Link to bioRxiv paper: http://biorxiv.org/cgi/content/short/2020.07.20.212175v1?rss=1 Authors: Sperber, C., Rennig, J., Karnath, H.-O. Abstract: Background: Hemiparesis is a common consequence of stroke to the primary motor system. Brain mapping and post-stroke outcome prediction studies suggest that damage to additional brain areas plays a causal role in occurrence and severity of hemiparesis and its recovery. Objective: The study aimed to evaluate if damage to areas outside of the primary motor system can play a causal role in the occurrence of hemiparesis, or if such brain-deficit relations constitute a mere correlation. Methods: In a sample of 102 patients with unilateral stroke, the neural correlates of acute and chronic upper limb paresis were mapped by univariate and multivariate lesion behavior mapping. Following the same approach, CST lesion biomarkers were mapped, and resulting topographies of both analyses were compared. Results: All mapping analyses of acute or chronic upper limb paresis implicated areas outside of the primary motor system. Likewise, mapping CST lesion biomarkers - that, by definition, are only causally related to damage of the CST - implicated several areas outside of the CST with high correspondence to areas associated with upper limb paresis. Conclusions: Damage to areas outside of the primary motor system might not play a causal role in hemiparesis, while still providing predictive value. This finding suggests that simple theory-based biomarkers or qualitative rules to infer post-stroke outcome from imaging data might perform sub-optimally. Algorithms using theory-driven feature selection do not consider the complexity of lesion data, and could be improved by utilizing high-dimensional models with data-driven feature selection strategies. Copy rights belong to original authors. Visit the link for more info