Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

A Landau-Ginzburg Phenomenology of Sleep-Stage Transitions

T0 review · 3 major / 5 minor · reviewed 2026-08-08 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Each sleep-stage boundary may be its own kind of dynamical transition: fold, crossover, or switch, all generated by one noisy neural-field potential.

desk verdict A carefully scoped Landau-Ginzburg taxonomy of sleep-stage boundaries; the math is standard and the honesty is refreshing, but the promised non-circular measurement model is the load-bearing piece and it is not yet fitted. read the letter →

arxiv 2608.03000 v1 pith:V3SEPI6Y submitted 2026-08-04 physics.bio-ph q-bio.NC

Alexander Poltorak This is my paper · ORCID
classification physics.bio-phq-bio.NC
keywords sleeparchitecturestagessleep-stagetransitionLandau-Ginzburgphasetransitionsstatecorticaldynamicsnonlinear
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper proposes that the boundaries between sleep stages are not one phenomenon but several: wake-to-N1 may be a fold-like loss of stability, N1-to-N2 and N2-to-N3 gradual ordering crossovers, NREM-to-REM a first-order-like desynchronizing switch, and a deep within-N3 regime a speculative mixed or tricritical-like transition. All four are presented as different parameter regimes of a single effective potential — a Landau-Ginzburg functional, an energy-like surface whose shape sets the stability and transition character of a noisy, spatially extended neural field — ordered by a latent cortical-ordering coordinate to be estimated from prespecified EEG/PSG features through a measurement model designed to prevent circularity. The paper's contribution is a testable organizational scheme: each boundary carries its own EEG/PSG signature (critical slowing, hysteresis and bimodality, correlation-length growth, or scoring artifact), and a synthetic classification experiment shows the six archetypes are separable above chance (cross-validated accuracy 0.49 versus a 0.17 baseline). The author is explicit that this establishes internal consistency and testability, not the taxonomy in human sleep; only the sleep-onset fold has prior external support, and the measurement model is specified but not yet fitted.

What carries the argument

The carrying object is the effective Landau-Ginzburg functional, $$\mathcal{F}[\psi] = \int_\$\Omega$ \left(\frac{a(\$\lambda$)}{2}\$psi^{2}$ + \frac{b(\$\lambda$)}{4}\$psi^{4}$ + \frac{c}{6}\$psi^{6}$ + \frac{\kappa}{2}|\nabla\psi|^2 - h(\$\lambda$)\psi\right)d^d r,$$ with $\psi = \varphi - \varphi_0$ the centered latent cortical-ordering coordinate, together with its stochastic relaxational dynamics, the time-dependent Ginzburg-Landau equation $\partial\psi/\partial t = -\Gamma(\delta\mathcal{F}/\delta\psi) + \eta$. The coefficients carry the operational meaning: $a$ sets the stiffness of the current state (recovery time, variance, lag-1 autocorrelation), $b$ the transition character (smooth ordering versus bistability), $c$ high-amplitude saturation, $\kappa$ the spatial coupling whose observable is the correlation length $\xi = \sqrt{\kappa/V''(\psi_{\rm eq})}$, and $h$ the biasing field that rounds sharp transitions into crossovers. The latent coordinate is estimated, not assumed, through the prespecified measurement model $\mathbf{y}(t) = \Lambda\varphi(t) + \varepsilon(t)$, with the requirement that features defining a boundary be excluded from $\varphi$ for that boundary's test. One Ginzburg-Landau family generates the fold, cusp-with-hysteresis, continuous-order, and tricritical signatures, which is what lets the paper assign different dynamical classes to different boundaries while keeping the description unified.

What would settle it

Run the paper's own proposed test on public polysomnography: estimate $\varphi$ from features that exclude the boundary-defining ones, estimate transition times independently of the scored labels, and compare smooth, switching, and mixed models at each boundary with subjects as the held-out unit. The continuous-crossover reading of N2-to-N3 is falsified if the order parameter shows a large jump and within-subject bimodality that survives jittered scoring boundaries and cycle-matched controls; the fold reading of wake-to-N1 is falsified if no reproducible tipping point with rising variance and autocorrelation precedes loss of wake stability; and the first-order reading of NREM-to-REM is falsified if forward and reverse paths overlap after matching sleep pressure, circadian phase, and prior sleep history.

