Pith. sign in

REVIEW 3 major objections 4 minor 1 cited by

Sign changes in empirically estimated species-interaction matrices can arise from the system's own relaxation dynamics in purely competitive communities with fixed ecological roles, so sign flips alone do not certify facilitation-competitio

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

Empirical interaction matrices are time-dependent; sign flips can arise from intrinsic dynamics alone, and short-time extrapolation separates direct from indirect interactions.

T0 review reviewed 2026-08-05 challenge →

load-bearing objection The short-time extrapolation protocol and the 1C1R sign-change demonstration are the real contributions; the general-N proof has a genuine row/column error, but the core message survives and the paper deserves refereeing. the 3 major comments →

arxiv 2508.19197 v2 pith:JBGH2XLO submitted 2025-08-26 q-bio.PE physics.bio-ph

Unraveling the temporal dependence of ecological interaction measures

Javier Aguilar , Samir Suweis , Amos Maritan , Sandro Azaele This is my paper
classification q-bio.PE physics.bio-ph MSC 92D2592D4037N25
keywords empirical interaction matrixpulse perturbationexperiment durationdirect and indirect interactionsconsumer-resource modelsign changestemporal scalesmodel inference
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

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 asks whether the fluctuating, sometimes sign-changing species-interaction matrices inferred from pulse-perturbation experiments reflect real ecological change or artifacts of how the experiment is run. Using synthetic time series from consumer-resource and other population models, the authors show the latter: the empirical interaction measure is itself a time-dependent object, and in purely competitive systems with fixed ecological roles it can oscillate and change sign during relaxation to equilibrium. This means apparent switches between competition and facilitation, often read as support for the stress-gradient hypothesis, can be intrinsic dynamical fluctuations. The paper then turns the effect into a tool: the short-time limit of the interaction measure isolates direct pairwise couplings, while longer durations add indirect, community-mediated feedback and experimental constraints such as batch, chemostat, or biotic resource renewal. A timescale-aware reading of interaction matrices, with short-time measurements or extrapolations, thus offers clearer experimental design and a route to infer generalized Lotka-Volterra models from many short perturbed time series.

Core claim

Empirical interaction matrices from pulse perturbations are functions of experiment duration. Expanding at small perturbation and sampling time but arbitrary duration yields M_{i,j}=G_i·E_j Δx; expanding in duration isolates direct couplings at zeroth order and indirect feedback at higher orders. Near a stable fixed point, the trajectory-sensitivity vector is a column of e^{Jt}, so interaction signs can flip in purely competitive consumer-resource systems with fixed roles. Consumer-consumer interactions vanish at t→0 and are negative at first order regardless of resource-renewal protocol; second-order terms depend on protocol. The short-time limit yields a model-inference scheme reducing to

What carries the argument

The carrying object is the empirical interaction measure M_{j→i} of Eq. (2), the difference in per-capita growth-rate estimates between a perturbed and a baseline trajectory in a PULSE experiment. Its small-perturbation, small-sampling-time, arbitrary-duration Taylor expansion factorizes as M_{i,j}=G_i·E_j Δx, with G_i the gradient of species i's instantaneous per-capita growth rate and E_j the evolution vector, i.e. the derivative of the whole trajectory with respect to the initial density of species j. Near a stable coexistence fixed point, E_j becomes a column of the matrix exponential e^{Jt}; comparing the damping timescale τ_D with the oscillation timescale τ_O predicts when interaction

Load-bearing premise

The load-bearing premise is that growth and density trajectories are smooth and differentiable enough for the Taylor expansion of the interaction measure and the t→0 limit to exist; the paper itself shows (Section III.C, Appendix G) that white-noise-driven growth has no such limit, so the short-time isolation result does not hold for discontinuous stochastic dynamics.

