If your hearing aids are professionally programmed but you still can’t make out the difference between “cats” and “cat,” or speech sounds sharp and garbled when people raise their voice, the problem might not be insufficient amplification. You may have cochlear dead regions: areas where the inner hair cells, or the nerve fibres connected to them, are no longer functioning, and where turning up the volume doesn’t restore clarity. It creates distortion instead.
This isn’t about needing a stronger hearing aid. It’s about needing a fundamentally different strategy.
What Actually Happens in a Dead Region
The cochlea contains about 3,500 inner hair cells (plus roughly three times as many outer hair cells), arranged by frequency like keys on a piano. High-frequency sounds (the hiss of /s/, the crispness of /f/, the distinction between “she” and “sea”) are coded by hair cells in the basal turn of the cochlea, nearest the oval window. When those cells and their connecting auditory nerve fibres die, you’re left with a frequency range that’s essentially offline.
Here’s where it gets counterintuitive. If we amplify sound at 4000 Hz and there are no functioning hair cells at 4000 Hz, the sound doesn’t just disappear. It spills over and gets picked up by neighbouring regions tuned to 3000 Hz or 2500 Hz: regions that are still alive but are now doing double duty. They’re trying to code both their own frequency and the off-frequency energy that’s washing over from the dead zone. The result is smeared, distorted sound. You hear something, but it’s not useful speech information. It’s noise.
This is why some people with severe high-frequency loss report that louder is worse. More amplification in a dead region doesn’t improve clarity: it muddies what’s left of the intact cochlea’s signal.
How Dead Regions Are Identified: The TEN Test
The most widely used clinical test for cochlear dead regions is the Threshold Equalising Noise test, or TEN(HL), developed by Brian Moore and colleagues. It’s elegant in its simplicity: the audiologist plays a specially calibrated threshold-equalising noise through the headphones while measuring your hearing thresholds again.
If your threshold in the noise is at least 10 dB above the noise level, and at least 10 dB worse than your threshold in quiet, a dead region at that frequency is likely. The noise masks off-frequency listening, so if your threshold shoots up, it tells us you were relying on neighbouring regions rather than the frequency-specific hair cells.
It isn’t needed for everyone, but if you have severe loss above 2000 Hz and poor speech clarity despite well-fitted aids, it’s worth asking about. The results directly change how hearing aids should be programmed.
Why High-Frequency Dead Regions Gut Speech Intelligibility
English grammar lives in the high frequencies. The plural /s/ in “dogs,” the possessive /s/ in “Sarah’s,” the past tense marker in “walked” (/t/): all depend on high-frequency sounds, and /s/ sits mostly between about 4000 and 8000 Hz. The fricatives /f/ and /th/ also carry much of their information in the higher frequencies. These aren’t just details. They carry meaning.
A child who can’t hear the /s/ in “two cats” may process it as “two cat” and find plural endings much harder to learn. An adult in a meeting might hear “the client want” instead of “the client wants,” losing subject-verb agreement and, with it, the confidence to track conversation in real time.
This is compounded by the fact that high-frequency sounds are low-energy. They’re quiet to begin with. If the cochlear real estate that’s meant to detect them is dead, traditional amplification, no matter how precisely fitted, can’t resurrect function that no longer exists.
Frequency Lowering: Moving Information to Where the Cochlea Still Works
If we can’t restore dead regions, the alternative is to move high-frequency information down into the frequency range where you still have functioning hair cells. This is frequency lowering, and it’s not a single technology: it’s a category that includes several very different approaches, each with trade-offs.
Nonlinear Frequency Compression (Phonak SoundRecover2)
This is the most widely researched method. SoundRecover2 compresses high-frequency sounds above a configurable cut-off frequency into a narrower range lower down. Sounds below the cut-off are left untouched.
The upside: it preserves naturalness in the mid-frequencies, where vowels and voicing information live. The downside: compression can create confusion when /s/ and /sh/ get squeezed too close together. We check using the Ling sounds (/m/, /oo/, /ee/, /ah/, /sh/, /s/) during the fitting to ensure the patient can actually discriminate them, not just detect them.
Frequency lowering shouldn’t simply be switched on by default. It’s worth considering when there’s a confirmed or suspected dead region, severe high-frequency loss, poor aided speech clarity, or a report that more high-frequency gain makes things worse rather than better.
Other approaches
Other manufacturers, including Oticon, Widex and Signia, use their own frequency-lowering methods. Some compress, like SoundRecover2. Others copy high-frequency content and add it into a lower region while leaving the original in place, giving the brain two chances at the same information. In practice, some people adapt well; others notice a metallic or slightly echoing quality, especially if some usable high-frequency hearing remains. Which approach works best depends on your audiogram, and every method needs careful verification and a few weeks of acclimatisation.
The algorithms differ, but the principle is the same: get critical speech information out of the dead zone and into living tissue.
What to Expect in the Fitting Appointment
Switching on frequency lowering isn’t a set-and-forget checkbox. At The Audiology Place, we verify it behaviourally during the session. We’ll use Ling sounds or filtered phoneme tests to ensure you can distinguish /s/ from /sh/, “she” from “sea.” If the settings are too aggressive, these collapse into the same percept. If they’re too conservative, you don’t gain audibility.
We also run real-ear measurement to confirm that the lowered energy is actually landing in a frequency region with sufficient cochlear function. There’s no point transposing a 6000 Hz /s/ down to 3000 Hz if you also have a dead region at 3000 Hz.
Adaptation takes time, often a few weeks. Your brain needs to relearn that the sound it now hears at 2500 Hz means /s/, not a mid-frequency hum. Some people adapt in days. Others need a month and a couple of follow-up tweaks.
When Frequency Lowering Isn’t the Answer
If the dead regions extend well into the mid frequencies, or if there’s severe auditory neuropathy, frequency lowering may help only marginally. There’s a floor effect: if there aren’t enough intact regions left to receive the transposed signal, moving deck chairs doesn’t save the ship.
In those cases, we talk candidly about cochlear implant candidacy. Referral is generally considered when speech understanding stays poor despite well-fitted hearing aids, and there are public and private pathways depending on your situation.
It’s a conversation worth having early, not after years of frustration.
Final Thought
Cochlear dead regions are one of the clearest examples of why amplification isn’t a one-size-fits-all solution. More volume can genuinely make things worse if it’s applied to the wrong frequency range. The right response, whether that’s frequency lowering, revised gain prescriptions, or a candidacy discussion for implants, depends on diagnostic evidence, not guesswork.
If you’ve been told your hearing is “too far gone” for aids to help, or you’ve been through multiple fittings without improvement, it’s worth asking us whether dead regions have been assessed and whether frequency-lowering strategies have been properly verified. There’s often more that can be done, but only if the clinical picture is actually understood.
