Michael T Rubens

At the moment, Michael is involved with the collection and analysis of fMRI data from several different studies that are currently being conducted in the Gazzaley Lab. These studies are investigating top-down modulation, a neural process that allows us to enhance relevant information and suppress irrelevant information, networks related to attention and memory, as well as neural differences related to short-term and long-term memory processes. His programming skills and knowledge of fMRI have lead to him becoming a primary contributor to the creation and maintenance of analysis tools used througout the lab for this modality. Aside from implementing standard univariate analyses commonly used throughout the field (statistical parametric mapping), he is also surveying multivariate approaches such as Independent Component Analysis, Beta Series Correlation (Rissman et al., 2004), and Multi-Voxel Pattern Analysis. Until recently, Michael was working on an fMRI-guided Transcranial Magnetic Stimulation (TMS) study with Electroencephalographic (EEG) recordings. The aim of this study is to see if performance on a low-level visual working memory task can be impaired by applying TMS to a frontal control region. His foray into EEG is expanded even further with his current academic endeavor. Presently, Michael is engaged in implementing a multi-modal imaging technique where EEG and fMRI data are recorded continuously and synchonously. His work involves development, testing, quality assurance, and eventually using the technique to answer complex experimental questions. Aside from assuring that data are collected across the same state/time, the technique allows for data from one modality to inform the analysis of the other. Though this is still a novel, groundbreaking technique, past research has used peak amplitudes and latencies from event related potentials (ERP) to weight regressors in the the general linear model (GLM), which directs multiple linear regression of fMRI data.