Contour integration in adults with a history of amblyopia

Abstract

Abstract: : Purpose: Our aim was to evaluate how abnormal binocular visual experience affects contour integration, a global visual processing ability. We varied both element cocircularity and background noise density in a contour integration task. Previous studies predict deficits in strabismic but not anisometropic amblyopes. Methods: We tested 16 adults; 6 had normal vision and visual experience; 10 had a history of strabismic or anisometropic amblyopia. We measured contour integration (CI) and contrast sensitivity (CS) over a range of spatial frequencies, using spatial 2AFC MOCS, testing each eye separately. The contour was a ring of Gabor patches in a field of randomly arrayed Gabor patches. To assess CI, we varied the background density of Gabor elements and the cocircularity of the elements comprising the contour. Results:Our measure of contour integration performance was "noise tolerance"; the maximum noise density (in patches/sq deg) yielding 75% correct was taken as threshold. Noise and cocircularity tolerance showed that CI deficits were consistent across cocircularity levels, regardless of a history of strabismic versus anisometropic amblyopia. Individuals with a history of strabismic amblyopia consistently showed CI deficits, whereas those with a history of anisometropic amblyopia had mixed CI results. Neither CS nor VA predicted CI. Conclusions: CI performance is not directly related to a history of anisometropic versus strabismic amblyopia. The degree of cocircularity within the range tested did not influence performance in strabismic or anisometropic amblyopia. We consider previously contradictory reports in light of these data.

Keywords: amblyopia • contrast sensitivity • shape, form, contour, object perception