Research | May 2015 Hearing Review
A look at how NAL, DSL, and proprietary first-fit algorithms have changed since 1998
Both generic and proprietary prescriptions have over time decreased their prescribed gain, and the reduction has been largest in a fairly broad frequency range around 1 kHz. Some of the changes in the proprietary prescriptions have taken place before the generic prescriptions changed, mainly by the use of special gain settings for first-time hearing aid users. It also appears that many hearing aid manufacturers are now providing default gain similar to NAL-NL2 and DSL v5. The amount of gain reduction for first-time hearing aid users varies among manufacturers.
The initial part of a hearing aid fitting is typically performed using a prescriptive approach. Calculations, most often based on hearing threshold levels, are used to derive hearing aid settings that are appropriate for a large number of hearing aid users with the same audiogram. This initial setting may later be fine-tuned to account for individual loudness growth, practical limitations, or individual preferences.
Prescriptions can either be generic, such as the NAL and DSL prescriptions, or proprietary and specific to a hearing aid manufacturer. The generic prescriptions are often based on an explicit theory, sometimes supplemented with empirical findings. Manufacturers’ proprietary prescriptions may also include device-specific considerations in their formulations.
The latest versions of the NAL and DSL prescriptions, NAL-NL2 and DSL v5 (adult version), provide less gain than their previous versions. The changes have been made partly based on empirical data.1-3 The previous versions of these generic prescriptions differed greatly in prescribed gain, with DSL v4 prescribing substantially more gain than NAL-NL1. The latest version of the two prescriptions have become much more similar,4 mainly because DSL v5 prescribes substantially less gain than DSL v4 for adult hearing aid users.¹
In the late 1990s, some hearing aid manufacturers introduced “gain adaptation” or “acclimatization” steps in their fitting software. The term acclimatization was originally used to describe a potential increase in a hearing aid user’s aided performance over time without any changes in hearing aid settings.5
At the beginning of 2000, the term was also used to describe the process whereby a first-time hearing aid user adapts to the aided condition. “Gain adaptation managers” or “acclimatization steps” reduce the prescribed gain and adjust other hearing aid parameters to facilitate hearing aid use for inexperienced users.6 With the introduction of NAL-NL2 in 2011, the NAL method now also prescribes lower gain for first-time users.
The described changes in generic prescriptions over the years prompted us to also examine gain changes in proprietary prescriptions. Specifically, we wanted to answer the following three questions:
1) How different are the proprietary prescriptions and have these differences increased or decreased over time?
2) Is the gain reduction seen recently for generic prescriptions mirrored in the proprietary prescriptions?
3) Do modern hearing aids implement gain reduction for first-time hearing-aid users in a way that is similar to the adjustments in the NAL-NL2 prescription?
An understanding of how the prescribed gain has changed over time might help clinicians interpret comments from long-term hearing aid users who might be used to something else than what default prescriptions in today’s hearing aids offer, or first-time hearing aid users, who might or might not need initial gain reduction depending on the prescribed starting point.
Summary of Measurements
On three occasions—in 1998, 2008, and 2013—we have measured hearing aid gain. At each point in time, various modern hearing aids of the type classified as top-end products were selected (Table 1).
Table 1. Hearing aids used in the three measurement studies. After each hearing-aid name, the two test settings, representing prescriptions for experienced and inexperienced hearing-aid users, are described where applicable.
They were all fitted according to the manufacturers’ suggested first fit, using the default prescriptive method, for a gently sloping audiogram (Figure 1). One hearing aid from each manufacturer was used and programmed for a monaural fitting. Hearing aid microphones were used in omnidirectional mode and all other features were used in their default settings.
After programming the hearing aids, coupler gain measurements were performed in a test box. The measured coupler-gain data were re-calculated to estimated insertion gain according to a method described by Smeds and Leijon.7
Various measurement signals were used in the three studies. These signals were selected to represent the following three listening situations with speech at various levels:
- A) Low-level speech in a quiet background;
- B) Average-level speech in a quiet or moderately noisy background of competitive talkers, and
- C) High-level speech in a noisy background of other talkers.
The NAL Prescriptions
All measurements are compared to the generic gain prescriptions developed at the National Acoustic Laboratories (NAL) in Sydney, Australia. At the time of the first measurements in 1998, NAL had not developed a prescription for non-linear hearing aids and the available prescription was NAL-R for linear hearing aids.8 At the time of the second measurements in 2008, NAL’s first version of a prescription for non-linear hearing aids (NAL-NL1) was commonly used.9 At the time of the third measurements in 2013, NAL’s second version of their prescriptive method for nonlinear hearing aids (NAL-NL2) was available.2,3
For speech at normal presentation levels (65 dB SPL), NAL-NL1 prescribes gain that is similar to the NAL-R prescribed gain. NAL-NL2 later introduced changes in gain-frequency response shape such that for the audiogram in Figure 1, the overall gain was reduced, mainly by reducing the gain between 700 Hz and 2 kHz. For frequencies above 3 kHz, NAL-NL2 prescribes slightly more gain than NAL-NL1. In NAL-NL2, the gain prescribed for an inexperienced hearing-aid user is decreased 3 dB compared to the gain prescribed for an experienced hearing-aid user (Figure 2).
