Comparing Two Connected Speech Tasks in Greek Speakers With the Logopenic Variant of Primary Progressive Aphasia and...

Different tasks have been used to elicit connected speech samples and analyze speech and language production in dementia. In this study, we explored whether there was a differential performance for individuals with Alzheimer's disease (AD) and the logopenic variant of primary progressive aphasia (lvPPA) on two elicitation tasks, picture description and story retell. Moreover, we investigated the relationship between connected speech measures and formal neuropsychological tests.

A total of 18 individuals participated in this study: six with lvPPA, six with AD, and six neurotypical adults matched for gender, age, and education. Connected speech analysis was based on a quantitative production analysis.

Differences between the two tasks were found for mean logarithmic frequency of open-class words, number of sentences, and mean utterance length. Individuals with lvPPA differed from individuals with AD and neurotypical adults on type-token ratio square root, mean utterance length, and sentence elaboration index. Significant intercorrelations between connected speech measures in the two tasks were found for speech rate and total dysfluencies. Finally, significant associations were detected between connected speech measures and formal neuropsychological tests.

Connected speech analysis can assist in diagnosing lvPPA and distinguishing it from AD. While dysfluencies and speech rate remain consistent across connected speech tasks, other variables vary. Story retell demands greater executive resources and encourages richer vocabulary and morphosyntactic production, more accurately reflecting real-life communication compared to traditional picture description tasks. The study highlights the value of using connected speech assessment in clinical evaluations and carefully selecting appropriate tasks.

https://doi.org/10.23641/asha.29114060.