Back

Striatal modulation supports policy-specific reinforcement and not action selection

Hodge, A. T.; Yttri, E.

2024-10-02 neuroscience
10.1101/2024.10.02.616336 bioRxiv
Show abstract

Two contrasting models dominate our understanding of basal ganglia function: action selection and reinforcement learning. Prolonged, indiscriminate stimulation of direct and indirect pathway striatal neurons produces effects consistent with the action selection; however this approach ignores the transient, movement-specific dynamics that characterize these cells. To determine how striatal subpopulations contribute to mouse behavior, we applied brief closed-loop optogenetic stimulation to modulate ongoing activity in a manner that directly dissociates the contrasting models: upon the detection of locomotor arrest or leftward turns. While action selection models predict that increased direct pathway stimulation should induce locomotion and turning contralaterally to the side of stimulation, selective stimulation biased behavioral policies towards more frequent locomotor arrest and leftward turns, regardless of the side of stimulation. Indirect pathway stimulation had the opposite effect. Behavior followed the policy associated with the change in striatal activity, providing a mechanism to enable the reinforcement a wide range of behavioral features to shape performance.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.