Spoken Discourse Biomarker of PD Cognitive Impairment

The purpose of this study is to identify unique profiles of speech and language changes that distinguish individuals with Parkinson's disease from adults without Parkinson's disease and individuals with Parkinson's disease with cognitive (e.g., memory, thinking skills) impairment from those without cognitive impairment.

Aim 1 will characterize PD-MCI (Parkinson's disease mild cognitive impairment), PDN (Parkinson's disease without cognitive impairment), and HA (healthy adult) spoken discourse, cognitive, and motor speech profiles. Phase 2 biomarker development requires robustly characterized cohorts in which to test candidate biomarkers. Using a standardized battery of cognitive, language, and motor speech tests PD participants will be assigned to PD-MCI (single/multi-domain) or PDN groups. The investigators propose collecting spoken discourse samples using standardized elicitation protocols. The same tasks will be extracted from the extant HA database. Researchers will transcribe, code, and analyze discourse samples. Group differences (including sub-analyses for single and multi-domain MCI subtypes), elicitation stimuli effects, and group x stimuli interactions will be examined using multivariate and mixed-design ANOVA procedures. Aim 2 will develop and evaluate the classification accuracy of an optimally weighted discourse classification function for PD-MCI and PDN. The investigators propose using discriminant function analysis to identify an optimized composite variable that best predicts PD-MCI, PDN, and HA group membership. Sensitivity/specificity analyses, positive/negative predictive values, and receiver operating characteristic curves will be used to evaluate the discourse classification function properties. The primary endpoint is an optimally weighted discourse function that can classify PD-MCI with \> 80% sensitivity/specificity.