What would settle it

Perform controlled pulse perturbations on a purely competitive two-consumer, one-resource system with known fixed couplings and measure M_{C→R} versus experiment duration. The mechanism predicts sign flips only when damped oscillations are present (consumer uptake efficiency above the threshold ε_TH of Eq. 14) and onset times scale with the oscillation timescale τ_O, while extrapolation to t=0 keeps the sign of the known direct coupling. Sign flips in the monotone regime, or a nonzero indirect interaction in the t→0 limit, would falsify the claim.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • Sign flips in empirical interaction matrices are not sufficient evidence for facilitation–competition shifts; the experiment duration relative to the system's relaxation and oscillation timescales must be checked first.
  • Interaction measurements taken at very short durations, or extrapolated to t=0, recover direct pairwise couplings; in consumer-resource systems, consumer-consumer interactions are zero at t=0 and negative at first order, revealing resource-mediated competition.
  • Experimental protocol—batch culture, chemostat, or biotic resource renewal—does not affect the leading short-time terms but changes higher-order terms, so protocol choice can create or suppress apparent sign changes.
  • Temporal integration over a time window does not remove the ambiguity: the dominant sign of the integrated interaction depends on model parameters.
  • Many short perturbed time series across initial conditions can replace long longitudinal data for model inference, and the inferred dynamics is generically generalized Lotka-Volterra when the coefficients are density-independent.

Where Pith is reading between the lines

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

  • The t→0 extrapolation offers a practical diagnostic: a sign flip that disappears under extrapolation to zero duration is a dynamical artifact of relaxation, not a change in ecological roles; the paper implies this but does not test it on noisy field data.
  • Because white-noise-driven growth has no well-defined short-time limit (Appendix G), a natural extension is to identify the noise correlation time setting the shortest reliable experiment duration, giving a lower bound for the validity of the method.
  • The inference scheme assumes the interaction vector field is conservative; checking its curl from data would test whether measured interactions are compatible with autonomous ODE dynamics or indicate missing variables.
  • The protocol dependence of second-order terms is directly testable in microbial co-cultures: with identical inocula and coupling constants, batch and chemostat setups should yield identical short-time interactions but diverging sign behavior at longer times.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The paper analyzes the finite-time, finite-resolution 'empirical interaction matrix' M_{j->i} obtained from pulse perturbation experiments. Starting from a general ODE model of per-capita growth, it derives Taylor expansions in perturbation size and sampling interval (Eq. 5), and in experiment duration (Eq. 20). The zeroth-order term is the derivative of the per-capita growth rate, while higher-order terms encode indirect and protocol-dependent effects. The authors show in a one-consumer/one-resource consumer-resource model that M can oscillate and change sign even though the underlying interactions are purely competitive, and they illustrate similar behavior in multi-species simulations, a yeast-galactose-ethanol model, and grasshopper data. They also show that the first two terms of the short-time expansion are independent of the resource renewal protocol, and propose a model-inference method based on short time series. The central conceptual claim is that temporal sign changes in measured interactions need not indicate facilitation-competition shifts, and that short-duration measurements can isolate direct couplings.

Significance. If the technical proof is repaired, this is a useful and timely contribution. It gives a clean analytical framework, supported by reproducible synthetic simulations, for a problem that ecologists often treat only heuristically: the dependence of empirical interaction matrices on experimental duration, sampling interval, and perturbation size. The demonstration that sign changes can arise from relaxation dynamics in a purely competitive system is an important caution for interpreting interaction shifts as role changes. The short-time expansion and the protocol-independence result in Eq. (23) are elegant and potentially of practical value for experimental design. The paper also makes code and data available, which strengthens confidence in the numerical results. However, the breadth of the claims currently exceeds what is rigorously established, because the general-N proof contains a transposition error and the stochastic/deterministic scope is not stated prominently enough.

