Abstract
Purpose :
We aimed to determine if Image Gradations (IG) can be preferred to Visual Acuity (VA) or Contrast Sensitivity (CS) measurements to differentiate the quality of vision obtained with various level of blur.
Methods :
We used an adaptive optics system to simulate different level of monochromatic aberration corresponding to different level of blur. Dynamic correction of subjects’ aberrations ensured that all subjects had the same images on their retinas. Ten subjects viewed 6 different combinations of monochromatic aberrations chosen for either being of different or identical RMS level (named from 1 to 6 and corresponding respectively to 0.25, 0.25, 0.34, 0.55, 0.55 and 1.03 µm RMS), as well as a control condition (without aberrations) through an artificial pupil of 4.5mm. We measured the high-contrast VA and 12c/deg CS. Subjects also scored the quality of the perceived images on a 5-items continuous grading scale for IG.
Results :
Intra-individual standard deviation (SD) was always smaller than inter-individual SD. Ratio between inter- and intra-individual variation was the highest for VA and lowest for CS (i.e. 2.17, 1.36 and 1.56 respectively for VA, CS and IG). Combinations 1 and 2 were never discriminated. Combinations 2 and 3 were always significantly different (Wilcoxon test, p=0.01, 0.01 and 0.02 for VA, CS and IG respectively). VA and CS were not able to provide a differentiation of the first and third combinations (p=0.44 and 0.24 respectively). On the contrary, IG allow such a differentiation (p=0.02). All other combinations were significantly different. To quantify the discrimination ability of the visual tests, we also calculated the ratio between the inter-individual SD and the difference between the minimum and maximum performance (i.e., averaged on the entire population, and across all the tested blur conditions excepted the control condition). It was respectively 28%, 26% and 12% for VA, CS and IG. Lower ratio means better discrimination ability.
Conclusions :
While VA had the best ability to discriminate subjects, IG seems more efficient than VA and CS to discriminate low levels of blur.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.