ON-OFF Pathway Function in Amblyopia

Abstract

"Purpose To understand the effect of amblyopia on ON and OFF pathway asymmetries. This pilot study used a novel, eye-tracking, saccadic-based perimetric test.

Methods 8 eyes of 4 amblyopic subjects (mean age 24 ± 2.4 years) and one eye of 4 age-matched control subjects (mean age: 27 ± 3.6 years) were tested using a novel perimetric test. We tested 4 Michelson contrast levels ranging from 15 to 5% at 3 annular eccentricities: 5- 10, 11-20, 21-30 degrees from fixation (2.5-degree radius circle). Each contrast level was comprised of 579 trials. Each test location was repeated 3 times for both light and dark stimuli, with 6 repeats in each of two blind spot positions. Stimuli, light or dark squares which varied in size as a function of eccentricity, were presented on a binary noise background. The angular difference between the target location and eye position following a saccadic movement to the target location and reaction time were measured for every trial. Percentage of undetected targets, the angular difference between the target location and eye position following a saccadic movement to the target location, and reaction time were measured for every trial. Hardware consisted of a head mounted display equipped with an eye tracker (HTC VIVE embedded Tobii) with a refresh rate of 90 Hz, a max luminance of 110 cd/m2, and a horizontal field of view of 60 degrees. Unity (version 2017) software was used to generate the stimuli.

Results At low contrasts (less than 10%), the amblyopic eyes of patients have a greater percent error rate (12.12 ± 17.56 %, p = 3.2 x 10-17, Wilcoxon tests for all statistical comparisons), detect fewer targets (6.98 ± 16.28 %, p = 8.5 x 10-8) and have longer reaction times (101.2 ± 193.8 msec, p = 1.13 x 10-11) when compared to fellow eyes. However, the fellow eyes of amblyopic subjects have greater percent error rate (8.03 ± 8.64 %, p = 0.0013), detect fewer targets (8.68 ± 7.2 %, p = 2.9 x 10-4) and have similar reaction times (29.8 ± 110.6 msec, p = 0.811) to the control subjects. At a low contrast (6%), the difference between amblyopic and fellow eyes across eccentricities was significantly larger for light than dark stimuli in percentage of errors (5.37 ± 10.43%, p = 0.031), undetected targets (4.91 ± 6.5%, p = 0.001) and reaction time (44.2 ± 96.7 msec, p=0.026). This difference was not apparent at 5% contrast due to the high rate of errors for both light and dark stimuli.

Conclusion These results provide further support to previous work indicating that amblyopia affects ON more than OFF pathways. However, given our very limited sample size, further evaluation is necessary. The use of the perimetric test in this study, along with continuing advancements, may enable this work to continue in a clinical setting."