4.2 Pure-Tone Air Conduction Audiometry & Transducers

4.2 Pure-Tone Air Conduction Audiometry & Transducers

Pure-tone air conduction audiometry is the universal clinical gold standard for determining auditory sensitivity across the acoustic frequencies vital for human speech communication. By presenting discrete sinusoidal pure tones through calibrated acoustic transducers directly into the external auditory meatus, air conduction audiometry tests the biological integrity of the entire auditory system—from the pinna and external ear canal, through the mechanical lever systems of the middle ear, into the sensory hair cells of the cochlear partition, and along the neural fibers of cranial nerve VIII into the primary auditory cortex.

---

Patient Preparation, Seating & Standardized Instructions

Accurate, reproducible audiometric thresholds depend heavily on proper patient physical positioning and clear psychoacoustic instruction before any transducers are placed on the head.

1. Seating and Spatial Orientation

• Viewing Angle: The patient must be seated inside an acoustically treated booth (or quiet testing environment meeting ANSI S3.1 permissible ambient noise standards) positioned at an angle of $90^\circ$ or $45^\circ$ relative to the clinician and audiometer console.
• Preventing Visual Cueing: The patient must never face the clinician directly. If seated face-to-face, the patient can detect subtle non-auditory cues—such as the clinician's hand moving toward the tone presentation bar, finger flexion on the attenuator dial, or eye gaze shifts toward the VU meter. These visual cues trigger false-positive responses.
• Clinician Line of Sight: Conversely, the clinician must maintain an unobstructed lateral view of the patient's profile and facial expressions to observe hesitation, listening effort, autonomic startle, or confusion.

2. Pre-Placement Physical Clearance

Before applying earphones, the specialist must instruct the patient to remove:

• Eyeglasses (temple pieces prevent supra-aural cushions from sealing against the mastoid and temporal bone, causing severe acoustic leakage and artificial low-frequency threshold depression).
• Large earrings or ear piercings (cushion pressure pinches jewelry against the lobule or canal, causing severe pain and acoustic slit leaks).
• Headbands, wigs, hats, and facial coverings that cross the periauricular space.
• Chewing gum or lozenges (mandibular movement alters canal volume and creates internal acoustic masking noise).

3. Standardized Patient Instructions

Instructions must be stated in clear, plain language, avoiding technical jargon, and must emphasize responding to the faintest audible whisper of sound:

"You are going to hear a series of beeping tones through these earphones, some high in pitch and some low in pitch. They will start out loud and get softer and softer until they are very faint. Your task is to press the response button (or raise your hand) the very instant you think you hear a tone, no matter how soft or far away it seems, even if you are only guessing. Release the button as soon as the sound stops. We will start with your better ear."

---

The Modified Hughson-Westlake Protocol (ANSI S3.21)

Threshold determination in clinical audiometry follows the standardized modified Hughson-Westlake ascending search procedure, formalized by Raymond Carhart and James Jerger in 1959 and codified in ANSI S3.21 (Methods for Pure-Tone Threshold Audiometry).

+-------------------------------------------------------------+
|         Modified Hughson-Westlake Search Staircase          |
+-------------------------------------------------------------+
|                                                             |
|   Tone Presented                                            |
|         |                                                   |
|         v                                                   |
|   Patient Hears Tone? (Response)  ====>  DECREASE by 10 dB  |
|         |                                ("Down 10")        |
|         v No Response                                       |
|   INCREASE by 5 dB                                          |
|     ("Up 5")                                                |
|                                                             |
|   * THRESHOLD: Lowest level heard on >=50% of ASCENDING *   |
|   *            runs (Minimum 2 out of 3 ascending trials) * |
+-------------------------------------------------------------+

1. Psychoacoustic Rationale for Ascending Search

Early audiometric protocols used descending presentations (starting loud and fading down). However, descending tones induce auditory adaptation and temporary fatigue at threshold levels. In contrast, the ascending method approaches threshold from complete inaudibility, providing a distinct, abrupt sensory onset that yields superior test-retest reproducibility.

2. Initial Familiarization Phase

1. Starting Ear: Begin testing in the patient's reported better ear (or the right ear if hearing is perceived as symmetrical or unknown).
2. Starting Frequency: Always begin at 1000 Hz (the psychoacoustic anchor frequency of the audiogram).
3. Starting Level: Present a pure tone at 30 dB HL (or 40 dB HL).
   • If the patient responds clearly, decrease intensity by 10 dB to begin the search.
   • If the patient does not respond, elevate intensity to 50 dB HL.
   • If there is still no response, elevate intensity in 10 to 15 dB increments until a definite response is elicited. This suprathreshold demonstration confirms that the patient recognizes the pitch and understands the task.

