The primary aim of Dr. Purcell’s research is to understand how the brain represents written language. He has a specific interest in neural changes associated with (re-)learning written language after brain injury. More broadly though, he is interested in novel statistical approaches for indexing and tracking neural changes associated with learning. Using functional magnetic resonance imaging (fMRI), he has developed a novel and generally applicable regression based algorithm (Local-Heterogeneity Regression) for quantifying learning induced changes in local neural representations.  In collaboration with other members of the MNC he is expanding on the application of this approach to a variety of populations and cognitive domains, and is working to develop further novel approaches for quantifying learning induced representational changes in the brain.