2Peripheral Generators — Cochlea, Synaptopathy, Deafferentation

FWhat the cochlea actually sends

The cochlea is not a microphone; it is a two-cell system. Outer hair cells (OHCs) act as a mechanical amplifier that sharpens tuning and boosts soft sounds, while inner hair cells (IHCs) are the true sensory receptors, converting vibration into the spike trains carried by type I spiral ganglion neurons [2019]. Roughly 95% of afferent fibres synapse on IHCs, so the integrity of the IHC ribbon synapse largely determines the quality of the signal reaching the brain.

Tinnitus pathophysiology begins by asking how this output is degraded. Loss of OHCs raises thresholds and widens tuning; loss of IHC synapses removes fibres without necessarily moving thresholds. Each pattern reduces afferent drive, the disturbance the central system reacts to [2004].

TCochlear synaptopathy — hidden hearing loss

The landmark insight is that the most vulnerable element is the IHC ribbon synapse, not the hair cell itself. Kujawa and Liberman showed that a noise exposure causing only a temporary threshold shift can permanently destroy up to half the synapses between IHCs and auditory-nerve fibres, with delayed neuronal death — all while the audiogram recovers to normal [2009]. This is ‘cochlear synaptopathy’ or ‘hidden hearing loss’.

The loss is selective: it preferentially removes low-spontaneous-rate, high-threshold fibres that code sound in noise. So a patient can have a normal audiogram yet a quietly thinned auditory nerve — the perfect substrate for a centrally generated tinnitus with ‘normal’ hearing .

TFrom animal model to the human ear

A reasonable objection is that synaptopathy was discovered in mice. Human temporal-bone work has since confirmed it: Viana and colleagues used confocal analysis of donor ears to show substantial loss of IHC synapses and auditory-nerve peripheral axons in ageing humans, often greater than the loss of hair cells themselves [2015]. Deafferentation can therefore be present and consequential even when surviving hair cells make the cochlea look intact.

Crucially, the degree of deafferentation tracks suprathreshold problems — speech-in-noise difficulty and tinnitus — better than it tracks the pure-tone audiogram [2017]. The standard audiogram is a blunt instrument for the lesion that matters.

CPatterns of deafferentation and the high-frequency edge

It is not only how much input is lost, but where. Cochlear damage from noise and ageing is biased toward the high-frequency base, creating a sharp boundary — an ‘edge’ — between a region of normal input and a region of reduced input. The central pathway reacts most vigorously at this border, and the perceived tinnitus pitch typically falls near the edge frequency of the hearing loss [2009].

This explains a familiar clinical pattern: a high-pitched tinnitus matched to the frequency where the audiogram (or extended high-frequency audiogram) begins to drop. The edge is the seam in the input map that the brain will later reorganise around [2010].

CThe peripheral signal that triggers central change

Why should a reduction of input produce a sound? Because the central pathway maintains its activity homeostatically. When deafferentation lowers afferent drive, downstream neurons down-regulate inhibition and up-regulate gain to compensate, increasing spontaneous and synchronous firing [2014]. The peripheral lesion does not contain the tinnitus; it removes a restraint, and the brain fills the gap.

This is why peripheral and central views are not rivals. The cochlear lesion is the necessary trigger and the determinant of pitch and laterality; the central response is the generator of the percept. Evaluating tinnitus therefore means looking for the hidden peripheral signal — with OAEs, extended high-frequency audiometry, and ABR wave I — not just reading the standard audiogram [2017].