Watch

Extended reading notes

Core claim

The central claim is that different sleep-stage boundaries may instantiate different forms of dynamical reorganization, and that these forms can be told apart empirically with tools adapted from nonequilibrium statistical mechanics. Concretely: wake-to-N1 is a local fold (saddle-node) whose loss of wake stability has external empirical support, with the open question of whether that fold is a spinodal of a globally bistable cusp showing hysteresis; N1-to-N2 and N2-to-N3 are continuous-like ordering crossovers, with N2-to-N3 the strongest candidate and the spatial Ginzburg term predicting correlation-length growth and local-to-global slow-wave recruitment; NREM-to-REM is a candidate first-order-like switch between a synchronized high-ordering NREM basin and a desynchronized low-ordering REM basin; and a consolidated within-N3 subregime is a clearly marked speculative hypothesis of mixed or tricritical-like character. These are not separate models: one time-dependent Ginzburg-Landau equation family, $\mathcal{F}[\psi] = \int_\Omega \left(\frac{a(\lambda)}{2}\psi^2 + \frac{b(\lambda)}{4}\psi^4 + \frac{c}{6}\psi^6 + \frac{\kappa}{2}|\nabla\psi|^2 - h(\lambda)\psi\right)d^d r$, with $\psi$ the centered latent cortical-ordering coordinate, produces all four signature classes as its coefficients move through fold, cusp, and crossover regimes. The paper claims the framework is internally consistent and testable, not that the taxonomy has been found in human sleep.

Load-bearing premise

The load-bearing premise is that one latent cortical-ordering coordinate, estimated from a prespecified linear measurement model, can capture the dynamics at each NREM boundary without using the features that define the boundary under test; the paper states plainly that this model is specified but not fitted and that fitting and invariance testing are deferred, so if that coordinate cannot be identified non-circularly, the taxonomy is unevaluable.

Editorial extensions

If this is right

  • Sleep-onset analysis shifts from scalar models to spatial ones: the Ginzburg term predicts that slow waves recruit larger cortical territories as N2-to-N3 is approached, testable as correlation-length growth in high-density or source-reconstructed EEG.
  • Each boundary gets a distinguishing battery: critical slowing and a reproducible tipping point license the fold reading of wake-to-N1; hysteresis and path dependence that survive matched-control comparison license the first-order reading of NREM-to-REM; absence of bimodality and jump argues against first-order behavior at N2-to-N3.
  • Scoring is itself an archetype: a scoring-induced discontinuity and a smooth crossover share the same latent dynamics and are the least separable pair in the synthetic experiment, so the validation pipeline must jitter or re-estimate scored boundaries and exclude boundary-defining features from the order parameter.
  • A within-N3 consolidated-SWS substate, if present, should show a continuous precursor plus a sharper derivative change, local-to-global nucleation, or late bimodality; if high-density EEG shows none of these, the tricritical-like hypothesis is to be weakened or rejected.
  • Hypnogram labels are demoted to temporal priors: transition times and transition classes are inferred from continuous EEG/PSG features, with subjects rather than epochs as the held-out unit in model comparison.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the taxonomy is validated, sleep-disorder phenotyping could be re-expressed as changes in effective potential coefficients — basin depth at sleep onset, barrier height between NREM and REM, spatial coupling $\kappa$ for slow-wave recruitment — rather than as changes in stage durations; the paper gestures at this direction but does not develop it.
  • A natural next experiment the paper does not run is to fit the time-dependent Ginzburg-Landau equation directly to EEG-derived $\varphi$ trajectories per subject and boundary, converting the qualitative taxonomy into estimated parameter sets (sign and scale of $b$, field $\tilde{h}$) that could be compared across ages or clinical groups.
  • The framework's 'one family, many regimes' structure suggests a sharper version of the tricritical test: high-density EEG nucleation maps — local patches of high $\varphi$ expanding as traveling slow-wave fronts — would distinguish a mixed continuous-plus-switch subregime from a smooth crossover in a way that scalp-spectral averages cannot.
  • Because REM frequently emerges from N2, pooling N2-to-REM and N3-to-REM transitions could manufacture the bimodality the first-order hypothesis predicts; the paper flags stratification by source stage, and a reader might go further and treat source-stage mixing as a built-in control in any cohort test.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. This paper develops a Landau–Ginzburg phenomenology for sleep-stage transitions. It introduces a latent cortical-ordering coordinate φ, a prespecified measurement model (Eq. 11), and a boundary-specific taxonomy: wake-to-N1 is a supported fold with a conditional cusp embedding, N1-to-N2 and N2-to-N3 are continuous-like crossovers, NREM-to-REM is a candidate first-order-like desynchronizing switch, and a within-N3 tricritical-like subregime is a speculative hypothesis. The Ginzburg term yields spatial predictions for correlation-length growth and local-to-global recruitment. The paper includes illustrative time-dependent Ginzburg–Landau simulations, a synthetic classification experiment distinguishing six archetypes (cross-validated accuracy 0.49 against a 0.17 baseline), and a detailed validation protocol with circularity controls. It explicitly states that the measurement model is specified but not fitted, and that the taxonomy is not validated in human sleep.

