Maximizing Sound Quality And Audibility With Hearing Aids - A Direct Approach To User Satisfaction
This paper introduces the theory and technology of how to directly utilize a client’s subjective experience in the fitting of hearing aids. A thorough knowledge of instrumentation and acoustics is a prerequisite. The method defines concepts of “natural,” context, auditory objects, clarity, etc. to extract useful information from the client’s report of their auditory experience. That information is then used to select and adjust hearing aids. A carefully monitored field evaluation period verifies the efficacy of the fitting and assists in establishing realistic expectations for the client. The goals of “natural sound quality” and optimal audiometric results are not mutually exclusive. This is demonstrated by both objective measures of successful fittings, as well as other measures of client satisfaction.
Recent surveys of user satisfaction with hearing aids indicated that about 50% of hearing aid users are dissatisfied with their hearing aids. Fitting hearing aids to create satisfied users is complex, both as an audiometric problem and as a human problem. Fulfilling calculated audiometric target requirements is often not enough to result in a satisfactory hearing aid experience. In addition to the audiological requirements, elderly clients present multiple human problems related to changes in their physical, auditory, visual, cognitive, emotional and social dimensions. Each individual presents unique capabilities, circumstances and requirements. Multidimensional problems can be solved by finding the most optimal solution to the variables presented. Finding such a satisfactory solution involves working with the individual most intimately involved with the problems… the client.
The indirect method of prescriptively fitting hearing aids is based upon the hypothesis that a satisfactory experience of speech and real world sounds can be extrapolated from the client’s responses to pure tone stimuli and discrete noise stimuli presented in controlled environments. Since the psychoacoustic experience of any particular sound is related to the context within which the sound is presented (Figure 1a), a more directmethod is presented here that uses the client’s responses to real world sounds, in real world contexts, as the stimuli for making electroacoustic adjustments. Understanding how this can be accomplished requires an appreciation of the experiential mechanisms involved in auditory perception for normal and impaired hearing.
Real world sounds can be looked upon as acoustic events coming from the physical objects that generate the sounds (Figure 2). People partially perceive (or “hear out”) some physical features of the acoustic event. These features become encoded into neural firing patterns by the cochlea and are used as cues to help recognize the object generating the acoustic event. These encoded cues, at several levels, are separated, evaluated, and integrated within the auditory system and are compared to memories of similar acoustic events. A dynamic interaction, between perceived cues and memory, acts to “fill in” the details of the object and establish a recognition of the acoustic event. If the cues perceived form a coherent match to an object in memory, an auditory object is experienced (which can be labeled: “the sound of a bird,” “an unusual sound,” etc.). The more cues available, the more knowledge in memory, the easier the recognition.
The activity to create auditory objects in our experience does not occur in isolation. The acoustic event attended to is imbedded within a context of other environmental acoustic events, as well as a visual context. The auditory system and associated brain areas must do the appropriate “figure/ground” separations in order to attend optimally to the acoustic events of interest. This again is accomplished by processes of separation, evaluation, and integration. The auditory and visual cue complex perceived is compared to memories of similar auditory object relationships. If the relationships are consistent with memories of similar events, the process of separating figures of interest from ground is more easily accomplished than if the relationships are unfamiliar.
While the mechanism of auditory perception is the same for individuals with hearing impairment, hearing- impaired individuals experience a loss of the subtle acoustic detail of the acoustic event. For example the soft high frequency sounds may be absent from the client’s experience. In addition, pathology in areas of the auditory processing system may interfere with the nervous system’s ability to encode, separate, evaluate, or integrate information in the acoustic clues into a format that is coherent with auditory memories.
Hearing aids that assist the processes of separation, evaluation, and integration, and minimize factors that impair the nervous system’s ability to perform these functions should be more acceptable by the auditory system and subsequently the client. Hearing aids can accomplish this by supplying the acoustic detail missing from the experience of the individual with a hearing loss, and by not contributing distortions that might impair nervous system functions.
