Tool use generalization and behavioral traditions in tufted capuchin monkeys (Sapajus sp)

Abstract

Customary tool use by wild tufted capuchin monkeys seems to constitute behavioral traditions, i.e., to be individually acquired by socially biased learning. Social influences apparently involve attentional/motivational effects (“stimulus enhancement”), rather than imitative learning. Persistent manipulation increases opportunities for trial-and-error learning. This process can take a long time, but once proficiently successful in a given context, the use of a type of tool can be quickly generalized to new problems. The use of percussive stone tools is common in savanna populations of tufted capuchins, and where tool material is abundant, as in Serra da Capivara, can be quite diversified. This is also the only wild population where customary use of probe tools has been observed. When exposed to molasses-filled problem-boxes, after brief (unsuccessful) attempts to break them with stone “hammers”, several males quickly succeeded in using twig probes to access the molasses through the top holes. Almost three years later, when exposed to a different Plexiglas problem-box, they promptly brought some twigs. In an experiment on the innovation and diffusion of probe tools’ use in a semi-free group, after the slow and partial diffusion of probing a molasses’ problem-box, the successful individuals quickly succeeded in extracting seeds from another box, even though the probing technique involved was different (pushing). Afterwards, a distinct sort of problem-box also elicited probing attempts. Tool use generalization can facilitate the establishing of behavioral traditions not only by increasing direct individual rewards, but also by enhancing the opportunities for socially biased learning by naïve, tolerated observers.