A NOVEL OLFACTORY PATCH FORAGING TASK TO STUDY DECISION MAKING IN HEAD-FIXED MICE

Flexibility in an ever-changing world is key to survival for all organisms. Patch foraging decisions are a natural example of this. Animals choose between exploiting local resources or leaving to explore potentially superior places (exploitation vs exploration). We developed an ethologically grounded olfactory patch foraging task to study flexible decision-making in head-fixed mice (Figure 1a). In our paradigm, thirsty mice foraged for clustered and depleting water reward patches on a closed-loop multi-sensory virtual track. Mice had to stop in odor sites to trigger potentials water rewards and run through a site to indicate a patch leaving decision (Figure 1b). This task was learned rapidly (~5 sessions) and required minimal behavioral shaping. In patches that depleted probabilistically (Figure 1c), a single leaving threshold for reward probability was observed across patch types (n = 23 mice; Figure 1d, left) despite requiring different number of stops and rewards collected for each (Figure 1d, middle and right). In addition, we showed that manipulations of global reward rate (Figure 1e) and travel cost (Figure 1f) impacted the local patch leaving threshold. Together, these results indicate that mice make local decisions that are sensitive to global reward rate. We discuss our findings in the context of the Marginal Value Theorem (Figure 1g) as well as alternative models. Ongoing and future work will use multi-site Neuropixels recordings to examine how interconnected neural populations compute and track patch value, leaving threshold, and global reward rate in dynamic environments.