Syntactic impairment in Parkinson's Disease : cross-task differences and relationship to cognitive impairment

Lowit, A. and Thies, T. and Roheger, M. and Kalbe, E. and Steffen, J. and Barbe, M. (2021) Syntactic impairment in Parkinson's Disease : cross-task differences and relationship to cognitive impairment. In: 18th Biennial Conference of the International Clinical Phonetics and Linguistics Association, 2021-06-22 - 2021-06-25, Strathclyde University.

[thumbnail of Lowit-etal-ICPLA-2021-Syntactic-impairment-in-Parkinsons-Disease-cross-task-differences]
Text. Filename: Lowit_etal_ICPLA_2021_Syntactic_impairment_in_Parkinsons_Disease_cross_task_differences.pdf
Accepted Author Manuscript

Download (270kB)| Preview


The presence of language difficulties in Parkinson's Disease (PD) is increasingly recognised. Less is known about the exact manifestation of these problems, their relationship to other cognitive domains, and how they vary across different task complexities. Particularly the latter is important in order to establish sensitive and clinically useful assessment procedures, which can then serve as the basis for developing effective treatment strategies. This study aimed to contribute to this knowledge base by investigating language performance in people with PD (PwPD) across a number of speaking tasks and to correlate this performance with cognitive skills. 22 PwPD and 22 healthy matched control participants performed a range of executive function and linguistic tasks. Four participants in each group presented with mild cognitive impairment (MCI), the rest were cognitively healthy. Language tasks ranged from complex grammar tasks to less structured sentence generation and picture description tasks. The investigation focused on grammatical accuracy and complexity in all three tasks, as well as speech characteristics (sentence generation and picture description) and information content (picture description only). Results indicated that PwPD produced more grammatical errors than their healthy controls across all language tasks. No significant differences were evident for other parameters such as grammatical complexity, false starts, pauses, concepts, etc. Error rate in the complex grammar was linked to set shifting ability whereas performance in the other tasks was independent of cognition. Syntactic performance did not correlate across the three tasks, indicating that each task presented with its own difficulties and error patterns. Our study highlights that language problems can occur at early stages of PD before speech or cognitive problems arise. In addition, the data demonstrate that at this stage only highly complex grammatical tasks depend on cognitive performance, whereas more natural language production appears to be decoupled from this skill. The findings that result from structured language tests cannot predict day to day communication performance highlighting that clinicians should focus on functional assessment in order to provide early effective intervention.