Eliciting evidence on linguistic perspectives for Subjective Cognitive Impairment, Mild Cognitive Impairment and Alzheimer’s Disease: A cross‐sectional study

Abstract Background Most descriptions of the clinical and cognitive features of Alzheimer’s Disease (AD) highlight deficits mainly in memory ability, while the in‐depth understanding of the language deficits of people with AD remains limited with most data reported as heterogeneous (see Eyigoz et al ., 2020; Kavé & Dassa, 2018). The overarching goal of DemLENS project (https://demlens.enl.auth.gr/) is to investigate the key linguistic characteristics of native speakers with cognitive impairment and the decline of language skills. Method The study integrates the clinical evaluation, the neuropsychological assessment and thorough linguistic profiling via language production data from 168 native speakers of Greek. Specifically, we examine the lexical, syntactic and content complexity of spoken language elicited through sentence repetition, single picture description, picture sequence narratives and personal narratives. We tested young healthy adults (yHC = 35), cognitively intact elder healthy adults (eHC = 31), speakers with Subjective Cognitive Decline (SCD = 32), speakers with Mild Cognitive Impairment (MCI = 33) and speakers with mild to moderate AD (AD = 38). Result The data analysis focuses on the annotated data at the microstructure level – e.g. POS, syntactic complexity, internal state terms – and at the macrostructure level – e.g. information content units, narration length and narration density and revealed that depending on the cognitive severity stage, language subsystems deteriorate at different stages in the progression of AD. Lexical access difficulties, reduced informational content (e.g. narration density: F (4,164) = 6.514, p <.001; yHC,eHC>SCI,MCI>AD) and syntactic complexity of speakers’ production appear to drive linguistic decline. Note also that sentence repetition was shown to have high discriminatory value in the development of AD (e.g. accuracy index: F (4,164) = 92.632, p <.001; yHC>eHC,SCI>MCI>AD (Figure 1). Conclusion Language performance exposes subtle early signs of progression to AD that can shed light on the early detection of cognitive decline, thus facilitating the clinical process. Consequently, a language‐based AD assessment is ecologically valid and one of the least intrusive ways to assess symptoms of AD. References Eyigoz, E., Mathur, S., Santamaria, M., Cecchi, G., & Naylor, M. (2020). Linguistic markers predict onset of Alzheimer’s disease. EClinicalMedicine , 28 , 100583. Kavé, G., & Dassa, A. (2018). Severity of Alzheimer’s disease and language features in picture descriptions. Aphasiology , 32 (1), 27‐40.