Experimental field study of problem-solving using tools in free-ranging capuchins (Sapajus nigritus, formerly Cebus nigritus)

• Published in: American journal of primatology Vol. 74; no. 4; pp. 344 - 358
• Main Authors: Garber, P.A., Gomes, D.F., Bicca-Marques, J.C.
• Format: Journal Article
• Language: English
• Published: United States Blackwell Publishing Ltd 01.04.2012; Wiley Subscription Services, Inc
• Subjects: Animals; black-horned capuchins; causal knowledge; Cebus - physiology; Cebus - psychology; cognition; Female; Learning; Male; Problem Solving - physiology; situation-bounded learning; tool use; Tool Use Behavior
• ISSN: 0275-2565; 1098-2345
• DOI: 10.1002/ajp.20957

Some populations of capuchins are reported to use tools to solve foraging problems in the wild. In most cases, this involves the act of pounding and digging. The use of probing tools by wild capuchins is considerably less common. Here we report on the results of an experimental field study conducted in southern Brazil designed to examine the ability of wild black‐horned capuchins (Sapajus nigritus) to use a wooden dowel as a lever or a probe to obtain an embedded food reward. A group of eight capuchins was presented with two experimental platforms, each housing a clear Plexiglas box containing two bananas on a shelf and four inserted dowels. Depending on the conditions of the experiment, the capuchins were required either to pull (Condition I) or push (Conditions II and III) the dowels, in order to dislodge the food reward from the shelf so that it could be manually retrieved. In Condition I, four individuals spontaneously solved the foraging problem by pulling the dowels in 25% (72/291) of visits. In Conditions II and III, however, no capuchin successfully pushed the dowels forward to obtain the food reward. During these latter two experimental conditions, the capuchins continued to pull the dowels (41/151 or 27% of visits), even though this behavior did not result in foraging success. The results of these field experiments are consistent with an identical study conducted on wild Cebus capucinus in Costa Rica, and suggest that when using an external object as a probe to solve a foraging problem, individual capuchins were able to rapidly learn an association between the tool and the food reward, but failed to understand exactly how the tool functioned in accomplishing the task. The results also suggest that once a capuchin learned to solve this tool‐mediated foraging problem, the individual persisted in using the same solution even in the face of repeated failure (slow rate of learning extinction). Am. J. Primatol. 74:344–358, 2012. © 2011 Wiley Periodicals, Inc.