_edited.png)
How Are Hearing Aids Programmed? A Guide to Prescription Formulas
Prescription hearing aids aren't off-the-shelf amplifiers. When we fit you at Ear Care Specialists in Sherwood, Nottingham, your hearing aids are programmed to a prescription worked out from your own hearing test, your ears and details about you (age, sex, previous hearing aid experience etc.), in much the same way glasses are made to an eye prescription. Patient's don't always know this.
Turning everything up evenly would make some sounds uncomfortably loud and still leave the quiet speech sounds you need out of reach. So we use a prescription, and then we check it in your own ear.
In this article we explain how we programme hearing aids, what a prescription formula is, and how the main formulas used in the UK differ: NAL-NL2 and the new NAL-NL3 from Australia's National Acoustic Laboratories, DSL v5 from Canada, and the manufacturers' own formulas from Phonak and ReSound. We'll also be honest about what these formulas can and can't do, because hearing aids help a great deal, but they aren't a miracle cure.
What information do we use to programme your hearing aids?
Your audiogram is the starting point. It is the chart from your hearing test that shows the quietest sound you can hear at each pitch, from low (250 Hz) to high (8,000 Hz), in each ear. Most age-related hearing loss affects the high pitches first, which is where consonants such as "s", "f" and "th" sit. That is why people often say they can hear others talking but cannot make out the words.
But the audiogram is not the whole story. Modern prescription formulas also take account of:
Whether your loss is sensorineural, conductive or mixed. Bone-conduction results tell us whether part of the loss is in the outer or middle ear, which changes how much amplification is needed.
Your age. Children and adults are prescribed differently, because children need to hear every speech sound clearly to learn language.
Your sex. NAL research on 187 adults found that women, on average, preferred about 2 dB less gain than men, so NAL-NL2 prescribes slightly differently for each (Keidser et al., 2011).
Your experience with hearing aids. New users tend to prefer less amplification at first, so the formula can build gain up gradually as your brain adapts.
One or two hearing aids. Sound from two ears adds together in loudness, so a two-aid fitting needs a little less gain per ear.
The language you speak. Tonal languages such as Mandarin carry meaning in pitch, and NAL-NL2 allows for this.
Your ears themselves. The size and shape of your ear canal, and whether you wear an open dome or a closed mould, change the sound reaching your eardrum.
Uncomfortable loudness levels (the point at which sound becomes too loud) and your own listening goals also shape the final settings. The British Society of Audiology's 2025 guidance on person-centred care puts your priorities at the heart of the rehabilitation plan (BSA).
What is a hearing aid prescription formula?
A prescription formula (also called a fitting rationale) is a research-based calculation that turns your hearing test results into precise targets: how much amplification, or gain, each pitch should receive for soft, average and loud sounds.
Why not simply "fill in" the hearing loss? With the most common type of hearing loss, damage to the inner ear means loud sounds can feel almost as loud as they do to anyone else, even though soft sounds are missed. Audiologists call this loudness recruitment. Professor Harvey Dillon explains in Hearing Aids, the standard textbook for audiologists, that a good prescription must balance two goals that pull against each other: making speech audible and understandable, and keeping overall loudness comfortable (Dillon, 2012).
This is why modern hearing aids use compression. Quiet sounds get a lot of help, average sounds get moderate help and loud sounds get very little. The prescription formula decides exactly how much, in each frequency band, for your hearing.
Fitting software turns your audiogram into prescription targets (dashed lines) for soft, average and loud speech. Real ear measurement (solid lines) shows what your hearing aid actually delivers, so we can match the two.
A prescription is a starting point, not the be-all and end-all
It's worth being clear about what a prescription formula actually gives us. It produces "theoretical" amounts of gain: the amplification that research predicts should help the average person with a hearing test like yours hear speech clearly and comfortably. That's a very good, evidence-based place to start, but it is still a prediction.
A few honest points we share with every patient:
Hearing aids don't restore normal hearing. With inner-ear hearing loss, the problem isn't only volume. Sounds can also be less clear, and background noise is harder to filter out. Hearing aids make speech more audible and easier to follow, but busy restaurants and group conversations can still be hard work.
Formulas are based on averages. Two people with the same audiogram can hear and prefer sound very differently. Your own experience, the shape of your ear and how your brain processes sound all matter.
Your brain needs time. If you've had a hearing loss for a while, everyday sounds can seem sharp or loud at first. Most people settle over a few weeks of regular wear, which is why we often build the amplification up gradually. Our guide to hearing aid acclimatisation explains what to expect.
Realistic expectations lead to better results. Research consistently shows that hearing aids improve communication and quality of life, but the best outcomes come from wearing them every day, good follow-up care, and sometimes extra help such as communication tactics or a remote microphone.
