Cortico-Striatal Beta-Oscillations as a Marker of Learned Reward Value

Abstract

Single neuron correlates of reward value have been observed in brain regions along the cortico-striatal pathway including ventral striatum, orbital, and medial prefrontal cortex. Brain imaging studies in humans further validate these findings and suggest that value is represented in a network of brain regions opposed to a particular area. Neural activity oscillates at periodic frequencies to coordinate long-range communication in widespread, dynamic networks. To explore how oscillatory dynamics across brain regions may represent reward value, we measured local field potentials of male Long-Evans rats during three distinct behavioral tasks, each probing a different aspect of reward processing. Our goal was to use a data-driven approach to identify a common electrophysiology property associated with reward value. We found that reward-locked oscillations at beta frequencies, in both single units and local field potentials, were markers of positive reward valence. More importantly, Reward-locked beta-oscillations scaled with expected reward value on specific trial types and in a behaviorally relevant way across tasks. Oscillatory signatures of reward processing were observed throughout the cortico-striatal network including electrodes placed in orbitofrontal cortex, anterior insula, medial prefrontal cortex, ventral striatum, and amygdala. These data suggests that beta-oscillations reflect learned reward value in a distributed network, and this may serve as a stable and robust bio-marker for future studies.