Tinnitus bothers some people more than others largely because of how they respond to it, not how loud it is: measured loudness predicts distress weakly or not at all, while fear, catastrophising, low mood and hyperacusis track it more closely.
That pattern has turned up repeatedly since the 1980s, and it shapes almost everything about how tinnitus is assessed and treated. It explains why the standard questionnaires score impact rather than volume, why the best-supported treatment makes no claim to quieten the sound, and why two people with what seems to be the same ringing can live such different lives with it.
Loudness measured in a sound booth barely predicts distress
Tinnitus loudness can be measured by matching: a clinician plays an external tone and adjusts it until it sounds as loud as the tinnitus. If distress were mainly about volume, that number should predict how badly people are affected. It mostly does not.
- More than 1,800 clinic patients, 1984. Patients rated their tinnitus severity from 1 to 10 and had their loudness matched against external sounds at the tinnitus pitch. There was no correlation between rated severity and matched loudness, and severity was not related to the type, quality or pitch of the sound either. What severity did correlate with, strongly, was sleep disturbance.
- 18 patients, 2003. The answer depended on how the match was expressed. Loudness in sensation level, meaning decibels above the person's own hearing threshold at that pitch, was not associated with self-reported severity. Loudness in hearing level, the masking level and the audiogram were, and the author concluded that severity relates to hearing thresholds while loudness in sensation level adds little clinically useful information.
- 293 patients, 2026. Handicap scores on the Tinnitus Handicap Inventory were not significantly associated with matched tinnitus intensity. They were independently associated with sound intolerance, tinnitus pitch and depression scores.
The literature is not unanimous, and it is worth saying so. A 2024 study of 139 patients with hearing loss found that, among the factors it tested, total Tinnitus Handicap Inventory scores were influenced only by tinnitus loudness. A 2019 study of 148 patients found significant correlations between handicap scores, a visual analogue rating and loudness matching. So booth-matched loudness is not irrelevant. It is a weak and inconsistent predictor, which is a long way from being the main one.
How loud it feels is closer, but only moderately
Asking people how loud their tinnitus seems gives a different number from matching it, and that number does relate to distress.
Two German surveys, each of about 5,000 members of the national tinnitus association (4,995 people in one, 4,958 in the other), used a self-rated loudness grading and a short distress questionnaire. Both found a correlation of 0.45 between loudness and annoyance. That is a moderate link: squared, it means rated loudness and annoyance share only about a fifth of their variation.
The second survey looked at the mismatch directly. Of those who described their tinnitus as very loud, about one third had only mild or moderate annoyance scores. They did not differ from highly annoyed people in age, sex or how long they had had tinnitus. Annoyance was higher when tinnitus came with hyperacusis (odds ratio 4.96), a history of neurological disease (3.16), vertigo or dizziness (1.94) or hearing loss (1.71). People with loud tinnitus but low annoyance were more often not feeling low or depressed, and did not see themselves as victims of their noises. The first survey found hyperacusis raised annoyance far more than it raised loudness, with odds ratios of 21.91 against 9.47.
A third survey, from a German group, of 4,705 people, separated the two more cleanly. Rated loudness went with being permanently aware of the tinnitus and hearing it in both ears. Distress went with depressivity, anxiety and the severity of other physical symptoms. The authors' conclusion was that the two should be assessed separately.
What predicts distress instead
Across studies, the factors that travel with distress are mostly about interpretation, emotion and behaviour.
- Catastrophic thinking and fear. In 61 ENT outpatients, catastrophising about tinnitus went with more fear of it and more attention towards it, and both catastrophising and fear were linked to poorer quality of life. Fear fully accounted for the link between catastrophising and quality of life.
- Anxiety sensitivity. In 1,705 tinnitus clinic patients, anxiety sensitivity was associated with annoyance and handicap scores but not with loudness or how much of the time people were aware of the tinnitus; hearing loss was linked to loudness. In 283 people from a national epidemiological survey, anxiety sensitivity correlated with tinnitus distress, and anxiety and depression symptoms did not explain that away.
- Avoidance. In that survey sample, avoiding situations fully accounted for the link between anxiety sensitivity and day-to-day functioning, and partly for its link with distress. In 373 clinic patients, fear-avoidance behaviour partly accounted for the link between anxiety sensitivity and the cognitive side of handicap.
- Depression. Among 436 clinic patients, the 121 with current depression reported no louder tinnitus than the 285 with no history of depression, yet scored higher on all 12 severity questions. Tinnitus and depression covers that relationship in both directions.
