. We argue that the cultural evolution of killer whale calls is not a random process driven by steady error accumulation alone: temporal change occurs at different speeds in different components of killer whale repertoires, and constraints in call structure and horizontal transmission often degrade the phylogenetic signal. We discuss the implications from bird song and human linguistic studies, and propose several hypotheses of killer whale dialect evolution.

One of the most conspicuous behavioural differences among great apes is the paucity of tool use among wild bonobos (Pan paniscus) in comparison to chimpanzees (Pan troglodytes) who are one of the most prolific and skilled tool users in the animal kingdom. This is in spite of the fact that bonobo tool use repertories are as large and diverse as chimpanzees’ in captive settings. In this study, we compared tool using behaviours and potential drivers of these behaviours in the Wamba bonobo population located in central Democratic Republic of Congo with the Goualougo chimpanzee population of northern Republic of Congo. The tool use repertoire of wild bonobos was comprised of only 13 behaviours, compared to 42 for chimpanzees. However, the number of tool behaviours observed in each study site was similar between bonobos and chimpanzees, and many types of tool use for social, self-grooming/stimulation, and comfort/protection functions were commonly used by both species. A marked difference is that 25 of 42 tool behaviours exhibited by chimpanzees are performed for feeding, in contrast to a single report of bonobos using a leaf sponge to drink water. We examined whether the differences in tool use repertoires can be explained by the necessity, opportunity, relative profitability, or invention hypotheses. We found that habitat composition and fluctuation of fruit production at these two sites were similar, particularly when compared with variation observed between sites within each species. Thus it was unlikely that the necessity hypothesis explains the lack of tool use for feeding in bonobos. Though further study at Wamba is needed, we did not identify any obvious differences in prey availability that would indicate differences in tool using opportunities between the sites. This study could not test the relative profitability hypothesis, and further research is needed on whether tool use is the most efficient means of calorie or protein intake for wild apes. Bonobos at Wamba formed much larger and stable parties than chimpanzees at Goualougo, which was contrary to the prediction by the invention hypothesis. Another explanation is that differences in tool use behaviour between bonobos and chimpanzees might not be explained by the current ecological or social conditions, but rather by circumstances during the Pleistocene Epoch. The observed species differences might also reflect divergent behavioural predispositions, rather than actual differences in cognitive abilities.

Visual communication has evolved independently in several families of frogs and is complex in diurnal stream frogs. In addition to isolated visual and acoustic signals, some species may emit multimodal displays characterized by the association of these signals. Through the analysis of focal recordings obtained in situ in southern Brazil, we investigated the visual and multimodal communication in the diurnal stream frog Crossodactylus schmidti. We found a complex visual repertoire of nine signals used in intraspecific communication, of which the signal Both legs kicking was described for the first time. Males, females and juveniles emitted visual signals, but males had a larger repertoire and used most signals in agonistic interactions. The four most common visual signals, Toe flagging, Limb lifting, Body jerking and Running-jumping, were emitted predominantly upon rocks in the stream bed. The emission of isolated visual and acoustic signals, as well as audiovisual displays with simultaneous or temporally coupled components, demonstrates that the communication in C. schmidti is complex and that the multimodal displays may have different functions. We discuss the historical, behavioural, ecological factors which may have influenced the evolution of the visual repertoire of C. schmidti, and how the ritualization of derived activities similar to visual signals may have led to the evolution of part of the repertoire. This study provides novel results on visual and multimodal communication for the genus Crossodactylus and demonstrates that C. schmidti represents an appropriate model for studies of complex displays.

Our understanding of the functioning of a species’ vocal repertoire can be greatly improved by investigating acoustic variation and using objective classification schemes based on acoustic structure. Here we used a syntactic approach to investigate the acoustic structure of the gorilla close distance vocalizations (‘close calls’), which remain as yet little understood. We examined 2130 calls of 10 mountain gorillas (Gorilla beringei beringei) from Bwindi Impenetrable National Park, Uganda, and 5 western lowland gorillas (Gorilla gorilla gorilla) from Bai Hokou, Central African Republic. We segmented calls into units using distinct acoustic features and employed model-based cluster analyses to define the repertoire of unit types. We then examined how unit types were combined into calls. Lastly, we compared unit type use between age–sex classes and the two study groups. We found that the gorilla close calls consist of 5 intergraded acoustic unit types which were flexibly but yet non-randomly concatenated into 159 combinations. Our results are in line with previous quantitative acoustic analyses demonstrating a high degree of acoustic variation in a variety of animal vocal repertoires, particularly close distance vocalizations. Our findings add on to (1) the recent argument that the common practice of describing vocal repertoires as either discrete or graded may be of little value as such distinctions may be driven by human perception and non-quantitative descriptions of vocal repertoires, and (2) recent studies indicating that flexibility in close range social calls can come about through combinatorial systems, which previously have been studied primarily in long distance vocalizations. Furthermore, our study highlights differences in the vocal repertoire of western and mountain gorillas, as expected given differences in environment and social behaviour. Our results offer opportunities for further in-depth studies investigating the function of the gorilla close calls, which will contribute to a more comprehensive understanding of ape vocal communication in general.

The classification and description of a species’ acoustic repertoire is critical to our understanding of broader behavioural patterns and provides data for future cross-species comparative studies. To date, our understanding of canid auditory communication remains limited as full acoustic repertoires have been compiled for only nine of 36 extant species. Dingoes (Canis lupus dingo) are apex predators in Australia, and while their ecology and life-history patterns have been extensively studied, their communication system remains poorly understood. Early studies noted four sound types, but whether this represented the dingoes’ full range of laryngeal and nasal sounds was unknown. We aimed to quantitatively and qualitatively describe the full acoustic repertoire of dingoes. We identified nine discrete vocalisations (i.e., laryngeal sounds) and two nasal sounds. Of these nine vocalisations, five were previously identified as common to other canid species. This study also revealed that dingoes possess a graded acoustic communication system, where the gradual change in acoustic characteristics of discrete vocalisations was noted. Dingoes also uttered ‘mixed sounds’, a finding in concordance with previous studies of social canids. Additionally, we established an ethogram to further our understanding of the contexts in which dingo acoustic communication occurs.

