Social influences affect individual responsiveness to key features of the environment, such as food. Galef (1993,Anim. Behav.,46,257-265) has predicted that social facilitation should affect food choice more powerfully when the food is novel than when it is familiar. This prediction was tested in monkeys. Eleven capuchin monkeys were tested individually (Individual condition) and in a group (Social condition) with eight familiar foods presented simultaneously (experiment 1). In experiment 2, the same subjects received 20 novel and four familiar foods presented singly in Individual or in Social testing conditions. More food was expected to be eaten in the Social condition than in the Individual condition, particularly when food was novel. In experiment 1, testing condition (Individual or Social) did not affect consumption of familiar foods. In experiment 2, capuchins ate more familiar foods than novel foods in both conditions. However, they were more interested in another individual's food when foods were novel than when they were familiar. Consumption of, and responses to, the novel foods were more frequent in social testing than in individual testing; testing conditions did not affect consumption of, or response to, the familiar foods. Nine of 10 individuals ate more types of novel foods in the Social condition that in the Individual condition. In short, social facilitation of eating was evident, but only with novel foods. These findings support Galef's prediction and suggest that social facilitation can have a role in enlarging dietary selection in capuchins.