Abstract
[Goal] Amblyopia is a neurodevelopmental disorder characterized by visual disparities between the eyes. Perceptual learning (PL) protocols, in which observers perform the same task across several days, improve amblyopic vision to a limited degree. Exogenous attention remains functionally intact in adults with amblyopia. Attention generalizes PL to untrained locations in neurotypical adults. Here, for the first time, we manipulate observers’ attention to investigate whether exogenous (involuntary) covert spatial attention facilitates and/or generalizes PL effects in adults with amblyopia. [Methods] Adults with amblyopia trained for 80 blocks (~10 sessions) on a 2-AFC orientation discrimination task using their amblyopic eye (dominant fellow eye patched). We monitored fixation with an eyetracker. We presented the Gabor stimuli at the contrast threshold (80% accuracy) of each eye for each observer. Observers were randomly assigned to train along one diagonal, either with peripheral attention cues (Attention group) or central cues (Neutral group). In sessions immediately before and after training, we measured training task performance for each eye along both diagonals. Observers were always presented with neutral attention cues during the contrast thresholding, pre-test, and post-test sessions. On the first and last days we also applied a test-battery to measure observers’ foveal contrast sensitivity, acuity, crowding and stereoacuity. [Results] For both groups, training improved performance at the trained diagonal, but more so for the Attention group than the Neutral group. Remarkably, only in the Attention group, learning transferred to the untrained diagonal in both the amblyopic and high functioning fellow eye. Both groups improved performance in the battery tests. [Conclusions] Exogenous attention facilitates PL at trained spatial locations in the trained amblyopic eye. Moreover, it generalizes these performance improvements beyond the specific training conditions—to the untrained spatial locations in the trained amblyopic and untrained fellow eyes. We discuss the translational implications of these findings for visual rehabilitation.