Role of Task-Specific Context in Sequential Motor Skill learning - PROJECT SUMMARY One of our brain’s most critical features is its ability for multi-task learning, e.g., allowing us to learn and generate countless motor skills throughout our lives, from playing the piano to mastering a tennis serve. Since the breadth of our behavioral repertoire largely depends on this essential ability, disruption of its underlying neural circuitry has devastating consequences. However, our limited understanding of these circuits and of fundamental mechanisms of learning and memory in general severely hampers our efforts to develop treatments for diseases that affect memory like Alzheimer’s or Parkinson’s Disease. In particular, how multiple similar memories are formed, organized and stored in the same neural substrates remains largely unclear. The relationships between memories are of critical relevance, since the brain has to continuously balance the separation of memories to avoid interference with taking advantage of existing knowledge. Here, I propose a multi-level research plan to investigate the neural mechanisms that enable the brain to achieve this feat. I will use sequential learning of multiple motor skills as a model for multi-task learning and aim to dissect the underlying neuronal circuits and mechanisms. I will therefore concentrate on the distributed motor network, which is crucial for learning individual skills, especially the dorsolateral striatum (DLS) and motor cortex (MC). In addition, I will test how the task- specific training context influences multi-skill learning. I hypothesize that more similar contexts push sequential learning towards adaptation of existing skill memories, while less similar contexts promote the development of distinct skill memories. In Aim 1, I will train rats sequentially on two versions of a motor skill task in which they develop complex, spatio-temporally precise kinematics. This will allow me to determine the relationship between sequentially learned skills and how they are influenced by the similarity of their training context. I will address the same questions on the level of the neural skill representations: I will perform 24/7 electrophysiological recordings of DLS activity, which represents the kinematics of individually learned skills, throughout multi-skill learning. In Aim 2, I will test the role of MC in sequential skill learning and whether this role is influenced by the training context. I will perform the same behavioral and recording experiments as in Aim 1, but I will chronically silence MC neurons that project to the DLS before learning of the second motor skill. This will reveal whether MC is necessary to infer context under ambiguous conditions also in sequential skill learning. This multi-level proposal promises advances toward an understanding of multi-skill learning, how it is influenced by context, and of memory formation and storage in general. My findings may also provide insights into learning and memory deficits in motor disorders and help the optimization and development of therapies, e.g., for movement rehabilitation. Finally, this proposal provides unique training opportunities for the PI in behavioral and electrophysiological experimentation and data analysis, invaluable skills on the path to independence as a well- rounded neuroscientist.
