This project has investigated the statistical problems involved in the measurement of experimental and clinical pain. (1) A study has been conducted to investigate the statistical technique used in deriving psychophysical measurements of pain. A report has been prepared for this work dealing with the interrelationship of sensory-decision-theory measures such as d' and Beta and nonparametric measurement indices, such as p(A), Hodo's percent bias and MacNicol's index of response bias, Beta. The investigation of this part of the work is completed. The report of this study was presented at the 4th World Congress on Pain in 1984. (2) A study of statistical quantification of the temporal characteristics of persistent, episodic pain such as migraine headache using a semi-Markov model has been investigated. An appropriate clinical data set on which to test this model is currently unavailable and no further work is planned in this area until such a data set is obtained.
