Most ongoing research to develop advanced hand and arm prostheses that allow an individual with amputation to perform highly dexterous movements and to sense features of objects in the world has separated the performance of movements from the sensing of objects. However, in individuals with normal ability, sensory and motor information are integrated. This project's goal is to create movement and sensation abilities in individuals with amputation that are more similar to that in normally enabled individuals. In particular, the goal is to understand how to best utilize all of the available motor and sensory information in the design of accurate neural controllers for future generations of forearm prostheses. This integrated approach to creating sensory percepts and interpreting motor intent will recreate the natural, minimally monitored use of an artificial hand in volunteers as they interact with their world. Successful completion of the proposed work will begin the process of creating a prosthetic hand that passes the 'Turing Test,' where the user perceives little difference between the artificial hand and their amputated biological hand.
The program of research involves implanting neural interfaces (for both stimulating and recording) in the major arm nerves of human volunteers and investigating use of these interfaces while the volunteers interact with a virtual world. In particular, the project has the goals of (a) identifying components in peripheral nerve signals that correlate with errors between the intended movement and the actual movement of the prosthetic arm; (b) developing accurate encoders and stimulation systems that provide punctate, naturalistic, multimodal, and graded sensory percepts; and (c) developing, characterizing, and evaluating advanced decoders that estimate movement intent from a combination of peripheral nerve signals and sensory information from the prosthesis. This third goal, an advanced decoder, will be evaluated with scenarios that provide a clear measure of performance and that represent activities of daily living. The integration of sensory percepts and motor intent is expected to create a natural, minimally monitored use of a prosthetic hand in order to interact more effectively with the world.
