Strategic intergroup alliances increase access to a contested resource in male bottlenose dolphins

Cooperation between allied individuals is ubiquitous in human societies. Our capacity to build strategic cooperative relationships across multiple social levels, such as trade or military alliances both nationally and internationally, is thought to be unique to our species. Here, however, we show that male bottlenose dolphins form the largest known multilevel alliance network outside humans, where the cooperative relationships between groups, rather than alliance size, increases male access to a contested resource. These results reveal that both dolphins and humans form strategic intergroup alliances between unrelated individuals, likely selecting for enhanced social cognition. This surprising case of convergence suggests that dolphin societies, as well as those of nonhuman primates, are valuable model systems for understanding human social and cognitive evolution.

Efforts to understand human social evolution rely largely on comparisons with nonhuman primates. However, a population of bottlenose dolphins in Shark Bay, Western Australia, combines a chimpanzee-like fission-fusion grouping pattern, mating system, and life history with the only nonhuman example of strategic multilevel male alliances. Unrelated male dolphins form three alliance levels, or “orders”, in competition over females: both within-group alliances (i.e., first- and second-order) and between-group alliances (third-order), based on cooperation between two or more second-order alliances against other groups. Both sexes navigate an open society with a continuous mosaic of overlapping home ranges. Here, we use comprehensive association and consortship data to examine fine-scale alliance relationships among 121 adult males. This analysis reveals the largest nonhuman alliance network known, with highly differentiated relationships among individuals. Each male is connected, directly or indirectly, to every other male, including direct connections with adult males outside of their three-level alliance network. We further show that the duration with which males consort females is dependent upon being well connected with third-order allies, independently of the effect of their second-order alliance connections, i.e., alliances between groups increase access to a contested resource, thereby increasing reproductive success. Models of human social evolution traditionally link intergroup alliances to other divergent human traits, such as pair bonds, but our study reveals that intergroup male alliances can arise directly from a chimpanzee-like, promiscuous mating system without one-male units, pair bonds, or male parental care.

Associates from infancy influence postweaning juvenile associations for common bottlenose dolphins (Tursiops truncatus) in Florida

In many long-lived mammalian species, association patterns between individuals have been found to influence sociality, behavioral traits, survival, and longevity. In common bottlenose dolphins (Tursiops truncatus), the early stages of development are of particular importance as associations experienced as dependent calves may influence future association patterns. While behavioral characteristics associated with the transition from a dependent calf state to an independent juvenile state have been documented, there are limited studies that examine associations between these time periods. This study aims to document association longevity for bottlenose dolphins as they transition from calves to juveniles and determine the extent to which kinship plays a role in the development of these associations. Using social network analysis, a generalized linear mixed model (GLMM), and a tiered association scale, we found 53.7% of associations were retained from the calf to the juvenile phase. GLMM results indicated that preferred associates (half-weight index [HWI] > 0.178) from the calf state were 3.6 times more likely to associate in the juvenile state (0.178 > HWI > 0) and 5.67 times more likely to be preferred associates in the juvenile state compared to nonpreferred calf associates. The majority of juveniles, 76.92%, maintained a low–moderate to moderate level association (0.089–0.54) with their mother, and a few retained their mother as their top associate. Kin were preferred associates in 46.15% of cases and found to be the top juvenile associate in 26.92% of cases. Identifying continuity in associations, particularly from the calving state to the juvenile state, is imperative as mammalian association patterns may influence community structure, disease transmission, reproductive success, and predict survival.

The emergence and early development of socio-sexual behavior in beluga calves (Delphinapterus leucas)

While the ontogeny of beluga (Delphinapterus leucas) socio-sexual behavior has been documented in animals between 37- and 108-months-old, the first 36 months of life are yet to be examined. This study investigated how socio-sexual behaviors emerge over the first three years of life in a group of belugas in managed care. The emergence of socio-sexual behaviors was relatively consistent among immature animals. More complex behaviors, like s-posture presentations, developed in a piecemeal fashion (i.e., simple components of the behavior sequence emerged before complex components). The presence of an adult male significantly predicted if immature belugas would initiate and participate in socio-sexual behavior. However, partner preferences changed with age. In the first year of life, belugas engaged in sexual behavior with their mother most frequently but preferred to play with older males if given a choice. By the third year of life, belugas engaged in socio-sexual behavior most frequently with other immature animals. These findings enhance our understanding of how belugas develop sexually and socially and have implications for social housing practices of immature belugas.

