Functional magnetic resonance imaging (fMRI), a non-invasive technique to capture brain
activity, is getting increasingly important in psychological research. Nowadays, instead of the whole brain, researchers are particularly interested in the activation pattern of very specific brain regions of interest (ROI). So-called localizer tasks, executed after the main task of interest, are used to define functional ROIs (fROIs) in each subject individually. The current data-analytical methods however tend to localize these fROIs inconsistently, focusing on avoiding detection of activity where there is none. The opposite, actually avoiding to misstrue activation is nevertheless an equally important focus in this context. The aim of this project is to provide a technique for defining an fROI that balances between these two focuses. This will be done by not only making conclusions based on information against the hypothesis of no activation but also on information against the hypothesis of true activation.
This will lead to larger reliability, stability and spatial accuracy, which is highly
recommended, since the reliability of the successive data analysis is build on the accuracy of the fROI. We also aim to expand the results of this procedure to group analyses, in which information of activation in the fROIs is summarized for each individual and aggregated across individuals to test group hypotheses.