Interview with Robert Sweetow Ph.D., Clinical Professor, Dept. Of Otolaryngology, Director of Audiology, University of California

Topic: Aural Rehabilitation and Neural Plasticity

AO/Beck: Hi Dr. Sweetow, as always, it's a pleasure to speak with you.

Sweetow: Good evening Dr. Beck

AO/Beck: Robert, how long have you been at UCSF?

Sweetow: It's been eight years.

AO/Beck: Who is on the UCSF team these days?

Sweetow: Our audiologists are Lisa Lamson, Becky Bingea, Andy Valla, Wileen Chang, and Erin Luckett. Toni Will heads up our newborn hearing program and Jan Larky is in charge of our cochlear implant division. On the medical team we have Rob Jackler heading up the otology group, Anil Lalwani whose primary interest is genetics and pediatrics, Steve Cheung and Michael Merzenich who are both involved in research in plasticity, and Robert Schindler. In general otolaryngology we have Andrew Murr, Andy Goldberg, Adite Mandpe, Mike Kaplan and Mark Singer.

AO/Beck: Sounds like a phenomenal group! All right, let me get into the topic at hand. Please tell me, what is neural plasticity as it relates to aural rehab?

Sweetow: Neural plasticity refers to the fact that the functional representation of the brain can be altered dramatically as a result of diminished or altered input and learning. In other words, specific neurons and areas of the brain that were 'wired' to perform certain functions, can actually be assigned new jobs This 'plasticity' of the brain can be facilitated by conditioning, as well. As audiologists, our job when providing amplification, is to teach patients to listen. In other words, we should be fitting brains, not merely fitting ears. We need to adapt the principles of neural plasticity into aural rehabilitation in order to take advantage of the ever-changing capacity of the brain We need to remember that aural rehab (AR) was the origin, and basis of our profession. However, audiologists have gotten away from doing AR because, basically it can be somewhat boring, and also because it historically wasn't very successful when using the traditional analytic approach to AR, such as lip-reading, or listening training based on phonemic distinctions. I think the reason it wasn't very successful in those cases was because we were trying to train people to listen through hearing aids that were ineffective. Now that we have much better hearing aids available, we have much better potential for greater audibility and comfort. If we couple a revised analytic approach with synthetic training (in other words, teaching the patient how to control his or her environment, to use repair strategies, and so on) we have a real opportunity to enhance the patient's listening abilities. However, keep in mind that even with new state of the art hearing aids, we are not seeing substantially greater satisfaction or objective benefit. I think a lot of this is because we spend so much time dealing with the electroacoustic and programming aspects related to fitting hearing aids in the year 2000, that we've stopped teaching people how to listen. I believe our results would be substantially better if we combined our hearing aid fitting skills with our AR skills.

AO/Beck: Robert, I wonder if you can comment on this.... I have found in my practice, and the literature seems to indicate, that patients do just about equally well with high quality digitally programmables and one hundred percent digital hearing aids. Why aren't we seeing a clear cut 'digital difference'? I have read some authors who state the digital difference is not going to be apparent as long as we use the same fitting formulas for digitally programmables and digitals. Specifically, they ask the converse. That is, why should you see a difference if both instruments are programmed identically using the same fitting formulas?

Sweetow: Let me give you a multi-part answer. First of all, I don't think it directly relates to the fitting formulas we are using. I am not sure that we haven't gone about as far as we can go with fitting formulas for a significant proportion of patients we serve. I think the formulas are well based and they do indeed need to be fine-tuned in order to work well for many patients. However, perhaps for a larger percentage of the patients, once sounds are audible, and once amplified sounds are comfortable, we need to retrain the brain in a manner such that it is capable of effectively utilizing this new input. Think about it this way, when you have a hearing loss, and you put new and improved (or at least, different) sound into the brain, the central auditory system may not know what to do with it! The central auditory neurons in the inferior colliculus and in the primary auditory cortex respond to changes in temporal and spectral characteristics differently with a reduced input than with an amplified input. So for some listeners neural plasticity occurring after having a hearing loss but before obtaining hearing aids can be beneficial and for others it can be deleterious. Either way, I believe we need to use AR to help the brain relearn how to interpret sounds provided through amplification.

