UC San Diego

Although rarely quantitatively considered together, extracranially recorded electric potentials, or their complementary magnetic gradients, and the underlying neuronal transmembrane ion flows form a unitary phenomenon. Because the conduction of these fields through the tissues of the head are for all biologically relevant intents and purposes instantaneous and additive, the variance and covariance of signals recorded by extracranial sensors are a weighted linear combination of the variance and covariance of current density fluctuations in the gray matter of the cortex. However, for a given sensor variance-covariance matrix, or complex cross-spectral matrix in the frequency domain, there are infinite number of valid cortical source configurations. In order to accurately distinguish between these possibilities assumptions must be made about the true patterns of correlativity, or functional connectivity, among sources which is in turn constrained by the anatomical connectivity of the cortex. In this dissertation, I investigated the structural and functional connectivity of the human cortex and developed computational models informed by these measures that quantitatively link intra- and extra-cranial activity.Chapter 1 examines relative human corticocortical structural connectivity using publicly available diffusion MRI data from the human connectome project. Chapter 2 refines these results by scaling relative connectivity in arbitrary units to absolute connectivity in physical units by taking advantage of the unique properties of the corpus callosum, and in which we find finds that the resulting connectivity is sparse. In chapter 3 the anatomical connectivity is used to inform a whole brain model of non-REM sleep capable of producing synthetic intra- and extra-cranial sensor activity, and in Chapter 4 the composited human invasive stereo-EEG data from epilepsy patients to estimate the spatial dependence of spontaneous cortical functional connectivity in a frequency-resolved manner.