What if the audiometer shows loss at 6kHz but not 4kHz?

This is atypical for NIHL (which usually hits 4kHz first). It might indicate age-related loss or other causes requiring specialist assessment.

Pure Tone Audiometry Explained: The Science Behind Audiometric Testing

Pure tone audiometry is the gold-standard method for measuring hearing in occupational health. Yet many WHS managers and HR professionals treat it as a black box—something that happens in a booth and produces a result, but the underlying science remains mysterious. Understanding the fundamentals helps you interpret results more meaningfully and appreciate why the testing protocol matters so much for validity.

This guide explains the science of pure tone audiometry in practical terms.

What Pure Tone Audiometry Is

Pure tone audiometry measures the threshold of hearing for pure sine wave tones at specific frequencies.

Breaking this down:

Pure tone: A single-frequency sound wave (no harmonics, no variation—just one frequency)
Sine wave: The mathematical shape of the wave (smooth oscillation, the simplest possible sound)
Threshold: The quietest level at which the person consistently hears the tone
Specific frequencies: Tested at 250, 500, 1000, 2000, 3000, 4000, 6000, 8000 Hz

The result is a hearing threshold measured in dB HL (Hearing Level) at each frequency. Plot these thresholds on a graph, and you have an audiogram.

The Frequencies Tested (250–8000 Hz) and Why They Matter

Audiometric testing covers eight specific frequencies, each chosen for its relevance to hearing and speech:

Low Frequencies (250–500 Hz)

These frequencies correspond to vowel sounds and the bass range. Hearing loss at low frequencies is relatively uncommon from occupational exposure and often indicates other conditions (ear wax, ear disease, age-related change). These frequencies are tested as baseline reference points but are less critical for detecting NIHL.

Mid Frequencies (1000–2000 Hz)

Critical for speech clarity. Loss at these frequencies noticeably affects conversation. Used as the primary frequencies for calculating hearing loss severity in many standards.

High Frequencies (3000–4000 Hz)

This is where noise-induced hearing loss characteristically appears first, especially at 4000 Hz. The 4kHz notch is the pathognomonic sign of occupational noise exposure. Testing these frequencies is crucial for detecting early NIHL.

Very High Frequencies (6000–8000 Hz)

Loss at these frequencies is common from noise exposure and age. These frequencies are less critical for speech but important for detecting overall trend of high-frequency loss. The 6–8kHz region often shows recovery in early NIHL (asymmetric pattern with the 4kHz notch).

Air Conduction vs Bone Conduction Testing

Pure tone audiometry uses two different pathways to deliver sound: air and bone conduction. This distinction helps identify the type of hearing loss.

Air Conduction Testing

Method: Sound is delivered through standard circumaural headphones (large over-ear headphones) directly into the ear canal.

Sound pathway: External ear → ear canal → eardrum → middle ear (ossicles) → inner ear (cochlea) → auditory nerve

What it measures: The entire hearing pathway from external ear to auditory nerve. If thresholds are elevated on air conduction, the problem could be in any part of this pathway.

Use in workplace testing: Air conduction is the primary test in occupational audiometry. It detects overall hearing change, which is what matters for occupational surveillance.

Bone Conduction Testing

Method: A bone conduction vibrator (a small oscillating device) is placed on the mastoid bone (the bony prominence behind the ear). Rather than sound traveling through the ear canal, vibrations pass directly through the skull bone to the inner ear.

Sound pathway: Mastoid bone → skull → cochlea (bypassing the external and middle ear)

What it measures: Sensorineural hearing (inner ear/cochlea function only). Because it bypasses the middle ear, bone conduction testing isolates the cochlea’s function.

Interpretation: Comparing air and bone conduction helps identify the type of hearing loss:

Conductive hearing loss: Air conduction thresholds worse than bone conduction (problem in outer/middle ear)
Sensorineural hearing loss: Air and bone conduction thresholds similar or both elevated (problem in inner ear)
Mixed hearing loss: Air conduction worse than bone, but bone also elevated

NIHL is sensorineural: Noise damages the cochlea directly. Both air and bone conduction thresholds are similarly affected. If bone conduction is normal but air conduction is elevated, it suggests a conductive problem (e.g., middle ear pathology) rather than NIHL.

The Threshold Determination Method

How does the audiometrist determine the exact threshold? The process is systematic:

The Bracketing Method

Step 1: Establish an Audible Level
Start with a clearly audible tone (e.g., 40 dB HL at 1000 Hz). The worker responds to confirm they hear it.

Step 2: Decrease Intensity in Large Steps
Reduce the level by 10 dB at a time (40 → 30 → 20 dB, etc.) until the worker stops responding (no longer hears it). This gives an approximate upper threshold.

Step 3: Increase Intensity in Small Steps
Now increase the level by 5 dB steps (or smaller) until the worker responds again. This narrowed the range.

Step 4: Confirm Reproducibility
Present the tone at the identified threshold level 2–3 more times to confirm the worker consistently responds at that level.

Result: A reliable threshold at that frequency, typically accurate to within 3–5 dB.

The Modified Hughson-Westlake Method

This is the formal name for the bracketing procedure, standardised in AS/NZS 1269.4. It ensures consistency across different audiometrists and locations. All compliant workplace audiometric testing uses this method.

What Determines a Valid Threshold?

For a threshold to be valid (reliable and representative of true hearing), several conditions must be met:

Reproducibility

The worker must consistently respond at the threshold level. If they respond inconsistently (sometimes yes, sometimes no at the same level), the threshold is unreliable. The test may be repeated to achieve reproducible results.

