Chance-level guessing forces threshold criteria above 50% to reduce false positives
Evidence
In some cases the value of the psychophysical function does not approach 0% at the lower end of the scale. This happens, for example, if the difference threshold is measured with the 2AFC method, where the chance of guessing correctly is 50% when the difference between the signal and the reference is below the threshold. In this case, the value of the threshold may be selected to be 71% or 75% of the maximum of the function. — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 914
Categorization
Category: Definitional Content
- XREF: Directly relevant to signal detection theory and the criterion/bias distinction in psychophysics.
Related
- Forced-choice guessing floors push threshold criterion above 50 percent detection (similarity 0.89)
- Thresholds are statistical detection probabilities, not absolute cutoffs (similarity 0.85)
- Thresholds are measured by presenting many stimulus levels and fitting a psychometric function (similarity 0.84)
- Response bias can contaminate any psychophysical threshold estimate (similarity 0.80)
- Accuracy and function-shape information trade off against speed in threshold measurement (similarity 0.80)
- Method-of-limits thresholds are distorted by subjects’ response tendencies (similarity 0.80)
- Absolute hearing threshold is a statistical construct, defined at 50% detection probability. (similarity 0.80)
- Thresholds can be measured by tracing stimulus change until perception flips (similarity 0.79)
- Forced-choice testing cancels listener response criterion effects in threshold estimation (similarity 0.79)
- Threshold estimates reflect both ear sensitivity and listener response bias (similarity 0.78)
- Absolute threshold is the minimum detectable sound level without external noise (similarity 0.78)