Wilkinson, G.S. & Boughman, J.W. (1999) Social influences on foraging in bats. In Box, H.O. & Gibson, K. (eds) Mammalian Social Learning: Comparative and Ecological Perspectives. Proceedings of Zoological Society of London 73. pp 188-204.
Bats represent a likely group in which to find evidence that social learning affects foraging for three reasons 1) many species form dense roosting clusters, thereby providing opportunities to roostmates for learning about the diet of conspecifics, 2) those species that use echolocation to capture prey can attend to conspecific calls in order to locate feeding areas, and 3) many species are long-lived so some individuals in roosting clusters are experienced foragers while others are more naive. Unfortunately, few studies to date have directly considered the degree to which social learning occurs in bats. Therefore, we begin by describing two learning problems – choosing a diet and finding resources – which appear to involve social learning in some species of bats. Bats can learn the location of food from others by eavesdropping or following successful foragers, and we describe examples from several temperate insectivorous species. Some species, including common vampire bats (Desmodus rotundus) and greater spear-nosed bats (Phyllostomus hastatus) appear to communicate actively with social group members to improve their ability to find and procure food.
Social foraging in bats is mediated by vocalizations when bats follow others to feeding sites, eavesdrop on actively foraging bats, defend feeding sites, or advertise resource location. The vocalizations themselves appear to be influenced by this social function, and in some cases may even be learned from social partners. Socially mediated learning of vocalizations used to coordinate foraging has been demonstrated in P. hastatus by reciprocal transfer experiments. Such call convergence is remarkable for a mammal and may facilitate social learning of diet preferences or feeding site locations.
We then use a recent model for animal social learning (Laland, 1996) to determine if variation in the duration of a feeding patch and in the foraging success of naive individuals can be used to predict differences in following behavior exhibited by evening bats (Nycticeius humeralis), greater spear-nosed bats (P. hastatus), and common vampire bats (D. rotundus).