Social security: social relationship strength and connectedness influence how marmots respond to alarm calls

Individuals vary in the number and types of social relationships they maintain. If beneficial, social relationships may reduce predation risk and thus increase an individual's sense of security. We tested this hypothesis by studying the responses of female yellow-bellied marmots (Marmota flaviventer) to broadcast alarm calls from unfamiliar individuals. First, we quantified affiliative interactions of animals in the field to calculate a set of social network measures. Because attributes of sociality are often correlated, we used principal component analysis to reduce our social network metrics to two unrelated factors and used the social network measure that accounted for the most variance for each principal component in further analyses. We then quantified the change in time allocated to vigilance and foraging following alarm call playback from baseline levels to the first 30 s and the second 30 s period (i.e., 31-60 s) following playback. We expected that if marmots with strong affiliative relationships felt more secure, they would forage more and allocate less time to vigilance after their immediate vigilance response to the broadcast alarm calls. Using mixed effects models that controlled for variation explained by a number of biologically important covariates and permutation tests to test the significance of social network variables, we found that marmots with a higher incloseness allocated significantly more time to vigilance in both the first and second 30 s after hearing a novel alarm call. Additionally, and while not significant (the observed parameter estimate fell between the 90 and 95% CI), marmots with a higher outstrength increased foraging in the second 30 s after hearing a novel alarmcall. If we assume that time allocated to foraging is a measure of security, then marmots with strong affiliative relationships reacted more to an alarm call from a novel individual and seemingly felt more secure than their counterparts. Our results, therefore, suggest that strong social relationships increase perceptions of security and illustrate an effect of social relationships on predation risk assessment. Significance statement In many species, including humans, there are benefits from maintaining good social relationships. These benefits include better health and greater longevity. We studied yellow-bellied marmots, a ground-dwelling squirrel that is notable because it has variable social relationships. Capitalizing on this social variation, we found that marmots that maintain strong affiliative relationships with other marmots may indeed forage more after hearing an artificially broadcast alarm call than marmots with weaker affiliative relationships. This result suggests that marmots that are more socially connected with others in their group may feel relatively more secure, possibly because they have others that can help them assess risk, and that a benefit of maintaining strong friendly social ties is that it permits animals to forage more. Thus, we suggest a new way that social relationships can be beneficial: they increase perceptions of security and this ultimatelymay facilitate foraging.