Teaching Statistical Analysis Of Fmri Data

Functional magnetic resonance imaging (fMRI) represents a new and important topic in biomedical engineering. Statistical analysis of fMRI data is typically performed using free or commercial software packages that do not facilitate learning about the underlying assumptions and analysis methods; these shortcomings can lead to misinterpretation of the fMRI data and spurious results. We are developing an instructional module for learning the fundamentals of statistical analysis of fMRI data. The goal is to provide a tool for learning about the steps and assumptions underlying standard fMRI data analysis so that students and researchers can develop insights required to use existing analysis methods in an informed fashion and adapt them to their own purposes. The module includes a simulation of fMRI data analysis that provides students with opportunities for hands-on exploration of the key concepts using phantom data as well as sample human fMRI data. The simulation allows students to control relevant parameters and observe intermediate results for each step in the analysis stream (spatial smoothing, motion correction, statistical model parameter selection). It is accompanied by a tutorial that directs students as they use the simulation. The tutorial guides students through the individual processing steps, considering multiple cycles of fMRI data analysis and prompting them to make direct comparisons, with emphasis on how processing choices affect the ultimate interpretation of the fMRI data.