Conference on Cognitive Computational Neuroscience

In some reward-learning contexts, more abstract stimulus representations can effectively guide behavior, whereas in others, people must use more finely grained, detailed representations to guide choice. Here, using a novel reinforcement learning task, we asked how children, adolescents, and adults flexibly adjust the specificity of the representations they use for learning across contexts, as well as how the specificity of the representations used during learning influences subsequent memory. We found that across development, participants up-weighted more detailed information when doing so was beneficial. Further, participants who placed greater weight on detailed information during learning also demonstrated enhanced mnemonic specificity for the stimuli they encountered.