The flow of reward information through neuronal ensembles in the accumbens

Abstract

SummaryReward information flows through neuronal ensembles in the nucleus accumbens shell (NAcSh), influencing decision-making. We investigated this phenomenon by training rats in a self-guided probabilistic choice task while recording single-unit activity in the NAcSh. We found that rats dynamically adapted their choices based on an internal representation of reward likelihood. Neuronal ensembles in the NAcSh act as dynamic modules to process different aspects of reward-guided behavior. Ensembles dynamically change composition and functional connections throughout reinforcement learning. The NAcSh forms a highly connected network with a heavy-tailed distribution and neuronal hubs, facilitating efficient reward information flow. Reward delivery evokes higher mutual information between ensembles and unifies network activity, while omission leads to less synchronization. Our recordings shed light on how reward information propagates through dynamically changing ensembles of neurons in the NAcSh. These functional ensembles exhibit flexible membership, dropping in and out and even shrinking in number as the rat learns to obtain (energy) rewards in an ever-changing environment.Graphical Abstract