Could tolerance to blur predict inter-individual variations of neural contrast sensitivity and/or loss of contrast sensitivity due to defocus or astigmatism?

Abstract

Purpose : Existing models of contrast sensitivity (CS) are not able to predict the inter-individual variations observed in the neural CS and the effect of an optical blur. The purpose of this study is to test the hypothesis that these variations could be predicted by a measurement of tolerance to blur.

Methods : We measured, using an adaptive optics device (crx1^TM) and through an artificial pupil of 4.5 mm size, CS (4-alternative forced choice sine-wave grating test) at 10 and 20 cycles per degree (cpd) in two conditions. In the first one, the observer’s wavefront aberrations and accommodation were dynamically corrected (perfect correction permitting to evaluate neural CS) whereas in the second, we also added a defocus (0.75 D) or an astigmatism (1D at 20°). While observer’s eye aberrations dynamically corrected, we measured CS as well as the tolerance to blur which was defined as the range of defocus for which the target (three 0.4 logMAR high-contrast letters) was still perceived acceptable (objectionable blur). Forty subjects aged between 20 and 60 years subdivided into four equal groups were involved. Their refractions ranged from -4 to +4D.

Results : Neural CS ranged from 1.03 to 2.38 u.log (2.10±0.27) and from 0.68 to 2.02 u.log (1.40±0.26 u.log) respectively at 10 and 20 c/deg. Loss of CS due to defocus ranged from 0.39 to 1.33 u.log (0.92±0.25) and from 0.21 to 1.29 u.log (0.54±0.20) respectively at 10 and 20 cpd. Loss of CS due to astigmastism ranged from 0.10 to 0.85 u.log (0.48±0.20) and from 0.05 to 1.01 u.log (0.44±0.20) respectively at 10 and 20 cpd. Tolerance to blur ranged from 0.37 to 1.32 D (1.04±0.20 D). Large inter-individual differences were observed.
Age and ametropia had no impact neither on the neural CS, nor on the effect of defocus or astigmatism on CS, nor on tolerance to blur. At 10 cpd, loss of CS due to defocus (r=0.75) was correlated to the neural CS but not the loss of astigmatism. At 20 cpd, loss of CS due to defocus (r=0.75) and astigmatism (r=0.73) were correlated to the neural CS. Tolerance to blur was not correlated to the neural CS or to the loss of CS due to defocus or astigmatism.

Conclusions : Large inter-individual variations of neural CS or loss of CS due to defocus or astigmatism cannot be predicted by tolerance to blur. However, the impact of defocus on CS seems to be linked to the neural CS.

This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.