Proceedings of the Annual Meeting of the Cognitive Science Society

During childhood and adolescence, sensitivity to the environment is often heightened, and individual differences in decision making emerge. However, it remains unclear which dimensions of experience produce these differences. Here, we focus on three candidate dimensions of early-life environments that theoretical work in reinforcement learning suggests should shape individuals' behavior. Environmental reward prevalence should govern learning from positive versus negative outcomes; controllability, reliance on reactive versus proactive control; and predictability, planning over a mental model. To test these predictions, we recruited a large sample of 10- to 25-year-olds to complete three reinforcement-learning tasks and a self-report measure of early-life experience. We found that behavior conformed to these predictions. Younger adolescents with less emotional support learned more slowly from positive outcomes, those with less control over their environment acted less proactively, and participants from less predictable environments planned more. Our results suggest that individual differences arise, in part, from rational adaptation to early-life environments.