When first-time, or experienced, elderly hearing -aid candidates were interviewed for their highest hearing aid expectations, they desired (1) to hear their best, and (2) to have everything sound “natural.” A classification of “natural” is a subjective experiential judgment and unfortunately an accepted taxonomy of first person experience has not, as yet, evolved. It is used here to separate it from earlier goals of hearing aid fittings, such as preferential (likes and dislikes) and normal (as occurs in physically unimpaired individuals). “Natural” will also be used as a communication device with the hearing-impaired client during the fitting process.
How do we categorize sounds as “natural?” The term “natural” can be understood as follows. If the sound from a physical object is distorted in its frequency-intensity spectrum or in its temporal characteristics, then normal hearing individuals will say that the object no longer sounds “natural.” The silver spoon falling on the glass table no longer sounds like a real spoon. Some psychoacoustical quality of the object will not fit the individual’s visual perception and conception of how that object should sound in reality. The “unnatural” acoustic event is experienced as an auditory object that lies outside the norm of the accumulated memories of similar acoustic events. That point where the experience of an acoustic event changes from “natural” to “unnatural” is known to both normal and to hearing -impaired individuals.
Individuals with normal hearing and common experience describe the “naturalness” of acoustic events in a similar manner. This is because our experience of “natural” is constrained by the consistency of the physical world and because our experience of “natural” is constantly tested through our physical and cultural interactions. Therefore, unlike preferences that vary from individual to individual and from time to time for the same individual, what is “natural” can be agreed upon by all individuals of similar experience. We may not agree that we both like the violin concerto, but when attending the live performance we would not disagree that the violins sound “natural.”
The hearing aid fitting method presented here targets (1) “natural” sound quality of real world objects (minimal distortion) and (2) optimal speech understanding in all environments. The potential incompatibility between “natural” and “optimal” creates a workspace within which a dynamic interactive fitting process can occur between the dispenser and the client.
The method consists of resolving these factors without sacrificing performance.
Definition Of Terms
(as used in the Guided Selection Method – Figure 1b.)
Natural: is defined for the client as “nothing artificial added.” Often a client would ask, “I’m hearing impaired. How am I supposed to know what ‘natural’ is like?” Although the client may not be able to imagine what certain natural sounds are like, they are able to recognize if a sound is consistent or inconsistent with their memory and current visual perception of the acoustic event. A particular acoustic event is experienced as “unnatural” if it has the wrong psychoacoustic qualities of loudness, pitch, timbre, or duration, to the object’s contextual appearance. When a sound is louder than expected, mechanically reproduced, or distorted in any way, it doesn’t sound like the real object. “Natural” is also highly contextually sensitive. A violin played outdoors sounds different than a violin played in a small room.
Threshold of Naturalness: This term is the focal point of the fitting procedure. It is defined as the level at which an increase in any electroacoustic parameter (gain, output, frequency response, compression, earmold modification, etc.) causes the experience of the acoustic event to be inconsistent with the client’s current visual perception and memory of similar naturally occurring acoustic events. More precisely, it is the normative surface function of the auditory object in multidimensional psychoacoustic space (Figure 2).
Common, Real Object of the World (CROW): is a familiar, present, physical object that can generate an expected psychoacoustically complex auditory experience. For example: a spoon & table, or a pair of hands, as opposed to a pure tone generator or a culturally unfamiliar musical instrument. CROWs will be used to create commonly described experiences for hearing aid fitting purposes.
Auditory Objects: When the sound of a silver spoon falling three inches onto a glass table is heard and observed at a distance of two feet, adults with normal hearing will all agree that the sound is appropriate to the particular spoon, table, drop height, distance from the listener, etc. There is a certain expected loudness, pitch, timbre, and duration of the sound based upon a recall of similar or comparable experiences. An auditory object is the bounded psychoacoustic experience caused by the dynamic interaction of the perceived cues and memory components. In normal experience the auditory objects are consistent with the commonly experienced acoustic event. In auditory illusions (analogous to Figure 1), however, or in hearing-impaired individuals, with pathology affecting proper encoding and processing of acoustic cues, the experienced auditory object is inconsistent with the acoustic event. Incorrect descriptions are given for the perceived objects. It is not uncommon that a hearing-impaired individual says with great certainty that the speaker said something other than what the speaker admits to.
