Abstract
“Purpose: Axial anisometropes have retinal images of different sizes in the two eyes. Nonetheless, stereopsis is optimized when the images approaching the eyes are the same size rather than when the image sizes on the retina are the same (Vlaskamp et al., 2009). Two adaptation mechanisms could account for this result. First, the retina may expand in proportion to axial length such that the number of cones sampling a given visual angle in the two eyes remains unchanged. Second, post-receptoral neural mechanisms may adjust for the differences in retinal image size prior to the stereoscopic processes. To determine which is the better explanation for adaptation to anisometropia, we used an adaptive optics (AO) ophthalmoscope to measure the linear and angular cone density in an axial anisometrope.
Methods: AO imaging was done using 840nm light with dynamic wavefront correction performed over a 6mm pupil. Resultant images of the cone mosaic were stabilized and averaged, and individual cones were identified using automated and manual methods. Axial length, corneal curvature, and anterior chamber depth were measured using the Zeiss IOLMaster. Those parameters were then entered into the Bennett-Rabbetts schematic eye to estimate image size on the retina and the relative magnification between the two eyes. Cone density estimates were computed via a Voronoi analysis by taking the inverse of the area of each Voronoi tile.
Results: The angular cone density (cone/deg2) in the longer eye was higher by 9% at a retinal eccentricity of 1 deg and increased nearly monotonically to 20% near the center of the fovea. The linear cone density (cones/mm2) was nearly the same between the eyes despite a difference in axial length of 1.08 mm. This means that images of the same visual angle are sampled by more cones in the longer eye. We could not resolve cones closer to the fovea than 0.25 deg. We are currently examining stereopsis in the retinal periphery of this observer where retinal imaging suggests that cone density is constant in angular rather than linear units.
Conclusions: An observer with different axial lengths has adapted to the differences in retinal image sizes via post-receptoral neural adaptation rather than via expansion of the cone mosaic in the longer eye.”