Control without cause: How covariate control biases our insights into brain architecture and pathology
Christoph Sperber, Laura Gallucci, Marcel Arnold, Roza M. Umarova
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.
bioRxiv · Apr 12, 2024 · PreprintNeuroscience