major comments (3)
  1. [Appendix C, Eq. (C6); Sec. V.B, Eq. (31)] The evolution vector is misindexed. Eq. (C6) states E_j = sum_k (e^{Jt})_{j,k} e_k, but from the linearized solution delta x(t) = e^{Jt} Delta x e_j one obtains (E_j)_k = (e^{Jt})_{k,j}, i.e. the j-th column, not the j-th row. Eq. (31) repeats this transposed expression. For the non-symmetric Jacobian of a consumer-resource system, row and column entries are not interchangeable. Since the arbitrary-N sign-change argument rests on Eq. (31), this is a load-bearing error. The equation should be corrected and the analytical curves in Fig. 3, which are claimed to follow from Eq. (31), must be rechecked against the corrected formula.
  2. [Sec. V.B and Sec. III.C] Even after correcting the transpose, the proof that sign changes occur for arbitrary N is incomplete. The statement that elements of e^{Jt} can be positive or negative does not imply that the weighted sum in Eq. (31) changes sign: the coefficients alpha and epsilon are positive, so cancellations are possible. References [53,54] concern Lotka-Volterra sign patterns, not signs of matrix exponential entries. The authors need either a concrete general proof (for example, exhibiting for each N a parameter regime in which one oscillatory mode has nonzero projection onto the relevant component) or a restriction of the claim to the demonstrated numerical cases.
  3. [Sec. III.D, Eq. (21); abstract] The headline recommendation that 'short-term measurements reliably isolate direct, pairwise species couplings' is too broad. Eq. (21) assumes differentiable deterministic dynamics and convergence of the empirical growth rate to the instantaneous growth rate. Appendix G shows that this limit does not exist for white-noise-driven processes such as Eq. (16): the self-interaction diverges as Delta t -> 0 (Eq. 17). The paper itself acknowledges this in Sec. III.C, but the abstract and the beginning of Sec. III.D do not carry the qualification. Please state the smoothness/regularity condition wherever the practical recommendation is made, and explicitly connect it to the Appendix G caveat.
minor comments (4)
  1. [Sec. V.C] In the yeast parameters, 'K1' is listed twice and 'nu1' twice; the second occurrences should presumably be K2 and nu2. Please check.
  2. [Appendix H] Eqs. (H3)-(H4) contain malformed superscript expressions involving the Ornstein-Uhlenbeck integral; they should read exp(v t + integral_0^t O_s ds).
  3. [Sec. III.C and Sec. V.B] Sec. III.C says 'Using dynamical system theory 42,43' for the arbitrary-N demonstration, but the detailed argument in Sec. V.B cites refs. [53,54]. Please align the citations and make clear which result is being used where.
  4. [Eq. (21) and Sec. III.D] The identification of the t->0 limit with 'direct interactions' is, to a large extent, a definitional consequence of taking direct interactions to be derivatives of the per-capita growth rate. This is a reasonable convention, but the paper should acknowledge that the empirical content lies in the higher-order terms and protocol dependence, not in the zeroth-order identification itself.

Circularity Check

1 steps flagged

Short-time 'direct interaction' identification is partly definitional, but protocol-dependence and sign-change results are independent.

specific steps
  1. self definitional [Section III.D, Eqs. (20)-(21) and following paragraph]
    "The zeroth-order in the expansion simply becomes the derivative of the per-capita growth rate evaluated at the perturbation point (see Appendix E for details): M(0)i,j(x) = ∂/∂xj gi(x) Δx. Therefore, the zeroth-order term in the expansion depends solely on the gradient vector Gi and captures the explicit couplings between species i and j present in the model used to generate the synthetic data. As such, this term represents the direct interactions between species."

    'Direct interactions' are not independently defined; they are identified with the zeroth-order Taylor coefficient ∂_j g_i. Since M^(0) is obtained from the same Taylor expansion (Eq. (5)) of the empirical measure, the claim that short-time measurements isolate direct couplings follows by construction rather than from an independent empirical test. The direct/indirect distinction is given substantive content by the separately computed first-order term (M^(1)), so the circularity is partial, not total.

full rationale

The paper's main dynamical claims — sign changes in purely competitive consumer-resource systems, sensitivity of measured interactions to resource-renewal protocol, and the model-inference method — are derived from explicit model equations and Taylor expansions, not from fitting the target conclusions. No load-bearing self-citation was found: refs. 37 and 26 involve overlapping authors but are used as sources of models/data, and the arguments do not reduce to those citations. The definitional step is the identification of 'direct interactions' with the t→0 (zeroth-order) term of the expansion; this is a mathematical identity rather than an independent discovery, so it partially builds the conclusion into the definition. The Appendix C transpose issue is a correctness concern (not a circularity), and Appendix G itself acknowledges the non-existence of the short-time limit for white-noise processes; these should be weighed separately.

