Nonlinear distortion in hearing aids: the need for reevaluation of measurement philosophy and technique.

Conventional pure-tone tests of harmonic distortion are of dubious sufficiency in expressing the nonlinear distortion (NLD) characteristics of hearing aids. This paper reviews the strategies and problems of intermodulation distortion. Spearman rank order correlations for a group of hearing aids varied widely as a result of changes in the selected frequency and/or level of input signal. Also, the reference criteria for expressing distortion was observed to contribute to the inconsistency of rank order impression. A technique is described which incorporated synthetic vowels as test signals. Hearing aid outputs were studied on a spectrum analyzer for evidence of NLD. It is argued that efforts should be directed at devising tests for hearing aids which are directed at comprehensive NLD including both harmonic distortion and intermodulation distortion.