Miguel A. Muñoz – University of Granada
Wednesday, July 31st, 10:45AM EDT
Quasi-universal scaling of brain activity: from real neurons to artificial intelligence
The brain is in a state of perpetual neural activity, even in the absence of specific tasks or stimuli. Shedding light on the origin and functional significance of such a dynamical state is essential to understanding how the brain works. An inspiring, albeit controversial, conjecture proposes that some statistical characteristics of empirically observed neuronal activity can be understood by assuming that brain networks operate in a dynamical regime with features, including the emergence of scale invariance, resembling those seen typically near phase transitions. In this talk, I present a data-driven analysis based on recordings of the activity of thousands of individual neurons in various regions of the mouse brain. These data are analyzed using techniques from Statistical Physics, including the theory of disordered systems and renormalization group approaches in a complementary way. The combined strategy allows us to uncover strong signatures of scale invariance that are “quasi-universal” across brain regions and experiments, revealing that all the analyzed areas operate, to a greater or lesser extent, near the edge of instability. I will discuss the meaning of this finding, paying special attention to its role for brain functionality. Finally, I will establish exciting and promising connections with the paradigm of ”reservoir computing” in machine learning and artificial intelligence.