That's why, at Ear Care Specialists, the prescription is only the first step. We then objectively verify it with real ear measurement, which is a measurement in your own ear rather than an estimate on a computer screen. After that we fine-tune it around what you tell us about real life.
The NAL prescriptions: from NAL-R to NAL-NL3
The National Acoustic Laboratories (NAL), the research arm of Hearing Australia, has produced the world's most widely used hearing aid prescriptions for nearly 50 years. Every version has shared the same aim: make speech as clear as possible while keeping overall loudness comfortable.
Year | Prescription | What changed |
2025–26 | NAL-NL3 | Unveiled in March 2025; first clinical fittings in March 2026. Built with machine learning on millions of real-world fittings, plus optional modules for specific needs. |
2011 | NAL-NL2 | Refined speech-intelligibility model built on about 1,000 participants; added sex, experience, language and one-vs-two-aid adjustments. |
1999 | NAL-NL1 | First NAL formula for non-linear (compression) hearing aids, built on about 300 participants. |
Late 1970s–1980s | NAL / NAL-R | Original linear formula for a single listening level. |
NAL-NL2: the current standard
NAL-NL2 was derived by computer modelling of which gain settings give the best predicted speech understanding for hundreds of audiograms. The results were then adjusted using a decade of real-world data from NAL-NL1 wearers. Compared with NL1, it gives relatively more gain in the low and high pitches and less in the middle. It also recommends that new users with more than a mild loss start lower and adapt upwards (Keidser et al., 2011).
NAL-NL2 is the prescription the British Society of Audiology recommends for adults alongside DSL v5 (BSA, 2018).
NAL-NL3: the rationale for change
NAL-NL2 worked well for typical sloping losses, but it was built on population averages. NAL's own researchers note it could over- or under-amplify less common hearing patterns, such as rising (reverse-slope), notched, "cookie-bite" and mixed losses (ENT & Audiology News, 2025).
NAL-NL3 addresses this in three ways:
A new core prescription trained on real-world data. Deep-learning models were trained on millions of hearing aid fittings, with updated modelling of speech perception. Before launch it was trialled at 30 Australian sites with more than 170 clients and 600 real ear measurements, showing better clinician usability and patient preference.
Modules for specific needs. NAL has moved from one formula for everyone to a modular system. The first two modules are:
Comfort in Noise, which adjusts gain and compression for people who find noisy places tiring or uncomfortable, while protecting speech clarity.
Minimal Hearing Loss, for people with a near-normal hearing test who still struggle to follow speech in noise. It gives just enough gain for directional microphones and noise reduction to work. NAL describes it as a support tool rather than a true prescription.
Room to grow. More modules are planned, so the system can be updated as evidence develops.
NAL-NL3 has been licensed to manufacturers covering more than 90% of the global hearing aid market (Hearing Australia, 2026). ReSound was first to add it to its fitting software in March 2026 (GN, 2026), and other manufacturers are expected to follow. Ask us whether it is available for your hearing aids. This is now the new default standard for many audology clinics.
DSL v5: the prescription designed with children in mind
The Desired Sensation Level (DSL) method was developed in the 1980s by Professor Richard Seewald and colleagues at Western University in Ontario, Canada. It was built for babies and young children who cannot tell us how hearing aids sound (Western University).
Its guiding principle is different from NAL's. Where NAL aims to maximise understanding at a comfortable loudness, DSL aims to make the full range of speech sounds audible at a comfortable level, because a developing brain needs that access to learn speech and language.
The current version, DSL v5, was published in 2005 by Professor Susan Scollie and colleagues. It has separate protocols for infants, children and adults (Scollie et al., 2005). For babies, it can work from estimated hearing levels measured during sleep with an ABR test. Adult targets are set lower than children's, by about 7 dB for a moderate loss, reflecting adults' preference for less loudness (AudiologyOnline). There is also a DSL v5 "Noise" setting for noisy places such as classrooms.
NAL vs DSL: what does the evidence say?
NAL-NL2 / NAL-NL3 | DSL v5 | |
Developed by | National Acoustic Laboratories, Australia | Western University, Canada |
Main goal | Best speech understanding at comfortable overall loudness | Audibility of the whole speech range for language development |
Typical gain | Lower overall, especially for soft sounds | Higher, especially for soft and high-pitched sounds |
Most common use in the UK | Adults | Babies and children (also has adult targets) |
The largest head-to-head trial is Australia's LOCHI study, which randomly assigned children to NAL or DSL at their first fitting. At three years, there were no significant differences in language, speech production or functional performance (Ching et al., 2013). At five years, DSL gave more audibility for soft sounds and parents rated everyday functioning higher, but speech and language scores were still similar (Ching et al., 2018).
