The purpose of epiretinal prostheses is to mimic natural vision by evoking neural activity in the retina using electrode arrays. Prototype devices use large, coarsely spaced electrodes that indiscriminately stimulate many cells simultaneously, producing large phosphenes and limited visual function in blind patients. In order to improve the quality of a future prosthetic device, we need the ability to selectively activate individual cells and groups of cells to accurately recreate the activity that occurs during normal vision. Recent studies from our lab show that we can stimulate individual cells selectively using safe current levels for specific cell densities. However, it is unknown whether selective activation is possible for cases where the density of RGCs is higher. In this proposal, we will test the hypothesis that we can improve selective activation in dense RGC regions by using patterned stimulation. We will exploit multi-electrode stimulating and recording techniques to perform the proposed experiments in isolated rat and monkey retinas.
Aim 1 will explore whether the distinct biophysical properties of neuronal somas and axons will allow selective activation of somas while minimizing unwanted axon activation.
Aim 2 will use empirical measurements combined with computational methods to empirically determine the best pattern for selectively activating a target cell.
Aim 3 will incorporate imaging to guide our understanding of the cellular features tha influence selective activation using the patterned stimuli developed in Aims 1 and 2. The results of this study will lead to a better understanding of whether patterned electrical stimulation can improve the selectivity of cell activation. Improved understanding of electrical stimulation techniques may guide the development of future, advanced retinal prostheses and other brain-machine interfaces.
The goal of this proposal is to develop methods to improve the achievable resolution of selective neuronal activation in epiretinal prostheses. We will approach this goal using an integrated stimulation and recording multi-electrode array capable of providing electrical stimulation patterns on one or more electrodes, while recording elicited activity on many electrodes and optically imaging the neurons being stimulated, in rat and monkey retinas. We expect the approaches developed in the course of this work will be used in future prostheses and other brain-machine interfaces.
