A Laboratory For the Study of Objective Measurement of Visual Acuity

Southern Methodist University, Dallas TX

Investigators

Abstract

0100820 Davila Visual acuity (VA) is a measure of how well the visual system can resolve closely spaced objects. The Snellen eye chart is the most common method of measuring VA in adults however it is useless when measuring VA in infants and other non-verbal patients. Preferential looking, optokinetic nystagmus, and photoscreening have been used to assess VA in infants, however all of these methods have been found to have their shortcomings. Diseases such as amblyopia, which occurs in 2-3% of the population, if not diagnosed by age 6 can lead to permanent visual loss, and this can lead to learning and behavioral difficulties. Standard pre-school visual screening programs can fail to diagnose ocular disease in as any as 2% of the population. The visual evoked potential (VEP), an electrical potential generated in the visual cortex of the brain in response to a visual stimulus, has been used to measure VA in infants. This is done by determining the maximum spatial frequency in the stimulus which elicits a detectable VEP. However, this method does not yield very repeatable measures and has a higher variability than psychophysical measures of VA. The PI's group has developed VEP detectors which are capable of detecting very low-level VEP's at much higher stimulus spatial frequencies than other detectors; nevertheless, long measurement times are still required. The goal of this proposal is to obtain the laboratory equipment necessary to study objective measurement of visual acuity based on a new model of the visual system. This model, if validated, will enable VA to be measured using the VEP by simply computing its temporal frequency spectrum (which is much faster than existing swept spatial frequency methods). The spacing of photoreceptors in the eye is not a good predictor of limits on spatial resolution. The average spacing of cones on the foveal of the retina would actually predict a much higher visual resolution limit than the often quoted value of 60 cycles/deg. The proposed model is based on the premise that the eyes are constantly in motion and hence the luminance falling on any given photoreceptor in the retina is constantly changing with time. The main hypothesizes is that resolution in foveal vision is determined entirely by the temporal dynamics of the visual system. The model for VA consists of a photoreceptor, a ban of temporal bandpass filters, and a detector at the output of the filter bank. The temporal bandpass filters have the same effect as the spatial frequency filters which have been used to model the visual system. A second hypothesis associates the temporal filter-based model for VA with the VEP. A number of psychophysical and electrophysiologic experiments are proposed which are designed to validate the two hypotheses. One experiment is based on the classical psychophysical adaptation experiments of Blakemore and Campbell but rather than adapting to a constant spatial frequency, we propose to adapt to a constant temporal frequency. This will be done by stabilizing the drift rate of the stimulus grating with respect to eye motion. By measuring the contrast sensitivity function (CSF) at two different drift rates and a fixed spatial frequency, the temporal filter model would predict a result which the spatial frequency channel model would completely fail to explain. Experiments co paring psychophysical and electrophysiologic spatiotemporal CSF's will also be performed which are designed to insure that the psychophysical results carry over to the electrophysiologic domain. The equipment needed for this research consists of a high accuracy dual Perkinje image eye tracker, an optical image stabilizer, a high scan rate, fast phosphor video monitor, and a psychophysical graphics generation system. The PI's institution has agreed to cost-share 30% of total project costs. This research will have important theoretical ramifications in understanding the basic processes in human vision. In addition, it will also impact the quality of health care by enabling an accurate method of electrophysiologically measuring VA in infants and non-verbal patients.