Axiom & Free-Parameter Ledger

3 free parameters · 4 axioms · 0 invented entities

The central claim rests on standard ODE population dynamics and differentiability assumptions; the paper introduces no new entities and fits no parameters to its main conclusions, though several illustrative simulations use hand-chosen parameters. The inference method additionally assumes the estimated interaction vector field is conservative, which the paper itself flags as likely violated by real data.

free parameters (3)
  • Consumer-resource simulation parameters (epsilon, r, K, alpha, d) = various, see Sec. V.C
    Hand-chosen to illustrate oscillatory and non-oscillatory regimes (Figs. 3, 4); not fitted to the target result, and the analytical claims hold for generic parameter values.
  • Yeast-galactose-ethanol model parameters from ref. 37 = listed in Sec. V.C
    Taken from an earlier fit to experimental data; used only to generate illustrative time series, not fitted in this paper.
  • Parabolic extrapolation coefficients in Fig. 4c = fitted to the discrete sample points
    Used to demonstrate the t->0 extrapolation trick; not load-bearing for the central claim.
axioms (4)
  • domain assumption Per-capita growth rates g_i(x) are differentiable and the dynamics are deterministic ODEs (Eq. 3).
    Required for the Taylor expansion in Eq. (5) and the t->0 limit in Eq. (21); the paper acknowledges stochastic extensions where the infinitesimal limit may not exist (Appendix G).
  • domain assumption The system admits a stable coexistence fixed point near which linearization is valid (Appendix C).
    Used to derive the exponential-matrix expression Eq. (31) and the sign-change argument.
  • domain assumption Consumer-resource models (Eqs. 9-10) with non-negative parameters and renewal functions capture the relevant ecology of microbial protocols.
    Basis for the protocol-dependence results in Section III.D; real systems may have other mechanisms.
  • ad hoc to paper For model inference, the empirical vector field F_i is conservative (Appendix F).
    The paper itself notes this is not guaranteed and that rotational components imply the data are incompatible with the ODE form or contaminated by finite-time errors.

reviewed 2026-08-05 · how reviews work

0 comments
Cite this review

Pith. "Pith review of Unraveling the temporal dependence of ecological interaction measures." pith.science (2026). https://pith.science/paper/JBGH2XLO

@misc{pith2026250819197,
  author       = {Pith},
  title        = {Pith review of: Unraveling the temporal dependence of ecological interaction measures},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JBGH2XLO}},
  note         = {Machine review of arXiv:2508.19197}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Identifying the network of species interactions is a fundamental step toward understanding ecosystem stability and biodiversity. However, the interpretability of empirical interaction measures remains a major challenge. Experimental estimates frequently exhibit puzzling temporal fluctuations, including sign shifts typically interpreted as transitions between competition and facilitation. Here, we analyze the temporal behavior of pairwise interaction measures to demonstrate that these fluctuations - and apparent shifts in ecological roles - can emerge intrinsically from standard population dynamics, without any underlying change in the actual ecological relationships. We show that inferred interactions are heavily distorted by experimental protocol choices, particularly the duration of observation and microbial growth constraints. By systematically evaluating interactions across timescales, we uncover a principled mechanism to mitigate these biases: short-term measurements reliably isolate direct, pairwise species couplings, whereas longer-term observations inevitably absorb indirect community feedbacks and systemic experimental constraints. By disentangling direct couplings from indirect network effects, our framework provides a robust, timescale-aware approach to interpreting empirical interaction matrices, offering critical quantitative guidance for experimental design and predictive ecosystem modeling.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

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

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

    q-bio.PE 2026-01 conditional novelty 6.0

    In a stochastic autotoxicity model, demographic noise creates a sharp threshold and an excitable-to-persistent transition, with analytic formulas for extinction-pathway probabilities.