Signal repertoires, such as the song type repertoires of many songbirds, are thought to play a role in male mating success, with females preferring larger male repertoires over smaller ones. Yet, in many songbird species, males sing in a way that does not readily reveal their repertoire sizes by repeating each song type multiple times before switching to the next. Here I describe a potential explanation for such signal redundancy, based on the predictions of signal detection theory as it applies to intersexual communication during mate choice. According to this idea, a female’s response threshold to male mating signals (i.e., her ‘choosiness’) should select for male signal features that elicit favorable female responses (i.e., that are more ‘detectable’). Males can increase the detectability of their signals by producing them with higher redundancy, as well as by increasing their intensity and distinctiveness. Thus, in species with relatively high female response thresholds to males, such as taxa in which the sexes associate only briefly during breeding, males are expected to produce mate attraction signals that are especially stereotyped and repetitive. High signal stereotypy is also expected to be associated with features within signals that are relatively extravagant. Phylogenetic studies of a songbird group with a wide range of mating patterns, the oropendolas and caciques (family Icteridae), provide evidence consistent with these evolutionary predictions. Singing modes in this group have become more repetitive in some lineages along with the evolution of polygynous mating systems, even as various features within songs have become more extravagant.

Song in birds is traditionally associated with two major functions: mate attraction and territory establishment/defence. In some birds, these functions are divided between different types or categories of signals, while in others the song repertoire is functionally interchangeable. Grasshopper Sparrows, Ammodramus savannarum, sing two distinct songs. Buzz song is prominent early in each breeding cycle, while warble song is typically more prominent later in the breeding cycle, and is only sung by paired males. To investigate the functions of these two song categories we did population-level song surveys, and performed a song playback experiment using buzz songs and warble songs as stimuli. Males singing buzz song responded to playbacks of both song categories almost exclusively with buzz song. Responses by birds singing warble or combined (buzz-warble) song were mixed, but these males responded to both song categories with buzz song at levels greater than chance. Our results suggest that buzz song is the principal category of song used for both the traditional intra- and inter-sexual functions associated with song. The function of the warble song class remains unclear. It may have predominantly inter-sexual functions such as social cohesion, reproductive synchrony within the pair, or extra-pair mate attraction. In a second playback experiment we examined song function further by testing the different components of buzz song. Results indicate that as in some other species, the initial segment of the Grasshopper Sparrow buzz song may act as an alerting component that enhances signal detection for long distance communication.

In most oscine bird species males possess a repertoire of different song patterns. The size of these repertoires is assumed to serve as an honest signal of male quality. The Eurasian blackbird’s (Turdus merula) song contains a large repertoire of different element types with a flexible song organisation. Here we investigated whether repertoire size in Eurasian blackbirds correlates with measures of body size, namely length of wing, 8th primary, beak and tarsus. So far, very few studies have investigated species with large repertoires and a flexible song organisation in this context. We found positive correlations, meaning that larger males had larger repertoires. Larger males may have better fighting abilities and, thus, advantages in territorial defence. Larger structural body size may also reflect better conditions during early development. Therefore, under the assumption that body size is linked to overall male quality, our results are in accordance with the hypothesis that repertoire size represents an honest signal in Eurasian blackbirds that has evolved in response to sexual selection.

The aim of this study was to assess whether geographic variation in wild orangutan behavior reflected the presence of innovations or merely variation induced by genetic or environmental differences. We improved upon previous attempts to answer this question in three complementary ways. First, to minimize the possible effects of differences in genetic composition or the physical, biological and social environment, we compared a population of Bornean orangutans (Pongo pygmaeus wurmbii) at a new site (Sungai Lading) with that of an established site (Tuanan). These two populations were physically separated but showed high genetic similarity and occupied very similar habitats, leaving different innovation histories as the only plausible source of significant differences in their behavioral repertoires, if these were found. Second, we used identical observation methods and overlapping observers at both sites, deliberately attempting to record the full behavioral repertoire. Third, we introduced statistical techniques to reduce the likelihood of false absences. The results showed several behavior patterns unique to the new site, as well as the absence of several patterns observed at the established site, most of them satisfying the criteria previously proposed for recognizing innovations. We, thus, confirm that orangutans produce behavioral innovations, and conclude that the geographic method for identifying innovations in wild populations is a valid, albeit conservative, approach that can complement other techniques. Finally, the results also strongly support a cultural interpretation of geographic variation in orangutan behavior.

Summary Song-type matching, a behaviour of some songbirds in which one individual replies to another’s song with a matching song type, has been studied primarily in birds that have small to moderately sized song repertoires (

Many animal signals have been produced by selection operating upon behavioural precursors that lack signal function, a process known as 'ritualization'. When the precursor remains in the repertoire, comparisons of its structure with that of the derived signal permit inference of the forces responsible. Selection for signal efficacy and recognition produce characteristic changes (e.g., increases in intensity and duration, and repetition). The tidbitting display of male fowl, Gallus gallus , has long been considered an example of ritualization. However, the changes associated with this evolutionary process have never been characterized, nor has there been any test of whether the signal is perceived as distinct from its putative precursor (self-feeding). To address this, we first analysed video recordings of males feeding and tidbitting. We found a significant increase in the intensity, duration and repetition of specific movements during the tidbitting display compared to feeding. These results suggest that tidbitting is a ritualized form of self-feeding behaviour, but the critical test requires assessment of receiver response. If females recognized the tidbitting signal as distinct from male feeding behaviour, we predicted that females would food search significantly more in response to tidbitting. The results of high-definition video playbacks support this hypothesis. This is the first empirical demonstration of the perceptual and functional predictions made by a classic ethological hypothesis.

