Biomedical Engineering

Neural Control of the Hand: From Sensorimotor Memory to Execution of Dexterous Manipulation

Printer-friendly versionSend by emailPDF version


Dr. Marco Santello, School of Biological & Health Systems Engineering, Arizona State University


Tuesday, May 21, 2013 - 11:00am to 12:00pm


102-D ENGR I


Neural Control of the Hand: From Sensorimotor Memory to Execution of Dexterous Manipulation Anticipatory control of movement has been characterized in motor tasks as a way through which the central nervous system can bypass delays associated with reflex-based control. We have been studying how humans learn anticipatory control of manipulation tasks to characterize the mechanisms underlying the transformation from multiple sources of sensory feedback to the coordination of multiple degrees of freedom of the hand. In our approach, we have removed constraints on digit placement to study how subjects explore and choose relations between digit forces and positions. The main difference between grasp control at constrained vs. unconstrained object locations is that anticipatory control of grasping in the former scenario can rely on sensorimotor memories of digit forces. In contrast, removing constraints on digit placement might result in trial-to-trial variability of digit placement, hence require more complex sensorimotor processes for the integration of digit position and force sensing. I will review recent work from my laboratory on the problem of digit position/force coordination using tasks that allow, or interfere with, the retrieval of learned manipulations. A key concept linking behavioral and theoretical considerations is that subjects high-level task representations that, when retrieved, allow control of manipulation in an effector-independent fashion. Interestingly, the extent to which these representations can be successfully used is highly sensitive to a number of factors, including the frame of reference in which manipulations are learned, time-dependent motor bias from based on most recent hand-object interactions, and potential conflicts that may arise between visual cues versus implicit knowledge of object dynamics. The theoretical framework that is emerging is that sensorimotor memories, when integrated with multiple sources of sensory feedback, can facilitate or interfere with the coordination of multiple degrees of freedom of the hand for dexterous manipulation. This framework is helping to identify the boundaries of the neural processes underlying learning and generalization of complex movements characterized by high-dimensionality in the sensory and motor domains.