Signal detection theory procedures are not equivalent when thermal stimuli are judged.

Publisher Website
Google Scholar
Abstract
Asked each of 4 Ss to discriminate between 2 thermal intensities differing by 25 mcal/sec/cm2. A total of 34 intensity pairs, ranging from detection to extremely noxious intensities (0-25 to 400-425 mcal/sec/cm2) at 2 a priori probabilities, were judged. Signal detection theory analysis revealed that 1-interval binary decision and 2-interval forced-choice judgments yielded higher values of sensory sensitivity, d' (p < .01, in both instances), than either the 1-interval confidence rating or the 1-interval sensory-magnitude rating procedure. The absence of a significant Methods * Intensity interaction demonstrated that the differences found among methods were the same for detection of warmth as for the discrimination of painful intensities. Analysis of the relationship between 1- and 2-interval procedures suggests that a variable criterion, rather than a noisy stored memory for the stimulus, is responsible for the lower values of d' obtained with the rating procedures. It is concluded that for the study of pain a combination of 1-interval sensory-magnitude estimation and 2-interval forced choice would be the most effective procedure. (18 ref.) (PsycINFO Database Record (c) 2006 APA, all rights reserved)

This publication has 0 references indexed in Scilit: