Proceedings of the Annual Meeting of the Cognitive Science Society

Young children rapidly infer causal relations by trackingcontingencies between causes and their effects, and cangeneralize these rules to novel instances of the same cause.However, this is distinct from the ability to make inferencesabout whether a particular cause is likely to produce noveleffects. Here, we investigate the development of two-, three-,and four-year-olds’ ability to recognize and use informationabout a cause’s variability to make predictions about othernovel outcomes it might produce. Experiment 1 finds thatchildren as young as two years of age infer that a cause thathas produced variable, rather than deterministic outcomes ismore likely to produce a novel, previously unobserved effect.Experiment 2 finds that four-year-olds, but not two- andthree-year-olds, infer that a higher variability cause is morelikely to produce a novel outcome than a lower variabilitycause.