Halai et al., 2017; Schumacher et al., 2019
Transdiagnostic Perspective on Aphasia: We advocate for understanding post-stroke language impairments through continuous, non-overlapping dimensions such as phonology, semantics, fluency, and executive function, rather than relying on discrete categorical classifications of aphasia.
Dimension Reduction Methods: We employ data-driven techniques like Principal Component Analysis (PCA) to identify core behavioral components across large patient groups, capturing individual variability within a unified model of post-stroke language deficits.
Integrated Language and Non-Language Networks: Recognizing the frequent co-occurrence of aphasia with non-verbal executive and attentional deficits, we investigate the behavioral and neural interactions between these separable dimensions during recovery and rehabilitation.
Behavior Models: We utilize both univariate and multivariate lesion-symptom mapping methods to establish links between behavior and recovery outcomes with specific brain regions, white matter tracts, and neural networks, aiming to elucidate complex symptom-brain relationships.
Predictive Value for Therapy: Identifying neural markers associated with behavior and recovery has the potential to guide personalized therapeutic interventions that target specific dimensions of impairment, moving beyond broad diagnostic categories.
