Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Does the language system reinvent itself in healthy and pathological ageing? MEG evidence of large-scale functional connectivity changes in older individuals and Parkinson’s disease patients.
Poster Session B, Friday, October 25, 10:00 - 11:30 am, Great Hall 3 and 4
Yury Shtyrov1, Rasha Hyder2,1, Mads Jensen1, Andreas Højlund1; 1Aarhus University, 2Cardiff University
Whereas many neurocognitive abilities decline with age, and even more so with age-related neurological disorders, changes in speech comprehension mechanisms in healthy and pathological ageing remain obscure, as the language function appears remarkably resilient to age-related degradation. We investigated the neural processing of spoken language in healthy older individuals and in patients with Parkinson’s disease (PD) — a neurodegenerative disorder which, apart from the motor system, adversely affects various cognitive functions. We devised a patient-friendly paradigm to probe language comprehension at different levels (lexical, semantic, morphosyntactic) without confounds related to active attention and overt motor responses that can be compromised in these populations. Participants were presented with spoken stimuli (action/abstract verbs, grammatically correct/incorrect inflectional forms, pseudowords) without any task, while their cortical activity was recorded using MEG. By applying machine learning-based classification algorithms on beamformer source estimates, we found that, for all linguistic contrasts, oscillatory activity patterns in bilateral distributed fronto-temporal networks diverged between older participants and younger controls across several frequency bands, suggesting multiple age-related changes in neurolinguistic circuits, possibly due to ageing as such and to compensatory processes taking place. These included a 200-ms delay (~100 vs. ~300 ms) in medium-gamma activation for the lexical (word/pseudoword) contrasts, a similarly delayed (~150 vs. ~400 ms) semantic activation accompanied by a beta-to-alpha shift in peak frequency, multiple changes for morphosyntactic contrasts, and overall, a more bilateral activation in older participants than in younger ones. Furthermore, a logistic regression classifier was able to classify newly diagnosed early-stage PD patients vs. healthy age-matched participants based on functional connectivity within large-scale temporo-fronto-parietal networks. The best classification results were achieved for responses to verbs and to incorrect inflections, indicating augmented involvement of both the core language cortices and the motor system in speech processing. These findings demonstrate quantifiable changes in cortical language-system connectivity in PD, which arise early, in the absence of overt clinically detectable cognitive or language deficits. Finally, we investigated effects of bilateral and unilateral deep-brain stimulation of the subthalamic nucleus (STN-DBS) on neural responses to spoken language in advanced PD. Compared to DBS-off state, both bilateral and right unilateral STN-DBS stimulation yielded significant dissociations in verb processing, with greater neuromagnetic responses for action verbs than abstract ones, potentially indicating restored engagement of the motor system in processing action-related semantics. For morphosyntax, only left unilateral stimulation yielded significant changes, with greater neuromagnetic responses to incorrect inflections than correct ones. This suggests that DBS can recover the normal incorrect>correct ELAN pattern, well-known from E/MEG studies in healthy adults. In sum, we show multiple changes in large-scale distributed neural networks underpinning spoken language processing, which take place in both healthy and neuropathological ageing. Furthermore, our results indicate that brain stimulation techniques are not only useful for ameliorating motor deficits in Parkinson’s disease but may also help restore normal neurolinguistic activity patterns. Finally, the present patient-friendly design combining task-free exposure to different linguistic stimuli with time-resolved neurophysiological recordings may in the future help develop non-invasive biomarkers of functional neurocognitive deficits in healthy and pathological ageing.
Topic Areas: Disorders: Acquired, Speech Perception