Significance. If the framework holds, it offers a principled way to classify sleep-stage boundaries by dynamical mechanism rather than by descriptive label, with concrete, testable EEG/PSG signatures. The paper's strengths are its explicit separation of fold, cusp, crossover, and scoring-artifact mechanisms (Sec. IVD); the prespecified non-circularity rules (Sec. IVE); the detailed decision criteria and confound controls (Secs. IX–X); the reproducible synthetic experiments with overlapping noise ranges and cross-validation inside folds; and the honest labeling of the within-N3 hypothesis as speculative. The main risk is that the empirical testability of the framework rests on the unverified identifiability of φ under the exclusion rule, which is not demonstrated in the manuscript.

major comments (3)
  1. [Sec. IVE / Eq. (11), Sec. XC] The paper's central testability claim (Sec. IVC) depends on the latent order parameter φ being identifiable from prespecified EEG/PSG features while respecting the non-circularity rule of Sec. IVE. For the N2-to-N3 boundary, which the paper identifies as the strongest continuous-ordering candidate, the AASM definition of N3 uses slow-wave activity; therefore slow-oscillation dominance cannot be used to construct φ for that test, and the paper must rely on residual indicators such as inverse complexity and spatial synchrony. The manuscript does not show, either empirically or in a synthetic identifiability analysis, that these residual features carry enough signal to distinguish the predicted continuous growth of correlation length from a scoring-induced discontinuity. Because Eq. (11) is specified but not fitted, and because Table V lists slow-oscillation dominance as a core indicator of NREM deepening, the falsifiable predictions for the NREM-deepening boundaries are conditional on an unverified premise. Please add a synthetic measurement-model analysis that estimates φ from a feature set that excludes the boundary-defining feature for each boundary test, and demonstrate that the predicted signatures remain detectable; alternatively, specify per-boundary feature sets that respect the exclusion rule and justify their sufficiency.
  2. [Sec. VII / Fig. 4, Sec. XC] The synthetic classification experiment of Sec. VII does not incorporate the jittered or shuffled-boundary null control that Sec. XC identifies as necessary to separate a scoring-induced discontinuity from a genuine smooth transition. In the confusion matrix of Fig. 4, the smooth crossover and the scoring artifact are the least separable pair, with smooth-crossover recall at 0.41. Since scoring-induced discontinuity is the principal confound for the continuous-like N1-to-N2 and N2-to-N3 hypotheses, the demonstrated pipeline cannot by itself falsify the continuous hypothesis against that alternative. The paper acknowledges this confusion as 'honest and expected,' but then uses the same experiment to support the 'testability' of the framework. To make the support valid, the experiment should include the jittered-threshold null model as an additional feature or classifier layer, or the paper should explicitly state that the current pipeline does not separate these two archetypes and that the null-model controls of Sec. XC remain to be implemented.
