Jackson, Pamela L.
M.A. (Master of Arts)
Department of Communicative Disorders
This study was conducted to investigate consonant lipreading performance at 0 degree and 90 degree angles of observation. Twenty consonants were placed in a consonant-vowel context to form 20 nonsense monosyllables. Each syllable was recorded on videotape ten times in random order for a total of 200 items. The speaker was simultaneously videotaped from a 0 degree angle and a 90 degree angle under controlled lighting conditions. Ten normally hearing graduate students in the Department of Communicative Disorders at Northern Illinois University served as viewers for this investigation. Small groups of viewers watched each of the tapes with no sound. Each received an answer sheet on which to record the consonant that was visually perceived for each stimulus item. Responses were recorded on individual confusion matrices for each subject at each viewing angle. A master matrix was tabulated for each angle of observation. The influence of the two viewing angles was assessed by performing a jt-test for related measures. In addition, confusion matrices were visually inspected in order to analyze the confusions that occurred in the identification of the stimulus items. An analysis of variance was performed in order to determine if a difference between angles existed. A Newman-Keuls post-hoc analysis was also performed to determine where significant differences existed. Final analysis of the data included computing the percentage of information transmitted by articulatory features relative to the maximum possible information. The features selected were voicing, nasality, affrication, duration, front place of articulation, and stop. Results indicated that consonants were significantly easier to identify at the 0 degree angle than at the 90 degree angle. Also, predictable confusions occurred among the consonants, and homophenous groups were formed. An analysis of variance and Newman-Keuls post-hoc analysis indicated that the groups /l,n/, /j/, and /Θ,[?]/ were identified correctly more often at the 0 degree angle than at the 90 degree angle. Results of the information transfer indicated that the place of articulation feature was transmitted most effectively while the features of voicing and nasality were transmitted poorly for both angles of observation.
Segriff, Carla H., "Consonant visibility from two horizontal viewing angles" (1986). Graduate Research Theses & Dissertations. 1927.
Northern Illinois University
Rights Statement 2
NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.