Residual inhibition: when masking sound briefly silences tinnitus afterward

Many people with tinnitus discover by accident that certain sounds make their tinnitus quieter or disappear, at least for a moment. A shower, a fan, music played at low volume: after the sound stops, the silence that follows is different from the silence before. The tinnitus that was present is briefly absent, reduced, or changed. This phenomenon has a clinical name, residual inhibition, and a research history stretching back to the 1970s. It is one of the clearest demonstrations that tinnitus is not simply a fixed output of a damaged cochlea but a dynamic signal that the auditory system can, under the right conditions, temporarily suppress.

Defining residual inhibition

Residual inhibition (RI) is defined as the temporary reduction or complete cessation of tinnitus following exposure to an external sound. The key word is temporary. RI is not a cure. It is a transient suppression that typically lasts from a few seconds to a few minutes before tinnitus returns to its pre-masking level.

RI was formally characterized by Terry and colleagues in 1983, in a standardized protocol that became the basis for subsequent research. Patients underwent audiometric assessment of their tinnitus pitch and loudness, then received a masking stimulus at a calibrated level for a defined period. After the sound ended, they reported on their tinnitus: was it gone completely, partially reduced, unchanged, or worse?

The clinical standard that emerged from this work presents broadband noise or a narrow-band masker centered near the tinnitus pitch at a level 10 dB above minimum masking level (the MML, the lowest level at which the masker covers the tinnitus) for 60 seconds. After the masker ends, the patient rates tinnitus on a simple scale: complete suppression, partial suppression, no change, or rebound (tinnitus temporarily louder than baseline).

Prevalence across tinnitus patients

Studies using standardized RI protocols show that roughly 80 to 90% of tinnitus patients experience some degree of RI. Complete suppression, where tinnitus disappears entirely for at least a brief period, occurs in approximately 50 to 60% of patients in controlled settings. Partial suppression is more common than complete suppression in clinical populations. A small percentage of patients report rebound, meaning the tinnitus is briefly louder after the masker ends.

These percentages vary substantially across studies depending on patient selection, masker type, and assessment timing. The most important variable is tinnitus etiology. Patients with primarily cochlear tinnitus, associated with audiometric notches consistent with noise-induced or age-related outer hair cell loss, show higher rates of RI than those whose tinnitus is believed to have a predominantly central origin.

What residual inhibition reveals about tinnitus mechanisms

The existence of RI is mechanistically informative. A purely peripheral tinnitus generator, like a damaged hair cell that fires spontaneously regardless of external input, would be difficult to suppress with an external sound after that sound ends. RI suggests that at least some component of the tinnitus signal is generated or sustained centrally, and that this central generator can be temporarily inhibited by auditory input.

Roberts and colleagues, reviewing the neuroscience of tinnitus in the Journal of Neuroscience in 2010, describe RI as evidence that the central auditory system actively participates in tinnitus generation, not merely in its transmission. Masking sounds that produce RI presumably saturate or suppress the neural populations generating the phantom signal, and the inhibitory effects outlast the stimulus by seconds to minutes.

The duration of RI correlates roughly with tinnitus severity in some studies: patients with milder tinnitus tend to show longer RI duration, while those with severe and persistent tinnitus often show very brief RI. This inverse relationship is consistent with the idea that more entrenched central generators are harder to suppress.

RI also provides a rough indicator of which frequency channels are involved in tinnitus generation. Maskers centered at or near the tinnitus pitch tend to produce better RI than those at distant frequencies. Narrow-band masking that matches the tinnitus pitch produces more focused RI in some patients, though broadband noise is often nearly as effective and easier to standardize clinically.

Residual inhibition as a diagnostic tool

In a tinnitus assessment battery, RI testing provides information that cannot be obtained from audiometry or tinnitus loudness matching alone. The presence, duration, and completeness of RI help characterize the tinnitus and inform treatment planning.

Patients who demonstrate complete RI with a simple broadband masker may be good candidates for sound-based therapies. The RI response suggests that auditory input can meaningfully affect their tinnitus generator, which is a prerequisite for sound therapies that depend on cochlear-cortical modulation.

Patients with absent or minimal RI do not necessarily fail to respond to all sound therapies, but the absence suggests that the sound therapy mechanism may be different, perhaps operating through habituation and attention modulation rather than direct inhibition of the tinnitus generator.

RI testing also establishes a baseline. If a patient shows RI of 30 seconds in an initial assessment, and the same test 6 months later shows RI of 3 minutes, that change can be an early indicator of improvement in the tinnitus mechanism, even before the patient reports subjective improvement.

The relationship to sound therapy

Residual inhibition is the conceptual foundation of several sound therapy approaches. If a specific type of sound can temporarily suppress tinnitus after it ends, the question for therapy is whether that suppression can be extended, deepened, or made cumulative.

Fractal tones and other structured auditory stimuli used in some sound therapy protocols are partly designed around the RI principle: they aim to deliver sound that engages the tinnitus-generating neural networks in a way that produces more sustained suppression than simple white noise. Evidence for specific protocols remains limited.

The minimum masking level protocol itself has therapeutic applications in patients who use masking for relief. Understanding that the masking sound continues to act for a period after it ends allows patients to plan their use of masking sounds, such as turning off a sound machine before a meeting rather than right at the start, to benefit from RI during a period when sound is not available.

What residual inhibition does not mean

RI is not a cure, not an indicator that tinnitus will resolve spontaneously, and not a guarantee that sound therapy will produce lasting benefit. It demonstrates suppressibility, nothing more. Many patients with complete RI continue to experience tinnitus that affects their daily function, because the moment-to-moment experience of tinnitus is distinct from a 90-second laboratory demonstration.

RI testing should be interpreted within a full audiologic and tinnitus assessment. Its most useful role is as one piece of a multicomponent evaluation that helps tailor treatment choices to the individual patient.

If symptoms persist or change, see an audiologist or physician.