In many animal species, chiefly birds and some marine mammals, the repertoire of acoustic communication signals represents learned features of the behavioural phenotype, acquired and transmitted across generations as cultural traditions. These signals often are found to vary spatially among sub-species or local populations (e.g., dialects), as well as among smaller social groupings (colonies, flocks, pods, clans). Moreover, within the whole repertoire of different vocal signals of a species there can be differences in the patterns of spatial variation from one type of vocalization to another. A previous study of birdsong, for example, showed that males had two different kinds of song, functionally differing and with discordant patterns of geographic variation. This, and other studies describing spatial distributions of different elements of a learned vocal repertoire, suggests a general hypothesis that functional differences among an ensemble of cultural traits may be manifest as differences in spatial variation. My extension of this hypothesis is that among such a suite of cultural traits those of similar function will tend to co-vary spatially. In this report, I use three vocalizations of a parrot species to test this hypothesis. I recorded vocalizations of the Australian ringneck ( Platycercus (= Barnardius ) zonarius ), at 49 sites across three subspecies distributions and quantified variation in the acoustic features of their vocal signals. Two of the vocal signals have similar function and, according to the hypothesis, should follow similar patterns of spatial variation, whereas the third signal has a differing function and is not expected to vary spatially in concordance with the other two signals. These predicted patterns were substantially upheld.

[Neighbour–stranger (hereafter N–S) discrimination in birds is common and has most likely evolved to settle repeated disputes more efficiently and without physical fights. We tested whether an oscine bird, the ortolan bunting (Emberiza hortulana), with a small repertoire size from a population with a local dialect and a high level of song sharing is able to discriminate between the songs of neighbours and strangers. We performed playback experiments with eighteen males to measure the response to a repeated single rendition of a single song type derived randomly from a repertoire of a neighbour or stranger. Subjects responded more aggressively to the songs of strangers than neighbours (faster approach, more calls and songs uttered), suggesting that ortolan buntings can discriminate between the songs of neighbours and strangers. Our results also suggest that cues for N–S discrimination may be based on individual within-song type variation regardless of song-type repertoire composition. We contrast our results with an earlier study where ortolan buntings from a non-dialect population were tested. Finally, we predict a positive relationship between the level of song sharing and within-song type variation, which may be maintained by selection for N–S discrimination in songbirds with small and moderate repertoire sizes., Neighbour–stranger (hereafter N–S) discrimination in birds is common and has most likely evolved to settle repeated disputes more efficiently and without physical fights. We tested whether an oscine bird, the ortolan bunting (Emberiza hortulana), with a small repertoire size from a population with a local dialect and a high level of song sharing is able to discriminate between the songs of neighbours and strangers. We performed playback experiments with eighteen males to measure the response to a repeated single rendition of a single song type derived randomly from a repertoire of a neighbour or stranger. Subjects responded more aggressively to the songs of strangers than neighbours (faster approach, more calls and songs uttered), suggesting that ortolan buntings can discriminate between the songs of neighbours and strangers. Our results also suggest that cues for N–S discrimination may be based on individual within-song type variation regardless of song-type repertoire composition. We contrast our results with an earlier study where ortolan buntings from a non-dialect population were tested. Finally, we predict a positive relationship between the level of song sharing and within-song type variation, which may be maintained by selection for N–S discrimination in songbirds with small and moderate repertoire sizes.]

Why do individuals in many songbird species sing multiple song types? Previous studies have often described the current utility of possessing a vocal signal repertoire, but this may not explain why repertoires evolve. We looked for repertoire function in black-capped chickadee (Poecile atricapillus) populations with a song type repertoire that is probably derived from the single song type found in most populations across the species' geographic range. Through observations of dawn singing, and natural and simulated territorial countersinging interactions, we tested several hypotheses on how a repertoire might facilitate improved vocal communication. We failed to find significant evidence supporting an adaptive origin for song type repertoires. Although we found that naturally countersinging males matched song type approximately twice as often as expected by chance, matching was not associated with conflict escalation in either natural or simulated contests and, therefore, the exact function of song type matching in these birds remains unclear. In addition, we found that novel song types frequently appear in small isolated populations, which suggests that a repertoire might evolve simply as the nonadaptive result of imperfect song-learning combined with geographic isolation.

[We studied conflict management and post-conflict affiliative interactions (reconciliation) in 3-to-6 year-old preschool children. We found development in reconciliatory proficiency with age for boys, but girls reconciled at similar levels throughout the age span. After peer conflict, 3-year-old boys were less likely to reconcile than same-aged girls, or as boys aged 4–6 years. Gender differences were not manifested in older same-age-groups (4–6-year olds). Younger boys (3–4-year olds) experienced higher reconciliation rates in conflicts with older boys (5–6-year olds) than in conflicts with same aged boys. We found no significant age or gender effects in distribution of reconciliatory repertoire or frequency of non-affiliative conflict management strategies (aggression and displacement activities, indicative of stress responses). However, 6-year-old children were more likely than younger children to mediate in conflicts in which they were not opponents, but had witnessed. Though not confirmed statistically, this phenomenon appeared more pronounced in girls than boys. Together, our results suggest that at 3 years of age, preschool children employ sophisticated behaviours to resolve conflicts, displaying remarkable receptiveness and responsiveness to reconciliatory behaviours exhibited by older opponents., We studied conflict management and post-conflict affiliative interactions (reconciliation) in 3-to-6 year-old preschool children. We found development in reconciliatory proficiency with age for boys, but girls reconciled at similar levels throughout the age span. After peer conflict, 3-year-old boys were less likely to reconcile than same-aged girls, or as boys aged 4–6 years. Gender differences were not manifested in older same-age-groups (4–6-year olds). Younger boys (3–4-year olds) experienced higher reconciliation rates in conflicts with older boys (5–6-year olds) than in conflicts with same aged boys. We found no significant age or gender effects in distribution of reconciliatory repertoire or frequency of non-affiliative conflict management strategies (aggression and displacement activities, indicative of stress responses). However, 6-year-old children were more likely than younger children to mediate in conflicts in which they were not opponents, but had witnessed. Though not confirmed statistically, this phenomenon appeared more pronounced in girls than boys. Together, our results suggest that at 3 years of age, preschool children employ sophisticated behaviours to resolve conflicts, displaying remarkable receptiveness and responsiveness to reconciliatory behaviours exhibited by older opponents.]

