From hawks and doves to self-consistent games of territorial behavior.

Explaining the "prior-residence effect" (automatic owner status of individuals who arrived first in an area) was one of the very first applications of game theory in animal behavior. These models, however, predict paradoxical solutions where intruders always win, with no satisfactory explanation for the absence of such cases in nature. We propose a solution based on new developments in evolutionary game theory. A self-consistent model with feedbacks between individual behavior and population dynamics produces qualitatively different frequency-dependent selection on intruders (floaters) than on territory owners. Starting with an ancestral population with no respect for ownership, the most likely evolutionary end point is complete or partial respect. Conventional rules of conflict resolution thus can rely on "uncorrelated asymmetries" without differences in resource-holding power or territory value, although they will be strengthened by such differences. We also review the empirical literature on animal contests, testing whether asymmetries in resource-holding power are required to explain the observations. Despite much empirical effort, results remain inconclusive, because experiments are often unable to distinguish between the motivation of individuals to fight and the behavioral outcome of a contest. To help arrive at conclusive answers, we suggest a standardized empirical approach to quantify prior-residence effects.