3.2 Speech Audiometry & Suprathreshold Testing

Speech Audiometry & Suprathreshold Testing

Introduction to Speech Audiometry

While pure-tone audiometry provides a graph of hearing sensitivity across discrete frequencies, it doesn't give a complete picture of how a patient functions in the real world. Human communication relies on complex, dynamic signals. Speech audiometry bridges this gap by using speech stimuli to evaluate hearing sensitivity, speech recognition abilities, and comfortable listening levels. It validates pure-tone results, helps determine the site of lesion, and is essential for hearing aid fitting and counseling.

Speech Recognition Threshold (SRT)

The Speech Recognition Threshold (SRT) is the lowest intensity level at which a patient can correctly repeat 50% of presented speech materials. The standard stimuli used for SRT testing are spondee words. Spondees are bisyllabic words pronounced with equal stress on both syllables (e.g., "baseball," "hotdog," "ice cream," "cowboy").

The primary clinical purpose of the SRT is to serve as a reliability check for the pure-tone audiogram. The SRT should closely align with the patient's Pure Tone Average (PTA)—the average of air conduction thresholds at 500, 1000, and 2000 Hz. In a reliable test, the SRT and PTA should be within +/- 6 dB of each other. A significant discrepancy between the two suggests a functional (non-organic) hearing loss, a misunderstanding of test instructions, or equipment malfunction. Because spondees are highly redundant and easy to guess, the SRT provides a reliable baseline measure of speech sensitivity.

Suprathreshold Measures: MCL, UCL, and Dynamic Range

Beyond thresholds, audiologists must assess how patients process speech at louder, conversational levels.

The Most Comfortable Loudness (MCL) is the intensity level that the patient finds most comfortable for listening to continuous speech. For individuals with normal hearing, MCL typically falls around 40 to 50 dB above their SRT. Establishing the MCL is particularly useful during the evaluation phase for hearing aid amplification, as it helps determine the target gain for conversational speech.

The Uncomfortable Loudness Level (UCL), also known as the Loudness Discomfort Level (LDL), is the intensity at which speech becomes uncomfortably loud or painful for the patient. Normal UCLs are typically around 100 to 110 dB HL. In patients with sensorineural hearing loss, the UCL often remains near normal levels despite elevated thresholds, a phenomenon related to loudness recruitment.

The difference between the SRT and the UCL is the patient's Dynamic Range (DR) for speech. Formula: DR = UCL - SRT A normal dynamic range is roughly 100 dB or more. Patients with sensorineural hearing loss often have a significantly reduced dynamic range. This "squeezed" range presents a major challenge for hearing aid fittings, as the device must compress a wide range of environmental sounds into the patient's narrow window of usable hearing without exceeding their UCL.

Word Recognition Score (WRS)

While SRT measures sensitivity (how soft one can hear speech), the Word Recognition Score (WRS)—formerly called Speech Discrimination—measures clarity (how well one can understand speech when it is loud enough). WRS testing evaluates the patient's ability to correctly identify single words presented at a suprathreshold level.

WRS uses Phonetically Balanced (PB) monosyllabic word lists. These lists (such as the W-22 or NU-6 lists) contain one-syllable words chosen so that the phonetic composition of the list represents the phonetic frequency of everyday English speech. Unlike spondees, PB words are not familiarized before testing.

The standard presentation level for WRS is either the patient's MCL or a set sensation level, typically SRT + 35 to 40 dB. Presenting at this level ensures the words are comfortably audible, theoretically allowing the patient to achieve their maximum possible score (PB Max). The results are recorded as a percentage of correctly repeated words.

Normal hearing or conductive hearing loss usually results in excellent WRS (90-100%). Patients with cochlear sensorineural hearing loss typically show reduced scores depending on the severity of the loss.

The Rollover Phenomenon

A crucial diagnostic indicator during speech audiometry is the rollover phenomenon. In typical cochlear hearing loss, as speech intensity increases, the word recognition score improves until it reaches a maximum plateau. However, in patients with retrocochlear pathologies (such as an acoustic neuroma or vestibular schwannoma affecting the VIIIth cranial nerve), increasing the intensity beyond the level of maximum performance causes the word recognition score to significantly decrease. This paradoxical decline in speech understanding at high intensities is a classic retrocochlear sign. When rollover is suspected, the clinician must test WRS at multiple intensity levels to calculate the rollover index, which helps differentiate between sensory (cochlear) and neural (retrocochlear) dysfunction.

Speech Audiometry Measures at a Glance