Cooperative Sentinel Behaviour and its Vocal Coordination in Meerkats

Abstract

A benefit of living in groups is the potential to evolve coordinated antipredator strategies. Sentinel behaviour is a form of coordinated vigilance behaviour, where usually one individual adopts a raised position and scans the environment for the presence of predators, while the rest of the group is foraging. All adult group members contribute to sentinel behaviour, however, the extent to which each individual acts as sentinel varies greatly. Understanding the specific underlying costs and benefits of cooperative behaviours to explain this variation among individuals, groups and populations has remained one of the major questions in the field of behavioural ecology. Another important aspect of cooperative behaviours is their coordination. Many animals evolved specific vocal signals to coordinate cooperative behaviours among group members. To fully understand the mechanism underlying the interaction between the producer of the signal and the receiver(s), it is fundamental to recognize the specific vocalisations involved, understand their function and information content as well as how this information is subsequently used by other group members to adjust their own behaviour accordingly. In my thesis, I focused on cooperative sentinel behaviour in meerkats (Suricata suricatta). Sentinel behaviour is well suited to investigate the underlying costs and benefits of cooperation because it represents a combination of costly aspects, i.e. foregoing foraging, as well as beneficial aspects such as adopting a safe position, or potential information gathering or reputational aspects, which are not yet clear in meerkats. Moreover, meerkat sentinels produce six distinct sentinel call types, but little is known about the underlying reasons for this comparatively large sentinel call repertoire, the information that can be encoded therein and the use of this information by group members. Combining both aspects together, individual variation in cooperative behaviour as well as its vocal coordination, I aimed to improve our understanding of how social and environmental factors promote or constrain this cooperative behaviour and how they affect the extent and mechanisms underlying its vocal coordination. The results suggest that the amount of sentinel behaviour an individual displayed was highly dependent on condition and predation risk. I found that during a drought - an extreme environmental condition with very low food availability - the extent of sentinel behaviour decreased significantly, especially in young individuals, small groups and groups with dependent young. The vocal coordination between sentinels and the rest of the group including the production of all six types of sentinel calls was already done by young meerkats when they first started to act as sentinels. Moreover, call rates and acoustic parameters showed little change with increasing age and experience. Quantifying whole sequences of sentinel calls produced during a sentinel bout revealed that the order of the call types was produced in a graded way and contained information about the identity of the caller. Testing whether the conveyed information about sentinel identity in calming sentinel calls was meaningful for the receivers resulted in clear discrimination among signallers, whereby receivers relied most on calming calls produced by the most experienced sentinels and littermates. Ecological conditions, too, specifically drought condition, resulted in a flexible adjustment of the behavioural response of foraging group members when hearing sentinel calls. The research presented in this thesis provides strong evidence that sentinel behaviour as well as its vocal coordination are driven by a strong trade-off between the costs of foregoing foraging and the risk of being predated. Comparison with other cooperative breeders inhabiting less-constraining environments will provide valuable insight into variation in cooperative tasks among individuals, as well as the influence of social and environmental variables on vocal complexity.