Abstract
Purpose :
Random Dot-Stereograms (RDS), used to test global stereopsis, have been proposed to be more sensitive in detecting stereo deficits in amblyopes, since local stereopsis tests (e.g. Wirt Circle) may exhibit monocular cues. These clinical tests only assess stereo acuity at a certain size, primarily at higher spatial frequencies (SFs) with small targets. Amblyopes are often diagnosed with a stereo deficit when assessed with these tests. It is unclear if the deficits occur across the entire spectrum of SFs, or only at high SFs, due to their reduced visual acuity (VA). We designed an RDS with disparity-defined optotypes to assess stereo in a range of SFs and contrast levels in normal and stereo deficient groups.
Methods :
We generated Sloan optotypes of various sizes (0.25-3.75cpd) with crossed disparity embedded in RDS. On each trial, a random optotype was presented for 1sec with a haploscope and LCD screens. Subjects (5 normal observers, 1 intermittent exotropia-IXT, 2 strabismic amblyopes) identified the letter. A staircase method was used to determine the minimum disparity needed to identify letters at each size. Stimuli were tested at 100%, 50% and 25% contrast. To simulate ocular imbalance, subjects also performed the task with an interocular contrast ratio (ICR) of 50% and 25%, with a lower contrast in the dominant eye.
Results :
The disparity-defined letter sensitivity function had a bandpass curve peaking at 1-2cpd and intersected the x-axis at a cut-off frequency of ~5cpd (disparity-defined VA, “dVA”=20/120). As the overall contrast decreased, the peak shifted downwards with a cut-off at lower SFs. However, only the middle and high SFs were affected, sparing the low SFs. Surprisingly, results for 50% and 25% ICR were similar to the outcome for binocular contrast reductions of 50% and 25%, respectively. The curve for the IXT subject with 100% contrast was even lower (dVA=20/300) than that of normal observers at 25% contrast, with greater loss at high SFs than low SFs. The strabismic observers were not able to perform the task even with the largest letters.
Conclusions :
We found that thresholds for identifying disparity-defined letters vary with the size of optotypes in a bandpass-shaped pattern. With a range of SFs, this test can help determine selective losses of stereo deficits at particular SFs, as well as identifying the high SF cut-off (dVA). Strabismic observers may need larger dots and longer duration.
This is a 2020 ARVO Annual Meeting abstract.