Quasi-universal scaling in mouse-brain neuronal activity stems from edge-of-instability critical dynamics

Abstract

The brain is in a state of perpetual reverberant neural activity, even in the absence of specific tasks or stimuli. Shedding light on the origin and functional significance of such activity is essential to understanding how the brain transmits, processes, and stores information. 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 near the edge of a phase transition. Moreover, the resulting critical behavior, with its concomitant scale invariance, is assumed to carry crucial functional advantages. Here, we present a data-driven analysis based on simultaneous high-throughput recordings of the activity of thousands of individual neurons in various regions of the mouse brain. To analyze these data, we construct a unified theoretical framework that synergistically combines cutting-edge methods for the study of brain activity (such as a phenomenological renormalization group approach and techniques that infer the general dynamical state of a neural population), while designing complementary tools. This unified approach allows us to uncover strong signatures of scale invariance that is “quasi-universal” across brain regions and reveal that these areas operate, to a greater or lesser extent, at the edge of instability. Furthermore, this framework allows us to distinguish between quasi-universal background activity and non-universal input-related activity. Taken together, the following study provides strong evidence that brain networks actually operate in a critical regime which, among other functional advantages, provides them with a scale-invariant substrate of activity in which optimal input representations can be sustained.

Dynamical stability and chaos in artificial neural network trajectories along training

“…A very similar hypothesis is that living systems exhibit self-organized criticality (Bak et al, 1988;Hidalgo et al, 2014;Watkins et al, 2016;Munoz, 2018), with the brain being an archetypical example (Chialvo, 2010;Moretti and Muñoz, 2013;Morales et al, 2023a). Connecting the apparent criticality of brain dynamics with the information processing advantages of artificial systems and neural networks at the edge of chaos (Carroll, 2020;Morales and Muñoz, 2021;Morales et al, 2023b) has invigorated this interdisciplinary research line even further. It is thus suggestive to relate this phenomenology to our findings in the so-called edge of stability: a region where the loss function is still converging to a minimum (i.e., the ANN learns) albeit in a non-monotonic and faster way.…”

Synchronization transitions on connectome graphs with external force

“…From the experimental directions the different behavior in modules of brains of the mouse [40], by phenomenological renormalization-group analysis of the spectrum of electrode spikes, and humans [62], via Hurst and β exponents analysis of fMRI; quasi-critical (off-critical) scaling like behavior has been shown. Here we attempt to model this using the Shinomoto-Kuramoto (SK) model on connectomes of the fruit-fly (FF) and humans.…”

“…More complex models than the two-state branching process, can also exhibit hybrid type of phase transitions, like threshold models [37], models with inhibitory nodes [38] or models with oscillatory units [39]. Subsystems can also show different scaling behavior and may be within different distances from criticality [40].…”