The hook
Saturday rehearsal. You spend an hour dialing the mix with the band pushing full volume, and it sounds huge. Rich low end, sparkling top, everything sitting perfectly.
Sunday morning, first song, the room half full and the level a notch lower — and the mix is thin. Where did the bass go? Why does the vocal suddenly poke out? Nobody touched the board overnight.
Your ears did this. Not because they're broken; because of how they're built. This is the one lesson in the course about the equipment you can't upgrade.
The lesson
The core fact: your ears do not hear all frequencies equally, and the imbalance changes with volume.
Ears are most sensitive in the 2 to 5 kHz range — the zone where speech consonants live. There's a physical reason: your ear canal is a little tube, and like any tube it resonates, boosting frequencies around 3 kHz by a large amount before they even reach the eardrum. Evolution tuned you to hear voices.
Away from that zone, sensitivity falls off, and it falls off hardest at low volume. At quiet listening levels, your ears massively under-report bass and, to a lesser degree, extreme highs. Turn the same music up loud and the ear's response flattens out: suddenly you hear the bass and the sparkle in nearly full measure.
Scientists mapped this decades ago; the resulting curves are called equal-loudness contours (you'll hear engineers say "Fletcher-Munson"). You don't need to memorize the curves. You need to internalize what they do to you at the board:
A loud mix flatters itself. When you mix at high volume, the ear hands you free bass and free treble. Everything sounds big and finished. Pull the level down and the ear takes its gift back. The bass collapses first, the mix turns thin and midrangey, and the balances you set no longer hold.
So what do working engineers actually do?
Habit one: mix at a moderate level, and check at extremes. Do most of your balancing at a comfortable conversational-plus level. Push it loud briefly to feel what the congregation feels on the big chorus. Then do the thing almost nobody teaches volunteers: pull it way down and listen quiet. A mix that still holds together at low volume, vocal clear and bass still present in the right proportion, is a mix that works at every level. Quiet listening is the truth serum.
Habit two: protect the equipment. From lesson 1.2: 85 dB SPL is the all-day line, and every 3 dB above it halves your safe time. FOH engineers live in front of loudspeakers for years. Hearing damage doesn't announce itself and doesn't heal. It takes the 3–6 kHz region first, exactly where your mixing judgment lives. A pair of musician's earplugs (the flat-response kind, not foam) in your pocket is standard professional equipment, not a sign of weakness.
There's one more ear behavior worth meeting now, because it quietly runs half of mixing: masking. A loud sound hides quieter sounds near it in frequency, and the hiding is lopsided. Low frequencies are much better at masking highs than the other way around. A thick, boomy low end does more than sound muddy. It erases the detail above it. This is why (spoiler for Level 4) engineers spend so much time cutting lows out of instruments that don't need them. Cut the boom and the vocal comes out of hiding.
Go deeperoptional
The frequency analysis happens mechanically, in the cochlea. The basilar membrane running down its coiled length is stiff and narrow at the base, loose and wide at the apex; each point resonates at a characteristic frequency, high at the base, low at the apex. Roughly 3,000 inner hair cells ride this membrane in tonotopic order: the cochlea is a mechanical spectrum analyzer that does in hardware what an RTA does in software. Outer hair cells add an active, electromotile amplification stage worth some 40–60 dB at low stimulus levels, which is where the ear's ~120 dB dynamic range comes from. Those amplifier cells are precisely what loud exposure destroys first. They don't regenerate.
Equal-loudness contours were first measured by Fletcher and Munson (1933) and are now standardized as ISO 226:2003. Levels on the curves are given in phons (a tone at N phons is as loud as an N dB SPL tone at 1 kHz). The practical readings: the curves bunch and flatten as level rises, and bass sensitivity improves the most; the 2–5 kHz dip in the contours reflects the ear canal's quarter-wave resonance (+15–17 dB near 3 kHz). Studio calibration standards exploit the flattest region: the K-System pins monitoring around 83 dB SPL, where equal-loudness curves are most level-independent and tonal judgments transfer best.
Masking has structure worth knowing. In the frequency domain, the cochlea's filters (critical bands — roughly 100 Hz wide below 500 Hz, about 20% of center frequency above) mean sounds within the same band interact strongly, and basilar-membrane excitation spreads asymmetrically toward the high-frequency side, giving low maskers their upward reach. In the time domain, a loud event masks quieter events for 100–200 ms after it ends (forward masking) and even ~20 ms before it starts (backward masking — the brain's assembly buffer at work). That 100–200 ms figure is one reason compressor release times around 100 ms so often sound "invisible": the ear wasn't going to hear that region cleanly anyway.
Temporary threshold shift (TTS) is the measurable dulling after loud exposure: elevated hearing thresholds that mostly recover in hours. Repeated TTS becomes permanent threshold shift, which starts as the classic notch near 4 kHz. An engineer's own audiogram degrading at 3–6 kHz is a professional injury in the most literal sense: it's damage inside the exact band where mix decisions are made.
Remember
- Ears hear the 2–5 kHz speech zone best, and under-report bass at low volume. The imbalance shrinks as level rises.
- Loud mixing flatters the mix with free bass and treble; the flattery vanishes when the level drops. Quiet listening is the truth serum.
- Loud listening also fatigues your ears within minutes, dulling them and pushing you to overbrighten.
- Masking: loud sounds hide nearby quieter ones, and lows mask upward hard. Controlling low end un-hides everything above it.
- 85 dB is the all-day line; damage lands first in the 3–6 kHz judgment zone and never comes back. Own real earplugs.
Your turn
- Do the quiet check on your next mix: master way down, thirty seconds, ask "vocal clear? bass still present?" Fix, then restore.
- Notice your own fatigue curve: after a loud rehearsal, step outside for five minutes, come back, and hear how different the same mix sounds.
- Play a bass-heavy song and toggle the low EQ down and up on the master while listening to the vocal, not the bass. Hear the vocal detail appear and disappear. That's masking.
- Buy (or at least price) a pair of flat-response musician's earplugs. Under $40 gets you protection that doesn't muffle the mix.