Evie Malaia

Current understanding of human brain function shows that neural activity is structured in time as finely as it is structured in space: oscillating cells communicate if synchronized in appropriate phase, creating temporary network connections. Component-based methods of EEG data analysis allow for multiple interpretations for similar components, making it increasingly difficult to interpret data framed by previous research findings. This is the largest impediment to the goal of human neuroscience - understanding how dynamic patterns of brain activity are transformed into cognition and action.

I propose to combine connectomics and miscrostate analysis approaches in developing a technique for EEG analysis in neurotypical and special populations, which could be used for development of fine-grained frequency-based models of cognition, and early diagnostics of neural disorders, such as autism and dyslexia. The knowledge gap this proposal seeks to address is that of the network-level interaction between local and global (distributed) neural processing as manifested by phase-amplitude coupling across frequency bands, and its relationship to behavioral performance in face recognition task, and resting state EEG in already available data from multiple research designs and systems.

Future application of spatiotemporal network analysis will allow us to investigate rich, high-dimensional dynamics in specific populations, and conduct vulnerability analysis in inter-network interactions for ASD, and other developmental brain disorders. The same methodology will be applicable to understanding learning strategies, including those in gifted children, leading to understanding of individual optimization of neural resource use, leading to vertical advancement in educational neuroscience. Use of resting-state EEG will allow to lower the age of participants to infancy, significantly advancing the toolkit of pediatric neuroscience.