Share article
Every aspect of OSHA’s hearing conservation standard — the 85 dBA action level, the STS formula using 2000/3000/4000 Hz, the characteristic audiogram pattern, even the reason workers don’t notice early damage — flows from a single physiological fact: the 3–6 kHz range is where cochlear noise damage begins. Understanding why this frequency band is so uniquely vulnerable to occupational noise explains both the biology of NIHL and the logic of the entire OSHA regulatory framework built around it.
Soundtrace provides audiometric surveillance that detects threshold changes in the 3–6 kHz danger zone at the earliest measurable stage — before damage spreads into the speech frequencies that workers notice.
The 3–6 kHz frequency range is where noise damage begins because three independent physiological factors converge to make it the most mechanically and metabolically stressed region of the cochlea during noise exposure. OSHA’s STS formula at 2000/3000/4000 Hz is not arbitrary — it is precisely calibrated to detect damage at the earliest point where audiometric surveillance can catch it.
The outer ear — the ear canal and pinna — functions as an acoustic resonator. Like all resonators, it amplifies sound energy most efficiently at its resonant frequency. The human ear canal has a resonant frequency of approximately 2.5–4 kHz, meaning that sound waves at this frequency range are amplified by 10–15 dB by the time they reach the tympanic membrane, compared to low-frequency sounds.
This amplification is acoustically useful: it’s part of what makes the human ear most sensitive to the frequency range most important for speech perception. But it has a direct consequence for noise damage: sounds in the 2–5 kHz range deliver more energy to the cochlea than lower-frequency sounds of equal external intensity. A 90 dBA broadband noise produces proportionally greater mechanical stress on the 3–5 kHz cochlear region than on the 500 Hz or 1000 Hz regions, simply because of the ear canal’s resonant amplification.
The middle ear contains a protective mechanism called the acoustic reflex (or stapedius reflex). When the ear is exposed to loud sounds, the stapedius muscle contracts reflexively, stiffening the ossicular chain and reducing the transmission of sound energy to the cochlea. This reflex provides approximately 10–14 dB of attenuation — but only for frequencies below approximately 2 kHz. Sounds above 2 kHz are transmitted to the cochlea with little or no attenuation from the acoustic reflex.
The practical consequence: low-frequency industrial noise (below 2 kHz) activates the acoustic reflex and receives partial protection. The 3–6 kHz range — precisely the range most amplified by ear canal resonance — receives no protection from the acoustic reflex. During sustained noise exposure, the 3–6 kHz cochlear hair cells are receiving amplified sound energy with no middle-ear attenuation. Low-frequency hair cells are receiving the same sound with middle-ear attenuation engaged.
This is why NIHL is almost never a pure low-frequency hearing loss. Even in workplaces with strong low-frequency noise components (diesel engines, compressors, HVAC systems), the acoustic reflex provides some protection to the 500–2000 Hz cochlear region. The 3–6 kHz region gets no such protection, and is hit first regardless of the industrial noise spectrum.
Inside the cochlea, different frequencies are processed at different locations along the basilar membrane — a tonotopic map. High frequencies (8 kHz and above) are processed at the base of the cochlea (nearest the oval window). Low frequencies (250–500 Hz) are processed at the apex. The 4 kHz region sits approximately one-quarter of the distance from the base.
This location has specific mechanical properties that contribute to vulnerability. The basilar membrane at the 4 kHz point experiences a particular pattern of shear stress during sound-driven motion that is especially damaging to outer hair cells (OHCs) during high-intensity noise exposure. The energy distribution of most broadband industrial noise spectra creates peak stress in this region. The OHCs at the 4 kHz location are the first to accumulate damage under sustained industrial noise, before damage extends to the 3 kHz and 6 kHz regions that bracket it.
The reason 4000 Hz is the most consistently and severely affected frequency in occupational NIHL is that all three vulnerability factors peak simultaneously at that frequency:
The 3 kHz and 6 kHz regions share most of these vulnerabilities, which is why the NIHL damage zone encompasses 3–6 kHz rather than being a pinpoint at exactly 4 kHz. But 4000 Hz consistently shows the worst thresholds because all three factors are maximally expressed at that frequency.
