Office: Muenzinger E039
Identifying regions of the brain associated with specific cognitive functions (e.g. visual processing) can be challenging. The common practice used in identifying the neural substrate associated with a specific function is to subtract activation responses to two different tasks which are presumed to vary in one discrete cognitive process. However, this subtraction often fails to isolate the neural substrates that support the targeted function due to the presence of modulatory factors. My research is focused on exploring machine learning algorithms that are commonly applied to functional magnetic resonance imaging (fMRI) data and their relative strengths and weaknesses in isolating parts of the brain while accounting for the modulation effect.