Threshold estimation in two-alternative forced-choice (2AFC) tasks: the Spearman-Kärber method.

Area of Science:

• Psychophysics
• Perception Science
• Statistical Modeling

Background:

• Two-alternative forced-choice (2AFC) tasks are common in psychophysics.
• Estimating thresholds and difference limens is crucial for understanding sensory perception.
• Existing methods like probit analysis often require assumptions about the psychometric function's shape.

Purpose of the Study:

• To introduce and evaluate the Spearman-Kärber method as a non-parametric estimator for difference limens.
• To compare the performance of the Spearman-Kärber method against traditional probit analysis.
• To assess the accuracy of the Spearman-Kärber method for estimating various parameters of the psychometric function.

Main Methods:

• Application of the Spearman-Kärber method to estimate difference limens in 2AFC tasks.
• Comparison with probit analysis, a standard parametric method.
• Evaluation of accuracy for estimating mean thresholds and dispersion.

Main Results:

• The Spearman-Kärber method provides a simple estimator for the difference limen and its standard error.
• This non-parametric approach does not require assumptions about the psychometric function's underlying shape.
• The method is highly accurate for estimating the mean and dispersion of the psychometric function.

Conclusions:

• The Spearman-Kärber method is a valuable, non-parametric tool for psychophysical research.
• It offers accurate estimation of absolute and difference thresholds.
• While excellent for mean and dispersion, it's less optimal for percentile-based parameters compared to other methods.