A UK study of 44 children with moderate or severe loss found that DSL's higher gain led to better detection and discrimination of soft speech sounds than NAL-NL1 (Marriage et al., 2018). The takeaway: both are valid, evidence-based prescriptions. What matters most is that the chosen target is actually met in the child's ear and checked regularly as they grow.
Manufacturers' own formulas, and where AI fits in
Alongside NAL and DSL, most hearing aid manufacturers have their own proprietary formula, which is usually the default when a hearing aid is first connected. These are tuned to each brand's sound processing and are typically designed for a comfortable, natural first impression. Two good examples are Phonak and ReSound.
Phonak: Adaptive Phonak Digital (APD) 3.0
Phonak introduced its own formula, Adaptive Phonak Digital, in 2005. APD 2.0 arrived with the Audéo Paradise in 2020 and added adaptive compression that switches speed depending on the sound around you; Phonak also brought its targets closer to NAL-NL2 (Phonak).
The current version, APD 3.0, launched with the Infinio range. It slightly boosts the mid frequencies around 1 kHz and eases back on 2–4 kHz compared with APD 2.0. It also uses a separate frequency curve for streaming, inspired by consumer headphone tuning. In Phonak's own research, 93% of first-time users preferred APD 3.0 to a competitor's first fit (Phonak).
Phonak's AI sits mainly around the formula rather than inside it. The Infinio Sphere range has a dedicated DEEPSONIC AI chip that separates speech from noise, and Phonak Target fitting software includes an AI-based dome recommendation drawn from thousands of past fittings.
ReSound: Audiogram+ and NAL-NL3
ReSound's own formula, Audiogram+, is based on neither NAL nor DSL. It grew out of ReSound's loudness-based fitting system of the late 1980s and was refined by analysing thousands of ReSound fittings alongside wearers' questionnaire responses. It generally gives less high-frequency gain than NAL-NL2. In a ReSound study, NAL-NL2's extra high-frequency emphasis did not improve speech-in-noise scores, and wearers showed no strong preference after everyday use (GN ReSound, via AudiologyOnline, 2025).
ReSound's AI story centres on its hearing aids. ReSound Vivia, launched in February 2025, has a dedicated deep neural network chip for separating speech from noise, and an Intelligent Focus feature that responds to where you are looking (GN, 2025). In March 2026 ReSound also became the first manufacturer to offer NAL-NL3, itself built with machine learning, together with automated real ear measurement (AutoREM) in its fitting software (GN, 2026).
Proprietary or generic: which is better?
Proprietary formulas can sound pleasant from day one, but most of their supporting evidence comes from the manufacturers themselves, and they differ from brand to brand. The BSA recommends verifying against validated generic targets (NAL-NL2 or DSL v5). It notes that proprietary targets often differ significantly from NAL and are often missing from verification equipment (BSA, 2018). At Ear Care Specialists, we can still use a manufacturer's sound features. We simply make sure the amplification reaching your ear meets a recognised prescription.
Why we check your prescription in your own ear: real ear measurement
Choosing a formula is only half the job. The fitting software predicts what the hearing aid will do in an "average" ear, so the targets on screen are still "theoretical". Your ear canal isn't average. Real ear measurement (REM) lets us objectively check what actually reaches your eardrum, and then match the hearing aid to your prescription in your own ear.
A soft, thin probe tube is placed in your ear canal alongside the hearing aid. We then play speech-like sounds at soft, average and loud levels and adjust the hearing aid until it matches your prescription targets. It is painless, takes around 15 minutes, and you can watch the results on screen.
The evidence for this step is strong:
When five leading manufacturers' first-fit settings were measured in real ears, most fell 7–10 dB below NAL-NL2 targets above 2,000 Hz, exactly where speech clarity lives (Sanders et al., 2015, summarised in Valente et al., 2018).
In a double-blind trial of 24 first-time users, hearing aids programmed to NAL-NL2 with REM gave better recognition of soft words, and 79% of participants preferred them to the manufacturer's first fit (Valente et al., 2018).
The British Society of Audiology calls real ear measurement the most direct way to verify a fitting. It recommends matching targets within ±5 dB from 250 to 6,000 Hz where possible. For babies and young children, a related test called the real-ear-to-coupler difference (RECD) is used instead (BSA, 2018).
For more detail, read our article on why real ear measurements are essential in modern hearing aid fittings.
What to expect at Ear Care Specialists
Full hearing assessment, including ear examination, pure-tone audiometry with bone conduction and, where useful, speech-in-noise testing.
A conversation about your goals, such as the situations where you struggle and what matters most to you.
Programming to a validated prescription, usually NAL-NL3, set for your age, sex, experience and hearing aid style. Depending on your hearing and what suits you, we sometimes use Phonak's own formula (Adaptive Phonak Digital) or ReSound's own formula (Audiogram+) instead. The manufacturers design their own formulas to work hand in hand with their AI speech-in-noise processing, and say this gives better results in noisy places, so for some people this is the better choice.
