Purchase this article with an account.
Jun-Yun Zhang, Cong Yu; Dichoptic de-masking learning in adult amblyopes and its mechanisms. Journal of Vision 2015;15(12):28. doi: https://doi.org/10.1167/15.12.28.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Amblyopia is characterized by poor visual acuity in the amblyopic eyes (AEs) and degraded stereoacuity. Previously we found that dichoptic de-masking training aiming at reducing the impact of interocular suppression could further boost stereoacuity, but not visual acuity, in adult amblyopes who had received extensive monocular perceptual training (Zhang et al., VSS2013). Here we further investigated whether dichoptic de-masking learning is a low- or high-level process. Eleven new adult amblyopes used AEs to practice contrast or orientation discrimination with two aligned vertical/horizontal Gabors (80% contrast). The stimuli were dichoptically masked by a band-filtered noise masker simultaneously presented in NAEs. A 1-interval 2AFC staircase procedure estimated the maximal tolerable noise contrast (TNC). Training of dichoptic de-masking for contrast discrimination (n=6) doubled maximal TNC, but maximal TNC was unchanged at an orthogonal orientation, showing orientation specificity. AEs were then exposed to the orthogonal orientation through irrelevant orientation discrimination also under dichoptic masking, which improved maximal TNC for contrast discrimination at the orthogonal orientation as much as at the trained orientation. A control experiment confirmed that improved maximal TNC for contrast discrimination did not result from de-masking training for orientation discrimination. (2) Reversely, orientation specificity in dichoptic de-masking learning for orientation discrimination (n=5) was also eliminated by exposure of the transfer orientation through contrast discrimination under dichoptic masking. (3) Training (15-hrs) improved stereoacuity by 61.3%, similar to previous 66% improvements after two-stage monocular and dichoptic training. The complete orientation transfer of dichoptic de-masking learning and the task specificity suggest that dichoptic de-masking learning is mainly a high-level process. The brain may learn to readout orientation or contrast signals from dichoptically presented noise, and this learning is rule-based to allow transfer to untrained orientations. The training may also reduce the impact of interocular suppression to improve stereoacuity.
Meeting abstract presented at VSS 2015
This PDF is available to Subscribers Only