Moebius syndrome: measures of observer intelligibility with versus without visual cues in bilateral facial paralysis

This study investigated observers' intelligibility for the spoken output of an individual with Moebius syndrome (MoS) with and without visual cues.

An audiovisual recording of the speaker's output was obtained for 50 Speech Intelligibility in Noise sentences consisting of 25 high predictability and 25 low predictability sentences. Stimuli were presented to observers under two conditions: audiovisual and audio only. Data were analyzed using a multivariate repeated measures model. OBSERVERS: Twenty students and faculty affiliated with the Department of Speech Pathology and Audiology at the University of Nevada, Reno.

ANOVA mixed design revealed that intelligibility for the audio condition only was significantly greater than intelligibility for the audiovisual condition; and accuracy for high predictability sentences was significantly greater than accuracy for low predictability sentences.

The compensatory substitutional placements for phonemes produced by MoS speakers may detract from the intelligibility of speech. This is similar to the McGurk-MacDonald effect, whereby an illusory auditory signal is perceived when visual information from lip movements does not match the auditory information from speech. It also suggests that observers use contextual clues, more than the acoustic signal alone, to arrive at the accurate recognition of the message of the speakers with MoS. Therefore, speakers with MoS should be counseled in the top-down approach of auditory closure. When the speech signal is degraded, predictable messages are more easily understood than unpredictable ones. It is also important to confirm the speaking partner's understanding of the topic before proceeding.