There is more to accommodation than simply maximizing retinal image contrast

Abstract

Purpose : We tested the hypothesis that accommodation operates as a contrast-maximizing negative feedback system.

Methods : Two experimental conditions were tested to determine whether blur of a target alone is what drives accommodation. In the first condition the subject viewed a target (monochromatic Maltese cross subtending 2° on a micro-display through a 4-mm artificial pupil) that was always imaged accurately on the retina while the Maltese cross itself was blurred by an amount that depended on the subject’s accommodative error (lead or lag). Thus the subject was rewarded with a less blurred target for correct accommodation, but punished by a blurrier target for incorrect accommodation. No vergence changes were present because the blurred target was always in focus on the retina. In the second condition, the target itself was always a clear Maltese cross, but the retinal image of the target was blurred by the subject’s accommodative error; thus changes in accommodation provided normal feedback from defocus blur. Three experiments were performed on nine subjects (ages 21-40) using these two conditions. In the first a stationary target was presented for 50 sec at −2D; in the second, the target stepped randomly from −2D to −1D or −3D during 50-sec trials; in the third, the target moved sinusoidally between −3D & −1D at 0.05, 0.1, 0.2 or 0.4 Hz. In all three experiments, astigmatism & higher-order aberrations were recorded & corrected with adaptive optics at 20 Hz.

Results : Accommodative error was similar for the 2 conditions for stationary targets, but standard deviations of accommodative responses were significantly smaller when there was feedback from vergence (second condition). Dynamic accommodative gain in the step & sinusoidal motion experiments was significantly greater when there was feedback from vergence. The figure is an example of results averaged over 6 trials for all 9 subjects at 0.2 Hz for the two conditions in the third experiment.

Conclusions : The visual system detects the sign of optical vergence even without monocular or optical cues. Recent models based on Stiles-Crawford effect & retinal blood vessels may explain how the eye extracts the sign & magnitude of defocus (Vohnsen, Biomed. Opt. Express 5,5, 1569, 2014; Lopez-Gil et al, ARVO 2016). Understanding these mechanisms of inferring optical vergence by retinal cone signals may help explain emmetropization & define new treatments for myopia control.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.