Proceedings of the Annual Meeting of the Cognitive Science Society

Many 4-year-olds give adultlike responses to questions about what "can" happen and the same responses to questions about what "has to" happen. Why? We argue that children repeatedly apply minimal representations of possibility to answer "can" and "have to" questions. This algorithm (a) only requires evaluating the single possibility each question asks about; (b) generates correct responses to "can" questions and (c) generates the same responses to "have to" questions. We show that half of 4-year-olds answer "have to" questions as adults do, correctly say what all the possibilities are and what the only possibilities are, and show sensitivity to multiple possibilities in nonverbal behavior. The other half make mistakes on "have to" questions, fail to quantify possibilities using "all" and "only", and deploy minimal representations of possibility on the nonverbal task. They make their decisions by evaluating a single possibility.