Real ear measurement to make sure the prescription is met in your own ear.
Fine-tuning and follow-up, because comfort and preference matter too, and your brain needs a few weeks to adapt.
Book a hearing assessment at Ear Care Specialists in Sherwood, Nottingham, to find out how a properly prescribed and verified fitting could help you. You can also see our hearing aid packages and prices, or call us on 0115 990 20 82.
Frequently asked questions
Are hearing aids programmed to my specific hearing loss?
Yes. A professionally fitted hearing aid is programmed using your audiogram and details such as your age, sex, hearing aid experience and ear canal. The settings are calculated by a prescription formula such as NAL-NL2, NAL-NL3 or DSL v5.
What is NAL-NL3?
NAL-NL3 is the latest hearing aid prescription from Australia's National Acoustic Laboratories, replacing NAL-NL2 from 2011. It was built using machine learning on millions of real-world fittings, handles unusual hearing loss patterns better, and adds optional modules for comfort in noise and for minimal hearing loss.
What is the difference between NAL and DSL?
NAL aims for the clearest speech at a comfortable overall loudness and is mostly used for adults. DSL aims to make all speech sounds audible for language development and is mostly used for children. Both are validated and recommended by the British Society of Audiology.
What is real ear measurement?
Real ear measurement uses a thin probe tube in your ear canal to measure the sound your hearing aid delivers to your eardrum. It confirms that your hearing aid meets its prescription. It is painless and takes about 15 minutes.
Can I just buy hearing aids online or over the counter?
Pre-set hearing aids are not programmed to a measured prescription or checked in your ear. For a hearing loss that needs more than mild help, a full assessment and verified fitting give you the best chance of clear, comfortable hearing.
Will hearing aids give me normal hearing again?
No, and it's important to be honest about that. Hearing aids make speech much easier to hear and follow, but they can't repair the inner ear, and noisy places can still be challenging. A prescription that has been verified with real ear measurement and fine-tuned to you, plus a few weeks of regular wear, gives you the best realistic result.
References
British Society of Audiology (2018). Practice guidance: verification of hearing devices using probe microphone measurements.
British Society of Audiology. Current guidance documents, including Guiding Principles for Person Centred Care in Adult Hearing Rehabilitation (2025).
Dillon, H. (2012). Hearing Aids, 2nd edition. Sydney: Boomerang Press / New York: Thieme. Chapter 10, Prescribing hearing aid amplification.
Keidser, G., Dillon, H., Flax, M., Ching, T. & Brewer, S. (2011). The NAL-NL2 prescription procedure. Audiology Research, 1(1), e24.
Bennett, B., Kitterick, P. & Edwards, B. (2025). Understanding the NAL-NL3 prescription system. ENT & Audiology News, 34(4).
Pang, J., Kitterick, P. & Edwards, B. (2025). Why do HCPs need NAL-NL3? Hearing Review.
Hearing Australia (2026). Hearing Australia first in the world to provide NAL-NL3.
Strom, K. (2025). NAL-NL3 and new modules introduced. HearingTracker.
Scollie, S. et al. (2005). The Desired Sensation Level multistage input/output algorithm. Trends in Amplification, 9(4), 159–197.
Ching, T.Y.C. et al. (2013). A randomized controlled comparison of NAL and DSL prescriptions for young children: outcomes at three years. International Journal of Audiology, 52(S2), S17–S28.
Ching, T.Y.C. et al. (2018). Hearing aid fitting and developmental outcomes of children fit according to either the NAL or DSL prescription. International Journal of Audiology, 57(S2), S41–S54.
Marriage, J.E. et al. (2018). Comparison of different hearing aid prescriptions for children. Ear and Hearing, 39(1), 20–31.
Valente, M. et al. (2018). Differences in word and phoneme recognition… between manufacturer first-fit and hearing aids programmed to NAL-NL2 using real-ear measures. Journal of the American Academy of Audiology, 29(8), 706–721.
Sanders, J., Stoody, T., Weber, J. & Mueller, H.G. (2015). Manufacturers' NAL-NL2 fittings fail real-ear verification. Hearing Review, 21(3), 24.
Phonak (2024). Phonak Infinio's new fitting formula, APD 3.0. Manufacturer publication.
Phonak (2020). Proprietary fitting formulas: APD 2.0. Manufacturer publication.
Groth, J. (2025). Is the ReSound Audiogram+ fitting rule based on NAL or DSL? AudiologyOnline.
GN Hearing (2025). ReSound Vivia launch. Press release.
GN Hearing (2026). World's first with NAL-NL3. Press release.
Western University. History of the DSL Method.
Comments