SummaryOne of the outstanding questions in evolutionary biology is the extent to which mutually beneficial interactions and kin-selection can facilitate the evolution of cooperation by mitigating conflict between interacting organisms. The indirect fitness benefits gained from associating with kin are an important pathway to conflict resolution [1], but conflict can also be resolved if individuals gain direct benefits from cooperating with one another (e.g. mutualism or reciprocity) [2]. Owing to the kin-structured nature of many animal societies, it has been difficult for previous research to assess the relative importance of these mechanisms [3–5]. However, one area that might allow for the relative roles of kin-selection and mutualistic benefits to be disentangled is in the resolution of conflict over territorial space [6]. While much research has focused on group-living species, the question of how cooperation can first be favoured in solitary, territorial species remains a key question. Using 22 years of data from a population of North American red squirrels, we assessed how kinship and familiarity with neighbours affected fitness in a territorial mammal. While living near kin did not enhance fitness, familiarity with neighbours increased survival and annual reproductive success. These fitness benefits were strong enough to compensate for the effects of aging later in life, with potential consequences for the evolution of senescence. We suggest that such substantial fitness benefits provide the opportunity for the evolution of cooperation between adversarial neighbours, offering insight into the role that mutually beneficial behaviours might play in facilitating and stabilizing social systems.Graphical Abstract

Interactions with conspecifics are an important aspect of an individual's environment. Although it is well known that the presence of conspecifics can have important effects on behaviour, in general it is also now acknowledged that the composition of the social environment can vary, and that this variation may have profound effects on individual behaviour and fitness. Using a wild population of North American red squirrels, Tamiasciurus hudsonicus, we investigated the importance of the composition of the social environment in a territorial species by assessing whether the risk of intrusion faced by territory owners varied with the degree of relatedness and familiarity in their social neighbourhoods. To test this, we conducted temporary removals of territory owners and observed the time until intrusion and the identity of intruding individuals. We found that individuals in neighbourhoods with low average familiarity faced a higher risk of intrusion and that unfamiliar neighbours were more likely to intrude. Surprisingly, we found that related neighbours also posed a higher risk of intrusion. The results from our study suggest that familiarity with neighbours may be an ecologically and evolutionarily relevant measure of the social environment, even in a species considered to be ‘asocial’. Future studies should consider the potential importance of the social environment, which has heretofore been mostly overlooked, as a relevant selective pressure in asocial, territorial species.

In many species, territory advertisement is thought to be one of the primary functions of acoustic communication. North American red squirrels are a territorial species in which 'rattles' have long been thought to be the principal signal communicating territory ownership. These vocalizations have been assumed to deter intruders, thus reducing energetic costs and the risk of injury associated with direct aggressive interactions. However, this hypothesis has not been directly tested. Here we used a speaker occupation experiment to test whether red squirrel rattles function to deter conspecific rivals. We studied 29 male squirrels and removed each individual from his territory twice in a paired design. During the experimental treatment, we simulated the owner's presence after its removal by broadcasting the owner's rattle from a loudspeaker at the centre of the territory once every 7 min. During the control treatment, the territory was left in silence following the temporary removal of the owner. We found that the presence of a speaker replacement reduced the probability of intrusion by 34% and increased the latency to first intrusion by 7%, providing support for the hypothesis that rattles play an active role in reducing intrusion risk. However, intrusions were not completely averted by the speaker replacement, indicating that for some individuals vocalizations alone are not a sufficient deterrent without other cues of the territory owner.