Hearing Aid Amplification, Psychoacoustics, and Perception of Ongoing Discourse: Interview with Wayne Staab, PhD
Douglas L. Beck, AuD, spoke with Dr. Staab about hearing aid amplification, mid-frequency amplification, frequency lowering, psychoacoustics, perception of ongoing discourse and more.
Douglas L. Beck (DLB): Hi, Wayne. It's great to speak with you and it's been a while since we last spoke! For the younger audiologists, I'd like to review a little of your career highlights. For example, I know you were a professor at University of North Dakota and Michigan State University back in the early days. Further, you were with Telex and Audiotone and when Audiotone was sold to Dahlberg, you worked for them as executive VP and then you started your own consulting company. I should also note you were one of the founders of the American Auditory Society in 1973 and you and the Board of the AAS created the peer-reviewed journal, Ear & Hearing. So it's been a pretty good ride and you've been involved in many creative and game-changing issues and technologies.
Staab: Yes, it really has been fun. I should note that I do enjoy being on the cutting edge of technology and new protocols, but once it's gone main stream, I get bored and move on.
DLB: And Wayne, one thing I want to clear up immediately…When was the first wearable digital hearing aid, and who made it?
Staab: Glad you asked! The first wearable digital was made by Audiotone, and was produced in 1983. Many people think it occurred a decade later, but Audiotone won that race, hands down. Further, to be clear, the first deep canal was the Philips XP peri-tympanic, deep canal…and that one required a whole new impression taking technique.
DLB: I recall taking your class on the XP and I'll never forget the trembling hands of the students (mine included) as we practiced taking ear impressions literally, all the way to the tympanic membrane!
Staab: That was an exciting time…and let me say you did a heck of a job taking my deep impressions! In fact, you'll be happy to know my ear canal finally healed a few weeks back!
DLB: Thanks, Wayne. And so moving on, let's talk about psychoacoustics and its significance with regard to modern hearing aid amplification and mid-frequency amplification.
Staab: That's one of my favorite topics, Doug. There are some key issues related to psychoacoustics and hearing aid amplification. When someone presents with a precipitous, sensorineural high frequency (i.e., ski-slope) hearing loss, it's not always necessary to give them back all of those sounds in the 6000 to 8000 Hz range.
In fact Margo Skinner wrote about this in detail in the 1980s and noted there's substantial redundancy in speech, and if one hears through the mid-tones pretty well, that it may often be sufficient. Of course, it's better to hear all the sounds, but psychoacoustics tells us that much of what you need to understand speech is below 4000 Hz. For example, think about your phone. Most people do fine and hear 100 percent of the phone conversation—yet their phone likely doesn't deliver any information above 3000/4000 Hz. Further, think about cochlear implant patients who are also likely to perform well with speech recognition in quiet—and again, it's not likely they perceive acoustic information above 3000/4000 Hz!
DLB: Would you please review the data relating to the long term average speech spectrum?
Staab: Sure thing. If you examine speech energy versus spectrum, as Dunn and White did in 1940, you'll see there's not much difference between male and female voices. Further, it's noteworthy that speech energy is essentially below 2000 Hz, and the greatest concentration of speech energy is most often between 400 and 500 Hz. Fletcher examined these issues in 1929 and related speech energy to intelligibility. Of note, between 1000 and 8000 Hz, the percent of speech power is only 5 percent, yet this spectral range does contribute some 60 percent of intelligibility, while at the same time, 70 percent of speech intelligibility resides between 500 and 2000 Hz.
DLB: Again, arguing for the tremendous importance of the mid-frequencies.
Staab: Absolutely. When we focus on simply delivering the very high information, the challenge is enormous. Importantly, the patient may not like the "tinny" sound quality of extreme high frequency amplification and of course, acoustic feedback becomes highly problematic with substantial gain from about 2000 Hz and upward.
DLB: I see your point and I tend to agree…although depending on the shape/configuration of the hearing loss, my thought is to give all the high frequencies possible to keep the sound as natural as possible, for as long as possible. So if the hearing loss in the 4000 to 6000 Hz range is up to 60 or 70 dB, I'd probably go for it, whereas if the hearing loss in that same spectral range is severe-to-profound, I think I'd focus on the area below 4000 Hz. Then again, with remote microphone technology, we do open up additional possibilities for amplification without distortion and without feedback.
Staab: Good points. Yes, there are limits and they vary with the patient presentation and their percepts, as well as the specific technology, and yes—remote mics are a game changer.
DLB: Wayne, getting back to psychoacoustics, please review the concept of "critical bands."
