Controlled studies of chimpanzee cultural transmission

Abstract

Following the first long-term field studies of chimpanzees in the 1960s, researchers began to suspect that chimpanzees from different African populations varied in their behavior, and that some of these variations were transmitted through social learning, thus suggesting culture. Additional reports of chimpanzee culture have since accumulated, which involve an increasing amount of behavioral variation that has no obvious ecological or genetic explanation. To date, close to 50 cultural variants have been reported, including subsistence behavior, tool-use, communication signals, and grooming patterns. Nevertheless, field studies lack the experimental controls and manipulations necessary to conclusively demonstrate that the observed variation results from differential invention and social transmission of behavior. This would require that behavioral variants have been learned from others, a question best addressed in a controlled experimental setting. The following chapter details a series of experimental studies at Yerkes National Primate Research Center of Emory University. In each case, the behavior of two captive groups (each N=12 individuals) was compared before and after the introduction of a novel foraging behavior by a trained conspecific "inventor." The studies were designed to investigate (i) the conditions under which chimpanzees learn from one another, (ii) how behaviors are transmitted, (iii) how cultures are maintained over generations. The results emphasize the importance of integrating both fieldwork and experimental approaches. Previous studies have reported deficits in chimpanzees' cultural capacities, but did so after testing them with human models, which are largely irrelevant to the problem at hand. A representative understanding of culture can only be gained when efforts are made to create a naturalistic learning environment in which chimpanzees have opportunities to learn spontaneously from conspecifics in a familiar social setting.