On the evolutionary stability of preferences for reciprocity

Recent theoretical work has incorporated assumptions that agents have preferences for reciprocity, leading them to behave non-opportunistically in social interactions. There is extensive experimental evidence for the existence of such preferences. This paper explains the emergence of preferences for reciprocity in a population consisting of two types: "reciprocators" and "opportunists". Unlike conventional evolutionary models, both types maximize their expected payoffs, given their preferences. The population proportions of these two types evolve from one generation to the next, according to their expected payoffs. It is shown that, under a variety of conditions, preferences for reciprocity can survive in the population.