On computing variance explained: comment on Zablocki, Xu, Fan & Thompson (2025)
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
In a recent paper in Developmental Cognitive Neuroscience, Zablocki et al. (2025) propose Bayesian Regularized and Annotation-Informed Integrative Analysis of Cognition (BRAINIAC). A promising method that estimates the total variance explained by all features for a given cognitive phenotype. However, the measures Zablocki et al. (2025) use to quantify the variance explained may yield misleading results. The aim of this comment is to caution against the use of potentially misleading R2 estimates and to promote the use of valid ones.