A feature of NIHL audiograms that often confuses reviewers is the “recovery” at 8 kHz — better thresholds at 8000 Hz than at 4000 Hz. This seems counterintuitive: if high frequencies are processed at the cochlear base and noise damages the base, why would 8 kHz be better than 4 kHz?
The answer is the cochlear geometry factor. The 8 kHz region, while also near the base, sits slightly differently within the basilar membrane’s stress pattern than the 4 kHz region. The 8 kHz hair cells receive somewhat different mechanical loading during typical industrial noise exposure. Ear canal resonance also attenuates above 5 kHz, so 8 kHz sounds are not amplified the same way 4 kHz sounds are. The net result is that 4 kHz hair cells accumulate damage faster than 8 kHz hair cells in early and moderate NIHL — producing the characteristic notch-and-recovery pattern.
▶ The diagnostic rule: if 8 kHz threshold is better than 4 kHz threshold, that is evidence of the NIHL notch pattern. If 8 kHz is equal or worse than 4 kHz, the pattern is consistent with presbycusis (age-related slope). See the NIHL vs. presbycusis guide for the full differential.
Conversational speech carries its most critical information in the 500–2000 Hz range. Consonant discrimination — the ability to distinguish “ship” from “chip” or “fine” from “vine” — extends into the 3–4 kHz range, but the core intelligibility of speech in quiet environments is carried at lower frequencies. A worker with a Stage 1 or Stage 2 NIHL 4 kHz notch — say, thresholds of 10 dB at 1000 Hz and 45 dB at 4000 Hz — will hear and understand normal conversation in quiet essentially perfectly. They will have no subjective sense that anything is wrong with their hearing.
This is the fundamental reason audiometric surveillance exists. Self-report cannot detect early NIHL. Workers are not malingering or failing to report — they genuinely have no functional experience of their high-frequency threshold loss in typical daily environments. The audiogram is the only instrument sensitive enough to detect the notch at Stage 1, when intervention prevents progression to Stage 3–4.
The Stage 1 NIHL notch — detectable on audiometry, asymptomatic, reversible in direction if exposure is controlled — represents the only window where audiometric detection can prevent functional hearing loss. By Stage 3, speech-in-noise difficulty is already present. By Stage 4, it is significant. The audiometric surveillance program is the mechanism that determines whether a worker’s NIHL is caught at Stage 1 or Stage 4.
With continued noise exposure, damage does not stay confined to 3–6 kHz. Once hair cells in the 4 kHz region are substantially depleted, noise energy is no longer absorbed there as efficiently — it travels further along the basilar membrane, exposing adjacent regions to higher stress. The notch widens. The 3 kHz and 6 kHz regions are affected first, then the 2 kHz region begins to show threshold elevation.
When 2000 Hz thresholds begin to rise, the NIHL has entered Stage 3. At this point, the STS average at 2000/3000/4000 Hz will show substantial shift from baseline. Speech-in-noise difficulty becomes functionally apparent. The OSHA 300 Log recordability threshold is often crossed. Workers begin to self-report hearing difficulty — but by the time self-report is possible, the damage has already spread well beyond its origin at 4 kHz.
OSHA’s STS definition is not an arbitrary selection of frequencies. The 2000/3000/4000 Hz average is calibrated to detect the early spread of NIHL from its 4 kHz origin into the adjacent 3 kHz and 2 kHz regions — the first sign that the notch is widening beyond its initial focal point. By averaging these three frequencies, OSHA catches threshold changes before they reach the 500–1000 Hz speech core while still detecting the beginning of damage spread that predicts future functional loss.
A formula using only 4000 Hz would catch NIHL too late — only after significant damage at 4 kHz. A formula using 500 Hz would be insensitive to early NIHL entirely. The 2000/3000/4000 Hz average is precisely placed to catch the spread of noise damage at its earliest audiometrically detectable point.
▶ Bottom line: OSHA’s hearing conservation framework is built on the biology of the 3–6 kHz danger zone. The action level, the STS formula, the 4 kHz audiometric signature, the reason for annual testing — all of it flows from the same physiological convergence that makes this frequency range the first casualty of occupational noise exposure.
Soundtrace audiometric surveillance flags STS at the earliest detectable point in the 2000–4000 Hz range, with documentation designed for OSHA compliance and WC defense.
Get a Free Quote