Anne Danneels¹, Ângela Gomes Tomaz¹, Dennis M Levi¹, Bilge Sayim²; ¹Herbert Wertheim School of Optometry & Vision Science, University of California Berkeley, Berkeley, United States of America, ²Laboratoire de Sciences Cognitives et Psycholinguistique, Département d'Études Cognitives (DEC), École normale supérieure (ENS), CNRS, Paris, France

Amblyopia is a neurodevelopmental disorder characterized by reduced visual acuity. Observers with amblyopia have increased spatial uncertainty and poor spatial localization. In redundancy masking (RM), observers perceive fewer items than presented (e.g., perceiving two lines when three were presented) predominantly with regular arrangements of items. RM is strongest along the horizontal meridian revealing a reverse horizontal-vertical anisotropy compared to crowding. Here, we explored the strength and pattern of redundancy masking in amblyopia. We hypothesized that RM would be reduced in amblyopia, although still strongest along the horizontal meridian, as accurate (relative) localization is required to detect stimulus regularity. Sixteen observers (5 amblyopes, 11 controls) were presented with arrays of 3, 4, or 5 lines at 5° and 10° eccentricity for 150 ms. Eight locations were tested (polar angles: 0° to 315° in steps of 45°). To ensure that the extent of the line array was not used as a cue to the number of lines presented, inter-line spacings were varied: 0.40°, 0.58° and 0.78° at 10° eccentricity (and scaled for presentation at 5° eccentricity). Observers’ minimum angle of resolution was smaller than the smallest spacing at each experimental eccentricity. Observers’ task was to indicate the number of lines perceived. Preliminary results showed a surprising trend for high accuracy (correct reporting of the number of lines presented) in control observers and overestimation (reporting more lines than presented) in observers with amblyopia. Along the right horizontal meridian, performance was more accurate (less overestimation and no RM) compared to all other locations, except the left horizontal meridian. Similarly, accuracy was higher along the left horizontal meridian but only compared to the left intercardinal locations. These results, together with previous RM studies, highlight possible individual differences in RM strength, and suggest potential group-level differences between individuals with amblyopia and controls.

Acknowledgements: Supported by a grant from the National Eye Institute awarded to Dennis M. Levi (R21EY030609) and a France-Berkeley Fund grant awarded to Dennis M. Levi and Bilge Sayim, as well as a fellowship of the Belgian American Educational Foundation awarded to Anne Danneels.