How Hearing Works: Sound Waves, the Cochlea, and Why Hearing Loss Is Permanent

A sound wave is a pressure fluctuation in air — nothing more. Yet within milliseconds of reaching your ear, those invisible mechanical ripples have been channeled, amplified, sorted by frequency with extraordinary precision, converted into electrical signals by cells smaller than a grain of rice, and dispatched to an auditory cortex that can distinguish a B-flat from a B-natural, identify a familiar voice in a crowded room, and localize a sound to within 1 degree of arc. The mechanism behind that transformation is one of the most elegant in biology — and its core components, once destroyed, cannot be replaced.

Sound: What the Ear Is Actually Detecting

Before explaining the ear, it helps to be precise about what sound is. Sound is a longitudinal pressure wave — alternating compressions and rarefactions of molecules moving through a medium (air, water, bone). What we call pitch corresponds to the wave's frequency, measured in hertz (Hz); what we call loudness corresponds to amplitude, measured in decibels (dB).

The human ear detects frequencies between roughly 20 Hz and 20,000 Hz, though the upper limit declines with age — most adults over 50 have significant loss above 8,000 Hz. For comparison, a conversation runs 500–3,000 Hz; musical instruments span roughly 27 Hz (bass guitar) to 16,000 Hz (highest harmonics of a violin).

Loudness is measured on a logarithmic scale: every 10 dB increase represents a 10-fold increase in sound pressure, or approximately a doubling of perceived loudness. A whisper is about 30 dB; normal conversation about 60 dB; a rock concert about 110 dB; a gunshot about 140 dB — above the threshold of immediate injury.

Source: National Institute on Deafness and Other Communication Disorders (NIDCD), NIH (2023) — "How Do We Hear?" — foundational account of sound physics, frequency range, and the anatomy of hearing.

The Outer Ear: Collection and Channeling

The visible part of the ear — the auricle or pinna — is an acoustic collector. Its asymmetric, ridged shape is not random. The specific geometry diffracts incoming sounds differently depending on their direction and elevation, providing cues the brain uses to localize sound vertically. Mammals with mobile ears can actively orient the pinna toward sound sources; humans largely lost this ability but retain vestigial auricular muscles.

Pinna — the external visible portion of the ear, made of cartilage covered in skin, which collects sound waves and introduces direction-dependent filtering cues used for sound localization.

The pinna channels sound into the external auditory canal — a roughly 2.5 cm tube ending at the tympanic membrane (eardrum). The canal acts as a resonance tube, amplifying frequencies around 2,000–5,000 Hz — exactly the range most important for speech intelligibility — by about 10–15 dB through resonance. This means the ear's outer structure is not neutral; it preferentially amplifies the frequencies of human voices before the signal reaches the eardrum.

The tympanic membrane is a cone-shaped membrane approximately 8–10 mm in diameter and 0.1 mm thick. Sound-induced pressure waves cause it to vibrate with displacements as small as the diameter of a hydrogen atom at the softest audible sounds — one of the most sensitive mechanical systems in biology.

The Middle Ear: Solving an Impedance Problem

After the eardrum, sound must transfer from air (low density, low resistance) into the fluid of the inner ear (high density, high resistance). Without intervention, approximately 99.9% of the sound energy would reflect at this air-fluid interface rather than crossing it — a loss of about 30 dB.

The middle ear solves this acoustic impedance mismatch through a mechanical lever system: three tiny bones called the ossicles — the malleus (hammer), incus (anvil), and stapes (stirrup). Together they are the smallest bones in the human body. The malleus is attached to the tympanic membrane; the stapes presses against the oval window of the cochlea.

Ossicles — the three smallest bones in the human body, linking the tympanic membrane to the inner ear, that amplify and transmit sound vibrations while matching the acoustic impedance between air and cochlear fluid.

The ossicles amplify sound through two mechanisms: lever action (the malleus-incus joint provides a mechanical advantage of about 1.3:1) and the area ratio of tympanic membrane to oval window (roughly 17:1 in surface area). Together these account for a pressure amplification factor of approximately 22-fold — recovering most of the energy that would otherwise be lost at the air-fluid transition.

The middle ear also contains two tiny muscles — the tensor tympani (attached to the malleus) and the stapedius (attached to the stapes) — that reflexively contract when exposed to very loud sounds, stiffening the ossicular chain and reducing sound transmission. This acoustic reflex provides about 15 dB of protection, but it has a latency of 25–150 milliseconds — too slow to protect against sudden impulsive sounds like gunshots, which peak and cause damage before the reflex can engage.

