The study of the human brain as a functional network has lead to tremendous insight into the function of the healthy brain and the different kinds of impairments that can lead to brain dysfunction in various neural disorders. Yet, an understanding of the performance of current connectivity measures and their methodological limitations remains lacking. A wide range of connectivity measures have been used to study functional connectivity in EEG and MEG. However, most such measures suffer from tremendous methodological limitations complicating the interpretation of connectivity results. I have recently assessed the performance of three of these measures in source space when combined with beamforming techniques in an effort to untangle methodological artifacts from true brain connectivity. Similarly, connectivity measures that are in wide use in fMRI such as Granger causality have been shown to suffer from various methodological limitations. Additionally, fMRI connectivity analysis techniques such as global signal regression have been shown to result in major systematic biases. I am interested in pursuing a better understanding of these methodologies and analysis techniques and in developing techniques that are better able to untangle true brain connectivity from methodological artifacts.

We are currently pursuing an understanding of various diseases at the network level including mild cognitive impairment and temporal lobe epilepsy using fMRI. Both studies have shown changes in connectivity between the posterior cingulate cortex and the hippocampus, and that these changes in connectivity are correlated with clinical and behavioral measures. We are also investigating the role of functional connectivity as determined from EEG time series in treatment resistant major depressive disorder after deep brain stimulation and how synchronization correlates with HAMD-17 scores. We have seen robust differences in asymmetry correlating with HAMD-17 scores in various frequency bands and we are currently quantifying these results and comparing connectivity results with the DBS electrodes on and off. In another EEG study, we will be assessing functional connectivity from patients with major depressive disorder who underwent electroconvulsive therapy compared to EEG recordings before treatment. We are also studying connectivity in temporal lobe epilepsy in EEG with the first round of this study submitted for publication.