No abstract

Three models concerned with the behavior of subjects estimating two successive durations are proposed. The accepted model assumes that the subjective total duration (sum of first and second durations) and the second duration are each accumulated in a separate sensory register. In a durationmatching experiment, for instance, the difference between the contents of the two registers is matched to the content of the second register. This model is accepted because (a) it does not include any memory, thereby eliminating certain difficulties connected with coding and storing of duration, (b) it copes with characteristic features of duration discrimination, and (c) it can satisfactorily explain data obtained in four scaling experiments, namely, magnitude estimation, matching, halving, and doubling of 10 durations between 1.3 and 20 sec. As a by-product, Stevens' power law is uniquely derived, and exponents are computed from matching data, thus eliminating the subjects' numerical behavior. The model also accounts for the time-order error for time.

No abstract

Six women and six men reproduced ten time intervals varying in logarithmic steps between 1.3 and 20 s. The durations were indicated by white noise of 10, 25, 40 and 55 dB SL, different sound intensities in different sessions. It was found that (i) greater sound intensity entails shorter reproductions, and (ii) reproductions by male observers are shorter than those by female, although for both (i) and (ii) there is an interaction with the standard durations. The data were treated in accordance with the "parallel-clock model" (Eisler, 1975), whereby the parameters of the psychophysical power function are determined from duration reproduction data. As in previous experiments (Eisler, 1975), the data showed a break in the function entailing two segments. The effect of sound intensity could be attributed to the exponent, which was lower for stronger noise, and the effect of sex to the weight coefficient of the upper relative to the lower segment of the psychophysical function, the coefficient being lower for men.

Eisler, H. Empirical test of a model relating magnitude and category scales. Scand. J. Psychol., 1962, 3, 88–96.—The function K =αlog(φ+ q/k) +β seems to describe the relation between category scale values K and subjective magnitudes φ. The additive constant q/k is obtained from the Sds of the magnitude estimates. The model was empirically confirmed for the loudness and softness of white noise scaled by the methods of magnitude estimation and category rating.