Presentation Information
[P1-23]Temporal Reward Prediction in the Visual Corticostriatal Circuit
*Rebekah Yidan Zhang1,2, Lianne Saussy1, Marshall Hussain Shuler1,2 (1. Johns Hopkins University (United States of America), 2. Kavli Neuroscience Discovery Institute (United States of America))
Keywords:
Reward timing,Corticalstriatal circuit,Neuropixels,Decision making,Reinforcement learning
Accurate prediction of reward timing is critical for adaptive decision-making, yet the neural mechanisms underlying temporal reward expectations remain poorly understood. We investigate how the visual corticostriatal circuit (VC>DS) encodes and transmits reward timing signals to guide time-investment behavior. While the visual cortex (VC) traditionally is regarded to simply processes sensory information, a growing body of work demonstrates its role in encoding both reward timing and action timing. Complementing this, the dorsal striatum (DS) is known to integrate motor timing and action valuation. We propose that VC transforms sensory cues into temporal reward predictions, which DS then translates into timed behavioral policies.To test this hypothesis, we developed a novel behavioral paradigm where head-fixed mice optimize waiting durations to maximize reward rates. Mice were divided into two groups and trained with different reward regimes of distinct background delays (1s vs. 5s), requiring strategic adjustment of wait times. Behavioral data reveal precise adaptation, with mice waiting significantly longer under longer background delays (3.84s vs. 1.95s; n=26 mice; 604,837 trials; p-value < 10e-28). Simultaneous neural recordings using Neuropixels 1.0 probes identified DS and VC neurons exhibiting wait time-dependent firing patterns, with peak activity prior to decision to end waiting scaling either positively or negatively with wait duration.Current findings support a model where VC computes reward timing expectations that DS utilizes to guide action selection. Future work will employ circuit-specific perturbations to test the causal role of VC>DS projections in timing behavior. This study provides mechanistic insights into how sensory-motor circuits integrate temporal information to guide decisions—a process impaired in Parkinson’s disease and addiction. By elucidating computational principles of the VC>DS circuit, we advance our understanding of predictive timing in adaptive behavior.