Dr. Ross on Hearing Loss
Some Comments on Hearing Aid Features
by Mark Ross, Ph.D.
Analog hearing aids, particularly the last few generations of them, were not the primitive, clunky devices they are often portrayed to be. Often, they included such sophisticated features as active frequency shaping, compression amplification (automatic gain control), and manually controlled directional microphones. Properly fit, they could ensure appropriate audibility right across the frequency spectrum for a hearing aid user. In fact, the amplified signal they provided was every bit as good as, if not better than, the signals delivered by the first generation of digital hearing aids.
The full import of digital signal processing was not realized until subsequent generations of digital hearing aids had been developed. Until then, consumers’ positive reactions to their digital hearing aids were apparently heavily influenced by their expectations of what the “computer age” could offer them. In one study, people compared their listening experiences with two presumably separate aids; in reality they were listening to exactly the same first generation digital aid, but they were told that one was digital and the other was a “conventional” aid. The subjects overwhelmingly preferred the digital aid to the ostensibly analog one, making sincere and specific comments about the superiority of the digital hearing aid over the “conventional” one. Clearly, their expectations of the superiority of digital signal processing influenced their judgments.
Since those early days, however, digital hearing aids have made enormous progress. What was only theoretically possible before is now each day becoming more and more a tangible reality. Hearing aid engineers and scientists seem limited only by the extent of their technical skills and imagination. With each new generation of hearing aids, we see creative features being introduced, usually accompanied by convincing descriptions of their value to consumers. Indeed, as I read these descriptions, it’s impossible not to be impressed. These new hearing aids are capable of performing operations that we could only dream about years ago (and many we couldn’t even conceive of). Still, it is important to keep in mind the lesson that can be drawn from the study referred to earlier. We should not be so beguiled by the appealing way some feature is described that we ignore the way it actually functions in real-life. Still, having expressed this caveat, there are some hearing aid features that have proven their value, while others appear beneficial on their face. It is on some of these features that I will now be commenting.
Noise Reduction (NR) Programs
One of the most common features included in modern hearing aids is the noise reduction (NR) program - for a very good reason. Most people with a hearing loss report difficulty understanding speech in noisy situations, and hearing aid manufacturers have been quick to respond when offered the possibilities inherent in digital signal processing. Each company has evidently developed its own proprietary method of reducing noise, and it is reasonable to assume that some are more effective than others. Unfortunately, there is no published research that directly compares the noise reduction features offered by different hearing aid manufacturers..
What they all have in common, however, is the goal of identifying speech signals spoken in quiet from those uttered in various kinds of noise. This process takes place in each hearing aid channel. When speech is the dominant signal in the channel, the degree of amplification (the gain) is either unchanged, or in some aids with another feature (Speech Enhancement) the gain may actually be increased. When noise is the dominant signal in the channel, the gain (for both the noise and the speech) is reduced. In making this decision, each company utilizes different formulas regarding the measured relationship between the speech and the noise.
A different type of NR program (called the “subtraction method”) acts to reduce noise signals occurring during pauses in speech. While this may not affect speech directly, it does reduce the overall perception of noise in the listening environment. Some noise reduction programs also target sudden, impulse sounds (transients) such as those produced by dishes clattering, doors banging, etc. This reduction takes place very rapidly, before the hearing aid user is exposed to its full impact. Another common noise reduction program focuses on wind noise, usually by reducing the low frequency response of the aid when it is present. We should know that when listening to speech in noise, no noise reduction feature is currently able to just “peel away” the noise from the speech when both are taking place at the same time in the same frequency band. But given the progress made to date, I wouldn’t take any bets on whether it won’t occur in the future.
In terms of effectiveness, the objective evidence suggests that a noise reduction program will not directly improve speech perception in noise (though it doesn’t negatively affect it either). Subjectively, however, an NR program will improve the sound quality and does make speech more comfortable to listen to, particularly for longer periods of time. It will also improve certain cognitive tasks undertaken in the presence of background noise (such as remembering words or responding quickly to complex visual tasks). These are not trivial benefits from the perspective of the hearing aid user attempting to communicate or learn new tasks in noisy places. (Or, as we’ve all said at one time or another: “Please keep the noise down. I’m trying to concentrate!”)
A directional microphone is the one hearing aid feature that can directly increase the speech to noise ratio (SNR), or the intensity level of a speech signal relative to the background sounds. In the classic directional microphone operation, the hearing aid reduces the intensity of the noises arriving from the listener’s sides and rear, leaving frontally arriving signals untouched. The result is an increase in the SNR. Therefore, in order to realize the full benefit of a directional microphone, listeners have to position themselves with the desired signal in front of them and potentially interfering noises to their sides or rear. Given the fluidity of real-life situations, it may not be realistic to expect someone to manually switch hearings aids to the directional mode, and then back again, when the listening situation changes. This may be one reason why the findings in controlled research of better speech perception in noise were not always paralleled by subjective preferences for the directional position. With the advent of automatic switching, however, the hearing aid itself has now assumed that responsibility, thus increasing the likelihood that the listener will derive benefits from directional microphones.
