DYNAMIC INFORMATION FLOWS IN BRAIN CORTICAL NETWORKS

Professor Vlasov and his research team are focused on understanding how information is flowing through brain circuits and how this flow enables complex behavior and cognition. The team’s goal is to reconstruct the cellular-level temporal functional anatomy of neuronal networks in the mammalian brain directly from the spike trains. The group will focus on discovery and analysis of millisecond-scale dynamics of interconnections in neuronal networks at a cellular level. Hence, this development will extend concepts of dynamic functional connectivity, traditionally considered for large-scale neuroimaging (e.g. fMRI), to the level of interacting individual neurons and will enable verification of different competing brain networks theories. Their work, therefore, is a fundamental shift from a traditional neuron-centric to a network-centric brain model that is firmly grounded on spiking activity of individual neurons and enables novel mechanistic interpretation of brain computations.

Vlasov’s group advances emerging experimental approaches for massive recording from and manipulation of thousands of individual neurons with single-spike time resolution in behaving brain loaded with cortex-rich behavioral task in a virtual reality. To analyze these large datasets, in contrast to traditional AI approaches, they develop explainable AI algorithms that automatically discover latent dynamic structures of these multidimensional dataflows and extract anatomically distinct interconnected networks of individual neurons, thus enabling detailed biologically-plausible interpretations.

The team’s approach provides a conceptual link between slow brain-wide network dynamics studied with neuroimaging and fast cellular-level dynamics enabled by modern electrophysiology that may help to uncover often-overlooked dimensions of the brain code. Integrated data-science-guided experimental design and data-driven AI analysis enable a host of new hypotheses on brain networks organization to be verified. Vlasov’s group envisions that their work will inspire development of brain-inspired computing systems that far exceed traditional computing approaches. Furthermore, their approach enables fast and reliable detection of the dynamic network patterns that are causally related to behavior. Hence, they make their analysis actionable for real-time intervention that can enable next generation of agile neural interfaces with potential revolutionary impact.