AO/Beck: What can you tell me about 'sound quality' and plasticity?

Sweetow: If you take patients with hearing impairment and adjust the hearing aids so they say it sounds good, their judgment is made based on what their status quo impression of the desired sound is. But we know their brain has changed since they last heard well, so the targeted sound probably will change as plasticity acts on it. For example, when patients with high frequency hearing loss wear aids with high frequency emphasis they frequently say 'it sounds horribly tinny'. And it probably does, but we know that if we're going to maximize their sound perception, we need for them to hear those tinny sounds, until they no longer sound tinny to them. So, sound quality will vary with plasticity. On the flip side, we've all seen patients who try new hearing aids and they say 'this is fantastic', and then 8 weeks later they say there's no difference between the new hearing aids and the old ones. Maybe that's plasticity without AR?

AO/Beck: Do you discuss neural plasticity with your patients?

Sweetow: Sometimes I do, it depends on the person. I usually tell patients that it will take time for the brain to undergo changes to maximally utilize new sounds. I tell them it will take time for the brain to recognize what is important and what is not. I often use the example about hearing one's own footsteps. New hearing aid wearer may say they can hear their own footsteps and they don't like that sound. I point out that people with normal hearing also hear their own footsteps, but because they are accustomed to the sounds, and because their brain recognizes the sound as relatively unimportant, they dismiss and ignore the sounds. Of course, if it's a novel stimuli, the brain will tune into the new, different or novel sound. I will tell some patients that the brain is constantly changing and that as a hearing loss occurs, and when hearing aids are later worn, that different parts of the brain assume different 'jobs' in processing sound.

AO/Beck: So again, we're fitting the brain, and not necessarily the ears.

Sweetow: Yes, I think that is absolutely correct. That's got a lot to do with why 5 people with the exact same hearing loss wind up with five different hearing aid programs. Changes occurring in the ears ultimately are translated and transferred into the brain as input, and the brain changes based on the input.

AO/Beck: This brings up the issue that many audiologists are currently addressing. Specifically, do we have too many circuit choices now? In other words, can disposables, or any other product with five, six or seven fixed circuits actually fit the vast majority of hearing losses in the mild to moderate categories?

Sweetow: For a significant part of the population, that may be a possibility. Again, if the sound quality from the disposables or any other hearing aid with limited flexibility is high, and if sounds are audible and comfortable, yes, it could be that these products are sufficient. They certainly stand a better chance of serving the patient's need if you train the brain to use the amplified sounds. Here's one way to think about this. If you buy new tires for your car, but the car also needs a tune-up, you're still going to have a lousy ride. Conversely, if the car is tuned-up well, but you have a flat tire, you're still going to have a lousy ride. So, if you have lousy hearing aids, you're gonna hear poorly. If you have great hearing aids, but your brain doesn't know what to do with the sound, you're going to hear poorly. This is probably representative of where we are as an industry and as a profession, we have fantastic hearing aids, but we forget or decided not to retrain the brains to use the available sound. In other words, if you have learned to listen poorly, and you use a bad or average or great hearing aid, it may not make much of a difference until you learn to listen well again. I want to stress however, that research has neither clearly proved or refuted the superiority of digital amplification. Therefore, I'm definitely not ready to give up on high quality custom fit hearing aids with greater fitting flexibility.

AO/Beck: So then, what is the most efficient and efficacious way to learn to listen well? Or better said, what is the best AR route?

Sweetow: I think we need to understand that we started 50 years ago with analytic approaches to AR, such as structured drills focusing on specific elements of speech and speechreading and auditory training, much of which didn't work because the hearing aids were horrible. Now we tend to concentrate on synthetic approaches, which focus on the larger units of meaning such as connected discourse and teaching patients to control their environment as it relates to listening. This too, while reportedly more successful, has not encouraged massive numbers of audiologists to incorporate these procedures into their practices. I think we need to develop new, computer-based combined analytic and synthetic approach and we need to take advantage of what we know about computers and learning theory to retrain the brain in order to enhance neural plasticity to better recognize meaningful speech through a system of rewards. I believe this is the most efficient way to facilitate neural plasticity to allow the brain to best utilize the incoming sound that high quality amplification provides.

