Abstract
Purpose:
People with central vision loss rely on their peripheral vision for reading. Previous studies have shown that peripheral reading can be improved with practice on character-based tasks. An optimal training paradigm should require little effort from patients to allow self-administration while providing efficient learning and prominent improvements. Heretofore, such a paradigm has been lacking. Here, we introduce a practical and effective training method that utilizes priming of stimulus identity to facilitate learning in the absence of a task.
Methods:
Seven normally-sighted young adults were trained at 10° in the lower visual field in a pre/post design. In the pre-test, we measured reading speeds using Rapid Serial Visual Presentation, visual-span profile (letter-recognition accuracy as a function of letter position) with trigrams (strings of three random letters), the spatial extent of crowding (letter separations yielding 80% recognition accuracy for the middle letters of trigrams), and isolated-letter recognition performance. To assess transfer of learning across print sizes, reading speed was measured at two print sizes (2.5° (trained), 1°). Training consisted of five sessions of viewing trigrams presented at various letter positions, a total of 7150 trials. Each training session had 26 blocks. Within each block, the middle letter of the trigram was always the same. The left and right letters were randomly selected resulting in differences, across trials, in the perceived appearance of the middle letter. The observer had no task, but rather was asked to attend to each trial and learn the known target letter with the option of repeating the same trigram stimulus for as many times as needed (averaged 0.4 times). The post-test was identical to the pre-test but in reversed order.
Results:
Averaged across observers, the five-day training reduced the spatial extent of crowding by 29% and enlarged the visual span by 1.6 characters. The improvements transferred to the untrained reading task for both the trained print size (reading speed increased from 132 to 298wpm) and the untrained print size (from 68 to 171wpm), which is beyond what is typically observed in traditional task-based training.
Conclusions:
Our findings demonstrate the effectiveness of utilizing priming and stimulus exposure in training to improve peripheral reading. Moreover, this non-task-based training is simple enough to allow self-training at home.
Keywords: 584 low vision •
650 plasticity •
672 reading