Dr. Schweitzer is Director, HEAR4-U International and Technical Consultant for IMH Corporation.
In a previous post, Dr. Schweitzer discussed accuracy and reliability of hearing threshold measurements, describing the inaccuracies and problems that exist with current methods. That hearing aids are fitted based on poor baseline information has led to the introduction of other tests and procedures to better determine in situ hearing aid performance. That we have not yet achieved desired results is explained further in this post, looking at the evolution of electroacoustic measures of hearing aids, and what they provide. WJS, Editor
Accuracy and Precision
Tests of hearing aid performance have relied properly on engineers and their well-developed desire for both accuracy and precision. Control of the level and frequency are specified in standards, such as ANSI S3.22-2014. These provide precise control tolerances for small deviations of both. Clinicians with well-equipped offices are able to conduct tests similar to those required of manufacturers. But it was long understood that there were multiple discrepancies between the coupling and impedance of an individual’s ears and those properties in controlled standard test device, such as a 2cc ear simulator. This naturally led to the development of instrumentation that produced in situ versions of these tests, and the proliferation of Real Ear Measures (REMs), and new standardization protocols for insertion gain and outputs (ANSI S3.46-2013). However, while undeniably important to the progress of clinical management of hearing aids, a case for looking past REMs can be made.
Looking Past REMs
First, it should be evident that measuring the sound pressure level at the tympanic membrane is still a measure of physical acoustics, not perceptual experience. It introduces an individual’s ‘organic coupler,’ which, to be sure, is preferable over a standardized machined one. Coupler-to-ear differences are unquestionably valuable to consider, whether by direct measure and calculation, or by average estimates, especially for children and low functioning hearing aid users. Some REM approaches may even incorporate body baffle effects, adding additional detail to what telecom engineers call the orthotelephonic response, i.e. the change of a sound pattern as it arrives at the eardrum (transfer function). To their credit, some systems also utilize recorded speech passages in an attempt to relate the audiogram to assumed audibility of those more relevant signals. But to extend any of these measures to preferred listening levels and how to best configure the various hearing aid parameters is still a curvy, and rather nebulous path. To apply pure tone audiograms with individual ear REMs of any signal type still requires numerous inferential leaps and assumptions.
The familiar flow pattern of the standard REM approach is illustrated in Figure 1. The wiggly arrows at the lower right indicate that attempts to use the prior measures to hit an uncertain “Target” in the listener’s mind is presumptive. The ‘target’ is a sound pattern that matches an individual’s personal preference for Comfortably Clear Listening levels (CLL) and a tonal bias for speech (complex signals) in normal (binaural in natural acoustic environmental space) conditions. Such settings presumably can only be known to the listener, although the advance of brain mapping instrumentation may someday close that gap. Even notable efforts by various manufacturers to forecast the presumed audibility of various speech components into insertion gain displays still rely on assumptions of audibility based on sound pressure levels at uncorrelated eardrum measurement points, and arguably still come up short of their intended use – to deliver a listener’s perceptual preferences into the basic settings of their hearing aids.
Numerous appeals to many of these arguments have been made in past years,1,2,3,4, but a clinically expedient and useful alternative has been lacking. One comprehensive approach did have moderate clinical utilization in Europe,5,6,7 but extensive equipment requirements and rapid changes in hearing aid properties blunted its appeal.
The next part of this paper will describe a simple, but robust, tablet application that introduces an easy way to move beyond Real Ear to a Real Hear method of hearing aid verification.
The introduction of REMs was an important and commendable step towards closing the gaps on hearing aid fittings and listener preferences. However, the assumption that proper verification of hearing aid fittings demands their use, deserves critical discussion. As with all progress, there comes a time to honestly examine the limitations of familiar procedures, and to consider what more can be done to advance the delivery of professional services. Whereas these comments began by contrasting ‘accuracy’ with ‘precision,’ the case will be made that in ‘validating’ hearing aid fittings, the contrast between message reception and acoustical signal delivery must also be considered.
- Cox, RM, and McDaniel, DM. (1989). Development of the Speech Intelligibility Rating (SIR) Test for Hearing Aid Comparisons”. J Speech Hear Res, 32, 347-352.
- Van Tasell DJ. Hearing loss, speech, and hearing aids. (1993) J Speech Hear Res. 136(2):228-44.
- Schweitzer, HC , Mortz, M & Vaughn, N (1999). Perhaps not by prescription, but by perception. High Performance Hearing Solutions (Hearing Review Supplement). 58-62.
- Schweitzer, HC & Donnelly, R (2013). Why it’s time to retire the audiogram (for hearing aid fittings). Hearing Health Matters.
- Schweitzer, HC & Haubold, J (2000). Fitting for an ‘Auditory Life’ (part 1). Hearing Review 7(9). 42-51,76.
- Haubold, J & Schweitzer, HC (2000). Closing the gaps on hearing aid acoustical satisfaction. Audiology Today 12(1). 18-19.
- Schweitzer, HC & Haubold, J (2000). Fitting for an auditory life. (part 2). Hearing Review 7(10). 68-88.