Auditory adaptation: notes on power and parabolic curve fitting across intensity: II

The classic Simultaneous Dichotic Loudness Balance (SDLB) procedure for the study of auditory adaptation was used by Weiler and Hood (1977) in a successful test of a model for predicting adaptation in db at intensities above the initial baseline values, using a linear approximation across levels. When Balzer et al (Brit. J. Audiol., 1984, 18, 49-50) fitted curves for simple baseline adaptation data from four studies based on SDLB data, they reported good fits for the Gompertz technique and for a parabolic formula, but did not test the relative goodness of fit for the linear formula, nor for power approximations as Stevens (1975) might have proposed. The present study makes use of the data used by Balzer et al but examines a linear fit as well as two approximations to Stevens' power law. The present finding that the linear approximation is the least efficient examined would suggest that the Weiler and Hood model should be re-examined to determine the degree to which the model can be improved by use of a curvilinear baseline fit. In addition, it might be noted that the curvilinear formulae, of necessity, involve two or more stages. It is the the current effort of this laboratory to determine whether this reflects a multifactor basis to the classic SDLB adaptation, as well as results from other adaptation techniques.