APE: Adaptive probit estimation of psychometric functions
- 256 Downloads
APE is a psychophysical procedure designed to estimate complete psychometric functions with maximum statistical efficiency in return for minimum subject labour. It is based on the classical Method of Constant Stimuli, but differs in that only certain stimuli from the whole set available are tested. APE selects those stimulus levels at which the most information concerning the psychometric function is to be gained, updating this selection in the light of recent response history.
The performance of APE is compared with that of the classical Method by Monte-Carlo testing and is shown to be superior in several respects. First, the standard errors obtained are all smaller (reliability is higher). Moreover, APE is shown to be considerably more tolerant of the necessarily arbitrary decisions concerning stimulus levels to be tested. In practice this would increase the accuracy of the estimates obtained.
APE has been used for a wide variety of tasks, and is available as a FORTRAN II computer subprogram from the authors.
KeywordsPsychometric Function Stimulus Level Constant Stimulus Stimulus Range Unstable Subject
Unable to display preview. Download preview PDF.
- Finney, D.J. (1971). Probit Analysis, 3rd ed. Cambridge University Press.Google Scholar
- Guilford, J. P. (1954). Psychometric Methods, 2nd ed. New York: McGraw-Hill.Google Scholar
- Taylor, M. M. & Creelman, C. D. (1967). PEST: efficient estimates on probability functions. Journal of the American Statistical Association, 41, 782–787.Google Scholar
- Wetherill, G. B. (1966). Sequential Methods in Statistics. London: Methuen.Google Scholar