Content area
Abstract
This work examines the extent to which human group behavior is explained by simple individual rules, which combine and interact to create emergent group behavior. Human collective behavior is examined through a set of three controlled, experimental tasks that explore behavior in competitive, indirectly cooperative, and directly cooperative group situations. The competitive task tests group distribution to resources in a spatial environment under a variety of perceptual, spatial, and food rate conditions, and a novel explanation for resource undermatching is proposed to account for the observed matching behavior, switching behavior, and wealth disparities. The competitive foraging task also shows a contingent use of public and private information according to the perceptual information available in each experimental condition. The indirect cooperation task examines the factors that affect spatial trail formation as humans navigate to their destinations. Individuals are shown to place special emphasis on canonical axes, and more importantly, individuals show a reluctance to deviate from a direct route when first traveling to a destination, and the same bias may affect individual and group behavior in more abstract goal-oriented tasks. The direct cooperation task examines the ability of group members to coordinate their contributions when only a group outcome is important, much like the challenges faced by teams and committees. Individuals appear to follow a simple, adaptive set of rules, and the greater task difficulty for large groups promotes differentiation and role specialization among its members. Agent-based models are described and analyzed for each of these tasks, providing more detailed explanations of the rules used by individual group members. The three experimental paradigms are also compared and contrasted with respect to observed group inefficiencies, use of public and private information, types of environmental information, and whether individuals exhibited specialization.
