Purchase this article with an account.
Soheil Mohammadpour-Doustkouhi, Philip RK Turnbull, Steven C Dakin; Effect of refractive error on optokinetic nystagmus. Invest. Ophthalmol. Vis. Sci. 2019;60(9):1816. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Subjective refraction is the gold standard procedure for prescribing spectacles but relies on patients being willing/able to articulate subjective judgments of the appearance of visual stimuli. In order to assess the feasibility of an alternative approach - that quantifies spherical equivalent refractive error (RE) using patients involuntary eye movements - we measured how RE changes optokinetic nystagmus (OKN) in response to moving patterns.
We used an infrared eye tracker (Eyelink 1000+) to measure the eye movements of 20 participants viewing horizontally drifting two-dimensional noise patterns (128, 2-second trials, 6 minutes). The spatial frequency (SF) of these patterns was restricted to a range centering on 3.8, 6.0, 9.6 and 15.2 c/deg. RE of our participants was +0.25 to -3.00 dioptre (D) measured using an open-field autorefractor. To allow us to compare eye tracking to behavioral approaches we had participants report the direction of drift during eye tracking. We then scored subjective responses and OKN signals as correct or incorrect identifiers of stimulus direction, and fit psychometric functions to estimate threshold. We correlated subjective- and OKN-thresholds to objective RE, and bootstrapped data to derive predicted RE with associated confidence (our outcome measures).
The correlation between RE and OKN thresholds, and between RE and subjective thresholds (both measured with drifting stimuli) were similar: r=0.872, p=0.000001 and r=0.830, p=0.000006 respectively. Note that these values are comparable to the correlation between RE and chart-based (static) acuity measures (r=0.871, p=0.000001).The mean deviation of OKN-based estimates of RE from the objective RE was 0.35D (95% confidence interval: 0.02 to 0.81D) and the mean deviation of the subjective judgment-based RE from objective RE was 0.41D (95% confidence interval: 0.005 to 0.89D).
Measuring refractive error using objective measures of OKN is feasible and potentially offers several advantages over subjective refraction that may make it particularly suitable for use with children. We are currently exploring the use of efficient psychophysical methods to maximize the efficiency of this procedure.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.
This PDF is available to Subscribers Only