It is your hearing turning down its own sensitivity when a sound stays steady for a while, so you stop noticing it.
Nerve fires hard at sound onset, then decays to a low plateau as the steady sound continues; a change spikes it again.
What it is
Your hearing dialling down its own sensitivity to a steady, unchanging sound so you stop noticing it.
Key facts
Mechanism: cochlear hair cells + auditory nerve fibres fire hard at sound onset, then firing rate decays to a lower plateau while the sound keeps going unchanged.
Fast adaptation: ~milliseconds to tens of ms. Slow adaptation: seconds to minutes. Recovery in seconds of quiet.
It is a NEURAL/synaptic effect (calcium-dependent vesicle depletion at the hair-cell ribbon synapse), NOT eardrum or muscle movement.
Different from the stapedius (acoustic) reflex: a MUSCLE contraction, latency ~50-150 ms, only ~10-20 dB attenuation, mostly below 1-2 kHz.
Different from TTS (Temporary Threshold Shift): real fatigue/damage needing hours of quiet; adaptation resets in seconds.
Human range 20 Hz to 20,000 Hz; most sensitive 2,000-5,000 Hz (ear-canal resonance ~3.4 kHz). 0 dB SPL = 20 micropascals at 1 kHz.
Loudness: ~+10 dB sounds 'twice as loud'; +3 dB doubles acoustic POWER; +6 dB doubles sound PRESSURE/voltage.
Smallest level change you reliably notice (JND) is ~1 dB; ~3 dB is a clearly audible step.
Safe exposure (AU/NIOSH 3 dB rule): 85 dBA for 8 h; every +3 dB halves safe time -> 88=4 h, 91=2 h, 100 dBA=15 min.
Speed of sound ~343 m/s at 20 C; inverse-square law: doubling distance from a point source drops level by 6 dB.
How it works
Sound starts: hair cells release a burst of neurotransmitter, nerve fires at a high spike rate.
Sound holds steady: synaptic vesicles deplete and the nerve's firing rate decays to a lower plateau.
Brain stops flagging the now-constant sound as 'new', so it fades from attention.
A CHANGE (new sound, level jump, sound stops) refills the synapse and firing spikes again - you notice instantly.
Stop the sound for seconds and sensitivity recovers; that is adaptation, not damage.
Real examples
The PA hum or air-con drone you forget is there until it cuts out.
Mixing for an hour: harsh top-end stops bothering you, so you quietly creep the level up.
FOH 'sounds great' at gig's end, then thin and harsh on fresh ears at next morning's playback.
A steady feedback ring you stop hearing while a punter across the room is wincing.
Walking outside after the show and your own voice sounding weirdly loud.
How it helps in live sound
Take a 5-10 min quiet break every ~45-60 min to reset your ears before key mix decisions.
Set vocal level and tonal balance in the FIRST 30 min while your ears are honest.
Trust a calibrated SPL meter, not your gut: ~96-100 dBA at FOH for live music, lower for spoken word.
Reference against a known commercial track at a fixed level to recalibrate your ears.
Use an RTA / spectrogram for the steady stuff (hum, room nodes, feedback) your ears have tuned out.
Mix at consistent moderate SPL; loud + long = fatigue + creeping level, and your 1 dB JND blurs.
Everyday analogy
Like walking into a room with a humming fridge: loud for ten seconds, then your brain mutes it until it switches off and the silence feels loud.
Watch out
Myth: 'it sounds fine now so it's mixed right.' Reality: that's adapted, fatigued ears - the fix is rested ears and a meter, not turning it up.
Fun fact
Adaptation is why total silence in an anechoic chamber feels unnerving: with no steady sound to ignore, you start hearing your own blood flow and heartbeat.
Key takeaways
Neural adaptation = your nerves turning DOWN sensitivity to steady, unchanging sound.
It's synaptic/neural (vesicle depletion), not the eardrum, muscle reflex, or hearing damage.
Recovers in seconds; TTS/fatigue and damage take hours to a lifetime.
It's why mixes drift loud and harsh - your ears lie after an hour.
Defence: scheduled breaks, an SPL meter, and a rested reference track.