Reference graph

Works this paper leans on

53 extracted references · 39 canonical work pages · cited by 1 Pith paper · 1 internal anchor

  1. [1]

    The indeterminacy of ecological interactions as perceived through perturbation experiments

    Peter Yodzis. The indeterminacy of ecological interactions as perceived through perturbation experiments. Ecology, 69: 0 508--515, 4 1988. ISSN 0012-9658. doi:10.2307/1940449. URL https://esajournals.onlinelibrary.wiley.com/doi/10.2307/1940449

    doi:10.2307/1940449 1988

  2. [2]

    Feedback between population and evolutionary dynamics determines the fate of social microbial populations

    Alvaro Sanchez and Jeff Gore. Feedback between population and evolutionary dynamics determines the fate of social microbial populations. PLoS Biology, 11: 0 e1001547, 4 2013. ISSN 1545-7885. doi:10.1371/journal.pbio.1001547. URL https://dx.plos.org/10.1371/journal.pbio.1001547

    work page doi:10.1371/journal.pbio.1001547 2013

  3. [3]

    Bertness and Ragan Callaway

    Mark D. Bertness and Ragan Callaway. Positive interactions in communities. Trends in Ecology & Evolution, 9: 0 191--193, 5 1994. ISSN 01695347. doi:10.1016/0169-5347(94)90088-4. URL https://linkinghub.elsevier.com/retrieve/pii/0169534794900884

    work page arXiv 1994

  4. [4]

    Callaway, R

    Ragan M. Callaway, R. W. Brooker, Philippe Choler, Zaal Kikvidze, Christopher J. Lortie, Richard Michalet, Leonardo Paolini, Francisco I. Pugnaire, Beth Newingham, Erik T. Aschehoug, Cristina Armas, David Kikodze, and Bradley J. Cook. Positive interactions among alpine plants increase with stress. Nature, 417: 0 844--848, 6 2002. ISSN 0028-0836. doi:10.10...

    work page doi:10.1038/nature00812 2002

  5. [5]

    Evolution of local facilitation in arid ecosystems

    Sonia Kéfi, Minus van Baalen, Max Rietkerk, and Michel Loreau. Evolution of local facilitation in arid ecosystems. The American Naturalist, 172: 0 E1--E17, 7 2008. ISSN 0003-0147. doi:10.1086/588066. URL https://www.journals.uchicago.edu/doi/10.1086/588066

    work page doi:10.1086/588066 2008

  6. [6]

    Bertness, and Andrew H

    Qiang He, Mark D. Bertness, and Andrew H. Altieri. Global shifts towards positive species interactions with increasing environmental stress. Ecology Letters, 16: 0 695--706, 5 2013. ISSN 1461-023X. doi:10.1111/ele.12080. URL https://onlinelibrary.wiley.com/doi/10.1111/ele.12080

    work page doi:10.1111/ele.12080 2013

  7. [7]

    Adams, Elizabeth M

    Amy E. Adams, Elizabeth M. Besozzi, Golya Shahrokhi, and Michael A. Patten. A case for associational resistance: Apparent support for the stress gradient hypothesis varies with study system. Ecology Letters, 25: 0 202--217, 1 2022. ISSN 1461-023X. doi:10.1111/ele.13917. URL https://onlinelibrary.wiley.com/doi/10.1111/ele.13917

    work page doi:10.1111/ele.13917 2022

  8. [8]

    Facilitation and competition deconstructed: a mechanistic modelling approach to the stress gradient hypothesis applied to drylands

    Rubén Díaz-Sierra, Max Rietkerk, Mart Verwijmeren, and Mara Baudena. Facilitation and competition deconstructed: a mechanistic modelling approach to the stress gradient hypothesis applied to drylands. Scientific Reports, 14: 0 2205, 1 2024. ISSN 2045-2322. doi:10.1038/s41598-024-52447-z. URL https://www.nature.com/articles/s41598-024-52447-z

    work page doi:10.1038/s41598-024-52447-z 2024

  9. [9]

    Banavar, and Amos Maritan

    Sandro Azaele, Samir Suweis, Jacopo Grilli, Igor Volkov, Jayanth R. Banavar, and Amos Maritan. Statistical mechanics of ecological systems: Neutral theory and beyond. Reviews of Modern Physics, 88, 7 2016. ISSN 15390756. doi:10.1103/RevModPhys.88.035003

    work page doi:10.1103/revmodphys.88.035003 2016

  10. [10]