In the following figures, the NAL prescriptions are used as “anchor points.” They are used for easier comparison of the measured gain between graphs. The NAL prescription should not be interpreted as providing the “correct gain” to which the proprietary prescriptions should be compared. Since the measurement conditions vary slightly between the three studies, level-appropriate NAL prescriptions were derived.
The DSL Prescriptions
In Figure 6 comparisons will also be made to the generic gain prescriptions developed at the University of Western Ontario, London, Ontario, Canada. At the time of the first measurements in 1998, DSL[i/o] v4 for wide-dynamic range compression hearing aids10 was available and implemented in one of the hearing aids included in the study (Table 1). At the time of the second and third measurements, 2008 and 2013, a new version, the DSL m[i/o] v5, was available.¹
For adult hearing-aid users, the prescribed gain was substantially reduced in the DSL v5 version, compared to the DSL v4 prescribed gain (Figure 2). The prescribed gain for infants and children was kept more similar to the DSL v4 gain.
The results from the 1998 measurements are shown in Figure 3. Six hearing aids were measured (Table 1). Generally, there was a large spread of almost 20 dB among the proprietary prescriptions. For a 65 dB SPL input, NAL-NL1 falls almost in the middle of the range of proprietary prescriptions for frequencies up to approximately 2 kHz. Higher than NAL-NL1 prescribed gain was provided for frequencies above 2 kHz. Considerably higher gains were provided by these hearing aids at 4 kHz. For the 55 and 76 dB SPL input signals, the measured gain resembles the NAL-NL1 prescribed gain over a larger frequency range.
The results from the 2008 measurements are shown in Figure 4. Ten hearing aids were measured (Table 1). The left panels show the measurement data when the hearing aids were programmed for an experienced hearing-aid user and the right panels show results of the measurements when the hearing aids were programmed for an inexperienced user (for the hearing aids that had that option).
Compared to the 1998 measurements, the spread in the data is slightly reduced, especially around 2 kHz. For the lowest input level (top panels), the gain prescribed by the proprietary methods was similar to NAL-NL1 prescribed gain when the hearing aids were programmed for an experienced hearing-aid user. Considerably less than NAL-NL1 prescribed gain was prescribed for an inexperienced hearing-aid user.
When the hearing aids were programmed for experienced hearing-aid users and the measurements were made using mid- and high-level speech (middle and lower left panels), the gain prescribed by the proprietary methods was similar to NAL-NL1 prescribed gain in the mid-frequency region, whereas more than NAL-NL1 prescribed gain was seen in the high-frequency range. When the hearing aids were programmed for inexperienced hearing-aid users (middle and lower right panels), the gain prescribed by the proprietary methods was similar to the NAL-NL1 prescribed gain in the high-frequency range, whereas the gain was lower than that prescribed by NAL-NL1 in the mid-frequency range.
The results from the 2013 measurements are shown in Figure 5. Six hearing aids were measured (Table 1). The left panels show the prescribed gain when the hearing aids were programmed for an experienced hearing-aid user and the right panels the results of the measurements when the hearing aids were programmed for an inexperienced user.
The spread in the data is smaller than in previous measurements. The gains prescribed for experienced hearing aid users (left panels) are generally below the NAL-NL1 prescribed gain for frequencies up to around 2 kHz and higher than NAL-NL1 prescribed gain above that. The gains prescribed for inexperienced users (right panels) were generally well below the NAL-NL1 prescribed gain, especially in a broad range around 1 kHz.
The gain prescribed by NAL-NL2 (for experienced and inexperienced hearing-aid users) was also added in these graphs for comparison. For experienced hearing-aid users (left panels), the proprietary prescriptions are similar to the NAL-NL2 prescription for low- and mid-level speech input. For inexperienced users, the proprietary prescriptions are very similar to the NAL-NL2 prescribed gain for a 65 dB SPL speech signal (middle, right panel), whereas lower gains are prescribed for the low-level input (top, right panel). For the highest input (bottom panels), the proprietary prescriptions provide more high-frequency gain than NAL-NL2, both for experienced and inexperienced hearing-aid users.
Comparison of Prescriptions Over Time
To facilitate a comparison of generic and proprietary gain prescriptions over time, the average gain in 1/3-octave bands around 0.5, 1, 2, and 4 kHz was calculated for each measured hearing aid. Based on these data, statistics were calculated across hearing aids. Comparisons were then made with both the NAL and DSL prescriptions. The results for a 65 dB SPL speech input are shown in Figure 6.
