Abstract
Purpose :
Virtual reality perimetry is an evolving diagnostic option for clinicians, yet there is little known on the effect of blur. As some of the devices utilize existing spectacles for distance correction that may have a bifocal or progressive lens design, this could impact the overall results of the test. Furthermore, from previous reports we would expect to see a modest decline in sensitivity with 3.00 diopters (D) of blur for a Size III stimulus and less effect with a Gaussian Blob. This study was aimed at determining the effect of blur in a virtual reality perimeter.
Methods :
A virtual reality (VR) headset driven perimeter (Olleyes, Inc. Summit, NJ) was modified to allow the choice of a Gaussian Blob (SD=0.25°) or a standard Size III stimulus. 12 control subjects aged 21-31 years-old were tested using the Normal Threshold 10-2 pattern with different levels of introduced blur for both stimuli (0.00D, +1.50D, +3.00D, +4.50D). For interleaving, the subjects were divided into four groups with either ascending or descending sequences of blur and either starting with the Size III stimulus or the Gaussian Blob. Mean sensitivities (MS) were calculated and a two-way ANOVA was generated.
Results :
The Gaussian Blob had a range in median MS from 32.8 decibels(dB) at 0 blur to 32.0dB at +4.50D. The Size III stimulus had a mean MS range from 32.5dB to 31.5dB. The ANOVA found that the effects of blur and stimulus were statistically significant (p<0.01) while there did not appear to be a significant interaction (p=0.85).
Conclusions :
The effects of blur and stimulus may be statistically significant, but the small effect may not have a clinical significance. The results here are unexpected and differ from what has been reported for bowl-style perimeters. A possible reason for the discrepancy could be a slight blur to the stimulus secondary to diffraction from the internal Fresnel lens in the VR headset. Another plausible cause could be the use of a mesopic background allowing for larger summation area that reduces the impact of blur.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.