  3. [Sec. XF vs. Sec. VII] The statement in Sec. XF that the synthetic result 'establishes that the pipeline can make that discrimination when the ground truth is known' overstates the evidence: the cross-validated accuracy is 0.49 against a 0.17 chance level, and per-class recall ranges from 0.41 to 0.61. The abstract's wording ('partially distinguished') is accurate. Please revise the later claim to match the actual performance and to note that the discrimination is partial and archetype-dependent, with the smooth-crossover/scoring-artifact pair being a known overlap.
minor comments (5)
  1. [Sec. IVE] Please display the lead-field generalization of Eq. (11) as a numbered equation, since it is used as the basis for the spatial tests in Sec. XE.
  2. [Sec. VII / Fig. 3] The caption of Fig. 3 reports onset exponents 0.58 and 0.26 while the text later gives seed-averaged values 0.56±0.05 and 0.25±0.01; please make the caption and the corresponding text consistent.
  3. [Appendix A, Table VII] The statement that the sign of b 'may be unrecoverable from short noisy trajectories' is an important caveat for distinguishing first-order from continuous transitions; please specify what trajectory length or ensemble size is needed, or indicate how Sec. X's model-comparison framework addresses this.
  4. [Sec. VIII, Ref. [49]] Please mark the related work by Passaro and Poltorak as a preprint in the reference list and state explicitly that it has not undergone peer review, since the text describes it as 'currently a preprint.'
  5. [Sec. IVC / Sec. XII] The terms 'self-organized criticality' and 'self-organized bistability' are used without definition; a sentence clarifying the distinction would help readers outside the active-matter community.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's taxonomy is an unfitted, internally consistent framework with explicit anti-circularity controls.

full rationale

The claimed derivation chain is self-contained rather than circular. The Landau-Ginzburg functional (Eq. 5) and TDGL equation (Eq. 8) are introduced as a phenomenological normal-form language, not derived from the sleep data, and the boundary-specific assignments in Sec. VI are explicitly posed as falsifiable hypotheses rather than as forced conclusions. The latent order parameter phi is not fitted anywhere: Sec. IVE states 'In this paper, the measurement model is specified but not fitted; fitting and invariance testing are part of the validation program,' and it imposes the anti-circularity rule that 'phi must not be constructed from the same features that define the boundary under test,' with the within-N3 slow-wave example given. The synthetic classification experiment (Sec. VII) uses known generative archetypes and is explicitly bounded as an internal-consistency check: 'These simulations are illustrative examples and an internal-consistency check. They are not a fit to empirical sleep data and do not validate the transition taxonomy.' The only self-citation, Passaro and Poltorak [49], is used solely as 'motivating evidence for the feasibility of transition-centered EEG analysis,' and the paper explicitly says spectral power alone cannot identify transition order, so the citation is not load-bearing. No fitted parameter is relabeled as a prediction, no uniqueness theorem from the authors' prior work is invoked to make the choice of normal form forced, and the empirical predictions in Sec. IXB are accompanied by confound controls and decision criteria. The main risk is identifiability of phi from prespecified features, but the paper assigns that to future validation, which is a conditionality limitation rather than circular reasoning.

