There is a rapidly growing brain-computer-interface (BCI) community working to develop methods of extracting and processing neural signals for directly controlling devices such as cmputers or prosthetic systems. A neuroprosthesis is a device that interfaces with the nervous system to restore function, usually through electrical stimulation. Researchers in the RNE Lab are currently using chronic multiunit neural recordings to examine the sensory feedback provided by peripheral mechanoreceptors during walking (Weber et al., 2003) and have shown that it is possible to decode limb trajectory information from small groups (10 – 30) of simultaneously recorded sensory neurons in the behaving cat. This approach may eventually be used to provide feedback for controlling a functional electrical stimulation (FES) system for walking and reaching.

Another goal of a sensory neural interface is to restore natural sensations of force, posture and movement (kinesthesia) by microstimulating the sensory pathways that normally provide this information to the brain. An important question is where in the sensory pathways is the best location for applying microstimulation to deliver the sensory feedback. The RNE Lab is exploring techniques for microstimulating primary afferents in the dorsal root ganglion. Delivering the stimulation early in the sensory hierarchy ensures tht the sensory information is delivered to the widely distributed sensorimotor regions in the brain and spinal cord that normally receive these sensory signals.


Figure 1: Neural Interfaces for Prosthetic Control
P3img1


Figure 2: Illustration of Goals Related to Artificial Sensory Feedback. In this case the person uses the sensory feedback to know where their hand is without looking at it. The person is able to grasp the cup without crushing it, and is able to know whether or not the coffee is too hot before potentially burning their lips.
armHandCartoon