3. The Step Rule: "Down 10, Up 5"

Once the initial suprathreshold response is obtained, the threshold search begins:

• After a Response (Tone Heard): Attenuate the stimulus by 10 dB ("Down 10"). Continue decreasing by 10 dB after each successive response until the tone becomes inaudible.
• After No Response (Tone Inaudible): Elevate the stimulus by 5 dB ("Up 5"). Continue increasing in 5 dB steps until a response occurs.
• Tone Duration: Present tones for a standardized duration of 1 to 2 seconds. Continuous uninterrupted tones should be avoided; pulsed tones (three pulses of 200 ms duration with 200 ms inter-pulse intervals) are clinically preferred because they improve threshold detectability by $1\text{ to }2\text{ dB}$ and dramatically help tinnitus patients distinguish the test stimulus from their internal head noise.

4. Operational Definition of Threshold

Threshold is defined as the lowest intensity level (in decibels Hearing Level, dB HL) at which a response is elicited on at least $50%$ of ascending presentations, with a clinical minimum requirement of two responses at the same level out of three ascending trials (or three out of five).

Trial Sequence Example at 1000 Hz:
1. 40 dB HL -> Response    (Down 10)
2. 30 dB HL -> Response    (Down 10)
3. 20 dB HL -> No Response (Up 5)      <-- Ascending run 1 begins
4. 25 dB HL -> Response    (Down 10)   [Count 1 at 25 dB]
5. 15 dB HL -> No Response (Up 5)      <-- Ascending run 2 begins
6. 20 dB HL -> No Response (Up 5)
7. 25 dB HL -> Response    (Down 10)   [Count 2 at 25 dB]

Threshold = 25 dB HL (Criteria met: 2 out of 2 ascending trials at 25 dB HL)

5. Retest Reliability Protocol at 1000 Hz

After completing threshold searches in the first ear through the high frequencies (1000, 2000, 3000, 4000, 6000, 8000 Hz), the clinician must retest 1000 Hz in that same ear before testing the low frequencies (500, 250 Hz):

• Acceptable Reliability: The retest threshold at 1000 Hz must agree within $\pm 5\text{ dB}$ of the original threshold.
• Discrepancy Action: If the retest threshold differs by $10\text{ dB}$ or more, test reliability is compromised. The clinician must pause, reinstruct the patient, and re-evaluate all prior frequencies. The more sensitive (better) of the two 1000 Hz thresholds is typically entered on the permanent diagnostic audiogram.

---

Standard Test Frequency Sequence and Inter-Octave Mandates

Testing must proceed in a standardized sequence to maintain cognitive focus and manage patient listening fatigue.

+-------------------------------------------------------------+
|              Standard Pure-Tone Test Sequence               |
+-------------------------------------------------------------+
|  1. Better Ear: 1000 Hz (Initial threshold search)          |
|  2. Ascending Highs: 2000 -> 3000 -> 4000 -> 6000 -> 8000 Hz|
|  3. Reliability Check: RETEST 1000 Hz (Must agree +/- 5 dB) |
|  4. Descending Lows: 500 -> 250 Hz (and 125 Hz if needed)   |
|  5. Contralateral Ear: Repeat identical sequence            |
+-------------------------------------------------------------+

Why 1000 Hz is Tested First

1. Lowest Test-Retest Variability: Human listeners exhibit the smallest psychophysical variance at 1000 Hz.
2. Center of the Speech Band: It anchors the mid-frequency boundary between low-frequency vowel energy and high-frequency consonant acoustics.
3. Familiarity: A 1000 Hz sinusoid has a clear, easily identifiable tonal pitch that does not easily trigger somatic vibration or acoustic confusion.

The Mandatory Inter-Octave Rule

Diagnostic audiometry evaluates octave frequencies: 250, 500, 1000, 2000, 4000, and 8000 Hz. In addition, inter-octave (half-octave) frequencies—specifically 750 Hz, 1500 Hz, 3000 Hz, and 6000 Hz—play a vital role in audiological diagnosis and hearing instrument programming.

The Inter-Octave Testing Mandate: Whenever a threshold difference of $20\text{ dB}$ or greater occurs between two adjacent octave frequencies (e.g., between 500 and 1000 Hz, or 1000 and 2000 Hz, or 2000 and 4000 Hz), testing the intermediate inter-octave frequency is mandatory.

Example:
- Threshold at 1000 Hz = 25 dB HL
- Threshold at 2000 Hz = 55 dB HL
- Threshold Difference = 55 - 25 = 30 dB  (>= 20 dB)
-> MANDATORY ACTION: Test inter-octave frequency at 1500 Hz!

Hearing Instrument Dispensing Significance

Modern prescriptive fitting algorithms—such as NAL-NL2 (National Acoustic Laboratories) and DSL v5 (Desired Sensation Level)—rely heavily on 1500 Hz and 3000 Hz thresholds. Because sensorineural hearing loss frequently exhibits steep "ski-slope" contours dropping precipitously between 1000 and 2000 Hz or between 2000 and 4000 Hz, omitting 1500 Hz and 3000 Hz forces fitting software to interpolate between disparate thresholds. This leads to substantial errors in calculating target gain and maximum power output (OSPL90), resulting in either speech under-amplification or acoustic feedback whistling.

