Mechanisms of Visual Perceptual Learning: From Artificial to Natural Stimuli

Visual perceptual learning (VPL) is a fascinating area of research that explores how our brains improve visual performance through training. This study aims to uncover the underlying mechanisms of VPL for both artificial and natural stimuli, with the ultimate goal of developing effective rehabilitation programs for individuals with damaged or deteriorating vision. The research focuses on two key aspects: the specificity of VPL to trained features and locations, and its generalizability to untrained stimuli encountered in everyday life.

The study employs Gabor patches, a type of artificial stimulus, to investigate the basic mechanisms of VPL specificity. According to prevailing theories, early visual processing (0-150ms after stimulus onset) involves feedforward signals, while late processing (150-300ms) involves recurrent processing. The research team will test whether early or late processing plays a more significant role in VPL specificity (Hypothesis 1) and whether excitatory or inhibitory signals are involved (Hypothesis 2). Two innovative methods will be used: backward masking to disrupt late processing, and a novel Rhythmic Synchronization Orientation Decoding Change (RSDC) method that analyzes EEG data to track changes in rhythmic synchronization bands after VPL training.

In the second phase of the study, researchers will examine how VPL applies to natural scenes (NS). Preliminary results suggest that VPL for dominant orientations in NS may generalize to other orientations, potentially due to higher-order statistics in natural images that create correlations between different orientation and spatial frequency channels. This finding could have significant implications for developing more effective vision rehabilitation programs that transfer learning to real-world visual tasks.

The study will involve 400 adult participants (ages 18-60) with normal or corrected-to-normal vision. Exclusion criteria ensure participants don't have eye disorders, use certain medications, or have conditions that might interfere with testing. Using a parallel assignment design with single masking, the research team will measure changes in visual task performance over a 2-week training period as the primary outcome.

This comprehensive investigation, led by Brown University with support from the National Eye Institute, promises to advance our understanding of visual learning mechanisms. The findings could lead to breakthroughs in treating visual impairments by clarifying how training effects transfer from laboratory settings to real-world vision. The study's innovative combination of behavioral measures, EEG analysis, and examination of both artificial and natural stimuli makes it a unique contribution to vision science research.