Source: Rosowski, J.J., in Springer Handbook of Auditory Research: The Cochlea (1996) — Detailed account of middle ear mechanics, impedance matching, and acoustic reflex physiology.

The Cochlea: A Frequency Analyzer Coiled in a Shell

The cochlea — from the Greek word for snail shell — is a coiled, fluid-filled tube in the inner ear that performs the most critical transformation in hearing: it separates incoming sounds by frequency and converts mechanical vibration into electrical signals.

Cochlea — a snail-shaped, fluid-filled bony structure in the inner ear containing the sensory organ of hearing (the organ of Corti), approximately 35 mm long when uncoiled, that performs frequency analysis and mechanosensory transduction.

The cochlea coils about 2.75 turns and is divided into three fluid-filled chambers along its length. The middle chamber (the scala media) contains the organ of Corti — a cellular mat sitting on the basilar membrane that contains the actual hearing cells.

The basilar membrane runs the full length of the cochlea, and its mechanical properties vary systematically: it is narrow and stiff at the base (near the oval window) and wide and flexible at the apex (the tip of the coil). This gradient creates a traveling wave in the fluid when the stapes vibrates — a wave that peaks at different points along the basilar membrane depending on frequency.

High-frequency sounds (high pitch) cause maximum basilar membrane displacement near the stiff base. Low-frequency sounds (low pitch) travel further and peak near the flexible apex. This is tonotopy — a spatial map of frequency along the membrane — and it is the cochlea's fundamental frequency-analysis mechanism.

Source: von Békésy, G., Journal of the Acoustical Society of America (1947) — Nobel Prize-winning work demonstrating the traveling wave and tonotopic organization of the basilar membrane.

Source: Ruggero, M.A., Current Opinion in Neurobiology (1992) — "Responses to sound of the basilar membrane of the mammalian cochlea" — experimental confirmation of frequency-place mapping in living preparations.

Think of the basilar membrane as a harp strung inside a fluid-filled tube, where the shortest, tightest strings are at the entrance and the longest, slackest ones are at the far end. Each note of a sound causes vibration to peak at the string tuned to that note's frequency. Except it is not strings — it is a continuous membrane with a gradient of stiffness that responds to the physics of wave mechanics.

Hair Cells: Converting Vibration to Electricity

Sitting atop the basilar membrane in the organ of Corti are approximately 15,500 hair cells — the sensory cells that convert mechanical vibration into electrical signals. They are named for the bundle of hair-like stereocilia projecting from each cell's top surface, arranged in rows of increasing height.

Hair cell — a mechanosensory cell in the organ of Corti that converts basilar membrane vibration into changes in membrane potential via stereocilia deflection, generating the neural signals that the brain interprets as sound.

The inner ear contains two types: roughly 3,500 inner hair cells (IHCs) arranged in a single row, and approximately 12,000 outer hair cells (OHCs) in three rows. Their functions are distinct.

Inner hair cells are the primary acoustic sensors — they are responsible for essentially all transmission of auditory information to the brain. Each IHC synapses with 10–30 auditory nerve fibers; the auditory nerve carries about 30,000 fibers total. Despite being outnumbered 3:1 by outer hair cells, inner hair cells account for 90–95% of the nerve fibers going to the brain.

Outer hair cells are electromotile — they can change their length in response to electrical signals, shortening and elongating tens of thousands of times per second. This active mechanical behavior amplifies basilar membrane motion selectively at frequencies the cochlea is already responding to, sharpening the frequency tuning by up to 50 dB at low sound levels. The outer hair cells are the cochlea's active amplifier — a biological analog of active noise cancellation in reverse.

Source: Brownell, W.E. et al., Science (1985) — "Evoked mechanical responses of isolated cochlear outer hair cells" — original discovery of outer hair cell electromotility.

How Stereocilia Convert Vibration to a Signal

When the basilar membrane vibrates, the stereocilia of the hair cells deflect against the tectorial membrane above them. The stereocilia are connected by tiny filaments called tip links running from the top of a shorter stereocilium to the side of the next taller one.

