Abstract
Purpose: The visual span refers to the number of adjacent characters that can be recognized in a single glance. In peripheral vision, the visual span for English is enlarged after perceptual training on a task requiring recognition of letter triplets. Here, we examined the transfer of training benefits from Korean to English characters. A lack of transfer would suggest a symbol-specific mechanism underlying training-related changes, e.g. better templates for recognition. Successful transfer would suggest a non-specific mechanism, such as a reduction in crowding, the inability to recognize objects in clutter. Method: Participants were 4 male and 5 female native Korean speakers (also proficient in English). In pre- and post-tests, we measured visual spans for single Korean letters (vowels and consonants), single Korean characters (composed of 2 or 3 letters), triplets of Korean characters, and English single letters and triplets. Training (1.5h * 4 days) consisted of repetitive visual-span measurements for Korean character triplets. Results: At baseline, the size of visual spans was negatively correlated (r=-0.96) with the perimetric complexity of the type of symbols, with visual-span size decreasing from single English letter = single Korean letter > single Korean character = English letter triplets > Korean character triplets. Training with Korean character triplets enlarged the visual spans for Korean single characters and character triplets. Such benefit also transferred to English letter triplets. Training-related enlargement was primarily due to a reduction of within-symbol crowding (for Korean characters) and between-symbol crowding (for triplets of Korean and English characters). Conclusion: The transfer of training benefits from Korean to English characters shows a non-symbol-specific mechanism underlying training. This suggests that the sensory limit underlying the visual span for script recognition is shared across languages. The generalization of training across types of symbols enhances the clinical potential of this training for individuals with central-field loss.
Meeting abstract presented at VSS 2017