Stroke is a leading cause of functional disability, with cognitive deficits—particularly action slowing and executive dysfunction—strongly linked to poor outcomes. This multimodal project aims to identify key brain regions that influence action speed and to develop personalized effective connectivity models using dynamic causal modeling (DCM) and high-density EEG to optimize targeted repetitive transcranial magnetic stimulation (rTMS) protocols for improving action speed in stroke patients. To investigate the role of seven brain regions involved in action speed modulation, we will model their interactions using DCM by disrupting ongoing activity through TMS-induced virtual lesions in 60 healthy individuals. Subsequently, we will evaluate the impact of offline intermittent theta burst stimulation (iTBS) on action speed, focusing on regions identified as critical to action speed modulation. Finally, we will assess the efficacy of iTBS in enhancing action speed in 20 stroke patients during the first three months post-stroke, using individualized EEG-based DCMs. By comparing the models of stroke patients to those of healthy controls, we aim to identify optimal regions for individualized iTBS targeting. This approach emphasizes personalized modeling to account for variability in brain damage among stroke patients, which is expected to improve the effectiveness of TMS-based rehabilitation. Anticipated outcomes include improved action speed in stroke patients, potentially reducing long-term disability and healthcare costs, while advancing knowledge in neuromodulation and rehabilitation.