arxiv: 2604.21071 · v2 · submitted 2026-04-22 · ⚛️ physics.bio-ph

A Critical Assessment of the Brain Criticality Hypothesis

Chesson Sipling , Yuan-Hang Zhang , Massimiliano Di Ventra This is my paper

Pith reviewed 2026-05-11 00:42 UTC · model grok-4.3

classification ⚛️ physics.bio-ph

keywords criticality hypothesisscale-invariant correlationsneural activitylong-range ordermemory resourcesbrain dynamicsneuroscience models

0 comments

The pith

Coupling neurons to slowly varying resources generates scale-invariant activity without criticality.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper argues that scale-invariant correlations in neural activity, commonly linked to the brain operating at a critical point, can arise instead from interactions between neurons and slowly changing resources that serve as memory. This memory-induced long-range order produces the observed patterns in a robust phase of activity. A sympathetic reader would care because the alternative avoids the need for precise tuning to a special state and offers a simpler account grounded in resource dynamics rather than criticality. The claim challenges a leading hypothesis by showing how memory effects alone can sustain the correlations seen in experiments.

Core claim

Coupling between neurons and slowly varying resources acting as memory is sufficient by itself to produce a robust phase of neural activity with scale-invariant correlations, offering a more natural explanation for existing experimental data than the criticality hypothesis.

What carries the argument

The memory-induced long-range order phase that emerges from neuron coupling to slow resources.

If this is right

Scale-invariant neural correlations can appear in non-critical regimes driven by resource memory.
Models of brain activity need to treat slowly varying resources as an essential component rather than an afterthought.
Experimental signatures previously interpreted as evidence for criticality may instead reflect memory effects from resource coupling.
Optimal information processing in neural systems does not require operation near a critical point.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Computational simulations that include explicit slow resource variables could reproduce experimental correlation patterns without parameter tuning to criticality.
The mechanism may generalize to other biological or physical systems where fast variables couple to slow memory-like degrees of freedom.
Targeted experiments that modulate resource recovery times in cultured neural networks would provide direct tests of the proposed phase.

Load-bearing premise

That the memory-induced long-range order phase explains the observed scale-invariant correlations without implicitly depending on critical dynamics or specific unstated assumptions about resource timescales.

What would settle it

Measuring neural activity in a preparation where resource variation is either eliminated or made as fast as neural firing and finding that scale-invariant correlations disappear would test the claim.

Figures

Figures reproduced from arXiv: 2604.21071 by Chesson Sipling, Massimiliano Di Ventra, Yuan-Hang Zhang.

**Figure 2.** Figure 2: FIG. 2. Avalanche size, [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗

**Figure 3.** Figure 3: FIG. 3. A plot of [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

read the original abstract

A major unresolved question in Neuroscience is: What is the origin of the observed scale-invariant correlations in neural activity? Many researchers support the ``criticality hypothesis,'' which proposes that the brain operates near criticality, optimizing various information processing functions. However, the nature and behavior of criticality in cortical systems are still unclear. Alternatively, this opinion paper highlights that the coupling between neurons and slowly varying resources (acting as ``memory'') alone may be sufficient to generate a robust phase of neural activity with scale-invariant correlations. This memory-induced long-range order phase could provide a more natural explanation of the existing experimental data than the criticality hypothesis.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Opinion piece floats slow resources as a memory-based alternative to brain criticality for scale-free activity, but stays purely conceptual with no derivations or checks.

read the letter

The core point is that this short opinion paper claims coupling neurons to slowly varying resources can produce long-range order and scale-invariant correlations on its own, without needing the brain to operate near a critical point. It positions this memory mechanism as potentially more natural than the criticality hypothesis for explaining existing neural data. The authors note that criticality in cortical systems remains unclear and suggest the resource coupling alone might suffice for the observed patterns. That is the main new angle: importing the idea of memory from slow variables into this specific neuroscience debate. It does a reasonable job of framing the open question about the origin of scale-invariant correlations and sketching why a resource-based explanation could sidestep some of the ambiguities around criticality. The writing is direct and the alternative is presented without overclaiming empirical results. The main limitation is that the argument never moves beyond qualitative reasoning. No equations, no model, no simulation, and no comparison to data appear to show that the memory-induced phase actually generates the right exponents or correlations generically. The claim that this phase is sufficient and more natural therefore rests on assertion rather than demonstration, which leaves the stress-test concern about implicit tuning or hidden critical dynamics unaddressed. Without that step, it is difficult to judge whether the proposal is truly distinct or just another tuned regime. This is aimed at people already following the brain criticality literature who might be interested in alternative mechanisms. It is not a quantitative study, so it will not change modeling practice on its own. I would still send it for peer review as a perspective. The topic is active, the authors engage the literature honestly, and referees could ask for the missing concrete support to make the case sharper.

