Abstract
Vision is generally considered in terms of phototransduction in rod and cone photoreceptors. However, a more recently discovered transduction pathway lies in retinal ganglion cells (RGCs) that express the photopigment melanopsin, making them intrinsically photosensitive (ipRGCs). Some studies have exploited the blind spot as a location where ipRGCs can be stimulated in the absence of rods and cones. A potential concern with this approach is that light from conventional sources is subject to spread, stimulating other areas. This study explores the use of a prototype augmented-reality device developed by Sony Group Corporation, that uses scanning lasers to project a video image into the eye. The collimated light minimizes stimulation beyond the intended target, and delivers an image that remains in focus regardless of lens accommodation. We stimulated the right eye while measuring its movements and pupil size. Observers (N=5) fixated a multi-wavelength reticule, whose component colors would diffract apart if the head/eye shifted. Tests began with blind spot mapping, and then presentation of stimuli inside and outside of the blind spot (red/green/blue = 643/522/445 nm; in fields, rings, dots, half-moons). When stimuli were presented in the blind spot, we found that fixational eye movements were biased to the right and downward, in their direction. We also conducted pupillary response tests based on Gamlin et al. 2007 and Miyamoto et al. 2015. We replicated the results in part, finding that blue light in the blind spot resulted in pupillary contraction, and a sustained post-illumination pupillary response. Other interactions were not replicated, but we identified procedural and technical factors that may have contributed. Future refinements will be required to confirm the results. This work presents an optically cleaner method of stimulating the blind spot, allowing direct attribution of responses to melanopsin activation. It also indicates that ipRGCs may implicitly direct spatial attention.