Assumptions & free parameters 5 free parameters · 5 assumptions · 2 invented entities

The framework rests on a small set of modeling axioms: the existence and sufficiency of a scalar latent order parameter phi for NREM boundaries, the treatment of neuromodulatory drives as slowly varying external control parameters, the adequacy of relaxational gradient dynamics with additive white noise, and the applicability of mean-field scaling. The coefficients are chosen by hand or by nondimensionalization for the illustrative simulations; the paper correctly notes that only scaled combinations such as the physical correlation length and recovery time are identifiable from data. No new physical entities are introduced beyond the latent coordinate and the speculative consolidated-SWS subregime, both currently without independent empirical evidence.

free parameters (5)
  • Landau coefficients a(lambda), b(lambda), c(lambda) and biasing field h(lambda) = Not fitted; illustrative values in Table IV
    Chosen by hand to realize the normal-form regimes; the taxonomy depends on signs, and only scaled combinations are identifiable (Appendix A).
  • Noise intensity D = 0.005 to 0.05 in simulations
    Set per panel and per generator archetype; the held-out robustness test increases it by 40 percent.
  • Gradient coefficient kappa and relaxation rate Gamma = Set to 1 by nondimensionalization
    Absorbed by rescaling; the physical correlation length xi = sqrt(kappa/mu) and recovery time tau = 1/(Gamma*mu) are the identifiable combinations.
  • Measurement loadings Lambda = Not fitted
    Eq. (11) is specified but not estimated; fitting and invariance testing are deferred to the validation program.
  • Tricritical scaling parameters beta0 and u_tri = Not fitted
    Enter via b_SWS(u) = beta0*(u-u_tri) in Eq. (24) for the speculative within-N3 scenario.
assumptions (5)
  • domain assumption A latent scalar cortical-ordering coordinate phi(r,t) exists and is sufficient to describe the collective state near each NREM boundary.
    Stated in Sec. IVA; the entire taxonomy is written in terms of phi, and the paper notes REM-to-wake needs extra dimensions.
  • domain assumption Neuromodulatory drives (S, C_circ, ACh, NE, GABA, G_thal) can be treated as slowly varying external control parameters with no fast feedback from the order parameter.
    The model is local by construction; Secs. I and XII explicitly exclude ultradian cycling and non-gradient flows.
  • domain assumption Relaxational gradient dynamics with additive white noise (Eq. 8) adequately approximate the effective cortical dynamics near transitions.
    Called a minimal Model-A-like approximation in Sec. XII; no derivation from a neural mass model is provided.
  • standard math Mean-field Landau-Ginzburg scaling (exponents 1/2 and 1/4, xi ~ sqrt(kappa/|a|)) applies to the effective field in transition windows.
    Used in Secs. VI.C and VII for the correlation-length and onset-exponent predictions; no non-mean-field corrections are incorporated.
  • domain assumption The measurement model can identify phi up to sign and scale without using boundary-defining features.
    Required by Sec. IVE; not yet demonstrated on data.
invented entities (2)
  • Latent cortical-ordering coordinate phi (centered as psi)
    purpose: Serves as the order parameter in the Landau-Ginzburg functional; aggregates slow-oscillation dominance, inverse complexity, and spatial synchrony.
    A latent variable introduced in Sec. IVA; the measurement model is specified but not fitted, so no independent empirical handle is provided yet.
  • Consolidated-SWS (N4-like) subregime within N3
    purpose: A hypothesized distinct deep-NREM substate with mixed or tricritical-like transition signatures, defined operationally by sustained slow-oscillation dominance, high spatial coherence, and low complexity.
    Explicitly marked speculative in Secs. VI.D and XII; the paper states it should be abandoned if high-density EEG and physiological data fail to identify it.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Landau-Ginzburg Phenomenology of Sleep-Stage Transitions." pith.science (2026). https://pith.science/paper/V3SEPI6Y

@misc{pith2026260803000,
  author       = {Pith},
  title        = {Pith review of: A Landau-Ginzburg Phenomenology of Sleep-Stage Transitions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/V3SEPI6Y}},
  note         = {Machine review of arXiv:2608.03000}
}
read the original abstract

