Comparable phosphene size and contrast properties, but differing detection reliability for optogenetic versus electrical stimulation of macaque V1

Cortical visual prostheses aim to restore vision by stimulating primary visual cortex (V1) to evoke artificial percepts (phosphenes). Electrical stimulation has long served this purpose, and optogenetic approaches promise higher spatial resolution via cell-type specificity. Recent investigations have demonstrated localized behavioral responses to optogenetic V1 stimulation (1–3), but the applied perimetry paradigms cannot dissociate a genuine percept from a reflexive oculomotor response, leaving the perceptual character of optogenetic phosphenes unresolved.

To address this gap, two macaques received optogenetic V1 injections guided by laminar electrophysiology and performed detection and discrimination tasks under interleaved electrical, optogenetic, and visual stimulation, enabling direct within-subject comparison.

Electrical stimulation reliably evoked detectable percepts (positive d′ in 100% of sessions), with thresholds scaling with current, resembling visual contrast-response functions. Optogenetic stimulation was far less reliable, with positive d′ in only ∼50% of sessions, despite confirmed V1 activation. Despite this reliability difference, optogenetically evoked percepts closely resembled electrically and visually evoked percepts in perceptual character: perceived size (∼1° of visual angle) was comparable across conditions and scaled similarly with eccentricity. Similarly, perceived contrast equivalents (5%-11% visual-contrast equivalent) were broadly comparable across stimulation approaches.

These findings indicate that successful optogenetic activation engages the same perceptual encoding mechanisms as electrical and natural vision, but that reliable detection may additionally depend on circuit-level mechanisms not consistently engaged by current optogenetic parameters. The comparable percept size further suggests optogenetic stimulation may not confer the enhanced spatial resolution often proposed as its key advantage.

Significance statement

Brain implants that stimulate visual cortex could restore vision to people blinded by damage to the eye or optic nerve, evoking flashes of light called phosphenes. Electrical stimulation, the currently used approach, activates tissue broadly; optogenetic stimulation, which uses light to activate genetically targeted neurons, has been proposed as a more precise alternative, but whether it produces a genuine visual sensation was unclear. Using monkeys trained to report artificially evoked flashes, we show that optogenetic stimulation evokes sensations resembling electrical stimulation in size and brightness but is detected far less consistently. These results show optogenetic stimulation engages the same perceptual pathways as natural vision, but does not yet deliver its assumed resolution advantage, informing the design of future optical visual prostheses.