Brain criticality beyond avalanches: open problems and how to approach them

Abstract

A homeostatic mechanism that keeps the brain highly susceptible to stimuli and optimizes many of its functions—although this is a compelling theoretical argument in favor of the brain criticality hypothesis, the experimental evidence accumulated during the last two decades is still not entirely convincing, causing the idea to be seemingly unknown in the more clinically-oriented neuroscience community. In this perspective review, we will briefly review the theoretical framework underlying such bold hypothesis, and point to where theory and experiments agree and disagree, highlighting potential ways to try and bridge the gap between them. Finally, we will discuss how the stand point of statistical physics could yield practical applications in neuroscience and help with the interpretation of what is a healthy or unhealthy brain, regardless of being able to validate the critical brain hypothesis.

Entropy, Economics, and Criticality

“…For example Zhu et al [ 51 ] studied neuromorphic nanowire networks using TE and active information storage, finding that information theoretical values peak when the networks transition from a quiescent state to an active state, illustrating the relationship between information theory as a measure of computational capacity and criticality in an artificial system. Likewise, other studies have shown that biological brains may be poised at or near a critical state [ 52 ] where it has been argued the brain is at a point of “self-organised criticality”, a term introduced by Bak [ 53 ], and see the recent critical review by Girardi-Schappo [ 54 ]. Others have argued that this may be a widespread property of many other systems as well, see, for example, the recent article by Tadić and Melnik [ 55 ].…”

Toward a Unified Analysis of the Brain Criticality Hypothesis: Reviewing Several Available Tools

“…When using the crackling noise scaling relation, many research often refer to Muñoz et al ( 1999 ) and Sethna et al ( 2001 ), where methods for derivation were proposed, but the relation was not explicitly written out or derived rigorously. There is a simple derivation proposed in Girardi-Schappo ( 2021 ), but it seems that the author presumed relation between the size and duration of avalanches, which may lack explanation. Here, we illustrate the derivation of the crackling noise scaling relation briefly based on the work of Sethna et al ( 2005 ).…”

“…However, the definition of experimental avalanches varies among research due to different recording techniques. A detailed illustration can be found in Girardi-Schappo ( 2021 ). Nevertheless, when dealing with discrete time series, a suitable bin size (see Levina and Priesemann, 2017 ) may be used to extract all the avalanches as long as there exists a clear separation of time scales, namely the time of quiescence between avalanches is much longer compared to the duration of any single avalanche.…”

Predicting future links with new nodes in temporal academic networks

“…Complex networks have emerged as a powerful modeling tool for understanding and describing the real-world systems from the traffic system [1,2] to the power system [3,4], from the neural system [5,6] to the ecosystem [7], from the economic system [8,9] to the social system [10,11], in which nodes are the entities of a system and links are the interactions or connections between entities [12,13]. Understanding the evolutionary mechanisms of real systems as to which generates a pattern governing the evolution is a problem of fundamental importance for network science [14].…”