Abstract
Activities with diminishing returns pose a unique computational problem for the brain, requiring a combination of outcome evaluation and temporal tracking. Deciding the right time to stop one pursuit and move to alternatives is an important part of effective time management, yet it is unknown how the decision-making circuits of the brain determine the moment to switch. The dorsomedial striatum (DMS) mediates both goal-directed decision-making and interval timing-two functions that converge during patch foraging, where animals must time when to exit patches to maximize reward rates. We recorded extracellular activity from neurons in DMS while freely moving mice performed a patch-foraging task. Mice employed a 'reward-reset' strategy, primarily basing their exit decisions on the time since the last reward, but with the patch residence time and environmental reward-rate context also contributing to the intended time of departure. Individual neurons in DMS underwent discrete firing rate transitions at characteristic delays following each reward. These transition delays were distributed across the population, creating a cumulative signal that reached a threshold coinciding with patch exit. This population activity pattern spanned the intended reward-to-exit interval, compressing or expanding in accordance with patch residence time and changing environmental conditions. Fiber photometry recordings revealed phasic dopamine signals in DMS encoding reward prediction errors that reflected the declining reward probability over time. Our results provide insights into how DMS integrates its dual roles in timing and action selection to guide time investment strategies during foraging.