Richard Legras

Assessment of just-noticeable differences for refractive errors and spherical aberration using visual simulation.

Purpose. The aim of this study was to evaluate the threshold levels of aberration change that a typical reference eye is able to detect. Methods. The method involved simulation of the foveal vision of a typical eye in polychromatic light through optics affected by different levels of the various chosen monochromatic aberrations. The reference eye had the following monochromatic wavefront characteristics based on the aberrations of a population of young adults: no spherical defocus, astigmatism −0.37 D oriented at 0°, coma −0.17 D/mm oriented at 270°, and spherical aberration −0.12 D/mm2. Average amounts of longitudinal and transverse chromatic aberration were assumed, and allowance was made for the Stiles-Crawford effect. The pupil diameter of the simulated eye was kept fixed at 6 mm. Three observers each compared, 100 times, a simulated image as seen through the standard reference eye with a variant “aberrated” image. The varying parameter was the value of a chosen additional aberration affecting the variant image in the reference eye. The test was repeated for varying amounts of spherical defocus, astigmatic defocus, and spherical aberration. For each of these aberrations and each observer, the discrimination probability as a function of the aberration level in the variant image was determined. The just-noticeable difference in aberration (JNDA) was derived from each discrimination curve as the difference between the aberrations corresponding to discrimination probabilities of 75% and 25%. The JNDA values obtained were expressed in the form of root mean square (RMS) wavefront error thresholds. Results. It was found that 0.04 &mgr;m of RMS aberration should be considered as the threshold of just-noticeable image change, in good agreement with the Maréchal criterion. Conclusions. The results imply that in normal viewing conditions (e.g., a 3-mm pupil size), optical corrections should be in the range of ±0.15 D in sphere and cylinder from the target prescription if perceptible change in the quality of the perceived images is to be avoided. The design of conventional soft contact lenses of high negative power or positive power should aim to produce −0.07 D/mm2 of spherical aberration, with a tolerated interval between −0.15 to +0.01 D/mm2 for a 6-mm pupil size.

A Method for Simulation of Foveal Vision During Wear of Corrective Lenses

Purpose. The aim was to simulate the visual appearance of images viewed through corrective lenses having known, arbitrary types and amounts of monochromatic aberration, so that the visual effect of changing the design parameters of the lens could be explored. Methods. We first calculate the optical response of the eye and any corrective lens using a numerical model eye. We then use this response as a filter, which we convolve with a selected original (unaberrated) image, to obtain an initial simulated retinal image. This image is then deconvolved by a second filter, which is calculated as the optical response of the eye of the observer who views the final image displayed on a video monitor. The originality of our approach to visual simulation is to take the aberrational characteristics of the observer’s eye into account in the calculation. We validated our simulation by comparing images degraded by simulated dioptric blur with real defocused images seen through corresponding optical lenses. Results. When using a small (2.5 mm) pupil size and a “typical” observer wavefront aberration model, there was a close resemblance between optical and simulated blurs. Although it was not necessary to consider the measured aberrations of the subject when simulating vision with a small pupil size, this requirement could not be ignored when vision through a larger pupil was simulated. With a 5.7-mm pupil diameter, use of Shack-Hartmann measurements of the ocular aberrations of the individual observers rather than “typical” levels of aberrations for the entire population gave excellent agreement between the effects of simulated and real defocus blur in monochromatic and polychromatic light. A Bland-Altman analysis of the differences between matching simulated and real blurs for a 5.7-mm pupil in polychromatic light with the model including allowance for individual measured aberrations gave mean differences close to zero and 95% confidence limits of about ±0.25 D over a defocus range of −2.00 to +2.00 D. Conclusion. The simulation technique can be expected to be a useful tool to evaluate the potential performance of an eye that wears various designs of corrective lens.

Effect of induced anisometropia on binocular through-focus contrast sensitivity.

Purpose. The aim of this study was to characterize binocular through-focus function in white light and to investigate the effect of induced anisometropia on binocular depth of focus. Methods. The subjects viewed sine-wave gratings generated on a monitor through a modified Badal system that produced gradual changes in target vergence ranging from −4.00 to +2.00 D. Binocular through-focus contrast sensitivity curves were obtained at a spatial frequency of 11 cpd and for different levels of induced anisometropia. Subjective depths of focus were derived from the through-focus curves. Results. An induced anisometropia lower than 1.00 D led to a monomodal through-focus curve involving a single depth of focus, whereas with higher anisometropia, the curves became bimodal indicating a lack of performance at intermediate distance. Binocular thresholds predicted by the quadratic summation model from our monocular measurements were well correlated to our binocular measurements. Predictions allowed us to estimate optimum levels of induced anisometropia.