Multimodal dataset linking wide-field calcium imaging to behavior changes in operant lever-pull task in mice

The link between comprehensive behavioral measurements during a behavioral task and brain-wide neuronal activity is an essential strategy to better understand the brain dynamics underlying the emergence of behavior changes. To tackle this, we provide an extensive, multimodal dataset that includes 15 sessions spanning 2 weeks of motor skill learning, in which 25 mice were trained to pull a lever to obtain water rewards. Simultaneous high-speed videography captured body, facial, and eye movements, and environmental parameters were monitored. The dataset also features resting-state cortical activity and sensory-evoked responses, enhancing its utility for both learning-related and sensory-driven neural dynamics studies. Data are formatted in accordance with the Neurodata Without Borders (NWB) standard, ensuring compatibility with existing analysis tools and adherence to the FAIR principles (Findable, Accessible, Interoperable, Reusable). This resource enables in-depth investigations into the neural mechanisms underlying behavior and learning. The platform encourages collaborative research, supporting the exploration of rapid within-session learning effects, long-term behavioral adaptations, and neural circuit dynamics.