Referee Report

2 major / 0 minor

Summary. The manuscript is an opinion paper arguing that coupling between neurons and slowly varying resources (acting as memory) may be sufficient by itself to produce a robust phase of neural activity with scale-invariant correlations, offering a more natural explanation for experimental data than the brain criticality hypothesis.

Significance. If a concrete model were to demonstrate generic emergence of long-range order from slow resource dynamics without fine-tuning, the proposal could meaningfully challenge the criticality framework and reframe interpretations of scale-free neural statistics. As presented, however, the conceptual argument lacks the quantitative grounding needed to evaluate whether the mechanism is distinct from or independent of critical dynamics.

major comments (2)

Abstract: the central claim that neuron-resource coupling 'alone may be sufficient' to generate scale-invariant correlations rests entirely on qualitative reasoning with no dynamical equations, stability analysis, or simulation results shown. This is load-bearing for the sufficiency and 'more natural' assertions, as the skeptic note correctly identifies that without an explicit derivation one cannot verify independence from parameter tuning or implicit critical behavior.
Abstract: the manuscript does not specify the functional form of the resource dynamics, the coupling term, or the relaxation timescale, leaving open whether the proposed memory-induced long-range order phase requires the resource timescale to be tuned to the observation window (which would reduce the mechanism to a form of effective criticality rather than a distinct robust phase).

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful reading and constructive comments on our opinion manuscript. We clarify below that the paper is intended as a conceptual discussion proposing an alternative mechanism, but we agree that additional specificity would strengthen the presentation and address the concerns about quantitative grounding.

read point-by-point responses

Referee: Abstract: the central claim that neuron-resource coupling 'alone may be sufficient' to generate scale-invariant correlations rests entirely on qualitative reasoning with no dynamical equations, stability analysis, or simulation results shown. This is load-bearing for the sufficiency and 'more natural' assertions, as the skeptic note correctly identifies that without an explicit derivation one cannot verify independence from parameter tuning or implicit critical behavior.

Authors: We acknowledge that the central claim relies on qualitative reasoning, as is appropriate for an opinion paper whose goal is to highlight a conceptual alternative rather than deliver a complete modeling study. The argument is that slow resource dynamics acting as memory can produce long-range correlations generically through timescale separation, without the fine-tuning required at a critical point. We disagree that this necessarily reduces to implicit criticality, but we accept that an explicit minimal model would help readers evaluate the distinction. In revision we will add a short section outlining a possible dynamical model (resource equation with linear relaxation coupled to neural activity) to illustrate the mechanism, while keeping the paper's opinion character. revision: partial
Referee: Abstract: the manuscript does not specify the functional form of the resource dynamics, the coupling term, or the relaxation timescale, leaving open whether the proposed memory-induced long-range order phase requires the resource timescale to be tuned to the observation window (which would reduce the mechanism to a form of effective criticality rather than a distinct robust phase).

Authors: We agree that the current text leaves the functional forms unspecified. The intended distinction is that any sufficiently slow resource dynamics (relative to neural timescales) will generate scale-free statistics across a wide range of observation windows due to the memory effect, without requiring the resource timescale to match the observation window exactly. This differs from criticality, which demands parameter tuning to a specific point. We will revise the manuscript to include example functional forms (e.g., resource relaxation dR/dt = -R/τ + coupling to activity, with τ much larger than neural timescales) and a brief discussion of why the separation of timescales renders the phase robust rather than tuned. revision: yes

Circularity Check

0 steps flagged

No circularity: conceptual opinion paper with no derivations or self-referential reductions

full rationale

The manuscript is framed as an opinion paper whose central claim—that neuron-resource coupling alone can produce scale-invariant correlations—is presented conceptually without equations, parameter fitting, or explicit derivations. No load-bearing steps reduce by construction to inputs, self-citations, or ansatzes; the argument does not invoke uniqueness theorems or rename known results. The provided abstract and structure contain no mathematical content that could exhibit the enumerated circularity patterns, making the derivation chain self-contained by absence of formal claims.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The central claim rests on the domain assumption that slowly varying resources can be treated as memory without additional critical dynamics; no free parameters or invented entities with independent evidence are specified in the abstract.

axioms (1)

domain assumption Neurons couple to slowly varying resources that act as memory sufficient to produce scale-invariant correlations
This is the core premise invoked to generate the alternative phase of activity.

