This investigation was undertaken to determine if economically deprived adolescents are more cooperative in a Prisoner's Dilemma game if their partner is of their own race. Other independent variables investigated were race and sex of Ss. METHODThe Ss were 24 economically deprived junior high school students. The Ss were distributed so that there were six of each combination of race (Negro and white) and sex.By definition, a Prisoner's Dilemma game (pDG) provides that if both Ss cooperate, they both gain moderately, but if one defects (i.e., competes), he gains extensively while the other S loses. If both defect, both lose to a moderate degree. In this experiment, the payoff matrix allowed ftve points each when both Ss cooperated; when only one S cooperated, he lost 10 points while the defecting S gained 10 points; when neither S cooperated, both Ss lost ftve points. The points were noncumulative.The Ss (visually separated from each other) played the games in like-sexed dyads. Each S had two levers (C fot cooperation, D for defection), a ready light and four outcome lights before him. A trial consisted of a ready light and each S's lever press followed by the appropriate outcome light. There were four possible outcome combinations including both Ss cooperating (CC), only one S cooperating (CD, DC), or neither S cooperating (DD).Each S played the game twice, always with a partner of ~ same sex, once with a member of the same race and once with a member of the other race. Each game consisted of 50 trials. Composition of the dyads was counterbalanced so that one-half of the Ss of each race played with a partner of his own race first while the other half played with one of the other race fmt. A similar balance within the sexes was arranged. All dyads were asked to try and gain as many points as possible and not to communicate during the game. RESULTS The data were the number of cooperative (C) responses made by each S. These were totaled over each 25 trials (half of the game). A 2 by 2 by 2 (Race of S, Sex of Dyad, Race of Partner) analysis of variance with repeated measures on the last variable revealed several signiftcant results. As represented in Fig. I, sex functioned as a major signiftcant variable (p < .05) in that female Ss produced more cooperative responses than the male Ss. When the signiftcant (p < .05)Race by Sex interaction is taken into consideration, however, it is obvious that Negro males and females do not differ from each other in cooperative responses. The signiftcant Sex effect is due entirely to the large difference between white males and females with the latter being much more cooperative.Psycbon. Sci., 1968, Vol. 13 (2) Therefore, when playing the game with a member of the same sex, white females are more trusting and cooperative than white males. This difference did not exist between Negro males vs Negro females.Possibly the most interesting significant (p < .05) effect was that of the partner's race in combination with S's race. The Ss of both races and of both sexes produced significantly...

No abstract