From staying in a shelter or exploring a new one, to keeping a job or searching for another one, we often have to decide between exploiting a resource or exploring the environment for new opportunities. Information about the environment and the animal’s state converges in the dorsomedial striatum, considered as an integrator of exploration/exploitation decisions. To study the role of the dorsomedial striatum in the exploration/exploitation balance, we designed a virtual reality foraging task. Head fixed mice explore a virtual linear track which consists of short rewarded areas (“patches”) separated by long unrewarded corridors. Inside the rewarded area, animals need to perform a sequence of licks to obtain water. Each consecutive reward requires exponentially more licks to be obtained. At some point, the animal reaches a breaking point when it decides to stop exploiting the current patch and run to the next one. We used an array of four chronically implanted tetrodes to record single unit and LFP striatal activity during the performance of the task. Preliminary data show that increasing the length of the unrewarded corridor raises the effort the animal invests in each patch. We aim to align the behavioral events with striatal neuronal activity to find correlates predicting decision making by using multiple logistic regression. We will also focus on striatal cholinergic interneurons known to encode reward-related cues.