How about that? Psycholinguistic characteristics of formulaic language that predict fluency in individuals with post-stroke aphasia

Purpose: Formulaic language is an under-explored area of research in the field of acquired language disorders as compared to propositional language. The primary purpose of this study was to explore the utility of a proposed theoretical formulaic language model (Van Lancker Sidtis, 2022) for individuals with post-stroke aphasia to inform research and clinical practice. Method: The dataset included previously described formulaic language extracted from Aphasiabank speech samples produced by 144 individuals with fluent and non-fluent aphasias. Formulaic language items were coded according to six psycholinguistic characteristics from the theoretical model. Between-group comparisons and regression analyses were run to determine whether particular psycholinguistic characteristics of produced formulaic items could predict speaker fluency. Results: Findings revealed formulaic language differences between fluent and nonfluent aphasias based on the theoretical model. Importantly, psycholinguistic characteristics of frequency and syntactic completeness along with presence of apraxia of speech predicted fluency status with high accuracy (88.4% of individuals with fluent and 70.3% with nonfluent aphasia). Conclusions: Findings in this study illustrate how theoretically-driven analyses of formulaic language production may enhance diagnostic and intervention practices in post-stroke aphasia.