Multiple testing correction over contrasts for brain imaging

The multiple testing problem arises not only when there are many voxels or vertices in an image representation of the brain, but also when multiple contrasts of parameter estimates (that represent hypotheses) are tested in the same general linear model. We argue that a correction for this multiplici...

Full description

Saved in:

Bibliographic Details
Published in:	NeuroImage (Orlando, Fla.) Vol. 216; p. 116760
Main Authors:	Alberton, Bianca A.V., Nichols, Thomas E., Gamba, Humberto R., Winkler, Anderson M.
Format:	Journal Article
Language:	English
Published:	United States Elsevier Inc 01-08-2020 Elsevier Limited Elsevier
Subjects:	Adult Brain - diagnostic imaging Brain imaging Contrast correction Data Interpretation, Statistical Humans Hypotheses Medical imaging Models, Statistical Multiple comparisons Multiple testing Neuroimaging Neuroimaging - methods Neuroimaging - standards Permutation tests Probability Psychological Distance Psychometrics - instrumentation Variance analysis Multiple testing Brain imaging Permutation tests Contrast correction Multiple comparisons
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	The multiple testing problem arises not only when there are many voxels or vertices in an image representation of the brain, but also when multiple contrasts of parameter estimates (that represent hypotheses) are tested in the same general linear model. We argue that a correction for this multiplicity must be performed to avoid excess of false positives. Various methods for correction have been proposed in the literature, but few have been applied to brain imaging. Here we discuss and compare different methods to make such correction in different scenarios, showing that one classical and well known method is invalid, and argue that permutation is the best option to perform such correction due to its exactness and flexibility to handle a variety of common imaging situations. •Correction for multiple testing across contrasts is necessary.•ANOVA followed by pairwise comparisons does not control the error rate in most cases.•Some well-known methods are conservative when a subset of the contrasts is tested.•Permutation performes well in all simulation scenarios.
Bibliography:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 CRediT authorship contribution statement Bianca A.V. Alberton: Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Writing - original draft, Writing -review & editing, Visualization. Thomas E. Nichols: Writing - original draft, Conceptualization, Validation. Humberto R. Gamba: Resources, Funding acquisition, Project administration. Anderson M. Winkler: Conceptualization, Methodology, Software, Validation, Formal analysis, Resources, Writing - original draft, Writing - review & editing, Supervision.
ISSN:	1053-8119 1095-9572
DOI:	10.1016/j.neuroimage.2020.116760