Worker Response Reliability

The worker must understand the task and follow instructions accurately. Some workers are poor responders (don’t press the button at the right time) or over-responders (press the button when they’re guessing rather than actually hearing). Experienced audiometrists recognise these patterns and can coach workers to improve reliability.

Background Noise Requirements

The testing environment must have low ambient noise. If background noise is audible, the worker may respond to the background rather than the test tone. This is why sound-attenuating booths are essential. The booth must meet specific ambient noise limits defined in AS/NZS 1269.4:

At 250 Hz: ≤70 dB SPL
At 500 Hz: ≤60 dB SPL
At 1000+ Hz: ≤55 dB SPL (approximately)

These tight specifications ensure that background noise doesn’t interfere with testing and that thresholds represent true hearing.

The Importance of 4kHz and 6kHz for NIHL Detection

In NIHL surveillance, the 4000 Hz and 6000 Hz frequencies are disproportionately important:

4000 Hz: The Noise Indicator

The 4kHz notch is the pathognomonic sign of noise exposure. It appears first, often before any other frequency is affected. If someone has occupational noise exposure, 4kHz will almost always be the first frequency to show deterioration.

This is why 4kHz is critical: it’s the earliest warning sign of occupational noise damage.

6000 Hz: The Recovery Indicator

In early NIHL, 6kHz often recovers (returns toward normal) while 4kHz remains damaged. This creates the characteristic pattern:

4kHz: 40 dB HL (notch)
6kHz: 25 dB HL (recovering)
8kHz: 20 dB HL (normal)

This asymmetric pattern is highly specific to noise exposure and confirms the diagnosis.

What the Test Can and Can’t Detect

What Pure Tone Audiometry DOES Detect

Threshold shift (change in hearing level)
Type of hearing loss (conductive vs sensorineural) when bone conduction is included
Frequency-specific patterns (e.g., high-frequency loss, 4kHz notch)
Bilateral vs unilateral loss
Severity of loss (mild, moderate, severe)

What Pure Tone Audiometry DOESN’T Detect

The cause of loss: The test shows there’s a hearing change but doesn’t identify whether it’s from noise, age, genetics, ototoxic chemicals, ear disease, or other factors. This is why specialist assessment (by an audiologist or ENT) is needed
Communication difficulty: A person might have “normal” audiometric thresholds but struggle to understand speech in noisy environments. Pure tone testing doesn’t evaluate real-world listening difficulty
Specific pathology: If loss is from a medical condition (tumour, infection, etc.), pure tone testing detects the hearing change but not the underlying condition—imaging or medical assessment is needed
Ototoxic chemical exposure specifically: The test detects hearing damage but can’t pinpoint whether it’s from noise or ototoxic chemicals (or both). Medical and exposure history must be considered

Equipment and Calibration: Why It Matters

The audiometer itself must meet strict standards:

Audiometer Type (IEC 61672, AS 3655)

Type 1 audiometer: Clinical-grade, highly accurate, used for occupational health testing. Required for AS/NZS 1269.4 compliance.

Type 2 audiometer: Less stringent accuracy, sometimes used for screening. Not compliant for WHS occupational testing.

Always confirm your provider uses Type 1 audiometers.

Calibration

Audiometers must be calibrated regularly:

Biological check (daily): Audiometrist tests their own hearing (or a reference worker’s) before client testing to verify equipment hasn’t drifted
Electroacoustic calibration (every 6 months): Equipment is tested for accuracy against known standards
Full NATA calibration (annually): External NATA-accredited laboratory tests the equipment comprehensively and issues a calibration certificate

Without proper calibration, thresholds can be off by 10+ dB, making results meaningless.

Frequently Asked Questions

Q1: Why test at exactly these frequencies and not others?

These frequencies were chosen because they cover speech range (important for communication) and the frequencies most vulnerable to noise (especially 4kHz). Other frequencies are less critical for occupational surveillance.

Q2: Can bone conduction testing detect something air conduction can’t?

Bone conduction primarily helps distinguish conductive from sensorineural loss. For NIHL (which is sensorineural), both air and bone are similarly affected. Bone conduction isn’t essential for NIHL surveillance but is helpful diagnostically.

Q3: What if someone fails the reproducibility requirement?

The test may be repeated, or the audiometrist may coach the worker on how to respond more reliably. Some workers are simply poor responders; in this case, results are marked as “unreliable” and testing may need to be repeated at a later date.

Q4: How accurate is pure tone audiometry?

When done correctly with calibrated equipment, pure tone audiometry is very accurate. Thresholds are typically reproducible to within 3–5 dB across repeated tests. Larger variations may indicate changes in hearing or inconsistent worker responses.

Q5: Can background noise affect the test results?

Absolutely. This is why ambient noise limits in AS/NZS 1269.4 are so strict. Testing in a noisy environment (without a booth) is invalid.

Q6: What if the audiometer shows I have hearing loss at 6kHz but not 4kHz?

This is atypical for noise-induced loss (which usually hits 4kHz first). It might indicate age-related loss, ototoxic exposure, or other causes. Specialist assessment would help clarify.

Q7: Can pure tone audiometry detect temporary hearing damage?

It can detect it if tested while the damage is present (within hours of noise exposure). However, temporary threshold shift resolves. Re-testing after adequate quiet would show recovery, confirming the shift was temporary.

Q8: Why is calibration so important?

Without calibration, the audiometer may present sound levels that aren’t accurately 40 dB, 30 dB, etc. This throws off all thresholds. An uncalibrated test is unreliable and potentially invalid for compliance purposes.