Skip to content

LESSON 189

NAL-NL2 and DSL v5

How do we decide how much gain a hearing aid should provide? In this lesson we explain the logic of prescriptive formulas and target gain, and compare the speech intelligibility focus of NAL-NL2 with the audibility focus of DSL v5, especially for children.

Watch on YouTube

Follow along

Hello, and welcome to the AIHEAR Audiology Academy.
In this lesson we will learn how the amount of gain a hearing aid should provide is determined.
We will compare the two most widely used prescriptive formulas, NAL-NL2 and DSL v5, and their goals.
A prescriptive formula is a set of rules that calculates the gain a hearing aid should provide from a person's hearing thresholds.
The aim is a scientifically defensible starting point, rather than leaving the fitting to trial and error.
The formula gives a separate target for each frequency, because hearing loss usually differs across frequencies.
Because today's hearing aids use compression, targets are calculated separately for soft, medium and loud input levels.
In other words, a soft sound needs more gain than a loud sound.
Besides thresholds, the calculation also uses information such as age, hearing aid experience and unilateral or bilateral fitting.
The value predicted by the formula is called the prescriptive target.
The best-known early example of prescription is the half-gain rule proposed by Lybarger in 1944.
By this rule, the hearing aid should give gain equal to about half the threshold loss at each frequency.
In later years, linear formulas such as NAL-R adjusted the balance across frequencies to the speech band.
But linear formulas were designed for a single input level; hearing aids with compression needed new approaches.
Australia's National Acoustic Laboratories published the nonlinear NAL-NL1 and, in 2011, NAL-NL2.
The DSL method, developed at the University of Western Ontario in Canada, reached version DSL v5 in 2005.
Today these are the two generic formulas most often met in the clinic.
This example shows a right ear loss sloping toward the high frequencies, from mild to moderately severe.
Compared with the normal range, it is clear that most high-frequency speech sounds are not heard.
When this audiogram is entered into a formula, the result is a target curve with more gain at high frequencies.
But for the same audiogram, NAL-NL2 and DSL v5 recommend different targets.
Even the same formula can give a different target when the person's age or hearing aid experience changes.
These differences arise because the two formulas are actually trying to answer different questions.
NAL-NL2 is a nonlinear formula developed by Australia's National Acoustic Laboratories.
Its main goal is to maximize speech intelligibility for a given sound input.
In doing so, it makes sure overall loudness does not exceed what a person with normal hearing would perceive.
The formula was derived using a speech intelligibility model together with a loudness model.
Empirical corrections from studies with hearing aid users were also added.
As a result, NAL-NL2 prioritizes the regions that contribute most to intelligibility, rather than making every sound audible.
Its targets can be shown as real-ear insertion gain or as output level in the ear canal.
An important feature of NAL-NL2 is that it limits gain at frequencies where the loss is very severe.
Research showed that in regions of severe loss, the ear extracts less information from sound even when it is made audible.
So giving a lot of gain in that region may not bring the expected benefit to intelligibility.
Moreover, extra gain increases loudness and the risk of feedback.
The formula strikes this balance with effective audibility: what matters is not hearing a sound, but its contribution to intelligibility.
This approach makes NAL-NL2 strong in balancing comfort and intelligibility, especially for adult users.
So at a frequency with a very severe loss, a target gain lower than the threshold loss would suggest is not an error.
NAL-NL2 adjusts the target by small amounts based on certain personal information.
For people new to hearing aids, especially with moderate and severe losses, starting gain is kept slightly lower.
For children, more gain is prescribed than adults prefer.
Studies found that female users on average preferred slightly less gain, so this is also taken into account.
With aids in both ears, less gain is needed than for a unilateral fitting, because the two ears sum loudness.
So every piece of information entered in the software must be correct for the target to be correct.
DSL stands for Desired Sensation Level.
Its main goal is to deliver speech across a wide frequency band at an audible and comfortable level.
The method was first developed for children, because a child who is learning language cannot learn a sound they cannot hear.
