Intelligibility is a multifaceted construct that is influenced by a myriad of variables such as linguistic-contextual knowledge of the listener, as well as the acoustic signal produced by the speaker. However, there has been little systematic investigation of the differential effects of listener-related and speaker-related variables on intelligibility or the extent to which these variables can be exploited to enhance intelligibility for speakers with chronic dysarthria. The long-term goal of this research is to understand the independent and interactive contributions of speaker-related and listener-related variables to speech intelligibility in individuals with chronic dysarthria of varying severity. The intervention strategy, alphabet supplementation (AS), in which speakers with dysarthria point to the first letter of each word while simultaneously speaking, is of particular interest in the present research. Preliminary studies suggest that AS has the potential to significantly enhance intelligibility of dysarthric speech; however; the reasons why intelligibility is improved are unknown. It is hypothesized that both speaker-related production changes and linguistic knowledge provided to listeners play a role. Two related studies will investigate the effects of these variables. Study 1 will examine the effects of intrinsic and extrinsic listener knowledge on intelligibility of habitual dysarthric speech through manipulation of semantic predictability and experimentally imposed alphabet cues. Study 2 will examine the effects of speaker-related production changes associated with implementation of AS through analysis of spectral and temporal acoustic differences between habitual speech and speech produced while implementing AS. Across both studies, 16 individuals with dysarthria secondary to cerebral palsy will serve as speakers and 256 non-disabled individuals will serve as listeners. Listeners will orthographically transcribe videotapes of speakers using habitual speech and using alphabet supplementation. Intelligibility data and acoustic data will be subjected to analysis of variance to answer research questions. The proposed studies will gather clinically-relevant efficacy data regarding AS as an intervention strategy within a theoretical framework. Outcomes of these studies will provide critical clinical information as well as important theoretical information that will begin to quantify contributors to speech intelligibility in dysarthria.