Sleep staging provides a reproducible clinical description, but it does not by itself explain why some boundaries are abrupt while others are graded, or why transition windows contain instability, synchrony, and apparent state coexistence. We develop a local Landau-Ginzburg phenomenology in which each boundary is represented by motion in an effective potential of a spatially extended, noisy, dissipative neural field. A latent cortical-ordering coordinate phi is inferred from prespecified EEG/PSG observables through a measurement model designed to avoid circularity. The canonical boundaries are treated separately. Existing data support a fold-like loss of wake stability at sleep onset; whether that fold lies on a globally bistable cusp with hysteresis remains open. N1-to-N2 and N2-to-N3 are posed as continuous-like ordering crossovers, NREM-to-REM as a candidate first-order-like desynchronizing switch, and a possible within-N3 mixed or tricritical-like regime as a speculative hypothesis. The Ginzburg term adds spatial predictions - growth of correlation length and local-to-global recruitment - that are absent from scalar sleep-onset models. We specify the evidence needed to distinguish bifurcation, coexistence, noise-driven escape, smooth crossover, and scoring-induced discontinuity. Illustrative time-dependent Ginzburg-Landau simulations reproduce the proposed signature classes. A synthetic classification experiment partially distinguished six archetypes (cross-validated accuracy 0.49 +/- 0.005; balanced baseline 0.17), with little change under a noise-regime shift. These analyses establish the internal consistency and testability of the framework, not the proposed taxonomy in human sleep. Transition-centered EEG validation is required before clinical or neuromodulation applications are pursued.

Figures

Figures reproduced from arXiv: 2608.03000 by the authors.

Figure 1
Figure 1. FIG. 1. Transition order is set by the control path, not by the static appearance of a single potential. [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. The boundary-specific taxonomy, drawn as a set of independent local transition windows rather than a single night [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Illustrative numerical signatures generated by the time-dependent Ginzburg–Landau family: (a) first-order-like hys [PITH_FULL_IMAGE:figures/full_fig_p014_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: FIG. 4. Signature-separability confusion matrices for six candidate archetypes. Left: five-fold cross-validation (accuracy 0.49). [PITH_FULL_IMAGE:figures/full_fig_p016_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5. Proposed validation pipeline with circularity controls. Standard hypnograms identify candidate transition windows only [PITH_FULL_IMAGE:figures/full_fig_p018_5.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Toward a Dynamical Taxonomy of Insomnia: A Multiaxial Framework for Sleep-State Transitions and Architectural Failure

    physics.bio-ph 2026-08 conditional novelty 6.0 of 10

    A multiaxial dynamical taxonomy for insomnia, using a Landau-Ginzburg phenomenological model, organizes failure by operation, stage, and causal status and makes two falsifiable predictions.

Reference graph

Works this paper leans on

57 extracted references · 54 canonical work pages · cited by 1 Pith paper

  1. [1]

    Government Printing Of- fice, Washington, DC, 1968)

    A.RechtschaffenandA.Kales,A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects(U.S. Government Printing Of- fice, Washington, DC, 1968)

    work page 1968

  2. [2]

    M. H. Silber, S. Ancoli-Israel, M. H. Bonnet,et al., J. Clin. Sleep Med.3, 121 (2007)

    work page 2007

  3. [3]

    M. M. Troester, S. F. Quan, R. B. Berry,et al.; for the American Academy of Sleep Medicine,The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications, Version 3(American Academy of Sleep Medicine, Darien, IL, 2023)

    work page 2023

  4. [4]

    P. Bak, C. Tang, and K. Wiesenfeld, Phys. Rev. Lett. 59, 381 (1987)

    work page 1987

  5. [5]

    J. M. Beggs and D. Plenz, J. Neurosci.23, 11167 (2003)

    work page 2003

  6. [6]

    J. M. Beggs, Philos. Trans. R. Soc. A366, 329 (2008). 23

    work page 2008

  7. [7]

    Di Santo, P

    S. Di Santo, P. Villegas, R. Burioni, and M. A. Muñoz, Proc. Natl. Acad. Sci. U.S.A.115, E1356 (2018)

    2018

  8. [8]

    K. K. Iyer, Front. Neurosci.12, 948 (2018)

    work page 2018

Show all 57 references
  1. [9]

    B. A. Pearlmutter and C. Houghton, Neural Comput.21, 1622 (2009)

    2009

  2. [10]

