7Masking and Residual-Inhibition Therapy

FCovering the sound: total versus partial masking

The oldest device-based approach to tinnitus is conceptually the most intuitive: present an external sound that obscures the phantom one. A total masker delivers enough broadband noise that the tinnitus disappears from awareness entirely, while a partial masker reduces its prominence without abolishing it. For decades many patients were fitted with ear-level noise generators set precisely to the level that just blanketed their tinnitus.

Total masking offers immediate, sometimes dramatic relief — but it has a cost. To mask tinnitus you may need a noise loud enough to interfere with speech and comfort, and the relief lasts only while the masker is running. Partial masking, by contrast, leaves the tinnitus faintly audible alongside the external sound, which turns out to matter a great deal once habituation enters the picture [2014].

TThe Feldmann tradition

Modern masking grew from the work of Hazel Feldmann, whose 1971 masking studies and later masking curves showed that the noise band and intensity needed to cover tinnitus vary enormously between patients. Feldmann classified patients by how their tinnitus interacted with external maskers — some were masked by sounds at any frequency, others only by noise near their tinnitus pitch, and a stubborn minority could not be masked at all.

This tradition gave us two enduring clinical measurements: the minimum masking level (MML), the lowest sound level that just covers the tinnitus, and the observation that the relationship between tinnitus loudness and masker level is frequently non-linear. The Feldmann masking curves remain a useful teaching tool even though pure masking is no longer the dominant therapeutic philosophy.

TResidual inhibition: the quiet after the noise

One striking phenomenon emerges from masking experiments: after a masker is switched off, many patients report that their tinnitus is reduced or even absent for a period — seconds to minutes, occasionally longer. This is residual inhibition (RI). It is not relaxation or distraction; it is a genuine, temporary suppression of the tinnitus signal that follows acoustic stimulation.

Roberts and colleagues mapped RI carefully and found that the sounds that produce the deepest, longest residual inhibition overlap with the tinnitus spectrum and with the region of audiometric threshold shift [2008]. In other words, RI is most effectively driven by stimulating the same deafferented frequency region that is generating the tinnitus — a clue that RI engages the very neurons responsible for the percept.

CUsing residual inhibition therapeutically

RI is fascinating, but is it treatment? Clinically it is used in two ways. First, as a diagnostic probe: a patient who shows robust RI demonstrates that their tinnitus is, in principle, suppressible, which is reassuring and can be motivating. Second, as a therapeutic target: protocols that repeatedly evoke RI — sometimes called residual-inhibition therapy — aim to extend the suppressed interval over a course of stimulation.

The honest clinical message is that RI is reliable as a phenomenon but inconsistent as a cure. The suppression usually fades, and durable lengthening of RI across sessions has not been convincingly established in large trials. RI therapy is therefore best framed to patients as a demonstration of suppressibility and a possible source of intermittent relief, not a permanent fix [2013].

CRelief versus habituation: masking and TRT are different philosophies

It is tempting to lump masking and Tinnitus Retraining Therapy together because both involve sound. They are philosophically opposite. Masking aims to remove the tinnitus from perception — relief is the goal and is delivered immediately. TRT deliberately keeps the tinnitus faintly audible (partial masking, never total) because the brain can only habituate to a signal it can still detect; covering it completely would deny the nervous system the repeated neutral exposure that drives habituation.

This is why the modern consensus favours partial, low-level, mixing-point sound over total masking for any patient whose goal is long-term adaptation. Total masking still has a legitimate role — for acute distress, for sleep, and for patients who simply want symptomatic relief — but clinicians should be explicit about which goal a given sound setting is serving [2014].