Pith. sign in

REVIEW 2 major objections 4 minor 4 references

Noise-induced excitability: bloom, bust and extirpation in autotoxic population dynamics

T0 review · 2 major / 4 minor · reviewed 2026-08-03 · deepseek-v4-flash

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

Pith's one-line read By adding demographic noise to a minimal autotoxicity model, the paper shows that boom-bust dynamics become a probabilistic threshold phenomenon, with analytic formulas for the probabilities of early extinction, excitable busts, and persist

desk verdict A solid stochastic-dynamics paper with a genuinely new extinction-pathway classification, but the central mechanism is only verified for a reduced one-noise model and the analytic p_e derivation has a boundary-condition blemish. read the letter →

arxiv 2601.20670 v2 pith:7MFO7UKE submitted 2026-01-28 q-bio.PE cond-mat.stat-mechnlin.AO

Pablo Moreno-Spiegelberg , Javier Aguilar This is my paper
classification q-bio.PEcond-mat.stat-mechnlin.AO MSC 60J7092D2560H10
keywords autotoxicitydemographicnoiseboom-bustdynamicsextinctionpathwaysexcitablesystemsabsorbingstatestochasticdifferentialequationsfirst-passageprobabilities
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

The paper proposes that the irreversible collapse of an autotoxic population—one that poisons its own habitat—can be explained without invoking any positive feedback such as an Allee effect. Starting from an individual-based model, the authors derive a two-variable stochastic differential equation for population and toxin densities, then reduce it to one noise source by arguing that toxin fluctuations are negligible because toxin molecules vastly outnumber individuals. In the reduced model they identify three distinct extinction pathways—short-lived, excitable, and persistent—and find sharp transitions between them as the noise strength D and the toxin-population timescale ratio ρ vary. The central quantitative contribution is a set of explicit formulas for the probabilities of each pathway, with a deterministic 'barrier height' η(ρ) that captures how close a deterministic boom-bust trajectory comes to extinction. The authors argue this provides a general route to understanding boom-bust-extirpation in invasive species, microbial fermentation, and other negatively regulated populations.

What carries the argument

The reduced model (Eq. 8) carries the argument: only population density x carries noise; toxin density y follows deterministic dynamics ρ(x-y), justified by the diffusion ratio α=√(ρ/r) with r≈10^13 for yeast. Trajectories are classified by winding number N_W around the coexistence point into three pathways. Two asymptotic limits yield analytic probabilities: ρ→0 gives p_sl = e^{-2x0/D^2}; small D gives p_e via an incomplete-gamma barrier height η(ρ). The exponential e^{-2·scale/D^2} converts the competition between noise scale and deterministic distances into sharp transitions.

What would settle it

Simulate the full two-noise model (Eq. 6) or the individual-based process with realistic parameters, and compare the extinction-pathway probabilities against Eq. (11); any significant deviation in the excitable regime would falsify the noise-reduction. Alternatively, measure extinction-time distributions in a well-mixed autotoxic culture and check for the predicted exponential forms and ρ-scaling.

Watch

Extended reading notes

Core claim

Central claim: noise-induced excitability, a regime where demographic noise replaces the fixed threshold of deterministic excitable systems, so identical initial conditions can lead either to a full boom followed by absorption in the first bust, or to a metastable persistent state. The three pathway probabilities are p_sl = exp(-2x0/D^2) (short-lived), p_e = (1-p_sl) exp(-2η(ρ)/D^2) (excitable), p_p = 1-p_sl-p_e (persistent), with η(ρ) a deterministic barrier height derived from the toxin-free flow. The first transition depends on x0/D^2 and not ρ; the second on η(ρ)/D^2 and not the initial condition. For any nonzero noise, small ρ forces the excitable regime.

Load-bearing premise

The reduction to a single-noise model assumes toxin fluctuations are negligible (α≪1); if the diffusion ratio is not small for a given system, the predicted probabilities and the excitable-persistent transition do not follow.

Editorial extensions

If this is right

  • Small populations (large D) are most likely extirpated before the first boom, so system size alone can determine whether an introduced species ever expands.
  • With slow toxin dynamics (small ρ), any nonzero noise makes the excitable regime dominant, so a population that survives the first boom is very likely to die in the first bust.
  • The excitable-persistent transition is sharp, so small changes in carrying capacity or toxin decay rate can flip a population's statistical fate from persistence to extinction.
  • No Allee effect or positive feedback is required; negative feedback plus demographic stochasticity suffice for boom-bust-extirpation.
  • The same probabilistic structure should transfer to other negative-feedback mechanisms (resource depletion, pathogen load, predation) when timescale and noise scales are matched.

