Speech Profiles and Cue Responsiveness After Left Hemisphere Stroke

摘要

ABSTRACT Speech production difficulties affect most of the approximately 2 million people with aphasia in the United States. Behavioral manifestations of speech impairment are heterogeneous and imperfectly categorized with current systems for differential diagnosis. Though customized treatment based on different speech profiles is widely believed to improve communicative abilities and quality of life, evidence-based practice is limited due to a poorly defined diagnostic process and uncertainty about which speech features distinguish among the core disorder categories (apraxia of speech, aphasia with phonemic paraphasia, unilateral upper motor neuron dysarthria), as they are defined behaviorally. In this project, we introduce a system for quantitative documentation and analysis of clinically relevant speech characteristics. The purpose of Aim 1 is to develop a new assessment for documenting speech profiles and estimating their severity. We define boundaries between normal and impaired speech production across key dimensions and metrics and identify a psychometrically sound and clinically feasible test for estimating severity and characterizing multidimensional speech profiles. In Aim 2, we develop a quantitative signature for speech profiles through empirically focused analyses. By replacing existing diagnostic checklists with empirically validated profiles, we expect project outcomes will resolve diagnostic confusion that has persisted for decades. The link between diagnosis and intervention is developed in Aim 3, in order to establish a foundation for precision intervention research. The project is expected to generate sound assessment and diagnostic procedures and remove fundamental barriers to scientific discovery concerning speech production difficulties after stroke.