Long-Term Spatial Memory and Learning Set Formation in Captive Capuchin Monkeys (Cebus libidinosus = Sapajus cay)

Abstract

Researchers have long suspected that nonhuman primates have long-term spatial memory for locating food. However, few empirical studies have assessed spatial memory for a period longer than 1 day in nonhuman primates in a foraging context. We used a modified version of the radial maze to test long-term memory for periods of 2 days or longer in two groups of Cebus libidinosus = Sapajus cay (N = 10; N = 6) in captivity (environment completely human constructed) and semicaptivity (relatively natural environment with few human-made structures) at La Plata Zoo and Botanical Park, Buenos Aires, Argentina, between 2007 and 2010. The experimental design included a set of three accessible baited feeders interspersed among three nonaccessible baited feeders. We tested monkeys in one initial exposure period, four periods of long-term memory (2 days, 76 days, 76 + 2 days, and 4 months of delay), and one period of inversion of the experimental set location. We used the latter to analyze the monkeys’ abilities to develop learning sets. Captive subjects appeared to remember sites with accessible and nonaccessible food for periods of 2 days, 76 days, and 4 months, and used learning sets to reduce relearning times when exposed to a change in the learned locations. Although semicaptive subjects also appeared to remember sites with accessible and nonaccessible food, their performance was less accurate. We conclude that capuchins used spatial memory to improve their foraging efficiency.