High false positive rates in common sensory threshold tests.

Area of Science:

• Psychology
• Neuroscience
• Psychophysics

Background:

• Sensory threshold variability is common in healthy populations.
• Genetics and daily fluctuations contribute to this variability.
• False positives are an underappreciated source of variability in threshold testing.

Purpose of the Study:

• To simulate and analyze false positive rates in various sensory threshold testing methods.
• To compare the performance of ascending and staircase methods regarding false positives.
• To provide guidance for researchers on selecting appropriate threshold testing methodologies.

Main Methods:

• Simulated responses using random number generation for different "stopping rules" in threshold methods.
• Evaluated ascending 2-alternative forced choice (AFC) and 3-AFC methods.
• Analyzed staircase 2-AFC and 3-AFC methods with varying parameters (e.g., reversals, correct/incorrect responses).

Main Results:

• Developed formulas for false positive rates in ascending methods.
• Generated performance curves for staircase methods.
• Staircase methods generally exhibited lower false positive rates but were more sensitive to the number of presentations.

Conclusions:

• High false positive rates are a concern across tested threshold methods.
• Researchers should consider multiple tests per individual or methods that correct for false positives.
• Fitting a logistic curve to response data is a potential strategy for error correction.