Allied male dolphins use vocal exchanges to "bond at a distance"

Vocal interactions are intrinsic features of social groups and can play a pivotal role in social bonding.^1,2 Dunbar's social bonding hypothesis posits that vocal exchanges evolved to "groom at a distance" when social groups became too large or complex for individuals to devote time to physical bonding activities.^1,3 Tests of this hypothesis in non-human primates, however, suggest that vocal exchanges occur between more strongly bonded individuals that engage in higher grooming rates^4-7 and thus do not provide evidence for replacement of physical bonding. Here, we combine data on social bond strength, whistle exchange frequency, and affiliative contact behavior rates to test this hypothesis in wild male Indo-Pacific bottlenose dolphins, who form multi-level alliances that cooperate over access to females.^8-10 We show that, although whistle exchanges are more likely to occur within the core alliance, they occur more frequently between those males that share weaker social bonds, i.e., between core allies that spend less time together, while the opposite occurs for affiliative physical contact behavior. This suggests that vocal exchanges function as a low-cost mechanism for male dolphins that spend less time in close proximity and engage in fewer affiliative contact behaviors to reinforce and maintain their valuable alliance relationships. Our findings provide new evidence outside of the primate lineage that vocal exchanges serve a bonding function and reveal that, as the social bonding hypothesis originally suggested, vocal exchanges can function as a replacement of physical bonding activities for individuals to maintain their important social relationships.

Assessment of animal management and habitat characteristics associated with social behavior in bottlenose dolphins across zoological facilities

Bottlenose dolphins are a behaviorally complex, social species that display a variety of social behaviors. Because of this, it is important for zoological facilities to strive to ensure animals display species-appropriate levels of social behavior. The current study is part of the multi-institutional study entitled "Towards understanding the welfare of cetaceans in zoos and aquariums" commonly referred to as the Cetacean Welfare Study. All participating facilities were accredited by the Alliance of Marine Mammal Parks and Aquariums and/or the Association of Zoos and Aquariums. Behavioral data were collected on 47 bottlenose dolphins representing two subspecies, Tursiops truncatus and Tursiops aduncus, at 25 facilities. The social behaviors of group related activity (group active) as well as interacting with conspecifics (interact with conspecific) were examined for their relationships to both animal management factors and habitat characteristics. The behavioral state of group active and the rate of interact with conspecific were both positively related to the frequency of receiving new forms of environmental enrichment. Both were inversely related to the random scheduling of environmental enrichment. Additional results suggested interact with conspecific was inversely related with daytime spatial experience and that males displayed group active more than females. Overall, the results suggested that animal management techniques such as the type and timing of enrichment may be more important to enhance social behavior than habitat characteristics or the size of the habitat. Information gained from this study can help facilities with bottlenose dolphins manage their enrichment programs in relation to social behaviors.

Cooperation-based concept formation in male bottlenose dolphins

In Shark Bay, Western Australia, male bottlenose dolphins form a complex nested alliance hierarchy. At the first level, pairs or trios of unrelated males cooperate to herd individual females. Multiple first-order alliances cooperate in teams (second-order alliances) in the pursuit and defence of females, and multiple teams also work together (third-order alliances). Yet it remains unknown how dolphins classify these nested alliance relationships. We use 30 years of behavioural data combined with 40 contemporary sound playback experiments to 14 allied males, recording responses with drone-mounted video and a hydrophone array. We show that males form a first-person social concept of cooperative team membership at the second-order alliance level, independently of first-order alliance history and current relationship strength across all three alliance levels. Such associative concepts develop through experience and likely played an important role in the cooperative behaviour of early humans. These results provide evidence that cooperation-based concepts are not unique to humans, occurring in other animal societies with extensive cooperation between non-kin.