Assessment of stereovision with digital testing in adults and children with normal and impaired binocularity

Abstract

New digital approaches allow stereovision to be assessed with greater precision than current clinical stereo tests. Those current tests present a relatively narrow range of stimulus disparities in coarsely quantized steps. With dichoptic treatments for amblyopia emerging, more accurate assessment of especially coarse stereopsis becomes increasingly important for verifying their aim to improve 3D vision. We used digital testing in subjects of a large age range (4-59 years), with groups having both normal (n=34) and impaired binocular vision due to unilateral amblyopia, with or without strabismus (n=27). Random-dot stimuli were presented on a 3D monitor with shutter glasses. The test applies adaptive procedures to measure psychometric functions and provides thresholds with associated confidence intervals. Digital thresholds for controls (range 11-160 arcsec) and stereodeficient subjects (range 43-911 arcsec) were compared to the TNO, a standard clinical test which uses similar random-dot targets presented with anaglyph glasses. Agreement between digital and TNO thresholds varied with the level of stereopsis. Stereoacuity was measurable in several subjects who failed on the TNO. With the digital test we found good repeatability for both groups, with the indication of a small learning effect for subjects with coarse stereopsis. Thus, assessment of all target groups for new tests is important, and repeated testing before therapy may avoid confusing learning and treatment effects. Our digital approach supplies a large range of accurate stereo data in children and adults; together with its associated measure of variability, it will be useful in longitudinal treatment studies.