Derivation of an Index of Discrimination From Magnitude Estimation Ratings
Beverly J. Cowart Monell Chemical Senses Center and Thomas Jefferson Medical College
As Gescheider ( 1988) has noted, the study of individual differences in psychophysical scaling may provide insight into the relative roles of sensory, cognitive, and perceptual factors in psychophysical judgment. A better understanding of these differences, their stability, and the factors that affect individual scaling performance is also necessary to determine the utility of scaling tasks in clinical settings (i.e., in the diagnosis of abnormality).
Experimental studies of individual differences have largely focused on differences in the slopes (exponents) of psychophysical functions. Although most studies indicate that individual differences in exponents are substantial and stable and tend to generalize across modalities, there are some conflicting results (see Gescheider, 1988). This chapter describes relationships among individual exponents for taste stimuli representing each of the four basic taste qualities, which were presented in a single session, as well as test-retest correlations among exponents for each taste in a subsample of subjects who were retested approximately a year later. In addition, however, an alternative analysis of individual scaling responses, which provides a measure of the reliability of repeated intensity judgments across a stimulus range and, thus, of discriminative performance, is also presented. The relationship (or lack thereof) between differential sensitivity/resolving power and the steepness of the psychophysical function has provoked strenuous arguments (e.g., Stevens, 1961; Teghtsoonian, 1971). Therefore, a major focus of this chapter is an exploration of relationships between the extent