Two hours into a session the mix that felt finished starts to come apart. The highs sound harsh, or the whole thing sounds flat and lifeless, and the obvious move is to start fixing it. Before touching a fader, it is worth knowing that two separate things degrade your hearing over a long session, and they get lumped together under one label: "ear fatigue." They have different causes, different recovery times, and different fixes. Applying the wrong fix is how a mix gets worse instead of better.
The physical one: temporary threshold shift
The first process is peripheral and physical. It is called temporary threshold shift, or TTS, and it happens in the cochlea.
Sound travels down the cochlea as a wave on the basilar membrane. The outer hair cells sitting along that membrane do not just sense the wave, they actively amplify it. This active amplification, often called the cochlear amplifier, is what gives human hearing its sensitivity and its fine frequency resolution. It is also metabolically expensive. After loud or prolonged exposure, the outer hair cells at the stimulated region become exhausted. They are stressed, not destroyed, but while they are exhausted your hearing threshold rises. You need more level to hear what you could hear an hour ago.
TTS hits the high frequencies hardest. That is the exact band an engineer leans on for detail, presence and air, which is why a tired mix so often drifts bright. Symptoms can include a slightly muffled quality and, in stronger cases, ringing.
Recovery takes time. The common figure quoted is 24 to 48 hours, but the real number scales with how much exposure caused it. A short stretch at a moderate level can clear in a few hours. A long day at 95 dB SPL can sit at the upper end of that range or beyond. The important point is that TTS is not permanent damage, but it is a stress signal. Repeated episodes are a recognised precursor to noise-induced permanent threshold shift. The ear is telling you something.
The neural one: loudness adaptation
The second process is largely neural, and it is called loudness adaptation. When a sound is continuous or repeats without change, the auditory system simply responds to it less. Perceived loudness drops even though nothing in the signal has changed.
The mechanism is well documented at the synapse. When an inner hair cell drives an auditory nerve fibre continuously, the readily available pool of neurotransmitter at that synapse depletes faster than it is replenished. Auditory nerve firing rates fall. On top of that, higher up the chain, the brain adapts its response to the unchanging statistics of the input. The combined result is the dull, flat, "I have heard this 200 times and it does nothing now" feeling after looping the same eight bars.
Crucially, loudness adaptation recovers fast. It clears on a timescale of minutes to a couple of hours, not days. That is why stepping away for ten or fifteen minutes genuinely resets your judgement of a section. You did not fix the mix during the break. You let the synapse refill.
Why mixing the two up makes mixes worse
Here is the trap. Both processes make the mix sound worse, so it is tempting to treat them as one thing and reach for one fix. They are not one thing.
The most damaging version of this is a feedback loop in the high end. Hours of loud monitoring cause TTS, which dulls your sensitivity to high frequencies. The mix now sounds less exciting, so you push the highs to bring the sparkle back. Your ears are still tired the next pass, so you push again. Each cycle adds brightness that is not really in the music, it is compensating for your own desensitised hearing. Played back on fresh ears the mix is harsh and fatiguing for everyone else.
Reaching for the wrong remedy is the other half of the trap. A ten-minute coffee break does a lot for loudness adaptation and very little for TTS. If the real problem is three hours at a high SPL, the break makes you feel briefly better while the underlying threshold shift is untouched, and you carry on making bright decisions on tired ears.
What follows from the two mechanisms
Once the two processes are separated, the working habits almost write themselves.
TTS responds to volume and total exposure, so the levers are level and rest:
- Keep monitoring SPL in check. The classic 85 dB SPL reference figure comes from large cinema rooms. In a small or bedroom-sized room a flatter and gentler working level sits closer to 79 dB SPL. Lower SPL directly slows the onset of TTS.
- Mix quiet most of the time and check loud only briefly. Loud playback is a tool for spotting problems, not a level to spend hours deciding at.
- Take real breaks and accept that heavy TTS will not clear inside a session at all. The honest fix for a tired-ear mix is a fresh listen the next morning.
Loudness adaptation responds to repetition and novelty, so the levers are different:
- Break the loop. Endless solo-looping of two bars is the fastest way to flatten your perception of them. Play the section in context and move around the arrangement.
- Rotate reference tracks so your ear keeps a fresh external anchor instead of adapting to your own mix.
- A short break, the ten to fifteen minutes that does little for TTS, is enough to undo adaptation. (The exact "ten minutes per hour" figure is a practical convention from engineers and plugin makers, not a measured result, so treat it as a habit rather than a law.)
One habit covers both. Put a C-weighted SPL meter, or a calibrated phone app, in your eyeline and watch your own behaviour. If you notice yourself nudging the monitor knob up over the course of a session, that creep is the fatigue signal. Your ears are losing sensitivity and you are compensating with volume, which only accelerates TTS. When the knob starts climbing, the session is over.
It also helps to spend your freshest hearing well. Make the irreversible tonal-balance and EQ calls early, while both processes are at their lowest. Leave editing, naming, routing and other mechanical work for late in the session when your judgement is already compromised.
The short version
"Ear fatigue" is two problems wearing one name. The flat, lifeless feeling from repetition is loudness adaptation, a neural recalibration that clears in minutes. The genuine loss of high-frequency sensitivity from hours of loud monitoring is temporary threshold shift, physical hair-cell exhaustion that can take a day or two to recover. A coffee break fixes one of them and barely touches the other. Knowing which one is degrading your judgement tells you whether to change the section, turn the monitors down, or simply stop for the day.