The audiogram is the foundational tool audiologists use to map a person’s hearing sensitivity across various sound frequencies. While pure-tone testing measures the softest sounds a person can hear, the Speech Recognition Threshold (SRT) provides a functional measurement of how well an individual hears speech. The SRT is defined as the lowest intensity level, measured in decibels hearing level (dB HL), at which a person can correctly identify speech material fifty percent of the time. This metric is part of a comprehensive hearing evaluation, offering a direct link between pure-tone results and real-world communication ability.
Defining the Speech Recognition Threshold
The purpose of measuring the Speech Recognition Threshold is to establish a reliable reference point for speech perception. This threshold acts as a cross-check for the pure-tone audiogram results, ensuring the patient’s behavioral responses are consistent across different stimuli. The SRT also establishes the appropriate starting intensity level for other speech tests, such as the Word Recognition Score.
The speech materials used to find the SRT are standardized and consist mainly of spondee words. Spondee words are two-syllable compound words, like “hotdog,” “baseball,” or “ice cream,” that have equal stress on both syllables. This equal stress distribution makes the words identifiable even at soft intensity levels and creates a steep psychometric function, meaning perception shifts rapidly from not heard to clearly heard over a small range of decibels.
The redundancy and familiarity of these words ensure the test primarily measures audibility rather than complex understanding or cognitive processing. The final SRT value, expressed in dB HL, is not plotted on the graphic grid of the audiogram like pure tones. Instead, it is recorded numerically in a dedicated section of the audiogram form, typically labeled as the speech audiometry test battery.
The Step-by-Step Measurement Procedure
The procedure for determining the Speech Recognition Threshold follows a modified descending technique, similar to the method used for pure-tone threshold testing. The first step is patient familiarization, where the audiologist presents a subset of the spondee words at a clearly audible level. This ensures the patient knows the words that will be used and understands the task of repeating them.
After familiarization, the audiologist selects an initial presentation level, typically starting 30 to 40 dB above the patient’s estimated hearing threshold, often predicted from the 1000 Hz pure-tone result. The test begins with the presentation of one spondee word at this supra-threshold level to confirm the patient can hear and repeat the stimulus.
The core of the measurement involves a descending bracketing approach. For every correct response, the intensity is decreased by 10 dB for the next word presentation. When the patient misses a word, the intensity is increased by 5 dB for the subsequent presentation. This “down 10, up 5” adjustment brackets the actual threshold, gradually narrowing the intensity range.
At each intensity step, a set of spondee words, usually two to six, is presented to the patient. The threshold is defined as the lowest intensity level at which the patient can correctly repeat fifty percent of the presented words. For example, if six words are presented at a specific level, the patient must correctly identify at least three for that level to be recorded as the SRT.
Using the Pure Tone Average for Verification
The Speech Recognition Threshold is considered a measure of reliability because it is compared against the Pure Tone Average (PTA). The PTA is a single numerical value that summarizes the patient’s overall hearing sensitivity in the speech frequencies. It is calculated by taking the arithmetic mean of the air conduction thresholds at three frequencies: 500 Hz, 1000 Hz, and 2000 Hz.
This comparison is known as the cross-check principle, a standard in audiology used to ensure test validity. For the results to be clinically consistent, the measured SRT must agree with the calculated PTA within a tight tolerance, typically ±5 or ±6 dB. Agreement within this range confirms the patient’s responses during both pure-tone and speech testing were reliable and accurate.
If the SRT and the PTA do not agree outside of this tolerance, it suggests an inconsistency requiring further investigation. A large discrepancy, defined as a difference greater than 10 dB, can indicate technical issues, such as equipment malfunction, or patient-related factors. Patient factors might include non-organic hearing loss, where the patient is exaggerating their hearing difficulty, or a specific hearing loss configuration, such as a steeply sloping high-frequency loss.
In cases of steeply sloping hearing loss, the standard three-frequency PTA may not accurately predict the SRT because the 2000 Hz threshold is poorer than the lower frequencies. For these configurations, a two-frequency PTA, using the average of the two best thresholds at 500 Hz, 1000 Hz, or 2000 Hz, is sometimes calculated to provide a more accurate verification of the SRT.