Proceedings of the Annual Meeting of the Cognitive Science Society

Research on the development of future hypothetical andcounterfactual thinking suggests that children as old as fivemay be unable to consider multiple, equally probablepossibilities simultaneously. Yet, a large literature on thedevelopment of causal reasoning suggests that much youngerchildren are able to generate, evaluate, and test causalhypotheses, often by integrating information about severalcandidate causes at once. The current research seeks to bridgethese two bodies of research. In three experiments, adults andtoddlers (18–30 months) observe a sequence of evidence thatis equally consistent with two hypotheses, each occupying adifferent level of abstraction (individual vs. relational).Results suggest that learners generate more than one potentialcause, hold both in mind, and flexibly apply the appropriatehypothesis to inform their inferences at test. Findingschallenge previous suggestions that much older children failto consider multiple, equally probable possibilities.