The hypothesis that the brain operates near criticality explains observations of complex, often scale-invariant, neural activity. However, the brain is not static– its dynamical state varies depending on what an organism is doing. Neurons often become more synchronized (ordered) during unconsciousness and more desynchronized (disordered) in highly active awake conditions. Are all these states equidistant from criticality; if not, which is closest? The fundamental physics of how systems behave near criticality came from renormalization group (RG) theory, but RG for neural systems remains largely undeveloped. Here we developed a temporal RG (tRG) theory for analysis of typical neuroscience data. We mathematically identified multiple types of criticality (tRG fixed points) and developed tRG-driven data analytic methods to assess proximity to each fixed point from relatively short time series. Unlike traditional methods for studying criticality in neural systems, our tRG approach is time-resolved, allowing us to track how distance to criticality changes on behaviorally relevant timescales. We apply our approach to recordings of spike activity in mouse visual cortex, showing that the relaxed awake state is closest to criticality. During both more active awake states and deep sleep, cortical dynamics deviate from criticality.
Contact: Sam Sooter, sooter@uark.edu
Department of Physics, University of Arkansas
Additional Authors:
Antonio Fontenele, Department of Physics, University of Arkansas, af080@uark.edu; Cheng Ly, Department of Statistical Sciences and Operations Research, Virginia Commonwealth University, cly@vcu.edu; Andrea Barreiro, Department of Mathematics, Southern Methodist University, abarreiro@mail.smu.edu; Woodrow Shew, Department of Physics, University of Arkansas, shew@uark.edu