When stereocilia deflect toward the tallest row, the tip links pull open mechanically gated ion channels at the tips of the shorter stereocilia. Potassium ions (K⁺) and calcium ions (Ca²⁺) rush in from the fluid filling the scala media (which is uniquely high in K⁺, unlike most biological fluids), depolarizing the hair cell. This depolarization triggers neurotransmitter release at the base of the cell, generating action potentials in the auditory nerve fibers.

When stereocilia deflect in the opposite direction, the channels close, the cell hyperpolarizes, and neurotransmitter release decreases. The hair cell essentially encodes sound as a pattern of enhanced and reduced firing — the beginning of the neural code that the brain will decode as pitch, timbre, and loudness.

Source: Hudspeth, A.J., Science (1985) — "The cellular basis of hearing: The biophysics of hair cells" — foundational work on tip link mechanotransduction establishing A.J. Hudspeth's Nobel Prize-winning contributions.

The Auditory Nerve and Brain: Decoding the Signal

The ~30,000 fibers of the auditory nerve carry signals from the cochlea to the cochlear nucleus in the brainstem — the first of several processing stations on the way to the auditory cortex in the temporal lobe. Each auditory nerve fiber retains the tonotopic identity of the hair cell it innervates, so the frequency map established in the cochlea is preserved all the way to the cortex.

Along the way, the auditory brainstem extracts timing information with microsecond precision — comparing signals arriving at the two ears to compute the direction of a sound source. A sound arriving 0.6 milliseconds earlier at the left ear than the right tells the brainstem the source is to the left. The superior olivary complex in the brainstem performs this computation, relying on coincidence-detecting neurons that fire only when signals from both ears arrive simultaneously.

Source: Grothe, B. et al., Nature Reviews Neuroscience (2010) — "Mechanisms of sound localization in mammals" — comprehensive review of binaural processing and sound localization circuits.

The primary auditory cortex (A1) in the superior temporal lobe maintains the tonotopic map and performs higher-level processing: recognizing patterns, extracting meaning from speech, and filtering relevant sounds from background noise. The auditory cortex is bidirectionally connected with memory and emotion systems — which is why music can trigger vivid memories and why certain sounds produce immediate emotional responses. For a broader look at how the brain encodes memory from sensory experience, see How Memory Works in the Brain.

Why Hearing Loss Is Often Permanent

Hair cells in the mammalian cochlea do not regenerate. This is the central and catastrophic fact of hearing loss biology.

When stereocilia are damaged by excessive sound pressure, ototoxic drugs (such as aminoglycoside antibiotics and cisplatin chemotherapy), or age-related degeneration, the hair cell eventually dies. No stem cell population in the adult mammalian cochlea can replace it. The frequency range corresponding to the lost hair cells is gone.

Source: Frontiers in Cell and Developmental Biology (2026) — "Hair cell loss — cause or consequence of hearing loss?" — current review of mammalian hair cell loss mechanisms and the state of regeneration research.

The pattern of loss is not random. High-frequency hair cells at the base of the cochlea are most vulnerable — they have larger mechanotransducer currents and smaller calcium buffering capacity, making them more susceptible to metabolic exhaustion from loud sounds. This is why noise-induced hearing loss preferentially affects high-frequency hearing first, typically showing a characteristic "4,000 Hz notch" on audiograms — a dip at the frequency range most affected by the acoustic resonance of the ear canal itself.

Outer hair cells are more vulnerable than inner hair cells. Because outer hair cells are the active amplification system, their loss first causes a 40–60 dB reduction in sensitivity and loss of frequency selectivity — sounds become not just quieter but harder to distinguish from each other.

The irreversibility stands in stark contrast to non-mammalian vertebrates. Birds, fish, and amphibians all regenerate cochlear hair cells after damage — restoring hearing within weeks. The molecular pathway for regeneration is present but suppressed in mammals, apparently as a trade-off for the specialized high-frequency hearing that required evolution of the rigid mammalian cochlea. Active research is targeting genes (including Atoh1, which drives hair cell differentiation) to reactivate this dormant regeneration pathway.

Source: Cell Reports (2024) — "Hair cell regeneration, reinnervation, and restoration of hearing thresholds in the avian hearing organ" — demonstrates complete hair cell regeneration and hearing recovery in birds post-damage, confirming the biological basis for potential mammalian therapy.

Source: Stanford Medicine Ear Institute (2023) — "Hearing and Hair Cells" — clinical overview of hair cell function, damage, and the status of regenerative therapy research.

