Manipulating the affiliative interactions of group-housed rhesus macaques using positive reinforcement training techniques.

Social housing, whether continuous, intermittent, or partial contact, typically provides many captive primates with opportunities to express affiliative behaviors, important components of the species-typical behavioral repertoire. Positive reinforcement training techniques have been successfully employed to shape many behaviors important for achieving primate husbandry goals. The present study was conducted to determine whether positive reinforcement training techniques could also be employed to alter levels of affiliative interactions among group-housed rhesus macaques. Twenty-eight female rhesus were divided into high (n = 14) and low (n = 14) affiliators based on a median split of the amount of time they spent affiliating during the baseline phase of the study. During the subsequent training phase, half of the low affiliators (n = 7) were trained to increase their time spent affiliating, and half of the high affiliators (n = 7) were trained to decrease their time spent affiliating. Trained subjects were observed both during and outside of training sessions. Low affiliators significantly increased the amount of time they spent affiliating, but only during nontraining sessions. High affiliators on the other hand, significantly decreased the amount of time they spent affiliating, but only during training sessions. These data suggest that positive reinforcement techniques can be used to alter the affiliative behavior patterns of group-housed, female rhesus monkeys, although the two subgroups of subjects responded differently to the training process. Low affiliators changed their overall behavioral repertoire, while high affiliators responded to the reinforcement contingencies of training, altering their proximity patterns but not their overall behavior patterns. Thus, positive reinforcement training can be used not only as a means to promote species-typical or beneficial behavior patterns, but also as an important experimental manipulation to facilitate systematic analyses of the effects of psychosocial factors on behavior and potentially even immunology.