Emergence of cooperation due to opponent-specific responses in Prisoner's Dilemma

Complex life would be impossible without cooperation at all levels of biological organization. However, Darwinian selection is commonly believed to favor selfish behavior, making societies of cooperators vulnerable to cheaters. A quintessential model of this behavior is the game of Prisoner's Dilemma in which cheaters always win, even though being cooperative results in greater rewards. Numerous scenarios have been proposed that allow for the evolution of cooperation in restrictive settings that postulate altruism between genetic relatives, explore mechanisms of direct and indirect reciprocity, focus on competition between groups, or impose spatial structure on the population. It is difficult to imagine how these scenarios would account for the evolution of cooperation in populations of organisms that lack sophisticated assessment mechanisms and have no spatial constraints. Here we demonstrate that it is possible to achieve high levels of cooperativity in the game of Prisoner's Dilemma without introducing any additional assumptions about genetic relatedness, population structure, or explicit reciprocal arrangements. The only requirement is that the willingness to cooperate varies depending on the opponent, for example in response to the opponent's physical appearance and patterns of behavior. This mechanism requires consistent opponent recognition during multiple encounters. Evolution of cooperativity due to opponent-specific responses may be the only available mechanism in many biological settings and may serve as a starting point for more sophisticated modes of cooperation observed in animal and human societies.