Training Novices to Discriminate Retinal Diseases Using Perceptual Learning

Abstract

Perceptual learning refers to the enhanced ability of users to pick up information from a class of stimuli and is achieved through rapid, repeated stimulus exposure with feedback. Classifying optical coherence tomography images of age-related macular degeneration (AMD) scans as either the wet or dry subtype is an extremely difficult perceptual task that typically requires years of training. The purpose of this study was to develop techniques to quickly train users to classify AMD retinal scan images as wet or dry. To determine whether gamification techniques might aid perceptual learning, half of the participants received a gamified feedback during training while the other half received informational feedback only. Results demonstrated significant improvement in AMD image classification for both the gamified and non-gamified groups, even on retinal scans never encountered during training. Participants in the gamified group, however, had higher levels of intrinsic motivation for the task.