Predicting subject-driven actions and sensory experience in a virtual world with Relevance Vector Machine Regression of fMRI data

► This works illustrates the approach of Maastricht Brain Imaging Center to PBAIC 2007 competition. ► The aim was to predict subject driven actions and sensory experience in a virtual world from fMRI data. ► We employed Relevance Vector Machine in combination with several pre-processing strategies, achieving high accuracy. ► Predictive maps associated with each predictor, include, but are not limited to, specialized regions consistent with previous studies.