Date of Award
Dissertation (Access Restricted)
Doctor of Philosophy
Comm Sciences and Disorders
Speech Language Pathology
Michael P Cannito
Shaum P Bhagat
Corinna A Ethington
Roger J Kreuz
Because motor speech disorders are typically characterized by some degree of impaired speech intelligibility, the measurement of speech intelligibility has become an essential part of assessment and treatment of speech disorders. When evaluating speech intelligibility, consideration should be given to factors related to the stimulus material (i.e., length and predictability), the listener (i.e., working memory capacity) and the experimental paradigm. A review of the literature provided some evidence for the potential influence of stimulus predictability, stimulus length, and response format (i.e., typed vs. orally repeated) on speech intelligibility percentages and word recognition scores. The majority of these studies did not incorporate verbatim transcription or repetition tasks (the type of task used in a typical speech intelligibility assessment). The purpose of this study was to examine the influence of certain listener- and stimulus-related factors on the assessment of speech intelligibility in dysarthric speakers (i.e., the levels of predictability of the sentences were examined in association with sentence length). The effects of written transcription vs. oral repetition on speech intelligibility were also examined. Forty sentences representing four categories (short length-high predictability, short length-low predictability, long length-high predictability, and long length-low predictability) were presented to 30 healthy female listeners in a quiet listening environment. Pink noise was presented simultaneously with each sentence. One group of listeners typed what they heard, while the other group of listeners orally repeated what they heard. All responses were scored and percentages of speech intelligibility were obtained for each listener across all sentence conditions. A repeated measures ANOVA was used to analyze the listeners' responses. This analysis revealed significant main effects for length and group, as well as a significant interaction effect for length and predictability. Specifically, short sentences yielded significantly higher intelligibility scores than long sentences and the typing group produced significantly higher intelligibility scores than the oral repetition group. In the long sentence category, high predictability sentences yielded lower intelligibility percentages than the low predictability sentences. These results support previous findings concerning the important influence of stimulus materials and the listener's working memory capacity while also providing evidence for the use of a typed response format.
Dissertation or thesis originally submitted to the local University of Memphis Electronic Theses & dissertation (ETD) Repository.
Beverly, Doriann Denise, "The Effects of Stimulus Predictability, Length, and Response Format on the Measurement of Sentence Intelligibility" (2011). Electronic Theses and Dissertations. 2285.