invented entities (1)

memory-induced long-range order phase no independent evidence
purpose: To explain scale-invariant neural correlations as an alternative to criticality
Postulated mechanism without a falsifiable handle provided in the abstract

pith-pipeline@v0.9.0 · 5396 in / 1234 out tokens · 28963 ms · 2026-05-11T00:42:11.469349+00:00 · methodology

Review history (2 revisions) →

discussion (0)

Reference graph

Works this paper leans on

52 extracted references · 52 canonical work pages

[1]

Emerge due to collective behavior,

work page
[2]

Possess long-range temporal correlations, and

work page
[3]

supporting information processing

Support information processing. Unlike Destexhe and Touboul’s counterexamples, memory-induced long-range orderdoessatisfy each of these physically motivated requirements. First, the long- range order in these systems emergespreciselydue to collective behavior in neural activity, which is induced by the presence of memory as provided by the resources. Addi...

work page
[4]

J. M. Beggs and D. Plenz, Journal of Neuroscience23, 11167 (2003)

work page 2003
[5]

Benayoun, J

M. Benayoun, J. D. Cowan, W. van Drongelen, and E. Wallace, PLOS Computational Biology6, 1 (2010)

work page 2010
[6]

J. M. Beggs, Philosophical Transactions of the Royal So- ciety A: Mathematical, Physical and Engineering Sci- ences366, 329 (2007)

work page 2007
[7]

Friedman, S

N. Friedman, S. Ito, B. A. W. Brinkman, M. Shimono, R. E. L. DeVille, K. A. Dahmen, J. M. Beggs, and T. C. Butler, Phys. Rev. Lett.108, 208102 (2012)

work page 2012
[8]

D. R. Chialvo, Nature physics6, 744 (2010)

work page 2010
[9]

J. M. Beggs and N. Timme, Frontiers in PhysiologyVol- ume 3 - 2012(2012)

work page 2012
[10]

J. M. Beggs, Frontiers in Computational Neuroscience16 (2022)

work page 2022
[11]

Destexhe and J

A. Destexhe and J. D. Touboul, eNeuro8(2021)

work page 2021
[12]

U. C. T¨ auber,Critical dynamics: a field theory ap- proach to equilibrium and non-equilibrium scaling behav- ior(Cambridge University Press, 2014)

work page 2014
[13]

W. J. Reed and B. D. Hughes, Phys. Rev. E66, 067103 (2002)

work page 2002
[14]

Mitzenmacher, Internet Mathematics1, 226 (2004)

M. Mitzenmacher, Internet Mathematics1, 226 (2004)

work page 2004
[15]

J. P. Sethna, K. A. Dahmen, and C. R. Myers, Nature 410, 242 (2001)

work page 2001
[16]

J. P. Sethna, K. A. Dahmen, and O. Perkovic, arXiv preprint cond-mat/0406320 (2004)

work page arXiv 2004
[17]

A. J. Fontenele, N. A. De Vasconcelos, T. Feliciano, L. A. Aguiar, C. Soares-Cunha, B. Coimbra, L. Dalla Porta, S. Ribeiro, A. J. Rodrigues, N. Sousa,et al., Physical review letters122, 208101 (2019)

work page 2019
[18]

Chan, T.-F

L.-C. Chan, T.-F. Kok, and E. S. C. Ching, PRX Life3, 013013 (2025)

work page 2025
[19]

J. K.-C. Sun, C. Sipling, Y.-H. Zhang, and M. Di Ventra, Phys. Rev. E112, 064401 (2025)

work page 2025
[20]

M. E. Fisher and M. N. Barber, Physical Review Letters 28, 1516 (1972). 7

work page 1972
[21]

Priesemann, M

V. Priesemann, M. H. Munk, and M. Wibral, BMC neu- roscience10, 40 (2009)

work page 2009
[22]

G. Hahn, T. Petermann, M. N. Havenith, S. Yu, W. Singer, D. Plenz, and D. Nikoli´ c, Journal of neuro- physiology104, 3312 (2010)

work page 2010
[23]

Yaghoubi, T

M. Yaghoubi, T. De Graaf, J. G. Orlandi, F. Girotto, M. A. Colicos, and J. Davidsen, Scientific reports8, 3417 (2018)

work page 2018
[24]

Z. Ma, G. G. Turrigiano, R. Wessel, and K. B. Hengen, Neuron104, 655 (2019)

work page 2019
[25]

J. G. Orlandi, J. Soriano, E. Alvarez-Lacalle, S. Teller, and J. Casademunt, Nature Physics9, 582 (2013)

work page 2013
[26]

Di Santo, P

S. Di Santo, P. Villegas, R. Burioni, and M. A. Mu˜ noz, Proceedings of the National Academy of Sciences115, E1356 (2018)

work page 2018
[27]

