Beyond the information centre hypothesis: communal roosting for information on food, predators, travel companions and mates?

Abstract

Communal roosting – the grouping of more than two individuals resting together – is common among animals, notably birds. The main functions of this complicated social behaviour are thought to be reduced costs of predation and thermoregulation, and increased foraging efficiency. One specific hypothesis is the information centre hypothesis (ICH) which states that roosts act as information centres where individuals actively advertise and share foraging information such as the location of patchily distributed foods. Empirical studies in corvids have demonstrated behaviours consistent with the predictions of the ICH, but some of the assumptions in its original formulation have made its wide acceptance problematic. Here we propose to generalise the ICH in two ways: (1) dropping the assumption that information transfer must be active, and (2) adding the possibilities of information exchange on, for example, predation risk, travel companions and potential mates. A conceptual model, inspired by shorebirds arriving at roosts after foraging on cryptic prey, is proposed to illustrate how testable predictions can be generated. The conceptual model illustrates how roost arrival timing may convey inadvertent information on intake rate, prey density, forager state (i.e. digestive processing capacity) and food quality. Such information could be used by naïve or unsuccessful foragers to select with whom to leave the roost at the subsequent foraging opportunity and thus increase foraging success. We suggest that inadvertent information transfer, rather than active information exchange, predominates in communal roosts.