Examining the Cognitive Linguistic Differences Between Adults with Mild Cognitive Impairment (MCI) and a Group of Healthy Controls using a Devised Cognitive Linguistic Assessment

Abstract

Background: Changes in language often precede other notable cognitive changes in people who go on to develop dementia, however, given their subtlety, these early-stage communication signs are frequently missed. Speech and language features have been widely cited as highly sensitive markers for early detection of cognitive impairment (Filiou et al., 2019; Thomas et al., 2020; Martinez-Nicholas et al., 2021; Robin et al., 2021; Sanborn et al., 2022) and for this reason, it is imperative we develop screening tools that incorporate detection of these subtle speech and language changes. While we have come quite a long way in identifying speech and language deficits associated with cognitive impairment, there is still a lot to unravel so that it can be utilised to its full potential. Examining the language profile associated with MCI compared with the language profile of healthy older adults can facilitate development of fast, easily accessible cognitive screening tools using cognitive linguistic tasks. Aim: This study aims to identify the value of commonly used cognitive linguistic assessment tasks in identifying those with mild cognitive impairment (MCI) compared with a healthy control (HC) group and assess if a quick, informal cognitive linguistic assessment completed in digital form and developed by the researcher, similar to ones that are used regularly by Speech and Language Therapists (SLTs), could be more targeted and sensitive in identifying those at risk for cognitive impairment. Methods: This study included 40 participants (20 HCs and 20 MCI). The study was a quantitative, prospective, cross sectional, observational design. Participants completed a cognitive linguistic assessment in digital form using a tablet that was designed by the researcher and wider research team. The cognitive linguistic assessment included picture naming, repetition, verbal fluency, reading, picture description, list learning, list recall, list recognition and digit span. Differences were examined across groups and across age/years of education. Cognitive linguistic scores were examined for differences across groups using the Mann Whitney U test. Results: There was a statistical difference across groups in the subtests of picture naming (p <.001), list learning (p <.001), list recall (p <.001), list recognition (p <.001) and repetition (p=.008). Conclusion: The results of this study indicate that picture naming, list learning, list recall and list recognition were the subtests which could successfully identify participants in the healthy control group versus those in the MCI group. Clinically, it can be difficult to objectively measure, as a SLT, subtle changes in someone?s cognitive linguistic profile, however this study has the potential to facilitate this type of assessment. Linguistic measures alongside acoustic, prosodic and voice measures could increase the predictive value of these types of screening tools in the future and this study is the first step in this type of research.