This project aims to generate a rich, large-scale dataset capturing the activity of defined neural populations during the IBL decision-making task, using a 2-photon random access mesoscope. By imaging calcium activity in excitatory neurons across the dorsal cortex of transgenic mice, we seek to characterize how population-level neural activity encodes key task variables such as stimulus, choice, and bias context. Imaging functionally connected regions simultaneously allows us to investigate interregional interactions and trial-to-trial dynamics, while repeated recordings across days will help assess the stability of these neural representations over time and their relationship to expert performance.

Complementing the IBL’s existing Neuropixels electrophysiology data, this project adds a critical new dimension by visualizing the spatial organization and long-term evolution of cortical population codes. To support this, we are developing and refining a technical pipeline for the entire calcium imaging workflow, from data curation and preprocessing to analysis and public dissemination. The mesoscope platform thus serves not only as a scientific tool but also as a testbed for evaluating and standardizing open-source methods that enable reproducible, scalable imaging research. This technical effort ensures that the dataset becomes a robust, long-term resource for understanding cortical function during decision making.