Psycho… what? Musings on hearing vs understanding speech

The following Hearing View was written by my Hearinghealthmatters colleague Dr. Holly Hosford-Dunn for the blog that she and Sharon K. Hopkins publish at their practice Tucson Audiologists/TAI, Inc. With Holly’s permission, I am re-publishing her thought-provoking essay on Hearing Views to give it the larger audience that it deserves. It also serves as an insightful contribution to my blog’s continuing series of Hearing Views related to the current proliferation of hearing aids being sold directly to consumers over the Internet.

David H. Kirkwood, Editor, Hearing Views


By Holly Hosford-Dunn

Hearing aids are not the same as glasses… It’s never been just about hearing sound; it’s also about understanding sound and separating it from background noise. We can help with microprocessors. Without them it would have been impossible.”

Andrew Oxenham, PhD, Psychoacoustician, University of Minnesota1 


Truer words were never spoken, much less repeated over and over again, to often “deaf” ears. By that I mean that engineers{{1}}[[1]]I’m including engineers-as-customers as a special group of the general public.  I’m not taking issue with the few, specially-trained engineers who work with acousticians and psychoacousticians in hearing aid manufacturing.[[1]] and the general public sometimes share common opinions that are typically expressed along these lines:

• Hearing tests are so simple to perform … they can be done by a monkey or a machine… it’s just a bunch of sounds presented at different loudness levels.

• Hearing aids are just amplifiers connected to microphones and receivers. Total cost is probably less than a few dollars, so why the gigantic mark-up?

Back in the day, I did a PhD in something called Psychoacoustics, under the tutelage of Dr. Earl Schubert, a brilliant, Renaissance thinker and an early advocate of the radical idea that successful hearing aid designs are not just about technologically advanced engineering.

Instead, he posited, successful hearing aid design hinged on incorporating the Human Condition for hearing and understanding speech into the design equation. Why did he think this? Because Psychoacoustics, as its name suggests, is all about taking sounds and environments (the acoustic variables) and figuring out how organisms with ears (the psych variables) perceive, manage, and react to those acoustics. It’s complicated, especially when you mix noise in with “signals” such as speech.



Dr. Schubert further believed that our modeling of inner ear function was far from complete, as it virtually eliminated or ignored challenging variables such as “monaural phase.” That is, he was not convinced that we had proven beyond the shadow of a doubt that people with only one hearing ear had no auditory means of figuring out the location of a sound source in space, leaving them helpless to parse out speech in noisy situations. As I said, it’s complicated.

I’d say that every one of the Sciences, hard or soft, adopts a set of models to test data and infer or deduce conclusions. I’d also guess that every sophisticated model has one or more variables or error terms for the human condition and/or things-that-happen-that-can’t-be-accounted-for-by-the-model.

There’s the rub. We can measure and model till the cows come home, but the Human Condition is not especially predictable, in good part because it is adaptive and always learning and changing. It can’t just be stuck off in an error term in an equation or, if it is, you can’t expect the end result–in this case a hearing aid–to produce satisfaction for everyone in every condition. It is more complicated than that.



But, you may ask, does all this scientific soul searching have any current relevance? I believe it does. That’s because new and exciting technologies have pushed hearing aids forward recently, while at the same time resurrecting some of the simplistic thinking expressed by the bullets above:

If we make a hearing aid that does lots of things and fits easily into (most) ears, then everyone will be happy and everyone will hear well, even if they measured their hearing loss themselves without regard for their speech processing ability.

Can this be right? Probably not. Even the most advanced automatic easy-fit hearing aids produce different results on different patients, even when in-depth hearing evaluation precede the fittings and real-ear measures accompany the fittings. This is readily demonstrated through a variety of divergent approaches. For instance:

1)  Satisfaction surveys repeatedly show that only about 75% of people are “satisfied” with their hearing aids.

2)  Don Schum, PhD, a psychoacoustician and VP at Oticon, spends a huge amount of time and effort educating audiologists on the importance of “cognitive processing” to successful hearing aid fitting, as the perceptual extension of psychoacoustic fitting. He’s a smart and busy guy and he wouldn’t spend so much time on the cognitive variables unless he thought it was critically important to the hearing aid fitting process. As I keep saying, it’s complicated.

In our practice, no one yet has managed to come in, pick up their new hearing aids, stick them in their ears, walk out, and never come back because they’re doing so well and are 100% satisfied. Maybe that happens in other practices, but we have yet to come across reports to that effect.

What we do see is new patients who were fitted with marvelous technology elsewhere, yet are bitterly disappointed because they received little or no pre-fit counseling and absolutely no post-fit follow-up.

When they stumble into our practice, shell-shocked over what they’ve spent and what they think they got for it, we spend a fair amount of time listening. So far, we’ve managed to “cure” their ills and convert them into satisfied hearing aid users–but not without lots of direct patient interaction.

Back to those great hearing aid advances and the simplistic assumptions: One can scarcely open one’s mailbox, computer, or newspaper without receiving offers to purchase hearing aids online, through insurance, whatever, at low prices. Look closely. The offers and the prices make no mention of personalizing the fitting in any way, much less providing instruction or expectations for using the hearing aids to improve speech understanding in challenging situations.

In short, the engineering is good, often gre3at, but the psychoacoustics and cognitive variables are too basic. That pesky Human Condition got lost in the packaging. That’s too bad, because somebody’s going to pay for ignoring human variability and adaptability. And that somebody is going to be any patient who pays for technology but doesn’t achieve the speech understanding they thought they would.

1. Oxenham AJ & Simonson AM: Masking release for low- and high-pass-filtered speech in the presence of noise and single-talker interference. J Acoust Soc Am 2009;125(1).


Holly Hosford-Dunn, PhD, who is in private practice in Tucson, has had a long and varied career in audiology and hearing science. After earning her doctorate from Stanford University, she did post-doctorate study at the Max Planck Institute (Germany) and the Eaton-Peabody Auditory Physiology Lab (Boston), then became director of the Stanford University Audiology Clinic. Entering private practice, she developed multi-office private practices in Arizona. She has authored/edited numerous text books, chapters, journals, and articles and has taught Marketing and Practice Management in a variety of academic settings. She is also editor of and of its blog Hearing Economics.