VEP measures of monocular attenuation in amblyopia are extremely reliable across sessions and predict perceptual responses
Poster Presentation: Sunday, May 18, 2025, 8:30 am – 12:30 pm, Pavilion
Session: Binocular Vision: Clinical, perception
Schedule of Events | Search Abstracts | Symposia | Talk Sessions | Poster Sessions
Kimberly Meier1 (), Mark Pettet2, Kristina Tarczy-Hornoch2, Geoffrey M Boynton2, Ione Fine2; 1University of Houston, 2University of Washington
Introduction: Binocular integration requires integrating information across the two eyes, while ensuring that the final percept is non-rivalrous. We examined binocular integration across non-rivalrous and rivalrous conditions, across both VEP and psychophysical measurements. Methods: Amblyopia and control observers (n=11 per group) viewed dichoptic gratings (orthogonal vs. same orientations, 2cpd, 7.5 Hz flicker) and reported perceived contrast via joystick while occipital VEP signals were recorded. Grating contrasts modulated independently in each eye (1/8 and 1/6 Hz). VEP and joystick responses were predicted using a model that captured attenuation (k), and a continuum between averaging and winner-take-all behavior (w), by finding the best-fitting values of k and w, using the equation (1-w)[mean(kA,F)] + (w)[max(kA,F)], where F=Fellow and A =Amblyopic eye. Results: Attenuation in the weaker eye (k) was similar across orthogonal and same orientation gratings for both VEP and joystick measurements. Amblyopic individuals had higher levels of attenuation than controls (p=0.030). In amblyopia, estimates of attenuation were highly consistent across the two visits, for both perceptual (r=0.77) and VEP (r=0.92) responses. This consistency across sessions and stimuli suggests that attenuation reflects a fundamental neural property that can be objectively and accurately assessed using VEP; offering a promising pathway for clinical assessment in populations unable to provide reliable behavioral responses. Measures of binocular integration (w) were similar across individuals with amblyopia and controls and did not differ for orthogonal vs. same orientation stimuli. Values of w were close to a mean rule for VEP responses (w=0.06), while perceptual responses were close to the max (w=0.66), suggesting that VEP responses reflect an earlier stage of binocular integration than joystick responses. Results suggest that a winner-take-all mechanism, that does not differ across individuals with amblyopia and controls, results in a moderate level of early attenuation having a disproportional effect on which eye dominates perception.
Acknowledgements: Knights Templar Eye Foundation, Research to Prevent Blindness