The boxes in Figure 6 illustrate the range of prescribed gain for the proprietary methods. The gray boxes illustrate the gain when the hearing aids were programmed for an experienced hearing-aid user. Over time, the median gain at 2 and 4 kHz has been fairly constant, but the spread in the data has generally been reduced. The largest change over time is seen at 1 kHz, where the gain has been substantially reduced in the 2013 measurements.
The white boxes in Figure 6 illustrate the prescribed gain for the proprietary methods when the hearing aids were programmed for an inexperienced hearing aid user. With this setting the median gain was reduced 5-6 dB both in the 2008 and the 2013 measurements except at 500 Hz, where the reduction was smaller, probably because the gain was lower. At some frequencies, the white boxes are wider than the gray boxes, indicating that the hearing aid manufacturers have different opinions on how much the gain should be reduced for first-time hearing aid users.
The generic NAL and DSL prescriptions are also included in the graph. The NAL prescriptions are illustrated by blue lines and the DSL prescriptions by red lines. A black diamond indicates when these generic prescriptions have changed.
DSL v4 prescribes very high gain values. DSL v5 was introduced in 2005 and the new prescription provides substantially lower gain. NAL-NL1, which was introduced in 2001, prescribes gain very similar to NAL-R for a 65 dB SPL speech input. When NAL-NL2 was introduced in 2011, a substantial gain reduction was seen at 1 and 2 kHz, whereas the gain at 4 kHz was increased. The proprietary prescriptions had, in median and at all measurements, provided gain at 4 kHz that was very similar to the higher gain now prescribed by NAL-NL2.
With the changes in the NAL and DSL prescriptions (for the audiogram and input levels used in this comparison), NAL-NL2 and DSL v5 today prescribe very similar gain at 1, 2 and 4 kHz, whereas DSL v5 prescribes higher gain than NAL-NL2 at 500 Hz.
These measurements and comparisons allow us to answer the questions we posed at the beginning of the paper. Again, it is important not to think of the generic prescriptions included as providing the “correct” gain to which the proprietary prescriptions should be evaluated, but rather as well-documented references.
Research Question #1: How different are the proprietary prescriptions? Generally, the proprietary prescriptions are more similar today than they were in 1998. There was almost a 20 dB gain difference between the hearing aids prescribing the highest and the lowest gain in 1998. For the default proprietary prescriptions in 2013, this difference was around 10 dB. This is a clinically important result since it indicates that there is today less difference in prescribed gain among manufacturers’ first fit than it was some years ago. However, this does not mean that the hearing aids are similar with respect to other features or performance.
In 2013, the gain prescribed by the proprietary methods for experienced hearing-aid users (Figure 5, left panels) showed, in median, a good fit to NAL-NL2 up to 5 kHz for both 55 and 65 dB SPL inputs, whereas all hearing aids gave higher gain than prescribed by NAL-NL2 for the 76 dB input signal (Figure 5, left bottom panel).
This result is somewhat in contrast to the results presented by Sanders et al,11 where they showed that measured real-ear gain for five hearing aids programmed using NAL-NL2 did not show good correspondence with independently derived NAL-NL2 targets. The two studies differ in design. Sanders et al performed real-ear measurements—which means that the natural variation among ears is included—whereas the current study used coupler gain measurements and re-calculated the results to estimated REIG.
Research Question #2: Is the gain reduction seen recently for generic prescriptions mirrored in the proprietary prescriptions? Yes, generally the default prescribed hearing-aid gain has been reduced recently. The gain reduction has been largest in a broad frequency range around 1 kHz. Prior to changing the default gain, hearing aid manufacturers have implemented fairly substantial gain reduction by introducing particular settings for first-time hearing aid users.
This has clinical implications. A first-time hearing aid user might not need to have the initial gain reduced as much as the clinician might think, based on previous experience with amplification. A long-term hearing aid user, on the other hand, might think that the default prescribed gain in a new hearing aid is too low, especially around 1 kHz.
Research Question #3: Do modern hearing aids implement gain reduction for first-time hearing aid users in a way that is similar to the adjustments in the NAL-NL2 prescription? No, the difference in prescribed gain for experienced and inexperienced hearing aid users in the proprietary prescriptions is generally larger than this difference in the NAL-NL2 prescription (Figures 2 and 6). This could potentially be the result of thinking about acclimatization or gain adaptation on different time scales.
When NAL-NL2 was developed and the difference in prescribed gain for inexperienced and experienced hearing-aid users was introduced, the developers referred to studies where a small gain adaptation effect was seen over several months, up to 1 year.³ There is reason to think that the first-time hearing aid users did some “acclimatizing” to their new aided condition quickly, over days or weeks, and that this initial acclimatization was not incorporated in the results.
