Our knowledge of dysarthria is largely based on neuropathophysiologic labels such as flaccidity, or hypokinesia, which are assumed to "explain" speech production characteristics (and associated intelligibility deficits). However, this theory has been recently challenged by experimental data reporting mismatches between neuropathophysiology and speech characteristics. In addition, the fact that the majority of dysarthria studies have been conducted on speakers of American English also contributes to the uncertainty about the link between neuropathology and speech output. To address this concern, this project proposes an examination of the effect of native languages on speech characteristics of a single, often-studied dysarthria. The assumption of this approach is that it can shed light on the distinction between language-universal aspects of dysarthria and language-specific aspects of dysarthria. First, we propose obtaining acoustic and intelligibility measures within two languages (American English and Korean) to yield a closely matched set of acoustic measures and construct acoustically-based predictive models of speech intelligibility, within each language. For this aim, acoustic analysis will be performed on utterances produced by English-speaking and Korean-speaking individuals with Parkinson disease and hypokinetic dysarthria. The predictive models of speech intelligibility will be compared between the two languages, with the hypothesis that the weights of the variables contributing to speech intelligibility variation will differ between languages. We also propose to investigate a second perceptual aspect of speech of Parkinson disease, as judged by listeners who respond to utterances both in their native and non-native languages. For this aim, listeners will be asked to rate multiple speech dimensions of speakers with hypokinetic dysarthria, while listening to utterances they understand (native language match between speakers and listeners) and utterances they do not understand (mismatch between native language of speakers and listeners). The results from this novel approach to dysarthria are expected 1) to extend our fundamental knowledge about the nature of dysarthria, and 2) to provide a theoretical basis as to why speech/language variations should be considered in clinical assessment and management of dysarthria.
The proposed project will address language-universal and language-specific characteristics of dysarthria in Parkinson disease, and advance a line of research to broaden our knowledge about the relationship between neuropathology and speech output. It is hoped that insights from this project will also lead to the clinical consideration of speech/language variations in diagnosis and management of speech disorders.