The ability to recognize specific individuals has profound implications for the evolution of complex social behaviors such as reciprocal altruism. Many researchers assumed that recognition of individuals, a phenomenon predominantly observed in laboratory studies of fish, might also operate extensively in free-ranging fish populations, where it could underpin these complex interactions. In fact, evidence of individual recognition in free-ranging fish populations is equivocal. The possibility exists that for many species, individual recognition observed in the laboratory might be an artifact of experimental designs, which enforce prolonged interaction between individuals and which prevent the diluting effects on social structure of immigration into and emigration from the shoal, factors that in nature would erode group stability and prevent the learning of individual identities.