Learning Visual Correlates of Fricative Production by Normal-Hearing Subjects: A Preliminary Evaluation of the Video Articulator

Abstract

This investigation provides a preliminary evaluation of the use of the video articulator, a phonemic recognition device for the hearing impaired. The subjects were five young adults with normal hearing and vision (corrected) who were matched with respect to age, sex, dialect, education, and phonological sophistication. Each subject received 150 min of programmed training to learn the video configurations of the eight English fricatives both in isolation and consonant-vowel contexts. Following the training period, the subjects were given a test to determine adequacy of learning and retention of the video configurations for the training stimuli, in the absence of auditory cues. The subjects' responses were analyzed using a common covariance measure. The results demonstrated generally low transmission values for consonants in isolation. Moreover, identification of consonants in context was less accurate. The subjects, as a group, had greater difficulty in recognizing the productions of other subjects when compared with recognition of their own utterances. The clinical implications of these findings are discussed.