Inter-subject pattern analysis A straightforward and powerful scheme for group-level MVPA

Abstract : Multivariate pattern analysis (MVPA) has become vastly popular for analyzing functional neuroimaging data. At the group level, two main strategies are used in the literature. The standard one is hierarchical, combining the outcomes of within-subject decoding results in a second-level analysis. The alternative one, inter-subject pattern analysis, directly works at the group-level by using, e.g. a leave-one-subject-out cross-validation. This study provides a thorough comparison of these two group-level decoding schemes, using both a large number of artificial datasets where the size of the multivariate effect and the amount of inter-individual variability are parametrically controlled, as well as two real fMRI datasets comprising 15 and 39 subjects, respectively. We show that these two strategies uncover distinct significant regions with partial overlap, and that inter-subject pattern analysis is able to detect smaller effects and to facilitate the interpretation. The core source code and data are openly available, allowing to fully reproduce most of these results.
Keywords : fMRI MVPA group analysis
Document type :
Journal articles
Complete list of metadatas

Cited literature [34 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02313840
Contributor : Sylvain Takerkart <>
Submitted on : Friday, October 11, 2019 - 3:27:16 PM
Last modification on : Tuesday, October 29, 2019 - 3:08:02 PM

File

Wang_ISPA_accepted_manuscript....
Files produced by the author(s)

Identifiers

Collections

Citation

Qi Wang, Bastien Cagna, Thierry Chaminade, Sylvain Takerkart. Inter-subject pattern analysis A straightforward and powerful scheme for group-level MVPA. NeuroImage, Elsevier, 2019, pp.116205. ⟨10.1016/j.neuroimage.2019.116205⟩. ⟨hal-02313840⟩

Share

Metrics

Record views

59

Files downloads

96