---

Acoustic Transducers: In-Depth Clinical Comparison

Selecting the appropriate acoustic transducer directly dictates test accuracy, masking requirements, and the presence or absence of physiological artifacts.

+-------------------------------------------------------------+
|               Three Primary Audiometric Transducers         |
+-------------------------------------------------------------+
|  1. Supra-Aural Earphones (Telephonics TDH-39, TDH-50)      |
|     - Diaphragm in MX-41/AR cushion resting on pinna        |
|     - IA = 40 dB; High risk of ear canal collapse           |
|                                                             |
|  2. Insert Earphones (Etymotic ER-3A, ER-3C, ER-5A)         |
|     - Transducer in box; sound tubed to foam eartip in EAC  |
|     - IA = 60-70 dB; Stents canal; 30 dB noise attenuation  |
|                                                             |
|  3. Circumaural Earphones (Sennheiser HDA-200 / HDA-300)    |
|     - Large earcups completely encasing the pinna           |
|     - Extended high-frequency testing (8 kHz to 20 kHz)     |
+-------------------------------------------------------------+

Clinical Advantages of Insert Earphones

1. Marked Increase in Interaural Attenuation (IA):
   • Supra-aural earphones make broad contact with the cranium and pinna; their acoustic energy easily crosses the skull to stimulate the contralateral cochlea at sound levels of $40\text{ dB HL}$.
   • Insert earphones decouple the transducer from the skull, transmitting sound down a small flexible tube to an expandable foam tip seated inside the cartilaginous EAC. Interaural attenuation increases to $60\text{ to }70\text{ dB}$ (and up to $75\text{ dB}$ at 250 and 500 Hz).
   • Clinical Impact: The need for contralateral clinical masking is reduced by over $80%$, expediting testing and preventing patient fatigue.
2. Complete Elimination of the Collapsed Canal Artifact:
   • In elderly patients, loss of subcutaneous elastin and cartilage structural rigidity causes the external canal walls to become flaccid.
   • The static spring tension of a supra-aural headband (exerting approximately $4.5\text{ to }5.0\text{ Newtons}$ of lateral force) compresses the auricle flat against the temporal bone, pinching the cartilaginous canal shut.
   • This artificial mechanical obstruction creates a false high-frequency conductive air-bone gap of $15\text{ to }30\text{ dB}$ (typically at 2000, 3000, and 4000 Hz) in an ear that actually has normal middle ear function.
   • Insert Earphone Resolution: The expandable open-cell foam eartip (e.g., ER-14) is compressed, inserted deep into the meatus, and allowed to expand. The expanding foam serves as an internal structural scaffold that stents the canal open, completely eliminating the collapsed canal artifact.
3. Superior Ambient Noise Isolation:
   • Insert foam tips provide $30\text{ to }35\text{ dB}$ of passive ambient noise attenuation in the low frequencies ($250\text{ and }500\text{ Hz}$), preventing elevated low-frequency thresholds when testing outside an ideal double-walled sound booth.

---

Patient Response Modes and Pitfall Management

Accurate audiometry requires continuous behavioral vigilance to detect and correct false responses.

+-------------------------------------------------------------+
|               Managing Patient Response Errors              |
+-------------------------------------------------------------+
|  1. False-Positive Responses (Patient responds without tone):|
|     - Etiology: Anxiety, rhythm anticipation, loud tinnitus |
|     - Correction: Reinstruct; vary intervals; pulsed tones  |
|                                                             |
|  2. False-Negative Responses (Fails to respond to tone):     |
|     - Etiology: Malingering (non-organic), timidity, fatigue|
|     - Correction: Reinstruct to guess on softest sound      |
+-------------------------------------------------------------+

1. False-Positive Responses (Type I Error)

• Presentation: The patient signals a response when no tone was presented.
• Underlying Causes: Hyper-anxious patients eager to "pass" the test, internal confusion caused by persistent high-pitched tinnitus that mimics audiometric beeps, or rhythmic presentation where the patient anticipates the timing.
• Remediation:
  • Reinstruct the patient: "Only press the button when you are sure you hear the beeping sound, not in between."
  • Strictly vary the inter-stimulus presentation interval (introduce irregular pauses of $2\text{ to }6\text{ seconds}$ between presentations).
  • Switch from continuous tones to pulsed tones (three distinct bursts), which provide a distinct rhythmic signature easily distinguished from constant tinnitus.
  • Introduce a silent trial ("catch trial") to confirm the patient is not guessing.

2. False-Negative Responses (Type II Error)

• Presentation: The patient fails to signal a response to a tone that is clearly audible above their physiological threshold.
• Underlying Causes: Excessive caution (the patient hears the tone but waits for it to become loud and clear before responding), listening fatigue, cognitive inattention, or deliberate non-organic exaggeration (malingering / functional hearing loss).
• Remediation:
  • Reinstruct the patient with emphasis on guessing: "Even if the beep sounds like a whisper or you feel like you are only imagining it, push the button right away."
  • Provide immediate verbal reinforcement through the talk-forward intercom when a hesitant response is observed.