Unlocking the Doors of the Prisoner's Dilemma: Dependence, Selectivity, and Cooperation

• Published in: Social psychology quarterly Vol. 62; no. 1; pp. 32 - 52
• Main Authors: Boone, R. Thomas, Macy, Michael W.
• Format: Journal Article
• Language: English
• Published: Washington, DC American Sociological Association 01.03.1999; American Sociological Association, etc
• Subjects: Behavioral Science Research; Behavioral Sciences; Betting; Biological and medical sciences; Choices; Computer Simulation; Cooperation; Correctional Institutions; Dependency; Dependency (Psychology); Dwelling value; Experiment design; Factors; Fundamental and applied biological sciences. Psychology; Game Theory; Games; Group Dynamics; Helping Relationship; Institutionalized Persons; Laboratory Experiments; Predators; Prisoner's dilemma games; Prisoners dilemma; Psychology. Psychoanalysis. Psychiatry; Psychology. Psychophysiology; Resistance (Psychology); Retaliation; Selection; Social interactions. Communication. Group processes; Social psychology; Socialization; Student Behavior; Undergraduate Students; Human; Choice; Social behavior; Social interaction; Cooperation; Prisoner dilemma game; Interpersonal relation
• ISSN: 0190-2725; 1939-8999
• DOI: 10.2307/2695824

Although the Prisoner's Dilemma game has been widely investigated, few studies have explored what happens when the doors to the prison are unlocked. Conventional game-theoretic analysis predicts greater cooperation in ongoing relationships than with occasional partners. This suggests that an option to exit decreases cooperation by reducing players' dependence on each other for help in the future. An alternative hypothesis is that cooperation increases with selectivity-that is, with the ability to play with a desired partner. This suggests that an exit option increases cooperation. We tested both hypotheses in laboratory experiments with human subjects, using "Trump," a new card-game representation of a Prisoner's Dilemma. In the first experiment, the exit option reduced dependence and increased selectivity, relative to a no-exit condition. We found that selectivity and dependence both increased cooperation, but for different reasons and with different players, leaving overall cooperation rates unchanged by an option to exit. In a second experiment, the exit option increased selectivity but not dependence, relative to a forced-exit condition. Here we observed higher cooperation with an option to exit, but only for subjects playing defensive strategies.