Visual Perceptual Learning for the Treatment of Visual Field Defect

This study will test the hypothesis whether visual field defect caused by chronic stroke could be improved by visual perceptual learning. The secondary hypothesis is that improvement of visual field defect would be accompanied by connectivity changes in visual networks. In addition, investigators will evaluate whether improvement of visual field defect could achieve improvement of subjective visual function.

Study design: A PROBE (Prospective, Randomized, Open-label, Blinded-endpoint) design Methods: 1. Visual perceptual learning (VPL) group 1. Visual perceptual learning (24 times for 2 months, 3 times per week) in visual field defect areas 2. At baseline and post-VPL evaluations (2 months after): Structural \& Functional MRIs, Visual field tests, Visual evoked potential, Visual cognitive tests, Subjective visual function questionnaires 2. Control group 1. At baseline and 2-months later: No VPL, Structural \& Functional MRIs, Visual field tests, Visual evoked potential, Visual cognitive tests, Subjective visual function questionnaires 2. Visual perceptual learning after 2-month follow-up evaluation (24 times for 2 months, 3 times per week) (The patients who were stratified by age (over 65 years old or not) and degree of visual field defect (quadrantanopia or complete hemianopia) will be randomly assigned into the learning and control group.) 3. Investigations 1. Improvement of visual field test scores between VPL group and control group 2. Analysis of brain network changes in MRI data 3. Correlation analysis of improvement of visual field defect and MRI data 4. Correlation analysis of improvement of visual field defect and visual evoked potential (VEP) data 5. Correlation analysis of improvement of visual field defect and neuropsychological data