Reading between the lines

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

  • Inference: The one-noise reduction is load-bearing; the paper does not simulate the full two-noise model (Eq. 6) or the individual-based process, so a direct numerical check is the natural next step.
  • Inference: The winding-number classification gives a practical time-series observable—counting rotations around the coexistence state—that could map empirical population data onto the p_sl/p_e/p_p diagram.
  • Inference: The exponential structure likely generalizes to other slow-manifold first-passage problems, with η(ρ) replaced by the analogous deterministic closest-approach distance for the relevant feedback.
  • Inference: The sharp transition suggests a control principle: increasing habitat volume or toxin degradation rate can move a population from the excitable to the persistent regime, informing both conservation and fermentation management.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 4 minor

Summary. The paper derives a mesoscopic stochastic differential equation model for autotoxic population dynamics from an individual-based birth-death process, then reduces it to a one-noise SDE in which toxin fluctuations are neglected. The authors analyze the deterministic skeleton, classify stochastic extinction pathways by a winding-number observable (short-lived, excitable, persistent), and propose parameter-free analytical approximations, Eq. (11), for the probabilities p_sl, p_e, and p_p. They identify a threshold-like 'noise-induced excitability' regime and a sharp transition between excitable and persistent outcomes. The analytical formulas are compared with Monte Carlo simulations of the reduced model and show good agreement in the displayed parameter slices.

Significance. If the central claims hold, the paper provides a useful parameter-free, analytically tractable picture of boom-bust-extirpation in autotoxic populations, with a clean classification of extinction pathways and a sharp excitable-persistent transition. Strengths include the microscopic derivation, the absence of fitted parameters, and the direct Monte Carlo verification of Eq. (11) at the shown slices. However, the paper's general conclusions about 'arbitrary autotoxic systems' rest on a one-noise reduction that is only supported by a single order-of-magnitude estimate, and the Appendix C derivation of the key quantity eta(rho) is not self-consistent as printed. The central idea is promising, but the manuscript needs correction and a broader validity check before the claims can be accepted.

major comments (2)
  1. [Appendix C, Eqs. (C1)-(C13)] The derivation of Eq. (12) and hence of p_e in Eq. (11) is not internally consistent. The boundary condition stated before Eq. (C1) is u(1,y)=0 for y<1, but the solution (C13) is matched to u(x,1)=0. More seriously, with nu(y) defined by Eq. (C11), nu(1)=0 and nu(y)<0 for y>1, so the claimed solution u=e^{-xi/nu} exceeds 1 and is not a valid absorption probability. Equation (C9) appears to have the wrong sign; reversing it would make nu(y)>0 and the exponential solution bounded. Because Eq. (12) is load-bearing for Eq. (11), the appendix must be corrected. The Monte Carlo agreement in Fig. 3(e-i) suggests the final formula may survive, but the printed derivation does not justify it.
  2. [Section II.B and Discussion] The reduction from the full two-noise SDE (6) to the reduced model (8) sets D_y=0 on the basis of alpha=sqrt(rho/r)<<1, supported by a single yeast parameter estimate (Appendix A). The full two-noise SDE or the original individual-based master equation (3) is never simulated. This is a load-bearing gap because near x=0, where extirpation is decided, the x-noise amplitude scales as D_x sqrt{x} while the y-noise amplitude is D_y sqrt{x+y}; hence y-noise can matter even when alpha is small at the deterministic fixed point. To support the Discussion's claim that the framework applies to arbitrary autotoxic systems, the authors should either simulate Eq. (6) (after correcting the typo in its x-noise term) or explicitly restrict the claimed domain of validity.
minor comments (4)
  1. [Eq. (6)] The x-noise is printed as sqrt{x(1-y)}, which becomes imaginary for y>1 and contradicts the Fokker-Planck equation (4) and the reduced model (8), where the term is sqrt{x(1+y)}. Please correct this typo.
  2. [Section III.A] The text contains a stray Spanish word, 'absorción', in the definition of theta_W in Eq. (10). Also, in Appendix A, 'to take our selfs an idea' should be 'to get an idea'.
  3. [Eq. (12)] The definition of y_c as 'the value of y at which the unstable manifold of S_0 cuts for first time x=1 with y>1' is terse; please clarify the geometric construction and state how y_c(rho) is computed in practice.
  4. [Fig. 3] Panels (e)-(i) are described as 'numerically computed' and 'approximated analytical'; please specify in the caption which panels are Monte Carlo and which are Eq. (11), to avoid ambiguity.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the analytic extinction probabilities are derived from the model's deterministic drift and noise with no fitted parameters; the one-noise reduction is an unvalidated scope assumption, not a circular step.

full rationale

The derivation chain is self-contained and parameter-free. The mesoscopic SDE (Eq. 6) is obtained from the explicit master equation (Eqs. 1–3) via a standard van Kampen expansion; the reduced one-noise model (Eq. 8) follows by setting D_y = 0 using the diffusion ratio alpha = D_y/D_x = sqrt(rho/r) (Eq. 7), with r estimated from yeast data in Appendix A rather than from the extinction probabilities themselves. The analytic extinction probabilities (Eq. 11) are computed in Appendices B and C from the deterministic drift and noise strength D: p_sl comes from an exact branching-process solution (Eqs. B1–B6), and p_e comes from a boundary-layer solution of the backward Kolmogorov equation (Eqs. C1–C13), with the barrier eta(rho) obtained from the deterministic unstable-manifold cut y_c(rho), not fitted to Monte Carlo results. Appendix D provides an independent asymptotic check. No target quantity is used as an input, and no load-bearing conclusion depends on a self-citation; the self-citations (e.g., refs. 2, 10, 12, 35) are contextual. The main caveats are that the toxin-fluctuation reduction is never checked against the full two-noise SDE or the individual-based model, and Eq. 6 contains a sign typo in the x-noise term; these are scoping and correctness risks, not circularity.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No fitted parameters. The central formulas involve the deterministic quantity η(ρ) via y_c(ρ), all computed from the drift field, not from extinction data. The main implicit assumptions are the reduction to one noise and the boundary condition in the singular-perturbation solution.

