Spatial-, Temporal-, and Contrast-Resolutions Obtainable With Retina Implants

Abstract

In blind patients with degenerated photoreceptors a large percent of inner retinal neurons remain histologically intact and stay functionally alive (1). Without visual input, the retinal ganglion cells continue to transmit spontaneously generated action potentials through the optic tract to the visual centers. They can also be artificially activated by trains of short electrical impulses and such stimuli were shown to elicit localized perceptions of light, called phosphenes ((2)–(5)). This finding opened the possibility to restitute basic vision with retina implants by delivering electrical stimuli to the retina. Two main types of retina implants are currently developed using either sub- or epiretinal electrode arrays (see the other reviews in this Volume). Subretinal devices are implanted between the pigment epithelial layer and the outer layers of the retina so that they can activate the outer plexiform layer and bipolar cells (6). Epiretinal implants are placed from the vitreous side onto the ganglion cell and nerve fiber layer of the retina (7). Epiretinal implants can evoke action potentials in retinal ganglion cells and/or their axons (8). Finally, both types of implants send visual information in the form of spike patterns through the axons of retinal ganglion cells along the optic nerve to central visual structures. One crucial question remains regarding the type of neural responses evoked in the visual cortex by stimulation with retina implants. Are the spatial, temporal, and intensity resolutions obtainable with current technologies sufficient, in order to provide useful vision for discriminating objects in a static environment and to perceive motion in dynamic scenes of everyday life? Data for estimates of perceptual resolutions achievable with retina implants are rare.KeywordsGanglion CellRetinal Ganglion CellRetinitis PigmentosaSpike RateRetinal NeuronThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.