Distortion: is viewed from the client’s perspective as any change in loudness, pitch, timbre, or duration that perturbs their experience of “natural.” Distorttions can effect (1) the sound of the object, (2) the acoustical context, or (3) their proportional relations.
Clarification: the increase in the resolution of subtle acoustical detail.
Amplification: the increase in the intensity of sounds the client already hears.
Optimal: means the hearing aid provides the clarification and amplification required for best understanding in all of the client’s environments.
The Guided Selection Method
Often, the client initially does not know how a proper hearing aid fitting should sound. instructed that “natural=good” (as opposed to “amplified=good”). A typical instruction is, “I’m going to be adjusting the hearing aid to you. I don’t want you to adjust to the hearing aid. You will have enough to deal with just getting used to all the new soft sounds and the greater distances. I don’t want you to have to get used to extra loudness or artificial sound quality.”
The Question Formats
- Does this sound natural?
- When does my voice no longer seem natural?
- Which sounds more natural: X or Y?
- Which is louder, my voice or the noise?
The Instruction Set
Although these instructions may appear trivial to normal hearing individuals, they reassure the hearing-impaired adult that their memories are valid, and will be validated by subsequent demonstrations with hearing aids. CROW sounds are presented that test all regions of the auditory domain.
The experience of speech as a “natural” CROW sound:
A whisper should sound like a whisper. A whisper at 3 feet should not sound like a whisper at 6 feet. A shout should sound like a shout, and not like a distorted or amplified shout.
The experience of CROW sounds as “natural”:
A dish should sound like a dish, no clank, sharpness, etc. A dropped spoon should sound like a spoon, not a clunk or clang or clink. A sheet of crumpling paper should sound like a sheet of crumpling paper, not like sheet metal, or excessively, surprisingly loud. Water in a sink, toilets flushing, doors slamming, plastic wrap wrapping, hands clapping, hands being rubbed together should sound exactly like what they are to the aided client.
The experience of noise as a “natural” CROW sound:
Twenty people in a large room should sound like 20 people in that specific room, and not like forty people, or like twenty people in a smaller room. A small truck across the street should sound like a small truck across the street, and not like a bus, or a large truck.
The experience of “natural” figure/ground or CROW/contextual relationships:
Everything should sound like what your eyes and mind say they should sound like. Everything should be in appropriate balance given the situation.
Interactive Fitting Period
- Speech and CROW sounds are presented. The electroacoustic parameters of the hearing aid are increased from the “off” position until an “unnatural” response is obtained. The settings are then backed off until a “natural” response is obtained. The earmold is similarly modified.
- The electroacoustic parameters are readjusted to maximize Articulation Index scores (as monitored by sound field tests or real ear device) without exceeding the client’s Threshold of Naturalness.
- Speech and CROW sounds are then presented in a variety of settings and the client is questioned to determine if the object is “natural” or “unnatural,” and to determine if the sounds are appropriate in each of the following settings:
- quiet – noise (temporally regular and irregular)
- reverberant – nonverberant
- close – distant
- right – left – front – rear
- Readjustments are made as necessary.
- Non-formal speech discrimination tests are performed in each setting as appropriate, with a variety of voices, speeds of speech, intensities, and word list formats.
At times, the dispenser may believe that the client would understand better with certain settings that may not be experienced by the client as “natural.” The client is then assessed for his/her willingness to participate in an A/B test. An A/B test is a comparison test of either different parameters of the same instrument, or of different instruments. Willingness to participate in such an experiment is often related to factors determined in the client assessment analysis performed earlier.
The goals of the protocol are (1) to provide a clear, distinct, “ordinary” experience of the world of sound, (2) to develop a relationship of communication and trust, and (3) to help the client achieve his/her highest level of performance with the hearing aids.