Automatic and adaptive directional microphones are designed to continually scan the acoustic environment and to automatically switch the directional microphones to the correct position. The hearing aid notes the positions of major noise sources and adapts its directional characteristics so as to reduce the intensity of the signals coming from those directions, rather than just from a single rearward direction. This should be advantageous in situations where noise sources often change directions. In one recent development by Siemens, two aids working as a team have increased the theoretical possibilities of directional microphones (the “SpeechFocus” feature) One example they give is when one aid focuses on a speech source from behind the listener (like someone talking from the backseat of a car), the other aid suppresses noise coming from other directions. Other companies, like Oticon, have also developed aids that “talk” to one another (via short-range radio links) with the eventual goal of offering unique binaural possibilities. These claims all “sound” good, though I have not seen any research that examines how this arrangement actually works in real-life.
While some analog hearing aids included some sort of feedback suppression, it was accomplished in a way that eliminated high frequency sounds, deleted a portion of the acoustic spectrum immediately adjacent to the feedback frequency, or reduced the gain of the hearing aid. In each of these methods, some potentially valuable acoustic information was omitted, which made these early feedback suppression circuits a bit counter productive. In the newer methods, made possible by digital signal processing, the feedback cycle is disrupted without modifying the frequency response of the hearing aid. According to what manufacturers reported at the last HLAA convention, it is possible to achieve 20-25 dB more gain from a hearing aid before an acoustic squeal (feedback) occurs. The benefit of this capability is self-evident.
Initially, it was thought that only people with severe or profound hearing losses, wearing powerful hearing aids, would find a feedback suppression circuit most useful. The sounds of acoustic squeal (not music!) seemed to accompany them wherever they went. We should know that the acoustic squeal caused by feedback is not a benign sound; it is not just disruptive, but it indicates that the individual is not receiving the amplified sounds that he or she should. An effective feedback suppression circuit permits people with severe to profound hearing losses to realize more amplified sounds from their hearing aids, by eliminating or reducing the occurrence of audible feedback.
Lately, the value of a feedback suppression circuit is being realized by people with mild or moderate hearing losses. We know that wearing an earmold in one’s ear can be a disagreeable experience, sometimes actually painful and often unpleasantly modifying the way a person hears his/her own voice. Many of these people would rather “get along” without a hearing aid than subject themselves to these unpleasant sensations. To resolve these problems, the hearing aid dispensers have typically employed large acoustic vents in the earmold or, with milder hearing losses, attempted to deliver the sound via a narrow tube inserted in the ear canal. But before the advent of an effective feedback suppression circuit, it was not possible to take these steps because of acoustic feedback. Currently, because this feature is available, the fastest growing segment of the hearing aid market is with people with mild and mild-to-moderate high frequency hearing losses. These people are now able to keep their ear canals unoccluded while wearing hearing aids appropriate for them.
Another hearing aid feature that should be considered is Frequency Lowering. While the first entry into this field took place about twenty years ago, with the introduction of the transposer hearing aid by AVR Sonovation, it never got much traction with the public. In recent years, several other companies have introduced their own versions. The rationale for this feature lies in the fact that much of the most important information in a speech signal can be found at the higher frequencies. All voiceless consonants, for example, have a predominance of their energy in the higher frequencies. People with extremely poor or absent hearing at the higher frequencies will be unable to perceive these high frequency sounds, no matter how good their hearing aids may be. This, indeed, is one major reason why people with such hearing losses complain of “hearing but not understanding.” What a frequency lowering feature will do is shift some of the high frequency energy of the speech signal down to lower frequencies, where the person still has some usable residual hearing. It will not sound “normal,” at least not for a while, but the acoustic information of the high frequency consonants will be perceptible.
At least two companies now offer this feature, the Widex with its AudibilityExtender and Phonak with its SoundRecover. Both companies have sponsored or engaged in research projects to examine the effectiveness of the feature, and both have demonstrated that it can improve speech perception for many people. However, it is not, in my opinion, the kind of feature that can be set in a hearing aid and then ignored. The listening experience it presents to a listener must be carefully monitored by the dispensing audiologist, and adjusted when necessary.
One other feature common in modern hearing aids is Wide Dynamic Range Compression (WDRC). What this circuit does is provide different degrees of amplification depending upon the intensity level of the input; the intent is to ensure that soft sounds can be heard without over amplifying louder input sounds. While all the above features have been described separately, they often (and usually should) co-exist in a single aid. The synergistic effect of combining them will exceed what is possible with each one separately. This indeed is what was found in the results of several studies sponsored or conducted by the Unitron Company. Modern hearing aids include many other features that I have not described above; what I’ve done is select the ones that seem to me to be the most useful. One other feature, simple and inexpensive, should be mentioned and, I believe, included in every hearing aid and that is a telephone (T) coil. The T coil can serve as an assistive listening device in addition to its original purpose, hearing on the telephone.
Finally, a consumer has to keep in mind that the more advanced features included in a hearing aid (the “premium” level), the more expensive the hearing aid is likely to be. Consumers, working together with dispensing Audiologists, have to weigh the cost/benefit ratio of any hearing aid loaded with features. Also, it is worth remembering that the “premium” aids of yesteryear may contain all the advanced features a person needs.