CRCNS Research Proposal: Brain-wide population dynamics and cross-region coordination for sensorimotor learning

People and animals have a remarkable ability to learn how to act based on what they sense. This project studies how the brain learns these associations and revises them as circumstances change, without disrupting previously learned skills or sensory abilities. Understanding how the brain strikes this balance between flexibility and stability could inform new ways to help people recover from brain injury or disease, and could inspire artificial intelligence systems that learn more like the brain. The researchers measure the activity of many thousands of individual neurons across the brain as they learn and update associations between sensory cues, actions, and outcomes. By tracking how brain activity changes and analyzing the measurements with advanced methods, the researchers aim to discover how different parts of the brain coordinate learning while keeping perception and movement reliable. The project tests the hypothesis that learning changes neuronal activity patterns in different ways across sensory regions, motor regions, and the intermediate regions that connect them, and that these changes are coordinated through shifts in how information is communicated between regions. First, the project analyzes large-scale brain recordings from mice as they learn in an operant decision-making task and a classical conditioning task. These analyses determine whether each brain region keeps its activity patterns fixed, reshapes them into new configurations, or keeps the same patterns but links them to different sensory cues during learning. Second, the project analyzes simultaneous recordings from pairs and triplets of brain regions to determine whether the multi-dimensional correlations between the regions change with learning and behavioral context. Third, the project plans to determine the order in which changes appear across the brain, trial by trial, as a mouse learns a new cue-value association. Altogether, the overall goal is to provide a brain-wide account of how neural populations reorganize during learning and how distributed brain activity supports flexible behavior. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.