A comparison of different connected-speech tasks for detecting mild cognitive impairment using multivariate pattern analysis

Background: It is common for the older adult population to have age-associated cognitive decline and develop neurodegenerative diseases such as dementia. Several studies have suggested that classification algorithms based on linguistic features may be useful for the early detection of mild cognitive impairment (MCI). Aims: The current study aimed to examine connected-speech performance in people with MCI and cognitively healthy controls (HC) using connected-speech tasks and tested whether patterns of lexical-semantic features extracted from these tasks could classify MCI and HC groups. Methods & procedure: We selected 16 English-speaking participants with MCI and 16 matched HC from the Delaware corpus. Four connected-speech tasks (a picture description, a story narrative, a story recall, and a procedural narrative) and eight lexical-semantic features were selected for analyses. Outcomes & results: Two-way ANOVAs results showed that there were group differences in revision ratio and core lexicon. Multivariate pattern analysis (MVPA) results demonstrated that the story recall task is the only task that can discriminate the two groups above chance. Conclusion: To conclude, the results provide extra evidence that connected-speech tasks can detect subtle language changes in people with MCI. It also showed that by using multivariate pattern analyses, the pattern of lexical-semantic features could significantly predict the class that a participant is associated with (MCI or HC). In addition, the story recall task could better discriminate participants with MCI than the other three connected-speech tasks.