The Field Evaluation Period
What is successful for the client in practice is more important than what is successful in the office. Given the complexity involved and individual differences, the field evaluation period affords the opportunity to test the success of the fitting.
After being trained as a discriminative listener, the client is ready to evaluate his/her entire acoustical world for a six-week period. They are asked to report (1) anything that does not sound 100% natural in their experience, and (2) anytime they feel they are not hearing as well as the normal hearing people around them. While these are unrealistic expectations for nearly all hearing-impaired individuals, they focus the client’s attention on any limitations of the current hearing aid fitting. In this manner, the client returns to the office with all of the problems that must be resolved or addressed for the client. The client and dispenser discover the extent to which these problems can be addressed given the limitations of both the instrumentation and the client’s auditory processing abilities. The result is realistic expectations.
Certain individuals (possibly 5% of the clients in the office) are not good judges of the point between “natural” and “unnatural.” With these individuals, the dispenser must assist the client with his/her experience and knowledge of appropriate electroacoustic settings. The field evaluation period, and an established basis of trust and communication, is of even more significance for these individuals.
A Hearing Aid Fitting Quality Card is provided to the client and serves to focus attention to eight points of interest and will be used to assess problem areas: understanding others, experience of own voice, experience of common sounds, control of loudness, control of background noise, comfort of earpiece, directionality, and appearance.
Questions & Answers
Can adult clients be trained to be credible reporters of complex experiential data? Yes.
Hearing-impaired individuals (after overcoming their initial insecurity) appear to make judgments of the Threshold of Naturalness as readily as normal-hearing individuals. If the real world sound cues heard are representative of the object of experience, then the client will experience the object as natural. A CROW sound is perceived as “natural” as long as the increase in the physical parameter increases the clarity (subtle acoustic detail) without distorting the experience of the CROW beyond the subject’s perceptual and conceptual estimation.
Can those responses be reliably used to determine electroacoustic hearing aid settings? Yes.
This Guided Selection Method, although subjective, is successful and efficient because it forces both the client and dispenser to select electroacoustic hearing aid settings that converge to a singular judgment of “natural”/”unnatural.” The fitting goals are usually accomplished in 10-30 minutes. Readjustments after the fitting period are generally minor. The hearing aid’s final settings are (1) repeatable, (2) approachable from different initial settings by the method above, and (3) do not change over years (without additional hearing loss).
Can hearing aids tuned by subjective methods result in objectively verified quality fittings? Yes.
The audiograms in Figure 3 demonstrate typical audiometrically determined results. The quality of the audiometric results is highly dependent upon the quality of the hearing instrument being fit. A/B tests with different instruments quickly show differences in the AI scores when fit to the same Threshold of Naturalness.
Non-formal audiometric tests in uncontrolled real world environments verify for the client, in their immediate experience, the improvement over their unaided performance, or over their old hearing aid’s performance.
Is the method presented here successful in creating satisfied clients? Yes.
The level of client satisfaction is dependent upon a complex balance of audiometric and human factors. This interactive fitting procedure attempts to optimize an individual’s performance with hearing aids given their personal limitations and current technical limitations. Should the client make decisions based on aid type, cost, size, etc., these choices are taken into consideration by the client in their estimation of the outcome of the final fitting. After the field evaluation period, each client feels they made the right choice. Individuals are generally willing to accept their limitations as long as we sensitively try to optimize their performance in accordance with their wishes and needs.
A measure of achieving satisfactory hearing aid fittings for a client can be demonstrated by their behaviors: client referral, their appreciation of better quality hearing instruments, their shorter adjustment periods, their subjective and objective improvement in hearing.
The results from this office are significantly outside the norm.
Measures Of Success And Satisfaction
- These results occurred using the Guided Selection Method in a small single-dispenser office with two employees (clerical and customer service) in a non-affluent community. 400-450 hearing aids per year were dispensed, as compared to 150-200 hearing aids per year for the average dispenser.
