Christoph Sperber

This study benchmarks predictive models for long-term stroke outcomes using a global consortium, revealing mechanistic diversity and key predictors across motor, cognitive, and emotional domains.

The paper investigates spatial neglect in subcortical stroke and finds that disconnection of long cortico-cortical association fibres predicts the condition, without requiring subcortical grey matter involvement.

The paper examines how covariate control in brain imaging analysis can bias results by systematically shifting associations between imaging features and covariates, without necessarily improving precision.

This study compares data-driven and theory-based biomarkers for predicting chronic motor outcomes after stroke, finding data-driven models perform better, especially when using structural disconnection measures.

The paper investigates brain disconnection in pusher syndrome using functional and structural measures in stroke patients, identifying thalamo-cortical network dysfunction as a likely cause.

This study examines how digital formats affect spatial neglect test behavior and finds the CoC measure is robust to digitization, while additional search behavior metrics can help differentiate neglect patients.

The paper maps structural white matter disconnections in apraxia by analyzing diffusion tensor imaging data from stroke patients, identifying key fronto-temporo-parietal and subcortical networks involved.

The paper questions the role of lesion size in cognitive deficits and argues that controlling for lesion size is not a valid covariate control.

This study evaluates PCA and factor analysis for identifying fundamental cognitive functions in neuropsychological data, finding they can measure underlying variables with high precision but face limitations due to data structure and factor rotation challenges.

The paper investigates how disconnections in left-hemispheric brain areas, particularly the angular gyrus, affect arithmetic fact retrieval following stroke.

The paper re-analyzes the anatomical basis of pathological line bisection using multivariate lesion-symptom mapping in stroke patients, finding multiple lesion locations and white matter damage contribute to the error.

The paper assesses whether lesion information beyond the primary motor system can predict motor outcomes after stroke, finding that such information improves prediction even when the biomarker is based on corticospinal tract damage.

This study confirms right-hemisphere predominance of lesions in lesional mania and identifies specific brain regions associated with mood regulation.

This paper examines the challenges of lesion-behaviour inference and highlights the limitations of multivariate models in capturing causal relationships in brain lesion data.

This study uses multivariate lesion-behaviour mapping to identify a common anatomical network for higher-order motor control in apraxia of pantomime.

This study uses multivariate lesion-mapping to identify a brain network associated with spatial neglect and attention in right brain damaged patients, showing benefits over univariate methods.

This study evaluates SVR-LSM for multivariate lesion behaviour mapping, finding that multiple comparison correction, sufficient sample size, and susceptibility to vascular misplacement are critical methodological considerations.