Stereo perimetry reveals a foveal impairment of stereopsis in amblyopia.

Amblyopia is associated with suppression in the central visual field. It has been proposed that in cases of anisometropia or small misalignment, there may be peripheral fusion that can serve as the basis for stereopsis (Siretanu & Fronius, 1981; Harrad, 1996). Here, we used a novel stereo perimetry method to test this conjecture.

We measured stereopsis in the fovea and the periphery using a method analogous to perimetry. The observer maintained gaze at a fixation point, while keeping nonius lines aligned, and detected whether a target was presented in front of or behind the fixation plane. The target could appear on the cardinal or diagonal axes, at eccentricities of 0, 1.25, 2.5, 5, and 10 degrees from fixation, with size m-scaled for eccentricity (1° X 1° at fovea). Targets were presented at a large disparity step (10 or 15 arc min) for 1 sec. The inter-trial-interval was 3-4 seconds. Viewing distance was 40 cm. The display was made up of full-field dynamic random dots updated every 500 ms to minimize monocular cues. We tested 4 amblyopic participants (1 with anisometropia, 3 with strabismus), 2 participants with micro- strabismus, and 4 controls.

For controls , the stereo perimetry map for depth discrimination showed high accuracy across the visual field. Our strabismic observers were unable to detect depth at any of the eccentricities we tested. For observers with anisometropia or micro-strabismus, stereo perimetry showed that depth discrimination was impaired specifically in the foveal and parafoveal regions (up to 5 degree eccentricity depending on the participant). Thus, our stereo perimetry technique demonstrates that when stereopsis is present in amblyopia, it is mediated by the periphery. This supports previous studies that suggest that coarser-scale mechanisms may survive small misalignment and the monocular blur associated with anisometropia (Giaschi et al, 2013; McKee et al, 2003).