Category: economics
Associated thinker(s): Merrill Flood; Melvin Dresher; Albert W. Tucker
Original publication or approximate period: Game developed around 1950; Flood’s RAND report (1952); Tucker’s explanatory story
Difficulty: Beginner
Type: model
Themes: cooperation; dominance; collective action
Related experiments: tragedy-of-the-commons; keynesian-beauty-contest
Two people independently choose to cooperate or defect. Use illustrative utility payoffs: if both cooperate, each receives 3; if both defect, each receives 1. If one defects while the other cooperates, the defector receives 5 and the cooperator 0. Assume each player prefers larger values and cannot make a binding agreement.
For either possible action of the other person, defection gives you a larger payoff. Yet both receive less from mutual defection than from mutual cooperation. The numerical labels are an editorial example; the defining feature is their ordering, not a specific prison sentence or currency.
How can individually compelling strategic choices produce an outcome each player ranks below another available joint outcome?
The game offers a precise model of a cooperation problem. It is useful because its assumptions can be changed systematically: repetition, monitoring, enforceable agreements, or altered motivations may transform the incentives. Whether a real dispute has this structure is an empirical modeling question, not something proved by giving it the game’s name.
One-shot dominance analysis recommends defection under the stated utility representation and standard individual decision assumptions. Repeated-game analysis asks whether future rewards or retaliation can support cooperation; conclusions depend on horizon, information, and payoff details. Moral or team reasoning questions whether individual maximization exhausts rational deliberation. Altering utility to include concern for others may remove the original payoff structure altogether.
A prisoner’s dilemma is not any disagreement or zero-sum conflict. Mutual cooperation benefits both relative to mutual defection. Communication alone does not create enforceable commitments. A strategy’s success in one repeated-game tournament does not establish a universally optimal strategy against all opponents.
Iterated games introduce history-dependent strategies. Public-goods and commons models generalize cooperation problems to many participants, but their payoff structures should be checked instead of assumed identical.
The scholarly entry distinguishes the Flood–Dresher game from Tucker’s prison narrative and notes uncertainty about the earliest printed label. The table-free payoff example above is original and designed to make the dominance argument easy to verify.
Verified on 2026-10-30. Source coverage: [1] Flood/Dresher’s game, Tucker’s story, payoff conditions, repeated games. Reflection questions and illustrative numerical examples are original editorial material.