    L. D. Landau, Zh. Eksp. Teor. Fiz.7, 19 (1937)

    1937

  3. [11]

    V. L. Ginzburg and L. D. Landau, Zh. Eksp. Teor. Fiz. 20, 1064 (1950)

    1950

  4. [12]

    M. L. Steyn-Ross, D. A. Steyn-Ross, and J. W. Sleigh, Prog. Biophys. Mol. Biol.85, 369 (2004)

    2004

  5. [13]

    M. L. Steyn-Ross, D. A. Steyn-Ross, J. W. Sleigh, M. T. Wilson, I. P. Gillies, and J. J. Wright, J. Biol. Phys.31, 547 (2005)

    2005

  6. [14]

    M. L. Steyn-Ross, D. A. Steyn-Ross, and J. W. Sleigh, Phys. Rev. X3, 021005 (2013)

    2013

  7. [15]

    J. Li, A. Ilina, R. Peach, T. Wei, E. Rhodes, V. Jaramillo, I. R. Violante, M. Barahona, D.-J. Dijk, and N. Gross- man, Nat. Neurosci.28, 2515 (2025)

    2025

  8. [16]

    S. M. M. de Mooij, T. F. Blanken, R. P. P. P. Grasman, J. R. Ramautar, E. J. W. Van Someren, and H. L. J. van der Maas, Comput. Methods Programs Biomed.193, 105448 (2020)

    2020

  9. [17]

    A. B. A. Stevner, D. Vidaurre, J. Cabral,et al., Nat. Commun.10, 1035 (2019)

    2019

  10. [18]

    Y. S. Perl, H. Bocaccio, I. Pérez-Ipiña, F. Zamberlán, J. Piccinini, H. Laufs, M. Kringelbach, G. Deco, and E. Tagliazucchi, Phys. Rev. Lett.125, 238101 (2020)

    2020

  11. [19]

    A. A. Borbély, Hum. Neurobiol.1, 195 (1982)

    1982

  12. [20]

    C. B. Saper, T. C. Chou, and T. E. Scammell, Trends Neurosci.24, 726 (2001)

    2001

  13. [21]

    C. B. Saper, T. E. Scammell, and J. Lu, Nature437, 1257 (2005)

    2005

  14. [22]

    J. Lu, D. Sherman, M. Devor, and C. B. Saper, Nature 441, 589 (2006)

    2006

  15. [23]

    P. A. Robinson, A. J. K. Phillips, B. D. Fulcher, M. Puck- eridge, and J. A. Roberts, Philos. Trans. R. Soc. A369, 3840 (2011)

    2011

  16. [24]

    Scheffer, J

    M. Scheffer, J. Bascompte, W. A. Brock,et al., Nature 461, 53 (2009)

    2009

  17. [25]

    Meisel, A

    C. Meisel, A. Klaus, V. V. Vyazovskiy, and D. Plenz, J. Neurosci.37, 10114 (2017)

    2017

  18. [26]

    Priesemann, M

    V. Priesemann, M. Valderrama, M. Wibral, and M. Le Van Quyen, PLoS Comput. Biol.9, e1002985 (2013)

    2013

  19. [27]

    Buendía, S

    V. Buendía, S. di Santo, P. Villegas, R. Burioni, and M. A. Muñoz, Phys. Rev. Res.2, 013318 (2020)

    2020

  20. [28]

    A. A. Borbély, S. Daan, A. Wirz-Justice, and T. Deboer, J. Sleep Res.25, 131 (2016)

    2016

  21. [29]

    Z. Hu, M. Aravind, X. Lei, J. N. Kutz, and J.-J. Aucou- turier, PLOS Comput. Biol.22, e1014246 (2026)

    2026

  22. [30]

    S. S. Cash, E. Halgren, N. Dehghani,et al., Science324, 1084 (2009)

    2009

  23. [31]

    De Gennaro and M

    L. De Gennaro and M. Ferrara, Sleep Med. Rev.7, 423 (2003)

    2003

  24. [32]

