Title: Thinking outside of the voxel: beyond mass-univariate analyses in functional MRI.

Abstract: Some critics of functional magnetic resonance imaging (fMRI) research have denounced the field by labelling it "modern phrenology”. This suggests that at best, fMRI might tell us where in the brain things happen, but not why or how;  and at worst, fMRI is a pseudoscience. Unfortunately, there may be some basis for such criticism in regards to the most commonly used “traditional” approach to functional neuroimaging analysis: the mass-univariate (general linear model) approach, which treats somewhat arbitrarily sized brain voxels as independent units. However, a whole host of more advanced analytic techniques are now available and widely used in fMRI research that can provide valuable insights into the nature of brain-behaviour relationships, including multivariate techniques such as Partial Least Squares, multivoxel pattern analysis, and graph theoretical network analysis. In this talk, I will provide an overview of how some of these analytic approaches have been implemented in fMRI research, including examples from my own work.