Contrast suppression and stereoblind zones in amblyopia

Last year we used a novel stereoperimetry technique to demonstrate that amblyopic observers with measurable stereopsis were functionally stereoblind at the fovea (Ghahghaei & Verghese, 2018). Here we set out to determine whether the size of the central stereo-blind zone is related to the depth of amblyopia as indexed by contrast suppression. Specifically, we measured contrast suppression locally using a dichoptic contrast-matching task and compared it with our map for stereopsis in the same individual. Participants included 5 controls, 3 with anisometropic amblyopia and 2 with micro-strabismus. In the contrast-matching task, observers viewed a full-field horizontal sinusoidal grating (0.5 cycle/deg, contrast 30%) presented to both eyes and a test patch with higher contrast presented to one eye. Observers used the method of adjustment to match the test contrast to the background. The test patch was presented at 0, 1.25, 2.5, 5 and 10° eccentricity along the horizontal and vertical meridians. The patch measured 1° in the fovea and its size was m-scaled with eccentricity. Eccentricity, meridian and the eye in which the test was presented were randomized. For each tested location, we measured the logarithm of the ratio of matching contrasts (RMC) for the amblyopic (non-dominant) to the fellow (dominant) eye. Overall, the amblyopic/deviating eye required greater contrast than the fellow eye, but the spatial mapping between zones with elevated RMC and impaired stereopsis differed between the two clinical groups. For anisometropic amblyopes, a within-subject comparison showed elevated RMC at the stereo-blind fovea. However, outside the fovea, there was no clear mapping between para-foveal locations with impaired stereopsis and contrast suppression. For micro-strabs, there was no systematic mapping between the central stereo-deficient zone and elevated RMC. Our results suggest that contrast suppression might contribute to stereo-deficiency but is insufficient to account for the spatial pattern of stereo loss.