    L. M. J. Fernandez and A. Lüthi, Physiol. Rev.100, 805 (2020)

    2020

  25. [33]

    Huber, M

    R. Huber, M. F. Ghilardi, M. Massimini, and G. Tononi, Nature430, 78 (2004)

    2004

  26. [34]

    Massimini, R

    M. Massimini, R. Huber, F. Ferrarelli, S. Hill, and G. Tononi, J. Neurosci.24, 6862 (2004)

    2004

  27. [35]

    Y. Nir, R. J. Staba, T. Andrillon,et al., Neuron70, 153 (2011)

    2011

  28. [36]

    V. V. Vyazovskiy, U. Olcese, E. C. Hanlon, Y. Nir, C. Cirelli, and G. Tononi, Nature472, 443 (2011)

    2011

  29. [37]

    Achermann and A

    P. Achermann and A. A. Borbély, Neuroscience81, 213 (1997)

    1997

  30. [38]

    Lempel and J

    A. Lempel and J. Ziv, IEEE Trans. Inf. Theory22, 75 (1976)

    1976

  31. [39]

    Costa, A

    M. Costa, A. L. Goldberger, and C.-K. Peng, Phys. Rev. Lett.89, 068102 (2002)

    2002

  32. [40]

    J. J. Iliff, M. Wang, Y. Liao,et al., Sci. Transl. Med.4, 147ra111 (2012)

    2012

  33. [41]

    L. Xie, H. Kang, Q. Xu,et al., Science342, 373 (2013)

    2013

  34. [42]

    L. M. Hablitz, H. S. Vinitsky, Q. Sun,et al., Sci. Adv.5, eaav5447 (2019)

    2019

  35. [43]

    N. E. Fultz, G. Bonmassar, K. Setsompop,et al., Science 366, 628 (2019)

    2019

  36. [44]

    Dagum, D

    P. Dagum, D. L. Elbert, L. Giovangrandi,et al., Nat. Commun.17, 715 (2026)

    2026

  37. [45]

    A. Miao, T. Luo, B. Hsieh,et al., Nat. Neurosci.27, 1046 (2024)

    2024

  38. [46]

    N. L. Hauglund, M. Andersen, K. Tokarska,et al., Cell 188, 606 (2025)

    2025

  39. [47]

    Tononi and C

    G. Tononi and C. Cirelli, Sleep Med. Rev.10, 49 (2006)

    2006

  40. [48]

    T. E. Scammell, E. Arrigoni, and J. O. Lipton, Neuron 93, 747 (2017)

    2017

  41. [49]

    A. D. Passaro and A. Poltorak, bioRxiv 10.1101/2024.12.16.628645 (2025)

    2025 doi

  42. [50]

    B. Kemp, A. H. Zwinderman, B. Tuk, H. A. C. Kam- phuisen, and J. J. L. Oberye, IEEE Trans. Biomed. Eng. 47, 1185 (2000)

    2000

  43. [51]

    Donoghue, M

    T. Donoghue, M. Haller, E. J. Peterson,et al., Nat. Neu- rosci.23, 1655 (2020)

    2020

  44. [52]

    J. D. Lendner, R. F. Helfrich, B. A. Mander,et al., eLife 9, e55092 (2020)

    2020

  45. [53]

    C.-K. Peng, S. Havlin, H. E. Stanley, and A. L. Gold- berger, Chaos5, 82 (1995)

    1995

  46. [54]

    J. A. Hartigan and P. M. Hartigan, Ann. Stat.13, 70 (1985)

    1985

  47. [55]

    Benjamini and Y

    Y. Benjamini and Y. Hochberg, J. R. Stat. Soc. B57, 289 (1995)

    1995

  48. [56]

    B. A. Mander, J. R. Winer, and M. P. Walker, Neuron 94, 19 (2017)

    2017

  49. [57]

    P. C. Hohenberg and B. I. Halperin, Rev. Mod. Phys.49, 435 (1977)

    1977

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.