Two experiments investigated the possibility that three year olds would do better on tasks in which belief cues were stronger than on standard false belief tasks, in which the children could reason backward to the belief from its effects. Findings provided strong support for the view that three year olds do not fully understand the representational nature of belief. (RH)