Agonistic support occurs when a bystander intervenes in an ongoing conflict. The consequences of agonistic support may differ when provided to victims or aggressors. Supporting victims may not only protect them, but also limit the escalation of aggression among group members. Our results onTheropithecus geladashowed that support was preferentially directed towards victims and high-ranking individuals provided the highest levels of support. Whereas the support towards the aggressor had no effect in reducing its renewed aggression, it increased the frequency of subsequent conflicts among fellows. The support towards victims significantly reduced subsequent aggression both towards the victim and among other group members. The support was biased towards victims who were unrelated and shared weak bonds with the aggressors. In conclusion, victim support may be a social tool, which intervenes when other mechanisms are less likely to occur such as the case when the opponents are not kin or friends.

Peaceful third-party interventions usually occur after an aggressive encounter and can be directed toward the victim or the aggressor. Macaca tonkeana, a cercopithecine species characterized by high levels of tolerance, frequently engage in consolatory contacts, which both calm the victim and reduce the probability of further attacks against him/her. Other post-conflict affiliative interventions such as reconciliation and quadratic affiliation are also common in this species. However, little attention has been given to contacts directed toward the aggressor. Here, we explore the role of bystander affiliative interventions toward the aggressor in influencing the affective state of the aggressor and the consequences of triadic interventions at group level. We found that triadic post-conflict affiliation occurred independently from the intensity of the conflict and that it was more frequent in absence of the conciliatory contact between the opponents (reconciliation). Bystanders showed a higher amount of post-conflict affiliation toward low ranking aggressors. Post-conflict triadic affiliation functioned as a tension reduction mechanism by lowering the arousal of the aggressor, which less frequently engaged in renewed aggression. All these findings suggest that post-conflict triadic contacts in Tonkean macaques can be considered as a strategic mechanism to calm the aggressor and reduce the risk of retaliatory aggression.

In several primate species, including humans, embracing predicts the level of affiliation between subjects. To explore the functional meaning of embracing we selected Theropithecus gelada as a model species. The basic level of gelada society is the one-male unit and the integrity of the group is maintained by the strong bonds between females. In our study group, we observed three different kinds of embracing: the Frontal and Side Embraces involving a face-to-face and chest-to-chest interaction and the Posterior Embrace which consists in putting the arms around conspecifics' back and posing a cheek on it. We verified whether the quality of relationships between subjects predicts the type of embracing. Frontal and Side Embraces were more frequent between females sharing strong bonds. Posterior Embracing was randomly distributed. We found a high level of female embracing among the mothers during the first months of lactation. This may improve female cohesiveness against males thus limiting the risk of infanticide, particularly high in geladas. Embracing seems also to act as an ice-breaker favoring grooming. In conclusion, female embracing could be an affiliative strategy which has evolved to maintain group integrity and high social cohesion among females, especially mothers, who mostly need it.

No abstract