For some hearing aid manufacturers, on the other hand, the goal is to facilitate initial hearing aid use for first-time hearing aid users. Some manufacturers provide automatic gain adaptation, where the gain is automatically increased typically over one month. These manufacturers usually reduce the initial gain substantially, but bring it up automatically over a short time.
However, it also seems as if some hearing aid manufacturers are compensating for the same type of long-term gain adaptation as the developers of the NAL-NL2 prescription. These manufacturers typically reduce the gain for first-time hearing aid users by only 2-3 dB, and they usually do not use automatic gain adaptation procedures.
Both generic and proprietary prescriptions have over time decreased their prescribed gain, and the reduction has been largest in a fairly broad frequency range around 1 kHz. Some of the changes in the proprietary prescriptions have taken place before the generic prescriptions changed, mainly by the use of special gain settings for first-time hearing aid users. It also appears that many hearing aid manufacturers are now providing default gain that is fairly similar to NAL-NL2 and DSL v5 prescribed gain. The amount of gain reduction for first-time hearing aid users varies among manufacturers.
The authors thank Francis Kuk, Arne Leijon, Florian Wolters, and Erik Schmidt for useful comments on earlier versions of this article.
Scollie S, Seewald R, Cornelisse L, Moodie S, Bagatto M, Laurnagaray D, Beaulac S, and Pumford J. The Desired Sensation Level multistage input/output algorithm. Trends Amplif. 2005;9(4):159-97. doi: http://dx.doi.org/10.1177/108471380500900403
Keidser G, Dillon H, Flax M, Ching T, Brewer S. The NAL-NL2 prescription procedure. Audiol Research. 2011;1(e24):88-90. doi: http://dx.doi.org/10.4081/audiores.2011.e24
Keidser G, Dillon H, Carter L, O’Brien A. NAL-NL2 empirical adjustments. Trends Amplif. 2012;16(4):211-23. doi: http://dx.doi.org/10.1177/1084713812468511
Johnson EE, Dillon H. A comparison of gain for adults from generic hearing aid prescriptive methods: Impacts on predicted loudness, frequency bandwidth, and speech intelligibility. J Am Acad Audiol. 2011;22(7):441-59. doi: 10.3766/jaaa.22.7.5.
Gatehouse S. The time course and magnitude of perceptual acclimatization to frequency responses: evidence from monaural fitting of hearing aids. J Acoust Soc Am. 1992. 92(3):1258-68. doi: http://dx.doi.org/10.1121/1.403921
Eberwein CA, Mack HA, Pallett S, and Lindley GA. Adaptation managers in hearing aid fitting software. Seminars in Hearing. 2001;22(2):199-207. doi: http://dx.doi.org/10.1055/s-2001-14982
Smeds K, Leijon A. Threshold-based fitting methods for non-linear (WDRC) hearing instruments–comparison of acoustic characteristics. Scand Audiol. 2001;30(4):213-22. doi: http://dx.doi.org/10.1080/01050390152704733
Byrne D, Dillon H. The National Acoustic Laboratories’ (NAL) new procedure for selecting the gain and frequency response of a hearing aid. Ear Hear. 1986. 7(4):257-65. doi: http://dx.doi.org/10.1097/00003446-198608000-00007
Byrne D, Dillon H, Ching T, Katsch R, Keidser G. NAL-NL1 procedure for fitting nonlinear hearing aids: Characteristics and comparisons with other procedures. J Am Acad Audiol. 2001;12(1):37-51.
Cornelisse LE, Seewald RC, Jamieson DG. The input/output formula: A theoretical approach to the fitting of personal amplification devices. J Acoust Soc Am. 1995. 97(3):1854-64. doi: http://dx.doi.org/10.1121/1.412980
Sanders J, Stoody T, Weber JE, Mueller HG. Manufacturers’ NAL-NL2 fittings fail real-ear verification. Hearing Review. 2015. 21(3):24-32. Available at: http://www.hearingreview.com/2015/02/manufacturers-nal-nl2-fittings-fail-real-ear-verification
Karolina Smeds, PhD, is director at Widex research group ORCA Europe, Stockholm, Sweden, where Martin Dahlquist, MSc, is a research engineer, Josefina Larsson, BSc, and Sofia Herzman, BSc, are research audiologists, and Sara Båsjö, BSc, previously worked as a research audiologist. Carsten Paludan-Müller, MSc, is an audiological research engineer at Widex A/S, Lynge, Denmark.
Correspondence may be addressed to [email protected]
Original citation for this article: Smeds, K et al. Proprietary Hearing Aid Gain Prescriptions: Changes Over Time. Hearing Review. 2015;22(5):16.