Coordinated, skillful movements are remarkable achievements of the human nervous system. While we have succeeded in building computer systems that can beat the best human chess player, it has been extremely difficult to develop an artificial system that can pick up and move the chess pieces about the board with anything resembling the grace and ease of the novice chess player. How the brain produces coordinated movements remains an important challenge for cognitive neuroscience. Part of the answer will come from an understanding of cerebellar function, a prominent component of the motor pathways. With support from the National Science Foundation, Dr. Richard Ivry of the University of California, Berkeley and Dr. Jörn Diedrichsen of the University of Wales, Bangor, will pursue a multi-faceted research program to further our understanding of the role of the cerebellum in coordination. Building on new theoretical insights from the field of robotics, the research program will center on evaluating two hypotheses: Is cerebellar function best understood as part of the control circuitry that shapes the motor commands to the muscles based on sensory and motor signals from other limbs? Or is this structure essential for generating predictions concerning the sensory consequences of our actions, a process known as state estimation? While it has long been recognized that cerebellar dysfunction leads to a loss of skilled movement, the symptoms of ataxia may be due to poor control, poor prediction, or both. To distinguish among these hypotheses, participants will be trained to perform novel bimanual movements in which the movement of one hand will depend on an accurate prediction of the state of the other hand. Functional magnetic resonance imaging will be used to determine if cerebellar activation is related to state estimation or control. The imaging work will be complemented by studies involving patients with unilateral cerebellar damage. Here the focus is on a comparison between conditions in which the ataxic limb is required for either state estimation or control, providing strong tests concerning the necessity of this structure to one or both of these functions.
The study of bimanual skills is important given that the vast majority of our daily actions -- typing, driving, cooking -- typically require coordinating the movements of our two upper limbs. The current project should serve as an important step in developing a general theoretical framework for studying coordination across multiple parts of the body (e.g., coordinating movements of the head, eyes, and arms), or even between different individuals (e.g., dancing, basketball). While the research plan focuses on the cerebellum, the basic ideas offer a general approach for exploring the neural mechanisms underlying skilled actions. This project will also provide research training opportunities in cognitive neuroscience for undergraduate, graduate, and post-doctoral researchers. The work will bring together students with interests in engineering, psychology, and the neurosciences, providing unique interdisciplinary training opportunities and advancing our understanding of the human motor system.
