Abstract
The foundation of modern behavior science is the selection of behavior by its consequences. So, behavior that benefits an organism should be maintained (through reinforcement) whereas costly behavior should not. Altruism—behavior that benefits others at one’s own expense—seems a clear counterexample to a reinforcement-based account of behavior. Despite the apparent paradox, five distinct (but mutually inclusive) reinforcement accounts are provided for instances of apparent altruism. These accounts include situations in which no reinforcement is observed following a prosocial act—due to subtle, delayed, or intermittent reinforcement. Also included is an account of how rules can occasion acts of self-sacrifice and an account of socially discounted reinforcement—through which benefits to others have a quantifiable benefit to an altruist. Any instance of apparent altruism can likely be accounted for through one or more of these five reinforcement-based accounts.