Estimation of Neural Code from the Electroencephalogram (EEG)

ended

The electroencephalogram (EEG) is one of the most important non-invasive brain imaging tools in neuroscience and in the clinic, but surprisingly little is known about the features of neural circuit activity that give rise to the EEG. The challenge is to explain the functional and anatomical configurations, e.g., the neural interactions among different classes of cells, that produce the diverse spatial, spectral and temporal EEG features linked to cognition and neurological diseases. By means of an interdisciplinary approach, combing advanced theoretical modeling with state-of-the-art multiscale neurophysiology and interventional techniques, I will address the above challenge. I will develop rigorous mathematical tools to disambiguate the EEG and robustly interpret it in terms of specific neural features (e.g., firing rate). Such features are key elements in determining the microcircuit configuration and have been documented to contribute to brain disorders such as schizophrenia and Autism Spectrum Disorders (ASD). First, I will develop neural network models that include the key components of cortical microcircuits. I will then turn these models into accurate EEG analysis tools by fitting them to empirical data to “invert”, or translate back, the EEG into an estimate of the neural parameters. In particular, I am interested in studying the relationship of the EEG with spiking activity and synchrony of excitatory and inhibitory populations. The experiments will record simultaneously EEG and intracortical neural activity in mice, combined with optogenetic tools that can determine the contributions of specific classes of cells and specific patterns of activity in cells to different EEG features. The analysis tools developed in this project will be used to infer neural circuit changes from EEG measures, which will produce substantial progress toward bridging the gap between EEG and neuron dynamics.