Summary Many species of fish use auditory cues as part of their reproductive repertoire but intended receivers must be able to localize sounds to make full use of this information. Specialized couplings between the ear and swim bladder are thought to be critical for acoustic localization, yet species without specialized connections use acoustic cues in reproductive displays. In an attempt to better understand mechanisms of acoustic localization, we used the round goby (Neogobius melanostomus), a hearing generalist, to assess responses to calls in the lab and field. The call used for playback was recorded in the field from an actively displaying male round goby and consisted of a series of low frequency pulses. In the field, playback of the call resulted in a significant enhancement of approaches toward, and entries into, an experimental arena as compared to when the sound was off. There was no effect on the amount of time spent near the speaker however. In the lab, males and females responded actively when calls were played and females showed a significant attraction to the playing speaker. Responses were highly directional with little angular deviation, suggesting true localization to the sound source. While the sensory mechanisms allowing round gobies to selectively respond to conspecific vocalizations remain unknown, it is clear that they do show highly directional responses to acoustic cues in both laboratory and field settings.

Summary It is believed that bird song has evolved as a reliable signal of quality of displaying individuals. Recent research has focused on costs of development of complex song. In the present paper we test if the acquired repertoire size is costly to maintain. We compared changes in song structure in male Whitethroats (Sylvia communis) after 48 h exposure to a stressor (5% body mass weight attached to the tail feathers) vs. changes observed within the same time interval in the control group. The strophe length was marginally significantly shorter in the handicapped males comparing to controls. However, the repertoire size (i.e., a measure of diversity of different song elements) remained intact in both groups. We concluded that the song repertoire in Whitethroats is a static secondary sexual trait. A review of literature has revealed no convincing examples of decreasing repertoire size in adult male songbirds. Further research is needed to improve our understanding of evolutionary and proximate mechanisms maintaining the stability of song repertoires.

Many songbird species use a repertoire of different song types in close-range interactions. In male blue tits, Cyanistes caeruleus, this includes songs with and without trill as well as scolding-like vocalizations. We here investigate using playback experiments whether these vocalization types are used at different stages during aggressive interactions, which may be reflected in differential responses of receivers. We predict that territory owning receivers will vary their strength of response with the vocalization type of a simulated intruder and that an aggressive territorial response may include song type matching, i.e. the use of the same vocalization type as the intruder. Supporting the different-signal value hypothesis, scolding-like playback elicited a stronger approach response of focal males than playback of songs with or without trill. There was no difference in song rate nor do songs without trill elicit a different strength of response than songs with trill. Males did not differ significantly in the amount of song type matching between the treatments but males that landed on the loudspeaker with playback of either songs without trill or scolding-like vocalizations did not match playback songs at all. This suggests a trade-off between approach and song type matching where song type matching is superfluous during very close approaches. Our results may suggest that in the blue tit threat display may be graded from songs with or without trill to scolding-like vocalizations and eventual attack of an intruder.

A study of the gargle call, an agonistic call of the black-capped chickadee Poecile atricapillus, for eight successive fall/winter seasons in three resident populations in Colorado (USA) revealed the following:(1) Individuals had repertoires of calls averaging 8.1 distinct types per bird (range 2-18).(2) Among individuals in a local population many calls shared some of the component syllables constituting the whole gargle calls; often the shared syllables occurred in different locations in the sequences of syllables (syntax) making up the calls.(3) Individuals also shared many of the same calls (identical or near identical syllable sequences); thus in a local sample the ensemble of call types in the population (the population repertoire) could be found by recording less than 10 birds.(4) In each population, the ensemble of gargle call types found in a given season consisted of a core group of about 10 kinds of gargle calls that persisted over all or nearly all the eight seasons of the study, but also the ensemble included a number of call types that were present for only one or a few seasons, or that occurred intermittently across time.(5) The great majority of call types that were lost from a population were explained by the loss of individual birds (carriers) by emigration or death.(6) The great majority of call types new in a population in any given season were explained by the occurrence of a new bird in the sample (local recruitment or immigration).(7) Both qualitative classification of gargle syllables and whole gargle calls by the method of visual inspection of sound spectrograms and automated quantitative analysis of acoustic features of syllables and calls were carried out in some examples, and these comparisons gave highly concordant results.Long-term studies of cultural evolution of non-song calls are infrequent but important because the sources and strength of social selection on non-song vocalizations are often likely to differ from those social factors influencing cultural evolution in territorial songs, the more often studied kind of bird acoustic signals. As a primary finding of this study we showed, using fine-grained measurements of acoustic variables, that a population retains a set of core gargle groups, each group representing a cultural tradition (behavioral lineage) exhibiting virtually no drift in detailed structure across eight consecutive seasons. In a small set of samples of core groups going back to 1988, there also was no evidence of change within a vocal tradition. These results suggest the hypothesis of strong canalizing social selection on the acoustic structure of gargles.

Summary White-browed sparrow weavers (Plocepasser mahali) are cooperatively breeding songbirds of eastern and southern Africa that live in small groups year-round consisting of a dominant breeding pair, and subordinate, non-breeding males and females. This species possesses an elaborate vocal communication system: solo songs are exclusively produced by the dominant male, duet songs are mainly produced by the dominant pair, and chorus songs, similar in syllable structure and temporal pattern to duet songs, are produced by all group members. We analysed songs of males and females with known social status recorded from different colonies in Zimbabwe and complemented these data by studying songs of captive birds at our institute. Solo songs are produced in one performance at dawn during the breeding season. Recordings of captive males on consecutive days revealed that males sing in a single solo performance 88.4 ± 4.1% of their total solo song repertoire. This suggests that dominant males recorded in Zimbabwe have a solo song repertoire of 67.0 ± 4.0 syllables, which is similar in size to those of captive males (58.3 ± 3.7 syllables). Repertoire sizes of both free-living and captive males are not correlated with the length of the solo song performance. Duetting is both antiphonal and in unison. Dominant males and females appear to have similar sized duet repertoires (51.9 ± 2.1 syllables). Recordings from captive pairs suggest that 7598% of the repertoire is shared with higher syllable sharing in more experienced pairs. Since all group members engage in duets and chorus singing, we estimate that each subordinate male and female shares the duet syllable repertoire with the breeding pair. For dominant males, the duet syllables are widely distinct from those of the solo songs; of their total syllable repertoire only 2.1% occur in both repertoires. Further, solo song and duet song differ in the temporal organisation.

