We summarize 20 years of empirical and theoretical research on causes and functions of social influences on foraging by animals. We consider separately studies of social influence on when, where, what and how to eat. Implicit in discussion of the majority of studies is our assumption that social influences on foraging reflect a biasing of individual learning processes by social stimuli rather than action of independent social-learning mechanisms. Our review of theoretical approaches suggests that the majority of formally derived hypotheses concerning functions of social influence on foraging have not yet been tested adequately and many models are in need of further refinement. We also consider the importance to the future of the field of integrating 'top-down' and 'bottom-up' approaches to the study of social learning. Copyright 2001 The Association for the Study of Animal Behaviour.