AO/Beck: Recently, there was an article which addressed -- Is it worthwhile to amplify sound in the regions of 6000 to 8000 Hz if the hearing loss at 6000 to 8000 Hz is worse than 70 dB or so? What are your thoughts on that?

Sweetow: I think these researchers indicated that if you lose the afferent input from the inner hair cells, that any sound delivered in that spectral range may be of questionable value. I think they're correct in that if you provide high frequency information to a brain that cannot use it, it's probably not worthwhile. However, if you can train the brain to use the information, you're probably better off in the long run. I think it may be dangerous to take their results and generalize them so that you don't even attempt to amplify sound in that region. This relates to what the brain has learned to do with its' impaired input. Some brains may learn to take the impaired input and suppress the impaired sound, while using the portion of the signal that isn't impaired. Other brains may learn to use the new amplified sound positively. So in essence, successful brains make a positive compensation and for them, the new sounds are worthwhile.

AO/Beck: This sounds similar to the situation that occurs after a patient has a stroke and the brain is able to rewire neurons a bit to accommodate the non-functional neurons.

Sweetow: Exactly right, that's why I would be hesitant to not amplify the 6000 and 8000 Hz region. If we don't give the brain the information, it has no chance of rerouting it into a useful signal. I prefer to give them the high frequency information, and let's see what the brain can do with it if we facilitate an effective AR program. Of course for some folks the high frequencies will cause feedback, and for some the sound quality will be aversive, but for all of them, it's worth trying if we provide the appropriate AR to maximize the brain's potential. We can always minimize the high frequency gain later, providing we have counseled the patient that this may be a possibility. That is one argument in favor of flexible programmability.

AO/Beck: Does AR and neural plasticity yield the same results as acclimatization?

Sweetow: No, they are different. I think of acclimatization as changes that occur without formal intervention. In my proposed AR model, formal brain training or AR is a critical part of the final result. Research data indicate that plasticity can be facilitated with conditioning from the neuronal level on up. We know that if you train a brain, neurons will change function much more quickly than they would without training. We can take advantage of this behaviorally to facilitate neural plasticity and get a better response to amplified sound.

AO/Beck: So how long does it take to train brains and how do we do it?

Sweetow: Frankly, I don't know. We're currently looking at interactive computer-based programs to train the brain to make distinctions in temporal, spectral and amplitude domains to see if improvements in detection and discrimination tasks will carry over to improvements in general speech perception. We have to develop time and cost effective procedures given the fact that reimbursement is sparse. We need to use outcome measures to prove to the payors that AR is effective in improving the overall outcome with amplification.

AO/Beck: So what does the future hold for us?

Sweetow: I think we'll quickly see more disposable hearing aids, and certainly less expensive custom hearing aids too. I believe we need to go back to our AR roots to help assure our survival by focusing more on AR and applying neural plasticity concepts to hearing aid dispensing. I hope we will begin to integrate our knowledge of learning theory with computer technology to make all these components a reality.

Suggested Reading on Neural Plasticity

Palmer CV, Turner CT, Lindley GA., (1998), The functionally and physiologically plastic adult auditory system. J Acoust Soc Am. 103:4,1705-1721.

Willott, JF, (1996), Physiological plasticity in the auditory system and its possible relevance to hearing aid use, deprivation effects, and acclimatization. Ear and Hearing 17:3,66S-77S.

Kilgard MP, Merzenich MM., (1998), Cortical map reorganization enabled by nucleus basalis activity. Science, 279:3, 1714-1718.

Juliano SL, (1998), Mapping the sensory mosaic. Science, 279:3, 1653-1654.

Harrison RV, Stanton SG, Nagasawa A, Ibrahim D, Mount RJ. (1993) The effects of long-term cochlear hearing loss on the functional organization of central auditory pathways. J Otolaryngology, 22:1, 5-11.