Multivariate approach toward classification of competition and collaboration: An fMRI study

Abstract

This functional magnetic resonance imaging (fMRI) study aimed to distinguish neural activation associated with competition and collaboration using multivoxel pattern analysis (MVPA). For each participant, a searchlight-based MVPA was applied to select informative voxels within training data. The support vector machine with a radial basis function kernel was used to obtain classification accuracy of the informative regions. As a result, within-individual maximum classification performance for the test data reached maximally 94.6%. Important regions classifying competition and collaboration were mainly found within prefrontal cortex (e.g., superior/middle frontal gyri) and visual area (e.g., calcarine sulcus and lingual gyrus). Furthermore, visual regions and dorsolateral prefrontal regions showed average accuracy around 70% across participants. In short, neural contribution during competition or collaboration was characterized as differences in multivoxel pattern with a high accuracy.