How Does Hearing Work? Air, Bones, Fluid and Twenty Thousand Hairs

The visible ear is a collector, and its irregular folds are functional rather than decorative: they alter the sound slightly depending on the direction it arrives from, which is how a person can tell whether a sound is above or below them with only two ears. The canal channels the wave to the eardrum, a membrane about a centimetre across that moves in and out with the pressure changes. Behind it sit the three smallest bones in the body, the malleus, incus and stapes, which transmit that movement to a membrane on the inner ear. Their job is impedance matching, which is the central engineering problem of hearing: sound passing directly from air into fluid would reflect almost entirely, losing around 99.9 percent of its energy, because the two media resist motion so differently. The bone chain solves this by concentrating force from the large eardrum onto a much smaller window and by acting as a lever, producing a pressure gain of roughly twenty times, which almost exactly compensates.

The cochlea is a fluid-filled tube coiled like a snail shell, and the structure running along its length, the basilar membrane, does the work of a spectrum analyser without any moving parts in the electronic sense. It is narrow and stiff at the base and wide and floppy at the apex, so different regions resonate at different frequencies: a high-pitched sound produces its largest displacement near the base and a low-pitched one near the apex. A complex sound therefore separates itself along the membrane, with each position responding to its own band, an arrangement called tonotopic organisation that is preserved all the way up to the auditory cortex, so the brain receives frequency information already sorted by location. Georg von Bekesy worked out the travelling wave behaviour of this membrane and received a Nobel Prize for it in 1961.

Sitting on the basilar membrane is the organ of Corti, containing around 15,000 hair cells per ear, each carrying a bundle of fine projections called stereocilia. The mechanism is direct and fast:

• •Membrane movement bends the stereocilia sideways, and tip links between adjacent projections physically pull open ion channels
• •Potassium flooding in depolarises the cell, which releases neurotransmitter onto an auditory nerve fibre
• •There is no chemical cascade between the movement and the channel, which is why hearing responds within microseconds while vision takes milliseconds
• •Inner hair cells, about 3,500 of them, do almost all the sending, carrying over ninety percent of the information to the brain
• •Outer hair cells, around 12,000, do the opposite: they change length in response to signals from the brain, actively amplifying quiet sounds and sharpening the frequency tuning, which is why hearing is far more sensitive and selective than a passive system could be
• •That active process leaks, so a healthy ear emits faint sounds of its own, which can be measured with a microphone and is the basis of newborn hearing screening

Human hearing runs from about 20 to 20,000 hertz in a young person, with the upper limit falling steadily with age, and sensitivity is greatest between 2,000 and 5,000 hertz, which is where speech consonants sit. Hair cells do not regenerate in mammals, although birds and fish replace theirs, which is why damage is permanent and why the search for regeneration is an active field. Noise-induced loss destroys hair cells at the base of the cochlea first, taking the high frequencies, which is why the early symptom is not quietness but difficulty understanding speech in a noisy room, since the consonants that distinguish words are high and quiet. The damage is cumulative and painless, with permanent loss beginning at around eighty-five decibels over sustained exposure, and the loudness scale is logarithmic so that every ten decibels is a tenfold increase in intensity. Tinnitus, a sound with no external source, commonly follows, and is now generally understood as the brain increasing its own gain in response to missing input rather than as a problem in the ear.

The auditory nerve carries the signal to the brainstem, where the first thing that happens is comparison between the two ears. Two cues are computed: the difference in arrival time, useful below about 1,500 hertz and resolvable down to a few tens of microseconds, and the difference in loudness, useful at higher frequencies where the head casts an acoustic shadow. From there the signal passes through several relay stations to the auditory cortex, which is organised tonotopically and which does the work people usually attribute to the ear, including separating one voice from many, which no microphone array does as well. That processing is why a cochlear implant works at all: it bypasses the hair cells entirely, stimulating the auditory nerve directly with an electrode array threaded into the cochlea and placed to exploit the tonotopic map, delivering a signal far coarser than a working ear and one that the brain learns to interpret, often well enough for telephone conversation.

The outer ear collects sound and its folds encode direction, three small bones match the impedance of air to fluid by concentrating force onto a small window and prevent almost all the energy being reflected, and the basilar membrane sorts frequencies by position because it is stiff at one end and floppy at the other. Hair cells convert movement into nerve signals by physically pulling open ion channels, with outer hair cells actively amplifying quiet sounds. They do not regenerate, which is why noise damage is permanent and takes the high frequencies first.