Multivariate adaptive testing applied to the prescriptive fitting of hearing aids

Abstract

Among the problems encountered in fitting of hearing aids are (i) measurements of performance are subject to significant random fluctuations; (ii) subjects typically adapt to the hearing aid resulting in gradual drifts in the optimum parameter values; (iii) variables interact, thereby presenting certain dangers if only one variable is manipulated at a time; and (iv) data obtained under carefully controlled conditions do not always correspond to performance in the field. The simplicial method [G. E. P. Box, Appl. Statist. 6, 81–101 (1957)] is a powerful adaptive technique designed for problems of this type and has been used as part of an ongoing study to develop a practical protocol for the prescriptive fitting of a wearable master hearing aid. Preliminary trials with five subjects showed superior performance for the fitted wearable master hearing aid than that obtained with the subjects' own aid under comparable conditions of speech in noise. [Research supported by PHS‐Contract No. NIH‐N01‐NS‐4‐2323.]