UC Berkeley

Visual resolution, the ability to see fine spatial detail, emerges from the capacities of both the eye and the brain. A great deal of insight into the anatomical and physiological basis of human visual resolution has been gained since Helmholtz first proposed his sampling theory of visual resolution. Anatomical, physiological and psychophysical investigations have revealed in great detail the properties of the biological structures underlying visual resolution and identified many of the optical, retinal, and cortical factors that govern the limits of visual resolution. However, technological limitations have long prevented researchers from examining both the structure and function of the visual system simultaneously in the living eye. The microscopic photoreceptors of the retina have been inaccessible to optical examination, preventing high quality measurements of both visual resolution and retinal anatomy from being obtained in the same eyes. The present studies investigated the relationship between the optical, retinal, and cortical factors that govern visual resolution in humans. These experiments employed adaptive optics scanning laser ophthalmoscopy (AOSLO) as a tool to study how these factors govern visual resolution in normal and diseased eyes. The AOSLO is an ideal tool for studying the limits of vision because of its ability to present complex stimuli to the retina that are of higher optical quality than the visual system has ever experienced, while simultaneously imaging the underlying cone photoreceptor mosaic on a microscopic scale.

Adaptive optics correction of ocular aberrations allowed observers to achieve immediate and significant improvements in visual resolution. Training was not required to achieve this benefit, which allowed the resolving capacity of the retinal and cortical visual system to be assessed unobstructed by the optics of the eye. Not all participants in these studies benefited to the same extent from AO correction. Visual resolution was found to be significantly poorer in low myopia as compared to emmetropia, despite the similar optical quality afforded by AO correction, showing that retinal and cortical changes in myopia caused the observed deficit. Simultaneous imaging and visual resolution testing determined the precise relationship between the spatial sampling limit of the cone mosaic and visual resolution across the human fovea. These studies revealed that the spatial sampling limit of the cone mosaic largely governs visual resolution at the center of the fovea for normal eyes, but that outside the foveal center visual resolution falls off at a greater rate than predicted by cone spacing and is governed by the spatial sampling limit of the mosaic of midget retinal ganglion cells.

Significant differences between otherwise normally appearing observers were revealed using AOSLO, showing the power of visual resolution testing after AO correction for detecting small changes in the visual system resulting from disease. Significant retinal changes were revealed in female carriers of a rare X-linked genetic mutation in the L and M opsin gene array that causes blue cone monochromacy (the loss of all L and M cone function) in affected males. Retinal findings from AO imaging provided insight into the development and function of the carrier retina. Although carriers had visual resolution within the normal range when tested clinically, visual resolution testing in AOSLO revealed significantly reduced visual resolution compared to normal eyes. Resolution testing across the fovea in normal and diseased eyes provided insight into the relationship between cones, ganglion cells and visual resolution across the visual field. Retinal imaging showed that carriers had fairly normal cone topography despite peak cone density that was ~50% lower than normal; showing that cones destined to express a non-functional photopigment degenerated early in development. Drastic reductions in visual resolution across the fovea in the carrier are best explained by ganglion cell loss that resulted from the loss of cones in the carrier. The relationship between resolution and the spatial sampling limit of the carrier cone mosaic was similar to what was predicted for normal eyes, suggesting an organizing principle in the visual system whereby cone density is the primary determinant of the retinal circuitry that governs visual resolution across the visual field. Together, these studies provide an important contribution to the understanding of the limits of human visual resolution.