We describe a stochastic model that simulates how the behavioural repertoires of animals living in complex and stochastic environments change in response to the changing environment. The model incorporates concepts derived from research in three main areas: (a) cognitive models of the environment, (b) the information-primacy hypothesis, and (c) approach/avoidance behaviours to novel and unfamiliar stimuli. The model animal is in an environment that stochastically changes, and in order to reduce primary needs it must perform behaviours that are appropriate to the changing environmental conditions. Different behaviours from its behavioural repertoire are performed and if the chosen behaviour matches the current environmental conditions the animal is able to reduce its primary needs with a probability that increases with the degree of match. A large mismatch between the performed behaviour and the environment simulates the situation where the behaviour does not meet the current environmental conditions, and hence there is a high degree of uncertainty and little chance of reducing primary needs. In such cases a plausible new behaviour may be created and added to the repertoire, and the unsuccessful behaviour can be deleted. Thus a history of small mismatch values indicates that performance of the behaviour has reliably met the environmental requirements; whilst a history of large mismatch values reflects far greater uncertainty associated with the performance of the behaviour. In the model a high priority to satisfy primary needs favours the selection of behaviours with histories of small mismatch values, and a high priority to reduce uncertainty favours behaviours with histories of high mismatch values. The degree of mismatch interacts with the current cognitive ability of the animal to process the information to create one of two types of negative affect; i.e. either boredom or anxiety. Boredom and anxiety interact with the need state to determine whether reliable or uncertain behaviours are performed. Simulations of low-variability environments result in a small behavioural repertoire, the repetitive performance of only a few behaviours, and an affective state that is neither markedly anxious nor bored. Whilst simulations of high-variability environments result in a large behavioural repertoire, the performance of many behaviours in different sequences, and an affective state that frequently shifts between high levels of boredom and anxiety. The time it takes after a switch to a novel environment for the behavioural repertoire to include behaviours successful in reducing primary needs in that novel environment is less when coming from a high-variability environment than when coming from a low-variability environment. The outputs of the model are compared to the behaviour of real animals living in environments with differing degrees of change, and the crucial role of uncertainty reduction in the creation of behavioural competence is discussed.

In most studies to date, innovations were studied if their origination was witnessed or if they arose in response to a pronounced environmental change, making it difficult to generalize. In this study, we use an operational definition developed by Ramsey et al. (MS) to design a procedure for recognizing the standing repertoire of innovations (in the sense of behaviors) in a natural population. The data were derived from an intensive field study of orangutans at Tuanan, Borneo. The main recognition criteria are (1) the incomplete geographic prevalence of the behavior, (2) identified causes of its absence in a population or an individual, and (3) comparison with the incidence of the behavior among captive orangutans. Using this procedure, we recognized 19 innovations at Tuanan and 43 for orangutans in general. Cumulative curves of number of innovations indicated that the total number of innovations observed at Tuanan remained stable after some 3,000 hours of observation, and is thus adequate for comparison with other studies. Additionally, an individual's repertoire size remained stable after ca 1,000 hours. The results showed that innovations are found in multiple domains (subsistence, comfort, and social communication), and that innovations that are performed more often are more likely to reach cultural status in a population. Across populations, innovations that increase comfort are less likely to become cultural than those that serve subsistence or are used in communication. Orangutan and chimpanzee innovation repertoires do not show significant differences across the three domains. Systematic comparisons across sites and with captivity will make it possible to validate the approach.

Male song has been shown to have two major functions, repelling rivals and attracting mates. Yet much debate still exists as to what extent male singing strategies have been influenced by selection pressures imposed by each sex. I use a combination of song recordings and playbacks to investigate how male competition and female choice may have shaped song structure and behaviour in migratory golden whistlers (Pachycephala pectoralis). Song output was low during territory establishment but increased dramatically when breeding commenced. Singing during the female fertile period may be an effective paternity guard against potential cuckolders. During this period, males decreased their song output when more neighbouring females were fertile, possibly reflecting an inability to concurrently guard a mate and pursue extra-pair fertilisations. However, high singing rates were maintained after the fertile period, suggesting that males may also sing to attract extra-pair copulations. Song repertoire size increased with male age and may signal male quality. Repertoires included two broad song-types: simple 'whistle' songs and more complex 'melodic' songs. Whistle songs were often sung by breeding males and may be more effective at broadcasting information at long-distances. In contrast, melodic songs were commonly sung in close-range male encounters and may contain more information about signaller quality. During vocal contests, males exhibited a range of countersinging behaviours, including song-type switching, song-type and frequency-matching, and song overlapping, which may reflect a hierarchical mode of signalling aggression. Each sex may have imposed different selection pressures on the vocal output of male golden whistlers.

Three processes, production, usage, and response, can be used to describe vocal ontogeny. They may develop independently of each other for a given vocalization and a given species as a result of the different selective pressures associated with each process. We have investigated vocal ontogeny in the swift fox Vulpes velox, using recordings and observations of captive foxes from the time of natal den emergence (age 3-4 weeks) to the time of natal dispersal in the wild (age 4-5 months). We first classified adult vocalizations used during the mating and pup rearing seasons into vocal types (19 types in total) and found that swift foxes have a vocal repertoire comparable in size and diversity to other canids. The repertoire of juvenile foxes contained 16 of the 19 adult-type vocalizations and one juvenile vocalization by age 10 weeks, after which no new types appeared by the end of the study period. Two of the 3 adult vocalizations not observed in juveniles appear to be associated with mating and possibly territoriality and the third is a high intensity alarm vocalization. Apart from 3 vocal types (1 alarm and 2 non-agonistic), once vocalizations had appeared in the juvenile repertoire, they did not seem to change in context over time. Juvenile responses to 5 vocalizations emitted by adults (3 alarm, 1 agonistic, and 1 non-agonistic) appeared to change with increasing age to approach adult-type responses. The emergence of these adult-type responses to 2 of the 3 alarm vocalizations coincided with their first appearance in the juvenile repertoire. The results indicate that there is variation in the degree (in terms of the number of vocal types) of modification over time among the three processes.

