Effects of friendship and disliking on cooperation in a mixed-motive game1

The amount of friendship between two people seems likely to affect their degree of cooperative behavior in a joint task. Though this hypothesis has occasionally been suggested by psychologists (May and Doob, 1937, pp. 17-18; Solomon, 1960, p. 224), it has never been tested in a carefully controlled empirical study. Instead, many studies have attempted to eliminate any effects of friendship by using subjects (Ss) who were unacquainted. Scodel, Minas, Ratoosh, and Lipetz (1959, p. 118) suggest that one reason for the low levels of cooperation frequently obtained in mixedmotive game research is that strangers act in a guarded fashion toward each other. A previous experiment by the present authors (Oskamp and Perlman, 1965) found