Staab: Critical bands are frequency bands, each making equal (5 percent) contributions to speech intelligibility. French and Steinberg (1947) identified 20 frequency bands from 250 through 7000 Hz and showed that bandwidth in frequency was not constant throughout the range. Lower frequencies had smaller bandwidths in Hz (e.g., 250-375 Hz) while higher bandwidths (e.g., 1720-1930 Hz) involved a wider frequency range for their 5 percent intelligibility contribution. Related to the previous discussion, one could argue that 75 percent of the critical band intelligibility would include mid-frequency bands from 505 to 4200 Hz.
And of course, that leaves 15 percent intelligibility contribution from above 4200 to 7000 Hz. So again, high frequencies matter, but one should not ignore the substantial contributions to intelligibility made by the mid-frequencies. Further, people don't actually listen to speech sounds in isolation, they listen to words and more importantly, words strung together in phrases and sentences. And when you consider cold running continuous discourse, there's even more information to be gleaned from the mid-tones based on research by Studebaker and his colleagues (1987).
DLB: And so if I may quote you from hearinghealthmatters.org, you summarized "It seems to suggest that greater concentration of the mid-frequencies, especially if amplified cleanly, may improve intelligibility as much, or perhaps even more, than unchallenged concentration on the high frequencies." And what does that indicate with regard to frequency lowering?
Staab: This requires a two-part answer. It is important to keep in mind that these comments relate to what psychoacoustics tells us. Once a hearing aid is placed on the ear and ambient noises are involved, additional factors come into play that might cause some to suggest that mid-frequencies are not all that good, especially if amplifying them results in masking effects (often referred to as the upward spread of masking). Regardless, having clean mid-frequencies is substantially more critical than applying distorted mid-frequency amplification. Your question about mid frequency amplification and the effect of frequency lowering is important. Based solely on psycho-acoustics, it suggests that there may rarely be a situation when one actually needs to employ frequency lowering (FL). This is based on the contributions that mid frequencies provide to make sense of speech…However, if you distort those mid-tones by adding high frequency energy to that same bandwidth, you're essentially distorting some of the same area that contributes substantially for clearly understanding speech.
DLB: I agree, and I think another issue which people forget is that the primary reason to employ FL is for speech in noise. Of course, the most common noise people have difficulty with is other people speaking. So then, if we lower the sound in a band such as 4000 to 6000 Hz, we also lower the sound of the noise that resided in that band, and so then we're moving the primary signal and the background noise for 4000 to 6000 Hz, and moving them to the mid-tones…which as you mentioned distorts the mid-tones, and the noise which we were trying to escape is still present and has piggy-backed into the mid-tones!
Staab: I don't suspect that there is too much noise in the frequency band you mentioned, but the fact that distorted amplification (FL) is now superimposed onto what may have been very good amplification, would appear to be problematic, These are all major issues associated with FL, and we just don't have the evidence that FL is a good strategy for most people with regard to speech in noise, and frankly, if the patients aren't doing better with regard to speech in noise secondary to FL, why bother?
DLB: Wayne, I want to get your thoughts on the new numbers from the Institute of Medicine (IOM) in 2014 that indicate 85 percent of all people with sensorineural hearing loss never seek help.
Staab: I'm pretty sure you know my response to that, as I published it decades ago, Doug. The thing is, for the vast majority of people with hearing loss, simply having hearing loss does not mean they want or need hearing aids. Having hearing loss and "feeling the pain" from hearing loss (not hearing clearly, difficulty communicating with others, isolation, loneliness, etc.) is a better indicator of who needs hearing aids, than simply an audiogram, which concentrates on the hearing that is lost, rather than on what remains.
Further, not only must the individual feel the pain, but they have to consider and accept the value proposition. That is, what are they willing to pay, and what are they going to get for that investment? So bottom line…the patient who feels the pain and accepts the value proposition is the person who seeks hearing aid amplification…and if the IOM numbers are correct, that's a small percentage of people with sensorineural hearing loss. So maybe the pain factor is what we need to focus on moving forward. The marketplace seems to be addressing the value proposition via the Internet and big box retail, regardless of our personal emotions! But to be clear, regarding the paucity of individuals being fit with hearing aids, if patients do not feel the pain, they don't have an issue needing to be addressed, from their perspective.
DLB: I agree and I think we can partially measure the pain factor with speech in noise tests, to actually get an objective measure of their ability to communicate in the most difficult situation. Further, if we can improve their communication in noise, and prove it objectively, that, too would impact the value proposition and be a game changer. That is, the value increases when you can see tangible benefits! Wayne, it's been a real treat to speak with you. You're a breath of fresh air and I appreciate your thoughts and insight.
Staab: Thanks, Doug. Great to speak with you, too!
Wayne Staab, PhD, audiologist and editor of Hearing Health Matters (www.hearinghealthmatters.org).
Douglas L. Beck, AuD, Board Certified in Audiology, is the Web content editor for the American Academy of Audiology and the director of public relations with Oticon, Inc.