Studies of Individual and Paired Interactional Problem Solving Behavior of Rats

Abstract

Two studies on the relationship between and social variables for SKINNER-box activities are reported. The first investigation dealt with the relationship between the maximum response rate achieved in solitary training and the assumption of worker-dependent roles in a paired situation. The second experiment concerned the differential effects of partial and invariable reward upon the assumption of these roles. Neither maximum response rate or kind of training produced more workers or dependents. However, further analysis of the data led to a distinction between role a s s u m p t i o n and role style, the way in which the role was carried out. The data suggest that for this latter feature of the paired behavior in the SKINNER-box, the kind of reinforcement schedule is significant. The behavior of partially reinforced animals who became workers was different from that of invariably reinforced animals who became workers. The animals, after meeting a criterion level in the paired situation, were then placed with different partners in a transfer situation. There was a reversal of roles in only two pairs. This was explained by the principle of generalization. Finally, there were some observations on problems arising from the procedures involved in introducing variable reward schedules.