Mice adaptively generate choice variability in a deterministic task

• DOI: 10.1038/s42003-020-0759-x
• Publication Date: 2020-01-21T11:02:51Z
• Authors: Marwen Belkaid; Elise Bousseyrol; Romain Durand-de ...; Malou Dongelmans; Etienne K. Duranté; Tarek Ahmed Yahia; Steve Didienne; Bernadette Hanesse; Maxime Come; Alexandre Mourot; Jérémie Naudé; Olivier Sigaud; Philippe Faure
• Keywords: Decision maker

Abstract Can decisions be made solely by chance? variability intrinsic to the decision-maker or is it inherited from environmental conditions? To investigate these questions, we designed a deterministic setting in which mice are rewarded for non-repetitive choice sequences, and modeled experiment using reinforcement learning. We found that progressively increased their variability. Although an optimal strategy based on sequences learning was theoretically possible would more rewarding, animals used pseudo-random selection ensures high success rate. This not case if animal exposed uniform probabilistic reward delivery. also show were blind changes temporal structure of delivery once they learned choose at random. Overall, our results demonstrate decision-making process can self-generate randomness, even when rules governing neither stochastic nor volatile.