Improving standards in brain-behaviour correlation analyses

• Main Authors: Guillaume A Rousselet, Cyril R Pernet
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2012-05-01
• Series: Frontiers in Human Neuroscience
• Subjects: Pearson correlation; outliers; robust statistics; multiple comparisons; Spearman correlation; skipped correlation
• Online Access: http://journal.frontiersin.org/Journal/10.3389/fnhum.2012.00119/full

Associations between two variables, for instance between brain and behavioural measurements, are often studied using Pearson correlation. However, Pearson correlation is not robust: outliers can introduce false correlations or mask existing ones. These problems are exacerbated in brain imaging by a widespread lack of control for multiple comparisons, and several issues with data interpretations. We illustrate these important problems associated with brain-behaviour correlations, drawing examples from published articles in mainstream high-impact and specialty journals. We make several propositions to alleviate these problems.