The Evolution of Social Dominance

A difference in dominance rank is an often-usedcue to resolve con icts between two animals without escalated Ž ghts. At the group level, adherence to a dominance convention efŽ ciently reduces the costs associatedwith con icts, but from an individual’s point of view, it is difŽ cult to explainwhy a low ranking individualshould accept its subordinatestatus. This is especially true if, as suggested by several authors, dominance not necessarily re ects differences in Ž ghting ability but rather results from arbitrary historical asymmetries. According to this idea, rank differentiationemerges from behavioural strategies, referred to as winner and loser effects, in which winners of previous con icts are more likely to win the current con ict, whereas the losers of previous con icts are less likely to do so. In order to investigatewhether dominance, based on such winner and loser effects, can be evolutionarily stable, we analyse a game theoretical model. The model focuses on an extreme case in which there are no differences in Ž ghting ability between individuals at all. The only asymmetries that may arise between individuals are generated by the outcome of previous con icts. By means of numerical analysis, we Ž nd alternative evolutionarily stable strategies, which all utilize these asymmetries for conventional con ict resolution. One class of these strategies is based on winner and loser effects, thus generating evolutionarily stable dominance relations even in the absence of differences in resource holding potential.