Take the 2-minute tour ×
Cognitive Sciences Stack Exchange is a question and answer site for practitioners, researchers, and students in cognitive science, psychology, neuroscience, and psychiatry. It's 100% free, no registration required.

The reason I ask is that I often hear of alternate measures of response bias besides β, but have never heard of an alternate measure of detection sensitivity to d'.

If there is no other measure of detection sensitivity, is that a matter of mathematical necessity?

share|improve this question
add comment

1 Answer 1

up vote 4 down vote accepted

Yes, there are several alternatives to d'.

First of all, why do we need alternatives to d'?

  1. d' assumes the distribution of internal representation of signal and noise are Gaussian. This assumption is common, but may not be true, and a measure that does not make this assumption is more robust.

  2. There are technical problems computing d' when the Hit Rate or the False Alarm Rate are 0 or 1 (d' becomes inf or -inf).

Proposed alternatives to d' include:

  1. Area under the ROC (Green & Swets, 1966)
  2. A' (Pollak & Norman, 1964), A'' (Smith, 1995) which is a correction of A', and A (Zhang & Mueller, 2005) which is a correction of A''.
  3. S' (Balakrishnan, 1998)

Balakrishnan, J. D. (1998). Some more sensitive measures of sensitivity and response bias. Psychological Methods, 3(1), 68.

Zhang, J., & Mueller, S. T. (2005). A note on ROC analysis and non-parametric estimate of sensitivity. Psychometrika, 70(1), 203-212.

Smith, W. D. (1995). Clarification of sensitivity measure A'. Journal of Mathematical Psychology, 39(1), 82-89.

Pollack, I., & Norman, D. A. (1964). A non-parametric analysis of recognition experiments. Psychonomic Science.

Green, D. M., & Swets, J. A. (1966). Signal detection theory and psychophysics (Vol. 1). New York: Wiley.

share|improve this answer
add comment

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.