What can modern statistics offer imaging neuroscience?

This paper presents a mixed-effects model, region-of-interest analysis of a longitudinal functional magnetic resonance imaging (fMRI) study of drug effects on human memory function. A key region of interest is the human hippocampus, affected by brain disorders such as Alzheimer's disease and schizophrenia. A brief section on human hippocampal cell microscopy complements the discussion of the macroscopic fMRI study. Statistical issues confronted in these two applications are then placed in a broader context for further discussion of the future roles of biostatisticians and our methods in the fertile intersection of applied mathematical abstraction and imaging neuroscience. Neuroscientific and fMRI background is provided for readers new to either area.