Trail-Traced Threshold Test (T4) With a Weighted Binomial Distribution for a Psychophysical Test

Area of Science:

• Ophthalmology
• Psychophysics
• Computational Statistics

Background:

• Clinical visual field testing uses psychophysical methods to assess differential light sensitivity (DLS).
• Current thresholding algorithms struggle with efficiency, accuracy, and stability.
• There is a need for improved methods in visual field testing.

Purpose of the Study:

• To introduce and evaluate a novel Bayesian perimetric threshold method, the Trail-Traced Threshold Test (T4).
• To compare the performance of T4 against established methods like ZEST and FT.
• To assess T4's robustness and potential for broader psychophysical applications.

Main Methods:

• Development of the Trail-Traced Threshold Test (T4), a Bayesian approach.
• Comparison of T4 with Zippy Estimation by Sequential Testing (ZEST) and FT.
• Evaluation metrics included presentation number, mean absolute difference (MAD), and measurement variability.

Main Results:

• T4 demonstrated superior accuracy and precision compared to ZEST, FT, and SWeLZ.
• T4 achieved these improvements with significantly fewer stimulus presentations.
• T4 showed greater robustness to initial threshold estimation errors.

Conclusions:

• The Trail-Traced Threshold Test (T4) offers significant improvements in visual field testing accuracy, variability, and efficiency.
• T4's robustness makes it advantageous for subjective psychophysical methods like clinical perimetry.
• T4 has potential applications beyond visual field testing in other psychophysical domains.