Intraindividual consistency on a cross-modality matching task

Abstract

Three observers matched tone bursts and flashes for stimulus magnitude. Matches were made by each observer during 14 sessions over a 2-month period. No intraindividual differences in power function exponents were found. Power function exponents for individuals seem to remain constant across many repetitions of this psychophysical task.Investigations of the stability of psychophysical power function exponents produced by an individual have inconclusive results. For example, Teghtsoonian and Teghtsoonian (1971), recording magnitude estimations of visual area, suggest that exponents from an individual on this task are not stable if the time period between testing is 24 h or longer. In contrast, Logue (1976) reports a correlation between the exponents produced by individuals in two sessions using magnitude estimation of loudness. McGill (1960) has shown that loudness functions for individuals can be reproduced rather well. In this study, subjects estimated loudness in two sessions by indicating the distance on a line that represented the loudness of a tone. Similarly, Luce and Mo (1965) present data which suggest a fairly high degree of stability in heaviness functions produced by magnitude estimation. Cross-modality matching studies suggest that individual observers can produce consistent exponents (Lilienthal & Dawson, 1976;Wanschura & Dawson, 1974). In each of these studies, observers were tested two to six times to evaluate the reliability of individual psychophysical function exponents. Although these studies often report a correlation between exponents produced by an individual, the extent of exponent stability is still questionable.Power function exponents for a single observer should remain fairly constant over time if exponent values are more closely related to sensory system properties than to non-sensory factors. Acceptance of individual power function exponents as accurate indications of sensory transduction requires that they be shown to endure within the individual over many sessions under constant conditions. The present study examined the performance of individuals on a cross-modality matching task. Fourteen power function exponents, obtained from data collected over a period of at least 2 months, were computed for each subject.

Consistency of individual exponents in cross-modal matching

“…Among factors on the stimulus side are variability in target values (Cain, 1977, demonstrated that, on the continuum of odorant concentration, large differences from the nominal level of the target may occur and are not ordinarily recognized by the experimenter), and differences in stimulus context in random presentation orders (Cross, 1973,and R. Teghtsoonian & M. Teghtsoonian, 1978 shown that the immediate context in which a stimulus occurs influences the judgment made); on the response side, a subject's choice of modulus (Engen & Ross, 1966) can influence the exponents, as may features of the response apparatus when the subject selects response values on some physical continuum. Walsh and Browman (1978) showed that, when care is taken to eliminate as many nonsensory sources of variability as possible, there is very little intraindividual exponent variability; unfortunately, their N of 3 does not permit us to estimate the interindividual variability under those circumstances. It remains a tenable hypothesis that very little of the variability in individual exponents is a reflection of enduring idiosyncratic characteristics, and that small individual differences in sensory systems may adequately account for that little.…”

Individual loudness functions determined from direct comparisons of loudness intervals

“…J. C. Stevens and Guirao (1964), using an estimationproduction combination, have shown that exponents obtained from individual subjects vary considerably. Even when direct numerical judgments are avoided, as in cross-modality matching experiments, individual differences still occur (Marks & Stevens, 1966;Rule & Markley, 1971;Walsh & Browman, 1978;Wanschura & Dawson, 1974) and sometimes are substantial. It would be interesting to determine whether the elimination of numerical judgments or concepts in the direct comparison of loudness intervals would reduce or eliminate large individual differences.…”

Stability of line-length estimates using the method of absolute magnitude estimation

“…Engeland and Dawson (1974) found subjective area and loudness with a l-week intersession interval to be stable (Kendall rank correlation, p < .01), as were cross-modality matches between these two continua. Logue (1976) reported individual exponents of loudness to be fairly stable (r = + .7 -50Ofo of variance) for up to 11 weeks, and Walsh and Browman (1978) reported stability over a 2-month period for a cross-modality matching procedure in vision and hearing. Hellman (1981) found individual exponent differences between tones of different frequencies to be stable for periods up to 6 weeks.…”