Abstract
Purpose :
To demonstrate computationally that the frequency-of-seeing (FOS) curves in patients with glaucoma can be modeled as a weighted sum of typical psychometric functions.
Methods :
The author prepared a set of psychometric functions that represent each threshold from <0 to 40dB, using zero function (<0dB) and cumulative gaussian function with the same slope (0~40dB) (Fig. 1A). Then, a weighted sum of these functions was calculated based on the following weights assigned to each threshold: λ (0<λ<1) for <0dB, (1-λ)φ(μ, σ) for 0~40dB, where φ(μ, σ) is the gaussian density function with mean μ and standard deviation σ (Fig. 1B).
Results :
The weighted sum resulted in a flat and shallow sigmoid curve, which was similar to the FOS curves at regions with glaucomatous damage (Fig. 1C). The decrease in maximum response probability corresponded to λ, and the steepness of slope was linked to σ. Furthermore, the luminance of inflection point agreed with μ.
Conclusions :
This result raises the possibility that a single FOS curve in patients with glaucoma may consist of multiple psychometric functions.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.