The Concise Language Paradigm in left and right hemisphere stroke: a longitudinal source-localized EEG study

Abstract

The mechanisms underlying language recovery after stroke remain poorly understood. Most previous studies have either focused on language comprehension or production and/or relied on hemodynamic measures, with limited complementary electrophysiological evidence. Here, we combined the Concise Language Paradigm (CLaP), assessing comprehension and production, with individualized electroencephalography (EEG) source reconstruction to investigate post-stroke language function and changes in hemispheric dominance over time. In two sessions, we tested stroke-free controls, individuals with left hemisphere (LH) or right hemisphere (RH) stroke. Participants with stroke were assessed two and seven months post-stroke. EEG data were analyzed using event-related potentials and time-frequency representations, followed by source reconstruction based on state-of-the-art individual MRI-derived forward models. Laterality indices were calculated to assess changes in hemispheric dominance between sessions. Behaviorally, the LH group showed low performance at two months followed by overall improvement, whereas the RH group mainly improved in functional communication. EEG activity was significantly modulated by task condition across comprehension and production contrasts. Source-level results in controls were consistent with established language-related networks, serving as regions of interest for the stroke analyses. In the stroke groups, hemispheric dominance differed between LH and RH stroke participants, with LH participants showing systematic shifts from left- to right-hemisphere dominance between sessions, if not already right-hemisphere dominant at two months. Our study provides evidence on comprehension and production performance after stroke using a single paradigm. These findings demonstrate dynamic changes in hemispheric dominance during post-stroke recovery, ultimately highlighting the value of individualized EEG source reconstruction to investigate post-stroke plasticity.