Dalla Porta and M

L. Dalla Porta and M. Copelli, PLoS computational bi- ology15, e1006924 (2019)

work page 2019
[28]

Haldeman and J

C. Haldeman and J. M. Beggs, Physical review letters 94, 058101 (2005)

work page 2005
[29]

H. R. Wilson and J. D. Cowan, Biophysical journal12, 1 (1972)

work page 1972
[30]

R. Kubo, J. Phys. Soc. Jap.12, 570 (1957)

work page 1957
[31]

Sipling, Y.-H

C. Sipling, Y.-H. Zhang, and M. Di Ventra, Physical Re- view E112, 014124 (2025)

work page 2025
[32]

M. C. Morrell, I. Nemenman, and A. Sederberg, eLife12, RP89337 (2024)

work page 2024
[33]

J. S. Marvin, A. C. Kokotos, M. Kumar, C. Pulido, A. N. Tkachuk, J. S. Yao, T. A. Brown, and T. A. Ryan, Proceedings of the National Academy of Sciences121, e2314604121 (2024)

work page 2024
[34]

H. Li, S. M. Foss, Y. Dobryy, C. K. Park, S. A. Hires, N. C. Shaner, R. Y. Tsien, L. C. Osborne, and S. M. Voglmaier, Frontiers in molecular neuroscience4, 10875 (2011)

work page 2011
[35]

A. J. Bower, C. Renteria, J. Li, M. Marjanovic, R. Barkalifa, and S. A. Boppart, Applied Physics Let- ters118(2021)

work page 2021
[36]

A. Bose, S. M. Epp, R. Belenya, K. Kurcyus, E. C. Dominguez, A. Ranft, E. S. Villa, M. Bursche, C. Preibisch, G. Castrill´ on,et al., bioRxiv , 2024 (2024)

work page 2024
[37]

Zhang, C

Y.-H. Zhang, C. Sipling, and M. Di Ventra, New Journal of Physics (2026)

work page 2026
[38]

Rabut, M

C. Rabut, M. Correia, V. Finel, S. Pezet, M. Pernot, T. Deffieux, and M. Tanter, Nature methods16, 994 (2019)

work page 2019
[39]

Demas, J

J. Demas, J. Manley, F. Tejera, K. Barber, H. Kim, F. M. Traub, B. Chen, and A. Vaziri, Nature Methods18, 1103 (2021)

work page 2021
[40]

Gutierrez-Barragan, N

D. Gutierrez-Barragan, N. A. Singh, F. G. Alvino, L. Coletta, F. Rocchi, E. De Guzman, A. Galbusera, M. Uboldi, S. Panzeri, and A. Gozzi, Current biology 32, 631 (2022)

work page 2022
[41]

D. J. Felleman and D. C. Van Essen, Cerebral cortex (New York, NY: 1991)1, 1 (1991)

work page 1991
[42]

R. J. Douglas and K. A. Martin, Annu. Rev. Neurosci. 27, 419 (2004)

work page 2004
[43]

K. D. Harris and G. M. Shepherd, Nature neuroscience 18, 170 (2015)

work page 2015
[44]

J. F. Mejias, J. D. Murray, H. Kennedy, and X.-J. Wang, Science advances2, e1601335 (2016)

work page 2016
[45]

C. G. Langton, Physica D: nonlinear phenomena42, 12 (1990)

work page 1990
[46]

R¨ am¨ o, S

P. R¨ am¨ o, S. Kauffman, J. Kesseli, and O. Yli-Harja, Physica D: Nonlinear Phenomena227, 100 (2007)

work page 2007
[47]

W. L. Shew, H. Yang, S. Yu, R. Roy, and D. Plenz, Jour- nal of neuroscience31, 55 (2011)

work page 2011
[48]

Bertschinger, T

N. Bertschinger, T. Natschl¨ ager, and R. Legenstein, Advances in neural information processing systems17 (2004)

work page 2004
[49]

Wilting and V

J. Wilting and V. Priesemann, Current Opinion in Neu- robiology58, 105 (2019), computational Neuroscience

work page 2019
[50]

Di Ventra and F

M. Di Ventra and F. L. Traversa, Journal of Applied Physics123(2018)

work page 2018
[51]

Di Ventra,MemComputing: Fundamentals and Ap- plications(Oxford University Press, 2022)

M. Di Ventra,MemComputing: Fundamentals and Ap- plications(Oxford University Press, 2022)

work page 2022
[52]

In this work, we will exclusively use the word “memory” in this context, rather than in the context of the storage or retrieval of information

work page