- Mood in the moment. In 658 users of the TrackYourTinnitus app, stress and how pleasant or unpleasant people felt partly carried the effect of loudness on distress. Loudness still had a direct effect of its own, which is a fair summary of the whole literature: it matters, and the emotional state it lands in matters too.
Almost all of these studies are cross-sectional snapshots. They show what goes with distress; they cannot show on their own which comes first, and tinnitus and anxiety sets out the evidence that anxiety and tinnitus feed each other in both directions.
Attention: why noticing it keeps it loud
A 2014 cognitive-behavioural model sets out the psychological account. Tinnitus becomes a problem when it acquires emotional significance through the way it is interpreted; those interpretations, and the changes in behaviour they lead to, create and maintain the distress. The model gives a central place to selective attention, and raises the possibility that attending to tinnitus distorts how it is perceived. The authors present it as a framework with testable hypotheses rather than a settled result.
The survey data fit it. In the German survey of 4,958 people, constant awareness of the noises, a reduced ability to ignore them and difficulty concentrating were reported often, even by people whose overall annoyance was comparatively low.
The neurophysiological model reached the same place
Jastreboff's neurophysiological model came from a different direction and arrived at a similar conclusion. As P. J. and M. M. Jastreboff summarised it in 2006, the model holds that the limbic and autonomic nervous systems are involved in all clinically significant tinnitus, linked to it by connections that follow the principles of conditioned reflexes. Tinnitus retraining therapy is built on it: counselling aims to reclassify tinnitus as a neutral signal, and the treatment aims to remove the reactions the signal triggers without trying to silence its source.
The psychological and neurophysiological models disagree on mechanism and method. They agree on the target: the reaction to the sound, not its volume. Brain-imaging work pointing the same way is covered in how tinnitus is generated.
What this means for treatment
Most people with tinnitus are not severely troubled by it. The 2020 Cochrane review of CBT puts tinnitus at up to 21% of adults, with an estimated 1% to 3% experiencing severe problems. That gap alone does not say what drives distress, but it fits the studies above: the same sound can be lived with very differently.
That is the logic behind cognitive behavioural therapy for tinnitus. The same Cochrane review included 28 randomised trials with 2,733 participants and found CBT may reduce the impact of tinnitus on quality of life: against a waiting list, 10.91 points lower on the Tinnitus Handicap Inventory, at low certainty. A randomised trial of 492 patients comparing a CBT-based stepped-care programme with usual care found that a fall in tinnitus-related fear explained part of the programme's benefit, which fits the fear-and-catastrophising picture above. The review was less sure about the mechanism in general: it judged that CBT probably reduces negatively biased interpretations of tinnitus compared with other active treatments, but was uncertain against a waiting list, and concluded that the evidence on this outcome overall is limited.
Guidance has followed. The American Academy of Otolaryngology–Head and Neck Surgery guideline recommends CBT for persistent, bothersome tinnitus, and the 2026 Nature Reviews Disease Primers volume names tinnitus-focused counselling and CBT as first-line treatments. CBT is not a volume treatment, and the treatment comparison lists its targets as distress and sleep, not loudness.
It also explains how tinnitus is measured. The Tinnitus Functional Index and the Tinnitus Handicap Inventory score impact rather than loudness, which is why how tinnitus is measured warns that a treatment can improve the scores a lot without changing the sound.
Loudness is not written out of care. The 4,705-person survey's authors suggest that where loud tinnitus is the main complaint and distress is mild or moderate, treatment should focus on auditory perception; where depression, anxiety or other physical symptoms are severe, those should be addressed first. In the 1,705-patient study hearing loss, not anxiety sensitivity, went with louder tinnitus, which is one reason hearing aids for tinnitus and sound therapy sit alongside CBT rather than competing with it.
If your tinnitus bothers you a lot
None of this says the sound is imagined, or that being troubled by it is a weakness. It says that the parts of tinnitus that vary most between people are the parts most open to change, and they are the reason habituation is possible even when the sound stays. Poor sleep went with higher severity in the 1984 clinic study, and sleeping with tinnitus covers what helps there.
Two practical points follow. Loudness and distress are worth tracking separately, because they can move independently; the symptom journal logs daily intensity and triggers. And if the tinnitus has changed suddenly, arrived with a drop in hearing, or pulses with your heartbeat, that is a different question from this one: the when to seek care checklist covers those signs. If distress about tinnitus has turned into thoughts of harming yourself, tinnitus and suicidal thoughts lists where to get help now.