Proceedings of the Annual Meeting of the Cognitive Science Society

To successfully minimize losses or maximize gains, individuals must acquire a profound understanding of the rules and regularities in their environment. The current project centers on the impact of the environment on exploration and exploitation behavior. Therein, we compare costly exploration in environments, in which it is only possible to win (even though the size of the gains differs), only possible to lose, and mixed environments, in which one can win and lose. Participants engaged in a Multi-Armed Bandit task in three such conditions. Notably, participants exhibited reduced exploration in the gain domain compared to the loss domain, with the mixed domain falling in between. Interestingly, participants performed best in the mixed domain. Computational modeling of participants' choice behavior revealed that individuals tended to underestimate outcomes of unchosen options in the gain domain and overestimated them in the loss domain. We discuss two explanations for this pattern of findings: Either, effects are driven by the absolute difference between gains and losses or by the relative difference that individuals experience in relatively better or worse environments compared to their expectations (e.g., compared to previous blocks).