The Numbers: Human Hearing in Data

Feature	Value
Human hearing frequency range	20 Hz – 20,000 Hz
Total cochlear hair cells	~15,500
Inner hair cells	~3,500
Outer hair cells	~12,000
Auditory nerve fibers	~30,000
Cochlea length (uncoiled)	~35 mm
Basilar membrane frequency map	~20,000 Hz (base) to ~20 Hz (apex)
Minimum audible displacement of basilar membrane	~0.3 nm (less than a hydrogen atom's diameter)
Safe continuous noise exposure (NIOSH)	85 dB for 8 hours
Exposure halving time per 5 dB	(every 5 dB increase halves safe exposure time)

Source: NIOSH (National Institute for Occupational Safety and Health), CDC (2023) — Occupational noise exposure limits and the NIOSH recommended exposure limit (REL) for hearing conservation.

Q: At what volume does hearing damage begin?
A: The NIOSH safe exposure limit is 85 dB for eight hours per day. Every 5 dB increase above that halves the safe exposure time — so 90 dB is safe for four hours, 95 dB for two hours, 100 dB for one hour. A rock concert at 110 dB exceeds the safe exposure time within about seven minutes. Damage occurs even without pain, since the cochlea has no pain receptors.

Q: Why does hearing loss make it harder to understand speech even with amplification?
A: When outer hair cells are damaged, the cochlea loses frequency selectivity — its ability to distinguish closely spaced frequencies. A hearing aid can amplify sound but cannot restore the sharpness of frequency tuning that outer hair cells provide. This is why people with cochlear damage often say they can hear speech but cannot understand it clearly.

Q: What is tinnitus and where does it come from?
A: Tinnitus — perceived ringing, buzzing, or hissing with no external source — results from abnormal spontaneous activity in the auditory pathway following hair cell damage. When hair cells in a frequency region die, the corresponding auditory cortex neurons lose their normal input and may become hyperactive. The brain interprets this spontaneous activity as sound. Roughly 15% of adults experience chronic tinnitus; it is the most common service-connected disability among military veterans.

Q: Can cochlear implants restore normal hearing?
A: Cochlear implants bypass damaged hair cells entirely, directly electrically stimulating auditory nerve fibers. They restore functional hearing — sufficient for speech understanding in most recipients — but do not replicate the frequency resolution of a healthy cochlea. Current devices use 12–22 electrode channels, far fewer than the ~3,500 effective channels of the intact cochlea. Recipients typically describe early cochlear implant sound as robotic or synthetic, improving with brain adaptation over months.

Q: Why does hearing decline with age?
A: Presbycusis (age-related hearing loss) results from cumulative hair cell loss at the cochlear base, stiffening of basilar membrane structures, degeneration of spiral ganglion neurons, and reduced blood supply to the stria vascularis (which maintains the high-potassium endolymph needed for transduction). High-frequency loss begins in early adulthood and progresses predictably — most people over 65 have significant loss above 4,000 Hz.

Q: Is hearing loss from headphone use a real risk?
A: Yes, and a well-documented one. Personal audio devices at high volumes deliver continuous noise exposure directly to the cochlea. The WHO estimates that 1.1 billion young people worldwide are at risk from unsafe headphone use. The 60/60 rule — no more than 60% maximum volume for no more than 60 minutes per session — provides a practical guideline, though the actual safe level depends on the device's maximum output and the listener's ear canal acoustics.

The Takeaway

Hearing begins as air pressure waves, amplified and impedance-matched by the outer and middle ear, then converted into a traveling wave in the cochlear fluid that peaks at specific locations depending on frequency. Roughly 15,500 hair cells — each positioned to respond to a narrow frequency band — convert that mechanical energy into electrical signals using tip link ion channels that are among the most sensitive mechanical detectors in biology. Those signals traverse 30,000 nerve fibers to an auditory brain that extracts pitch, location, and meaning with microsecond precision. The whole system is extraordinary, and its failure is largely irreversible — which is why protecting the hair cells you have, through noise avoidance and safe listening practices, is the only currently available strategy for preserving the hearing you will want decades from now.

---

Read next:

• How the Eye Works: Vision Explained
• How the Brain Works: Neurons, Regions, and the Mind
• How Memory Works in the Brain
• The Hodgkin-Huxley Squid Axon Experiment