Social learning strategies for nut-cracking by tufted capuchin monkeys (Sapajus spp.)

The spontaneous use of stone tools for cracking nuts by tufted capuchin monkeys, now known to be habitual among wild populations in savanna environments, was first described in a semifree group living in the Tiet Ecological Park (SP, Brazil). Nut-cracking at TEP was first observed by our team in 1995 (Ottoni and Mannu in Int J Primatol 22(3):347-358, 2001), and its ontogeny and associated social dynamics, with inexperienced observers highly interested in the activities of proficient individuals, greatly tolerant to scrounging, support hypotheses about social biases on tool-use learning. Here we further analyze the social learning biases, better characterizing: the social context of nut-cracking in which observation by conspecifics occurs, the quality of the nut-cracking behavior itself and whether scrounging may be the motivation behind this behavior. We confirm that the choice of observational targets is an active one; monkeys do not simply observe those who they are socially close to. We investigate social learning strategies, describing how young capuchins choose to observe older, more proficient and dominant individuals during nut-cracking bouts. Monkeys with higher productivity rates were also more frequently targeted by observers, who were tolerated scroungers, further supporting the scrounging hypothesis. Finally, based on changes of the demographic patterns of tool use and observation, we set to retrace data from 14 years of continuous studies. We argue that we have followed the dissemination of the behavior (Transmission Phase) almost from its beginning, when juveniles were the most frequent nutcrackers, to a more common pattern where adults are the most active tool users (Tradition Phase).