Summary By combining different call types, such as barks with screams, individuals have the potential to vastly increase the range of information that can be decoded by listeners. Few animal studies, however, have examined the information content of call combinations compared with the information conveyed by each call singularly. We examined several aspects of call combinations in the repertoire of wild chimpanzees in the Tai Forest, Ivory Coast, including the types of combinations, the contexts and possible functions and compared these with the use of single calls. Almost half of all vocalisations produced by the Tai chimpanzees occurred in combination with other vocalisations or with drumming. A total of 88 different types of combinations were used. Single calls and call combinations were each produced in specific contexts. The contexts in which six of ten combinations were produced differed from the contexts of at least one of the component calls. The contexts in which the combinations were produced varied from the component calls in five different ways, three having potentially novel functions. Two of the three may have an additive function, such that two pieces of information can be conveyed simultaneously, increasing message complexity. This analysis clearly shows that call combinations are an important part of chimpanzee communication. The implications with regard to the evolution of human language are discussed. Testing of the information conveyed by single calls and of call combinations, using playback experiments, is advocated.

Common nightingales (Luscinia megarhynchos) are among those bird species that possess an extremely large repertoire and perform it in a versatile singing style. Thereby, repertoire size, composition, and performance differs considerably among individuals. In this longitudinal field study, we investigated the long-term stability of these differences in the song characteristics of free-ranging nightingales. We determined the repertoire characteristics for nine adult male individuals in two successive years (three of these individuals were investigated over the course of three years) and compared these to similar measurements obtained from comparisons of song samples of different birds. Comparisons revealed remarkable differences among males, but we did not find systematic differences in the song performance of birds in successive years. Instead, song characteristics were remarkably stable within successive years. The long-term persistence of individual song characteristics suggests that they are not related to dynamically changing individual attributes, but may reflect long-term storage of information during song acquisition as juveniles. In addition, we found that the repertoire performance of adult nightingales allows fine-tuned vocal interactions among several neighbouring males.

Songbirds with large vocal repertoires often show individually distinct profiles in the quantity of using their different song types. Factors involved in determining how often a song type is used have been identified in the social domain, e.g. the singing routines of territorial neighbours. In this study, I investigated the role of song development in the formation of individual performance profiles. I examined whether and how the amount of motor practice during vocal ontogeny would account for profiles of song type use in the adult singing. The study was carried out on hand-reared nightingales (Luscinia megarhynchos) which are particularly appropriate subjects for this kind of approach. A clear positive relationship was discovered between the ontogenetic 'age' of imitations (i.e. time of emergence) and the performance rate of these song patterns. Also time of emergence played a role in whether song patterns were discarded from the adult repertoire. I assume that vocal practice plays a role in selectively consolidating song type memories, whose respective strengths then contribute to differential performance observed in the adult birds.

Studies of variation in bird song within a species have tended to concentrate on repertoire size and on song rate, while other aspects of song delivery have received little attention even though they are also likely to play an important part in the signalling process. We here examine the pattern of delivery of individual males' repertoires in the chaffinch (Fringilla coelebs) to identify variation in a range of temporal parameters of song delivery. Coefficients of variation differed between song units and across hierarchical levels. Basic units (syllables) were delivered with great temporal precision. The larger building blocks of a song (the different phrases and the flourish) showed the highest variability while variation for song duration itself was intermediate. Most variables showed consistent differences between males, but there were also consistent differences between song types within a male's repertoire, as well as between the same song types as sung by different males. This suggests an interaction between a male's phenotype and song production, but because of consistent differences between song types within repertoires, there is potential for selection to act on aspects of song delivery both of the individual and of the song type (i.e. the culturally transmitted unit).

African wild dog (Lycaon pictus) pup vocalizations were studied in Hwange National Park, Zimbabwe for weeks 3 through 7 of the socialization period. Here we present the vocal repertoire, including the use of repetitive and mixed sounds, and investigate the extent to which the emerging sound system of Lycaon conforms to predicted design features of Morton's (MS) motivation-structural rules. Features of the pup sound system are highlighted by comparison with adults and other social canids. Data were collected at three den sites (litter sizes: 8, 8, and 9) of two study packs. A total of 1903 vocalizations were classified, and eight vocal classes and seven subclasses were identified. Although all sounds identified persist into adulthood, observations indicate a delayed onset in some vocal classes, including both the lowest (i.e. rumbles) and highest (i.e. twitters) frequency sounds. As predicted by the (MS) model, pups invested heavily in high frequency, harmonic care/social soliciting sounds (91%, N = 1586 unmixed vocalizations), however, no clear association between acoustic structure and sound repetition was found. Significantly more repetition was heard in all vocal classes with the exception of moans and barks. Intra-pack aggression is generally muted in this obligate social carnivore suggesting that repetition may be a low cost strategy to induce social outcomes and obtain food. The patterning of mixed vocalizations (N = 317) was consistent with the (MS) model. Given the high degree of cooperation necessary for individual survival, the predominant use of cross-mixed sounds may serve to minimize conflict as pups begin to form relationships with littermates and adults. Noisy/noisy sounds were exceptionally rare. Comparative data suggest a relationship between the early patterning of mixed sounds and species-specific social organization in canids.

Abstract Adult pale-winged trumpeters produce a varied repertoire of vocalizations: apart from one loud 'tremolo song' we recorded 11 structurally discrete close-range calls, one of which (the 'mew') was individually distinct. There was significant variation across vocalization type with respect to the identity and behaviour of the caller and the response of the receiver(s). It was possible to group vocalizations into six broad contextual classes: alarm, recruitment, social, contact, feeding and territory defence. On detection of danger, trumpeters gave two acoustically different calls, one for aerial predators, and another for terrestrial predators or conspecific intruders. They also produced distinct calls on detection of large prey items such as snakes. These (alarm and snake-finding) call types seemed to evoke different responses by receivers and therefore appeared to be functionally referent. Vocal behaviour was positively correlated with dominance rank and at least two other calls had important roles in mediating social interactions within the group. Finally, the 'mew' call was only given when a trumpeter was separated from, and usually out of visual contact with the rest of the group. This call was functionally referent, eliciting a vocal response from receivers: they produced a loud 'grunt' call, which was also unique to this situation. This is the first experimental demonstration in a bird of the proximate factors motivating production of an individually distinct contact call.

