Summary
High-fidelity sensory neural implants must be calibrated to precisely activate specific cells via extracellular stimulation. However, collecting and analyzing the required electrical stimulation data may be difficult in the clinic. A potential solution is to infer stimulation sensitivity from intrinsic electrical properties. Here, this inference was tested using large-scale high-density stimulation and recording from macaque retinal ganglion cells ex vivo. Electrodes recording larger spikes exhibited lower activation thresholds, with distinct trends for somas and axons, and consistent trends across cells and retinas. Thresholds for somatic electrodes increased with distance from the axon initial segment. Responses were inversely related to thresholds, and exhibited a steeper dependence on injected current for axonal than somatic electrodes. Dendritic electrodes were largely ineffective for eliciting spikes. Biophysical simulations qualitatively reproduced these findings. Inference of stimulation sensitivity was employed in simulated visual reconstruction, revealing that the approach could improve the function of future high-fidelity retinal implants.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
↵10 Lead Contact