D5 Vocal Audiometry: Intelligibility and Central Function

Topic

Instead of presenting pure tones—as in pure-tone audiometry—the speech audiometer presents the listener with standardized word lists: fixed sets of words validated beforehand to ensure that all words within a given list are equally easy or difficult to recognize. To measure different aspects of hearing, speech audiometry employs two types of standardized word lists: one composed of spondaic words—two-syllable words pronounced with equal stress on both syllables, making them easy to repeat even at very low volumes—and another composed of monosyllabic words, which are much more challenging to repeat correctly because even a slight distortion in perception can lead to confusion with a similar-sounding word.

When using spondaic words, the speech audiometer lowers the presentation intensity until it identifies the lowest level at which the listener can still correctly repeat half of the presented spondaic words; this level is known as the speech reception threshold. Because spondaic words are easy to repeat even with imperfect hearing, the speech reception threshold depends primarily on audibility—simply whether the listener can hear the spondaic word—and thus typically aligns closely with the thresholds previously established by pure-tone audiometry for that same ear.

In contrast, when using monosyllabic words, the speech audiometer does not seek a threshold; instead, it presents the words at a comfortable listening level—well above the speech reception threshold, where audibility is no longer an obstacle for the listener—and records the percentage of these monosyllabic words that the listener repeats correctly. This figure is known as the word recognition score. Since monosyllabic words are presented at a comfortable listening level where audibility is not an issue, the word recognition score ceases to reflect how much the listener hears and instead reflects the fidelity with which their auditory system transmits—without distortion—the monosyllabic words they do hear clearly.

When the word recognition score is significantly lower than what would be expected based on the listener's speech reception threshold and pure-tone audiometry thresholds, the discrepancy cannot be attributed to a lack of audibility, as the monosyllabic words were presented at a comfortable level where audibility was not the problem. Instead, a disproportionately low word recognition score indicates retrocochlear dysfunction—a problem with the auditory nerve, beyond the cochlea—or impaired central processing; in both cases, the signal that successfully passes through the peripheral ear arrives distorted or poorly synchronized before being recognized as a word.

By repeating the word recognition measurement while adding background noise, speech audiometry further requires the listener to separate the monosyllabic words from the accompanying background noise—a task that relies far more heavily on central processing than the same measurement performed in silence. A word recognition score that drops markedly with the addition of background noise—more than it would for a listener with the same speech reception threshold but without central processing impairment—confirms that the problem lies not merely in audibility or basic transmission fidelity, but specifically in the central processing system's ability to separate monosyllabic words from background noise. With these four measurements—the speech reception threshold, the word recognition score in quiet, the same score with added background noise, and a comparison of all three against pure-tone audiometry thresholds—the speech audiometer featured in this model can determine, using only standardized word lists, whether a listener's difficulty in understanding speech stems from simply not hearing it loudly enough, from the peripheral ear or auditory nerve transmitting a distorted signal, or from central processing failing to separate the speech from surrounding background noise.

ID:4251