- 65% of the aids dispensed over the past three years were programmable instruments (nearly 1,000 ReSound units personally dispensed by the dispenser in his facility) vs. 3-4% nationally.
- In the first quarter 1994, the number of units dispensed increased 8% over the similar period last year, this even as the dispenser reduced his workload to a four-day week. (Nationally, hearing aid sales declined 12% last year and continued to decline through that first quarter of 1994.)
- 73% of the hearing aids dispensed in 1993 were behind- the-ear units. At a time when people are conditioned to believe that small canal aids are best, these clients change to, and refer their friends to get, behindthe- ear units.
- In 1992, 96% of clients obtaining hearing aids were either repeat users or mentioned the name of a referring customer.
- All clients critically evaluated the eight areas on the Hearing Aid Quality Card and signed contracts declaring that they would keep their hearing aids only if satisfied. “100% satisfied, or 100% refunded” is the office policy.
- Only 7% of clients returned their hearing aids after the trial period (1992), as opposed to nearly a 20% return rate for programmable hearing instruments nationally.
- In light of the proposed new FDA regulations, 46% of clients (752 of 1642 clients in a single congressional district) contacted by letter, took the time to write their congressman acknowledging the quality of service by this office.
It has been traditional in the hearing aid industry to fit hearing aids by first applying an audiometrically determined formula (predetermined by the manufacturer, or a dispenser’s chosen prescriptive target), followed either by readjustment of the settings to the client’s preferences and/or by aural rehabilitation or counseling.
Difficulty in dealing with the complexity of a client’s subjective preference responses has led to an aversion to asking questions that elicit subjective responses and inspires attempts to find better objective fitting methods.
Merely providing an objective measure of sound in a given circumstance does not ensure either the client’s understanding in real world situations or the client’s satisfaction with the quality of their hearing experience.
Recently issues of “understanding in noise,” “the quality of hearing,” and “user satisfaction” have become significant. These are subjective issues. Objective performance standards are inadequate. When the dispenser uses primarily objective standards without significant direct client input the dispenser determines only what he believes will be satisfying for the client. This will not necessarily result in satisfaction from the point of view of the client.
Given that hearing satisfaction is subjective, it becomes imperative to utilize data from the client’s experience in order to obtain both audiometrically reasonable and subjectively satisfying hearing fittings. This paper reintroduces a method (Magilen, 1991) which focuses on dispenser-client dialogue. This dialogue, centered around the category “natural,” avoids the problems and complexity of preference judgments and achieves its desired result of natural sound quality and optimal audibility.
As we age we are introduced to changes that challenge our lives. How we face those changes strongly influences our experience of personal satisfaction. Loss of hearing separates individuals from the ones they love at a time when they need them more. Hearing aid fitting procedures for the elderly must support the entire individual. It must respect the individual’s personal and auditory needs and accomplishing this requires direct interaction and communication with the client.
Boothroyd, A. (1993). Speech perception, sensorineural hearing loss, and hearing aids. In G. Studebaker & I. Hochberg, (Eds.), Acoustical Factors Affecting Hearing Aid Performance (pp. 277-299). Boston: Allyn and Bacon.
Bregman, A.S. (1990). Auditory Scene Analysis – the perceptual organization of sound. MIT Press.
Light, L.L., Burke, D.M. (1988). Language, Memory, and Aging. Cambridge University Press.
Lockhead, G.R. (1992). Psychophysical scaling: Judgments of attributes or objects? Behavioral and Brain Sciences, 15, 543-601.
Magilen, G., (1990). A Need for Knowledge, Responsible Hearing System Specialists, Hearing Centers Network Publication.
Magilen, G. (1991). The Guided Selection Method of Hearing Aid Fitting, Audecibel 40 (1): 16-20.
Massaro, D. W. (1987). Speech Perception by Ear and Eye. Lawrence Erlbaum Associates.
Moore, B.C.J. (1989). An Introduction to the Psychology of Hearing. Academic Press.