Abstract It is predicted that differences in mammalian alarm call systems may be explained relative to the complexity of their habitat, with species inhabiting three-dimensional habitats classifying predator types (externally referential), and those living in two-dimensional environments indicating the level of risk (urgency-based). We tested this prediction in a two-dimensional environment for a small African rodent, Brants' whistling rat, Parotomys brantsii. Colony members were presented with predator models of a raptor and puff adder, as well as a human observer, to investigate their alarm call repertoire. Alarm calls consisted of simple, singlenote vocalizations, often repeated in non-rhythmic bouts. Virtually no variation was detected in the structure of alarm whistles between the calls elicited by the two model predators and humans, indicating that P. brantsii did not identify different predator types by means of vocalizations. However, note duration was dependent on the reaction of the caller: when the caller bolted towards safety, the whistle was significantly shorter than when it remained stationary. A snake and far-off human elicited relatively long-duration calls and the caller would remain above ground, signifying a low-risk situation. High-risk encounters with a nearby raptor or human provoked short calls before the whistling rat bolted underground. We conclude that P. brantsii's alarm call repertoire represents a graded 'urgency-based' system, indicating perceived threat level rather than predator type. Our study supports the prediction that mammalian alarm call systems in two-dimensional environments primarily provide information indicating the level of risk and not predator type.

Abstract Mountain gorillas use elaborate, multi-stage procedures for dealing with plant defences. This paper investigates the use of mathematically-inspired, informational measures to gauge the complexity of one of these tasks, eating thistle Carduus nyassanus, from field observations of 38 adults and juveniles. Behaviour was analysed at two levels, a detailed, movementbased description of the form of actions, and an organizational description of techniques that were composed of a series of many actions. Complexity, as measured by counting the sizes of behavioural repertoires, correlated at the two levels. Repertoires were shown to be incomplete, but the rates of cumulative increase in actions differed between tasks. Thistle eating was the most complex, and apparently involved many more actions than even chimpanzee tool-use. Techniques were highly selective arrangements of actions, so that their organization (sequence, bimanual coordination, hierarchical structure) reflected cognitive capacity. Although ideally it would preferable to estimate complexity of task organization, this may seldom be feasible, and was not in this case. Instead, the length of a regularly occurring sequence of actions may be the best practical estimate of an underlying complexity of mental process. Confidence in this measure will be increased if it broadly agrees with other, independent estimates of task complexity; in the case of gorilla plant processing, both the size of repertoire of functionally distinct actions and the degree of lateral specialization were, like sequence length, greater for thistle processing than for other tasks studied to date.

Experience can have both instructive and selective effects on vocal development in song birds. Learning by instruction occurs when one male imitates the song of another. Learning by selection occurs when a male chooses one or more songs to retain in his repertoire based on interaction with other individuals. These models of learning make different predictions about the degree of microgeographic variation in song present in wild populations of birds. If males are instructed by their immediate territory neighbors, then the songs of territory neighbors should be more similar than are the songs of non-neighbors. In contrast, if males select a song for retention that was learned elsewhere in the dialect, the songs of neighbors should be no more similar than are the songs of non-neighbors sampled from the same dialect. We compared the songs of males sampled in two sedentary populations and four migratory populations of four subspecies of the white-crowned sparrow, Zonotrichia leucophrys. In the two sedentary nuttalli populations, males on neighboring territories sang very similar songs, indicating that males are instructed by their territory neighbors after they disperse short distances to their breeding territories. Learning by selection during territory establishment after natal dispersal appears to predominate in the four migratory populations: the songs of territory neighbors were no more similar than were the songs of non-neighbors. We conclude that the sedentary/migratory distinction in the annual cycle determines the form of vocal learning that occurs between territory neighbors.

From 1992-1998, a study of up to six radio-collared packs of free-ranging African wild dogs ( Lycaon pictus ) was conducted in Hwange National Park, Zimbabwe. Wild dogs were primarily observed during the day and data were collected across seasons on vocalizations and social context. An analysis of spectral properties, along with an examination of context and function, is presented. A vocal repertoire consisting of 11 vocal classes and 18 subclasses was identified. The sound system of this socially obligate carnivore is among the most complex in Canidae. Several vocal classes, i.e. twitters, begging cries, and rumbles, appear to be unique. Heavy investment in high frequency sounds relative to other social canids is offset by a greater variety of low frequency sounds. More kinds of barks are found in its reportoire than have been reported for other canids. Sound mixing is an integral component of wild dog vocal communication and occurs via succession or superimposition of noisy and harmonic sounds. Mixing by succession occurs for all classes, while mixing by superimposition occurs for yelp/squeals, whines, moans, growls, and barks. The Lycaon vocal repertoire is highlighted by comparison with other social canids ( i.e. wolves, bush dogs, dholes, and domestic dogs). Finally, attention is drawn to features in the wild dog sound system that are both consistent and inconsistent with the motivation-structural rules of Morton's model. While most sounds conform to the model, there appear to be anomalies between the expected design features of certain vocalizations ( i.e. twitters, barks, and rumbles) and the contexts in which they are given. Some of these anomalies appear to reflect the highly cohesive social system of the wild dog.