So DSL places importance on bringing as much of speech as possible above threshold.
Targets are usually given as sound pressure level in the ear canal, in dB SPL.
Because children's ear canals are small, the measured real-ear-to-coupler difference, the RECD, is taken into account.
DSL v5 also has separate targets for adults, and these are lower than the targets for children.
Separate targets can also be defined for quiet and noisy environments.
This drawing shows the region that speech sounds occupy on the audiogram.
Sounds below threshold, such as the high-frequency consonants s, sh and f, are not heard without amplification.
Missing these consonants makes it harder to tell similar words apart, and harder for language to develop.
DSL v5 aims to bring as much of the speech band as possible above threshold and to a comfortable level.
NAL-NL2, by contrast, looks at the same picture in terms of the regions that will add the most intelligibility.
So DSL v5 generally prescribes more gain than NAL-NL2, especially for children and soft inputs.
Now let us put the two formulas side by side: NAL-NL2 was developed in Australia, DSL v5 in Canada.
NAL-NL2 centers on speech intelligibility, while DSL v5 centers on speech audibility.
NAL-NL2 tries to keep overall loudness at or below normal; DSL v5 places sound at a comfortable level.
NAL-NL2 targets can be given as gain or output; DSL v5 targets are sound pressure levels in the ear canal.
As a general trend, DSL v5 prescribes more gain than NAL-NL2.
Knowing these differences helps you interpret the target curve you see on the measurement screen correctly.
This is also why it is not surprising that the target curve shifts noticeably when you change the formula in the software.
Nonlinear formulas give targets separately for at least three different input levels.
For soft speech, an input of 50 to 55 dB SPL is usually used.
For average speech, 65 dB SPL is usually used.
For loud speech, an input of 75 to 80 dB SPL is usually chosen.
The difference between the soft and loud targets shows how much compression the hearing aid needs to apply.
In addition, the hearing aid's maximum output must be limited so that it does not exceed the person's discomfort level.
Getting close to target at all three levels is important for the aid to work consistently in quiet and in noise.
Both formulas are generic, research-supported formulas that are independent of manufacturers.
DSL v5 is widely used for children, especially in North America; NAL-NL2 is also used for children, notably in Australia.
Comparative studies have not shown a consistent advantage of one formula over the other in overall outcomes.
Manufacturers' own first-fit settings, however, can differ from generic formulas and often fall below target.
So the first fit you see in the software does not mean the prescriptive target has been reached.
Whatever formula is chosen, it must be verified with real-ear measurement that the target is actually met in the ear.
The target is a starting point; fine-tuning based on the user's feedback comes later.
A tip from the field: the prescriptive target is only as accurate as the information you enter in the software.
If the date of birth is entered wrongly, an adult may get a child target, or a child an adult target.
If bone conduction thresholds are not entered, the target may be miscalculated for a loss with a conductive component.
Which earphone the thresholds were measured with, and whether the fitting is unilateral or bilateral, also change the target.
Reviewing these fields with a short checklist before fitting shortens the fine-tuning time later.
In this lesson we saw that prescriptive formulas calculate targets from thresholds for each frequency and input level.
NAL-NL2 aims to improve speech intelligibility while keeping loudness normal.
DSL v5, especially for children, aims to make speech audible and comfortable across a wide band.
Whichever formula is chosen, the target must be verified with real-ear measurement.
Test yourself: why does NAL-NL2 limit gain at frequencies where the loss is very severe?
The correct answer is B: because in these regions, increasing audibility contributes less to intelligibility.
Using effective audibility, the formula directs gain to the regions where it will help most.
In the next lesson we will see how real-ear measurement checks whether the prescriptive target is actually met in the ear.
Subscribe to the channel so you never miss a new lesson. See you next time!

Keywords

Click a keyword to search all lessons for that term.

Prerequisites

LESSON 188 Who Is a Hearing Aid Candidate? Needs Assessment Coming soon
LESSON 181 Compression in Hearing Aids: WDRC, Kneepoint, Ratio Coming soon

Comments

Questions and comments appear after approval.

No comments yet. Be the first to comment.