Modelling group foraging: Individual suboptimality, interference, and a kind of matching

In this article a series of agent-based models support the hypothesis that behaviors adapted to a group situation may be suboptimal (or ¿irrational¿) when expressed by an isolated individual. These models focus on two areas of current concern in behavioral ecology and experimental psychology: the ¿interference function¿ (which relates the intake rate of a focal forager to the density of conspecifics) and the ¿matching law¿ (which formalizes the observation that many animals match the frequency of their response to different stimuli in proportion to the reward obtained from each stimulus type). Each model employs genetic algorithms to evolve foraging behaviors for multiple agents in spatially explicit environments, structured at the level of situated perception and action. A second concern of this article is to extend the understanding of both matching and interference per se by modeling at this level.