Highlights Parkinson’s disease (PD) affects motor system and cognitive skills such as language. We used machine learning to study language network connectivity in early-stage PD. Connectivity was probed by classifying MEG responses to different spoken words. PD vs controls were successfully classified using verbs and incorrectly inflected nouns. Early-stage PD exhibits altered language network connectivity despite normal cognition.

Neural processing of language is still among the most poorly understood functions of the human brain, whereas a need to objectively assess the neurocognitive status of the language function in a participant-friendly and noninvasive fashion arises in various situations. Here, we propose a solution for this based on a short task-free recording of MEG responses to a set of spoken linguistic contrasts. We used spoken stimuli that diverged lexically (words/pseudowords), semantically (action-related/abstract), or morphosyntactically (grammatically correct/ungrammatical). Based on beamformer source reconstruction we investigated intertrial phase coherence (ITPC) in five canonical bands (α, β, and low, medium, and high γ) using multivariate pattern analysis (MVPA). Using this approach, we could successfully classify brain responses to meaningful words from meaningless pseudowords, correct from incorrect syntax, as well as semantic differences. The best classification results indicated distributed patterns of activity dominated by core temporofrontal language circuits and complemented by other areas. They varied between the different neurolinguistic properties across frequency bands, with lexical processes classified predominantly by broad γ, semantic distinctions by α and β, and syntax by low γ feature patterns. Crucially, all types of processing commenced in a near-parallel fashion from ∼100 ms after the auditory information allowed for disambiguating the spoken input. This shows that individual neurolinguistic processes take place simultaneously and involve overlapping yet distinct neuronal networks that operate at different frequency bands. This brings further hope that brain imaging can be used to assess neurolinguistic processes objectively and noninvasively in a range of populations.

Assessing the brain activity related to language comprehension is required in a range of situations. Particularly in cases when subjects’ cooperation with instructions cannot be guaranteed (e.g., in neurological patients), a protocol is needed that could be independent from attention and behavioral tasks. In this study, we aimed at designing a novel approach for neuromagnetic recordings of brain activity which could allow for probing the neural foundations underpinning three key levels of speech comprehension: lexical, semantic, and (morpho)syntactic, without requiring active attention on speech input or any active task, while keeping the recording session duration as short as possible. To this end, we designed two different auditory paradigms using the same set of single word‐based lexical, semantic, and syntactic contrasts: a modified version of the multifeature oddball paradigm and an equiprobable design. Combined magnetoencephalography/electroencephalography data were recorded form young, healthy participants, presented with these stimuli while watching a silent movie. Data from the equiprobable design yielded significant activations in temporal and inferior frontal areas associated with the lexical, semantic, and morphosyntactic contrasts. In turn, neural dissociations observed in the multifeature paradigm emerged mainly in temporal cortices, and were confined to the lexical and semantic conditions with a striking lack of any statistically significant effects for syntactic violations. Our findings indicate that, by employing the equiprobable design, a comprehensive range of key linguistic processes could be assessed in a passive, attention‐free manner within a relatively short time (here, ~27 min), thus making this paradigm a time‐efficient and patient‐friendly tool.

Objectives In this study, we investigated the effects of bilateral and unilateral deep brain stimulation of the subthalamic nucleus (STN‐DBS) in PD patients on neural responses associated with two aspects of spoken language processing: semantics of action‐related verbs and morphosyntactic processing. Materials and methods Using a passive unattended paradigm to present spoken linguistic stimuli, we recorded magnetoencephalographic (MEG) responses in three PD patients in four DBS conditions: left unilateral STN‐DBS, right unilateral STN‐DBS, bilateral STN‐DBS, and no STN‐DBS. To ensure that any observed effects of DBS on the neuromagnetic responses could be attributed to the linguistic context per se and were not merely induced by the electrical stimulation, we assessed the effects of STN‐DBS on linguistic contrasts within each stimulation condition. Hence, we contrasted the processing of action vs. abstract verbs as well as the processing of correct vs. incorrect morphosyntactic inflections within each DBS condition. Results The results revealed that, compared to the DBS‐off state, both bilateral and right unilateral stimulation of the STN yielded significant dissociations in the processing of action and abstract verbs, with greater neuromagnetic responses for action verbs compared to abstract verbs. For morphosyntax processing, only left unilateral stimulation yielded significant dissociations (relative to the DBS‐off state), with greater neuromagnetic responses to the incorrect inflections compared to the correct inflections. Conclusion The results reflect differential effects of unilateral and bilateral STN‐DBS on neuromagnetic responses associated with the processing of spoken language. They suggest that different specific aspects of linguistic information processing in PD are affected differently by STN‐DBS.