[In this paper we describe the organisation of the song of the willow warbler and its pattern of singing, with the objective of providing robust measurements of song complexity that can be related to the functional and evolutionary aspects of song. Songs are composed of distinct individual elements, which can be reliably identified. These elements are not randomly assembled within each song: most songs start always with the same element, and thereafter elements are produced in a highly predictable way. The organisation of elements corresponds to a hierarchically branching pattern, where elements either connect predictably to a second one, or constitute a decision point where two or more transitions are possible. This branching pattern pervades the whole length of the sequence of elements, and suggests that the concept of song type is not appropriate in this species. The singing pattern could be characterised by a random delivery: consecutive songs are not more similar from each other than if they were delivered at random. Individuals differ greatly in several song characteristics. A principal component analysis performed in a comprehensive set of song complexity measurements produced 5 meaningful components: Song repertoire size (PC1); Song length (PC2); Song versatility (PC3); Element rate (PC4); and Repertoire fixation (PC5). The first component - song repertoire size — showed the highest variation between individuals and a high sameyear repeatability. This makes of song repertoire size a likely target for sexual selection in this species., In this paper we describe the organisation of the song of the willow warbler and its pattern of singing, with the objective of providing robust measurements of song complexity that can be related to the functional and evolutionary aspects of song. Songs are composed of distinct individual elements, which can be reliably identified. These elements are not randomly assembled within each song: most songs start always with the same element, and thereafter elements are produced in a highly predictable way. The organisation of elements corresponds to a hierarchically branching pattern, where elements either connect predictably to a second one, or constitute a decision point where two or more transitions are possible. This branching pattern pervades the whole length of the sequence of elements, and suggests that the concept of song type is not appropriate in this species. The singing pattern could be characterised by a random delivery: consecutive songs are not more similar from each other than if they were delivered at random. Individuals differ greatly in several song characteristics. A principal component analysis performed in a comprehensive set of song complexity measurements produced 5 meaningful components: Song repertoire size (PC1); Song length (PC2); Song versatility (PC3); Element rate (PC4); and Repertoire fixation (PC5). The first component - song repertoire size — showed the highest variation between individuals and a high sameyear repeatability. This makes of song repertoire size a likely target for sexual selection in this species.]

Abstract Some birds with song repertoires sequentially associate (or cluster) songs of different types. That is, certain song types may occur together repeatedly, even on different days. We determined whether clustering of meadowlark songs correlated with repertoire size. We also tested whether clustered songs reflect either their structural similarities, or dissimilarities. Our data were obtained from recordings of free-living individuals of two meadowlark species, eastern, Sturnella magna , and western, S. neglecta . Eastern meadowlarks have approximately 10 times more song types per bird than do westerns. Therefore, if clustering is related to repertoire size, we predicted that there should be (1) proportionately more song clusters in eastern meadowlarks than in westerns, and (2) a similar correlation across individual birds within a species, especially so in easterns, which have a broader range of repertoire sizes. All 14 easterns examined showed clusters whereas only 5 of 11 westerns did so, and the easterns had proportionately more per bird. Many of the same clusters occurred in different recordings of individual easterns. In easterns, the extent of clustering as measured by a PCA analysis correlated strongly with estimated repertoire size. Clustered songs were neither more similar in structure, nor less so, than randomly-paired songs.

Abstract The songs of male canaries are composed of different phrases; a phrase is defined as a segment of song composed by the repetition of the same syllable. When singing alone or in the presence of a male or female conspecific, the number and duration of the songs of male domestic canaries are similar. Despite these similarities, there are differences in the production of phrases between confrontations with conspecifics and singing alone. The presence of another canary, either male or female, induced songs which were characterized by longer strings of A phrases. We have previously demonstrated that this particular phrase elicited many more copulation sollicitation displays in female canaries than other kinds of phrases in the canary repertoire. Thus, the phrase A may be used by the singing males so as to affirm their readiness to interact. By singing A phrases male canaries may communicate both intrasexually (e.g. challenge to male competitors) and intersexually (e.g. courtship of the female). Thus two functions can be linked together in the same signal, the male or female receiver either having different perceptions or giving different senses to the same vocalization.

The purpose of this study was to analyse the structure, social contexts, and individual variation of South American sea lion (Otaria flavescens) airborne vocal communication during the breeding season. This is the first comprehensive analysis of the vocal repertoire of this species. Since breeding colonies are complex acoustic environments, we expect that male, female, and pup vocalizations would show individuality, and that individual variation would rely on several acoustic traits. We identified four calls for adult males: high pitched call or HPC (aggressive interactions, such as attack and retreat displays, and fights), bark (aggressive interactions, territory establishment, and usually combined with growls), growl (male-female interactions), and exhalation (after agonistic encounters); and two for adult females: mother primary call or MPC (female-pup interactions, such as after birth, during pup separation, and pup development), and grunt (agonistic interactions between females). Juveniles vocalized yearling primary calls or YPC (searching for their mothers or after being threatened by females), pups gave pup primary calls or PPC (in response to MPCs, when hungry, searching for their mothers, if nursing was interrupted, or when trying to nurse). Our results suggest that male's HPC and bark, female's MPC, and pup's PPC have acoustic features that support individuality. These calls are associated with contexts in which recognition of neighbours (HPC and bark) or relatives (MPC and PPC) may confer a relative advantage in reproductive performance. Individual variation depended on frequency, temporal, and intensity traits; such multi-trait individuality could be regarded as an adaptation to colonial life to overcome the high noise levels within pinniped breeding colonies.

The pygmy marmoset is a small South American primate with a complex social system based on cooperative breeding. Infant pygmy marmosets are extremely vocal; most of their calling is a repetitive pattern of mixed call types that is babbling-like. In a longitudinal study of vocal development in 8 infant pygmy marmosets, we recorded more than 750 calling bouts which occurred in a wide range of behavioural contexts. The infants used 16 different call types that we grouped into three categories: Adult-Like (acoustic structure consistent with that of adult calls), Adult-Variant (acoustic structure with some adult features and some variable features), and Infant (absent from the adult repertoire). The calling bouts were highly conspicuous in their duration (ranging up to more than 6.5 min/bout), complexity (up to 10 different call types/bout), and call rate with nearly 3 calls/s. When the infants were older, their call rate slowed and they shifted to using several of the Adult-Like calls with greater frequency, and used fewer Adult-Variant types. The infants did not use the Adult-Like call types appropriately when compared to the typical adult usage of those types. Caregivers were significantly more likely to respond to an infant when it was vocalizing than when it was not.

Female European starlings prefer to mate with males that have large song repertoires, and those males will, on average, be older birds. We tested the hypothesis that female European starlings (Sturnus vulgaris) that choose males with large song phrase repertoires benefit by obtaining more experienced mates that will provide better parental care. Contrary to the hypothesis, males with large repertoires did not spend more time incubating nor did they make more feeding trips to the nestlings than did males with small repertoires. After controlling for the effects of brood size and age on feeding rates, there was no correlation between the nestling feeding rates of males and females. The possibility that the benefit of mating with large repertoire males is that these males are of higher genetic quality is discussed.