The low dimensionality of post-stroke cognitive deficits: it's the lesion anatomy!

This article investigates the impact of stroke lesion anatomy on the estimation of fundamental cognitive functions using principal component analysis (PCA) in neuropsychological data. Through an in silico study simulating independent cognitive deficits based on lesion imaging from 300 stroke patients, the authors demonstrate that lesion anatomy alone induces statistical associations between deficits, leading PCA to underestimate the true dimensionality of cognitive functions by 62–70%. The findings highlight that low-dimensional structures identified by PCA may reflect lesion patterns rather than genuine cognitive architecture, challenging common interpretations of PCA-derived factors as fundamental cognitive processes. The study suggests that more theory-driven approaches and consideration of dissociation patterns are necessary to refine dimensionality estimates and improve the interpretation of cognitive deficits in post-stroke research.