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

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 respectively 15 and 39 subjects. 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.

[1]  Shin Ishii,et al.  Zero-shot fMRI decoding with three-dimensional registration based on diffusion tensor imaging , 2018, Scientific Reports.

[2]  Leonie Koban,et al.  Representation, Pattern Information, and Brain Signatures: From Neurons to Neuroimaging , 2018, Neuron.

[3]  Doris Y. Tsao,et al.  Single-Unit Recordings in the Macaque Face Patch System Reveal Limitations of fMRI MVPA , 2015, The Journal of Neuroscience.

[4]  Rainer Goebel,et al.  Information-based functional brain mapping. , 2006, Proceedings of the National Academy of Sciences of the United States of America.

[5]  Satrajit S. Ghosh,et al.  Prediction as a Humanitarian and Pragmatic Contribution from Human Cognitive Neuroscience , 2015, Neuron.

[6]  Bryan R. Conroy,et al.  A Common, High-Dimensional Model of the Representational Space in Human Ventral Temporal Cortex , 2011, Neuron.

[7]  Joachim Gross,et al.  The early spatio-temporal correlates and task independence of cerebral voice processing studied with MEG. , 2013, Cerebral cortex.

[8]  Martin A. Lindquist,et al.  Group-regularized individual prediction: theory and application to pain , 2017, NeuroImage.

[9]  R. Adolphs,et al.  Neural predictors of evaluative attitudes toward celebrities , 2016, Social cognitive and affective neuroscience.

[10]  Thomas E. Nichols,et al.  Nonparametric permutation tests for functional neuroimaging: A primer with examples , 2002, Human brain mapping.

[11]  Joachim M. Buhmann,et al.  Variational Bayesian mixed-effects inference for classification studies , 2013, NeuroImage.

[12]  Oluwasanmi Koyejo,et al.  What's in a pattern? Examining the type of signal multivariate analysis uncovers at the group level , 2016, NeuroImage.

[13]  Tyrone D. Cannon,et al.  Predicting risky choices from brain activity patterns , 2014, Proceedings of the National Academy of Sciences.

[14]  Emanuele Olivetti,et al.  Bayesian hypothesis testing for pattern discrimination in brain decoding , 2012, Pattern Recognit..

[15]  John-Dylan Haynes,et al.  Valid population inference for information-based imaging: From the second-level t-test to prevalence inference , 2015, NeuroImage.

[16]  Krzysztof J. Gorgolewski,et al.  The human voice areas: Spatial organization and inter-individual variability in temporal and extra-temporal cortices , 2015, NeuroImage.

[17]  Jonathan D. Cohen,et al.  Confounds in multivariate pattern analysis: Theory and rule representation case study , 2013, NeuroImage.

[18]  Bertrand Thirion,et al.  Graph-Based Inter-Subject Pattern Analysis of fMRI Data , 2014, PloS one.

[19]  C. Summerfield,et al.  Visual Prediction Error Spreads Across Object Features in Human Visual Cortex , 2016, The Journal of Neuroscience.

[20]  Jing Wang,et al.  Identifying Core Affect in Individuals from fMRI Responses to Dynamic Naturalistic Audiovisual Stimuli , 2016, PloS one.

[21]  Yi Chen,et al.  Statistical inference and multiple testing correction in classification-based multi-voxel pattern analysis (MVPA): Random permutations and cluster size control , 2011, NeuroImage.

[22]  C. Keysers,et al.  Is Brain Activity during Action Observation Modulated by the Perceived Fairness of the Actor? , 2016, PloS one.

[23]  E. Koechlin,et al.  Broca's Area and the Hierarchical Organization of Human Behavior , 2006, Neuron.

[24]  Joset A. Etzel,et al.  MVPA Permutation Schemes: Permutation Testing for the Group Level , 2015, 2015 International Workshop on Pattern Recognition in NeuroImaging.

[25]  Gaël Varoquaux,et al.  Cross-validation failure: Small sample sizes lead to large error bars , 2017, NeuroImage.

[26]  Lucila Ohno-Machado,et al.  Logistic regression and artificial neural network classification models: a methodology review , 2002, J. Biomed. Informatics.

[27]  A. Ishai,et al.  Distributed and Overlapping Representations of Faces and Objects in Ventral Temporal Cortex , 2001, Science.

[28]  J. S. Guntupalli,et al.  Decoding neural representational spaces using multivariate pattern analysis. , 2014, Annual review of neuroscience.