assumptions (4)
  • standard math Van Kampen system-size expansion yields a valid Fokker-Planck/SDE description for large volume V (small D).
    Used in Sec. II A to derive Eq. (5)/(6) from the master equation; standard asymptotic expansion, but requires V large.
  • domain assumption Toxin concentration fluctuations are negligible: α = D_y/D_x = sqrt(ρ/r) << 1.
    Sec. II B; empirical estimate for yeast gives r~10^13 (Appendix A), but no full-model simulation checks this reduction.
  • domain assumption The analysis is restricted to the oscillatory deterministic regime ρ < 4 and initial conditions x0<1, y0=0.
    Sec. III A; the boom-bust classification depends on the deterministic trajectory dipping near x=0 after the first cycle; outside this regime the scenario changes.
  • ad hoc to paper Absorption probability boundary condition: u(x,1)=0 for x>0 (a trajectory returning to y=1 is considered safe during the first bust).
    Appendix C; the paper states u(1,y)=0 for y<1 but constructs a solution satisfying u(x,1)=0; this replacement is not justified.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Noise-induced excitability: bloom, bust and extirpation in autotoxic population dynamics." pith.science (2026). https://pith.science/paper/7MFO7UKE

@misc{pith2026260120670,
  author       = {Pith},
  title        = {Pith review of: Noise-induced excitability: bloom, bust and extirpation in autotoxic population dynamics},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7MFO7UKE}},
  note         = {Machine review of arXiv:2601.20670}
}
read the original abstract

Species populations often modify their environment as they grow. When environmental feedback operates more slowly than population growth, the system can undergo boom-bust dynamics, where the population overshoots its carrying capacity and subsequently collapses. In extreme cases, this collapse leads to total extinction. While deterministic models typically fail to capture these finite-time extinction events, we propose a stochastic framework, derived from an individual-based model, to describe boom-bust-extirpation dynamics. We identify a noise-driven, threshold-like behavior where, depending on initial conditions, the population either undergoes a ``boom'' or is extirpated before the expansion occurs. Furthermore, we characterize a transition between an excitable regime, where most trajectories are captured by the absorbing state immediately after the first bust, and a persistent regime, where most populations reach a metastable state. We show that this transition is governed by the noise strength and the ratio of environmental-to-population timescales. This framework provides a theoretical basis for understanding irreversible transitions in invasive species, plant succession, microbial dynamics, and the elimination of cancerous tumors.

Figures

Figures reproduced from arXiv: 2601.20670 by the authors.

Figure 1
Figure 1. FIG. 1 [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2 [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

4 extracted references · 2 linked inside Pith

  1. [10]

    A way of making Europe,

    Our results suggest that such behaviors can instead emerge from the interplay be- tween negative feedback and noise, reducing the number of assumptions required to model excitable ecological dynamics. In this paper, we utilize the standard Van Kampen’s system expansion technique to derive a mesoscopic Langevin descrip- tion from individual-level first pri...

    2020

  2. [37]

    Boom-bust dynamics in biological invasions: towards an im- proved application of the concept,

    And therefore: ν(y)∼ −1 2 a1−aΓ(a)∼ r πa 2 e−a (D8) Notice that using Stirling’s approximation for the gamma function for big values ofa, i.e.Γ(a)∼ √ 2πa a−1/2 e−a, and substituting in (D6) we found: ν(y)∼ − r π 2ρ e−1/ρ ,(0<ρ≪1,y>1).(D9) Using this expresion in Eq. 11 we obtain that forρ≪1: ln(ˆpe/(1−ˆpsl))∼ r 2 π x D2 ρ1/2e(ln(y)−y+1)/ρ ,(D10) which con...

    arXiv 2017

  3. [2010]

    Wkb theory of epidemic fade-out in stochastic populations,

    p. 379–524. 33B. Meerson and P. V . Sasorov, “Wkb theory of epidemic fade-out in stochastic populations,” Physical Review E80(2009), 10.1103/phys- reve.80.041130. 34Forx>y(resp.x<y) one necessarily hasdy/dt>0 (resp.dy/dt<0), implying that the dynamics aroundS p must be counterclockwise. 35J. Aguilar, J. W. Baron, T. Galla, and R. Toral, “Sampling rare tra...

    doi:10.1103/phys- 2009

  4. [2026]

    boom-bust-extirpation

    Species populations often modify their environment as they grow. When environmental feedback operates more slowly than population growth, the system can undergo boom-bust dynamics, where the population overshoots its carrying capacity and subsequently collapses. In extreme cases, this collapse leads to total extinction. While deterministic models typicall...

    arXiv 2026

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.