Six just-noticeable differences in retinal image quality in 1 line of visual acuity: toward quantification of happy versus unhappy patients with 20/20 acuity.

PURPOSE: To determine the number of just-noticeable differences in wavefront blur necessary to induce a 1-line loss of corrected distance visual acuity (CDVA).
SETTING: Visual Optics Institute, College of Optometry, University of Houston, Houston, Texas, USA.
DESIGN: Evidence-based manuscript.
METHODS: The 3.0 mm wavefront error of a well-corrected average eye was scaled to yield 9 small steps of blur quantified in units of log visual Strehl (logVS). For each logVS value, 10 unique 3-line acuity charts were generated. Using a temporal forced-choice paradigm, subjects compared each test chart to a reference acuity chart and indicated which chart was blurrier. The difference between 80% and 50% on the psychometric function defined a just-noticeable difference. The CDVA was measured up to fifth-letter miss for several aberrated logMAR charts for 6 logVS values. The number of just-noticeable differences necessary to lose 1 line of acuity was defined as the change in logVS necessary to lose 1 line of acuity divided by the 1 just-noticeable difference in logVS.
RESULTS: Linear regression showed that logVS = -2.98 × (logMAR acuity) - 0.31 (R(2) = 0.961). The mean just-noticeable difference was 0.049 logVS ± 0.012 (SD), resulting in a mean of 6.1 just-noticeable differences per line of logMAR acuity.
CONCLUSIONS: The retinal image quality metric logVS was highly correlated with logMAR acuity. The 6 just-noticeable differences in logVS before 1 line of acuity was lost may provide an objective explanation for the distinction between patients with 20/20 CDVA who are happy and patients with 20/20 CDVA who are unhappy and other aberration-related clinical complaints when acuity is near normal.