    Macroecological laws describe variation and diversity in microbial communities

    Jacopo Grilli. Macroecological laws describe variation and diversity in microbial communities. Nature Communications, 11: 0 1--11, 2020. ISSN 20411723. doi:10.1038/s41467-020-18529-y. URL http://dx.doi.org/10.1038/s41467-020-18529-y

    doi:10.1038/s41467-020-18529-y 2020

  11. [11]

    Ubiquitous abundance distribution of non-dominant plankton across the global ocean

    Enrico Ser-Giacomi, Lucie Zinger, Shruti Malviya, Colomban De Vargas, Eric Karsenti, Chris Bowler, and Silvia De Monte. Ubiquitous abundance distribution of non-dominant plankton across the global ocean. Nature Ecology and Evolution, 2: 0 1243--1249, 8 2018. ISSN 2397334X. doi:10.1038/s41559-018-0587-2

    work page doi:10.1038/s41559-018-0587-2 2018

  12. [12]

    The stability–complexity relationship at age 40: a random matrix perspective

    Stefano Allesina and Si Tang. The stability–complexity relationship at age 40: a random matrix perspective. Population Ecology, 57: 0 63--75, 1 2015. ISSN 1438-3896. doi:10.1007/s10144-014-0471-0. URL https://esj-journals.onlinelibrary.wiley.com/doi/10.1007/s10144-014-0471-0

    doi:10.1007/s10144-014-0471-0 2015

  13. [13]

    Generalized model of island biodiversity

    David A Kessler and Nadav M Shnerb. Generalized model of island biodiversity. Physical Review E, 91 0 (4): 0 042705, 2015

    work page 2015

  14. [14]

    Complex systems in ecology: a guided tour with large lotka–volterra models and random matrices

    Imane Akjouj, Matthieu Barbier, Maxime Clenet, Walid Hachem, Mylène Maïda, François Massol, Jamal Najim, and Viet Chi Tran. Complex systems in ecology: a guided tour with large lotka–volterra models and random matrices. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 480 0 (2294): 0 20230284, 2024. doi:10.1098/rspa.202...

    arXiv 2024

  15. [15]

    Generalized lotka-volterra systems with time correlated stochastic interactions

    Samir Suweis, Francesco Ferraro, Christian Grilletta, Sandro Azaele, and Amos Maritan. Generalized lotka-volterra systems with time correlated stochastic interactions. Physical Review Letters, 133 0 (16): 0 167101, 2024

    work page 2024

  16. [16]

    John H. Lawton. Are there general laws in ecology? Oikos, 84: 0 177, 2 1999. ISSN 00301299. doi:10.2307/3546712. URL https://www.jstor.org/stable/3546712?origin=crossref

    work page arXiv 1999

  17. [17]

    Banavar, Stephen P

    Igor Volkov, Jayanth R. Banavar, Stephen P. Hubbell, and Amos Maritan. Inferring species interactions in tropical forests. Proceedings of the National Academy of Sciences, 106: 0 13854--13859, 8 2009. ISSN 0027-8424. doi:10.1073/pnas.0903244106. URL https://pnas.org/doi/full/10.1073/pnas.0903244106

    work page doi:10.1073/pnas.0903244106 2009

  18. [18]

    Identifying keystone species in the human gut microbiome from metagenomic timeseries using sparse linear regression

    Charles K Fisher and Pankaj Mehta. Identifying keystone species in the human gut microbiome from metagenomic timeseries using sparse linear regression. PloS one, 9 0 (7): 0 e102451, 2014

    work page 2014

  19. [19]

    Foster and Thomas Bell

    Kevin R. Foster and Thomas Bell. Competition, not cooperation, dominates interactions among culturable microbial species. Current Biology, 22: 0 1845--1850, 10 2012. ISSN 09609822. doi:10.1016/j.cub.2012.08.005. URL https://linkinghub.elsevier.com/retrieve/pii/S0960982212009335

    work page doi:10.1016/j.cub.2012.08.005 2012

  20. [20]

    Beyond competition: the stress gradient hypothesis tested in plant herbivore interactions

    Pedro Daleo and Oscar Iribarne. Beyond competition: the stress gradient hypothesis tested in plant herbivore interactions. Ecology, 90: 0 2368--2374, 9 2009. ISSN 0012-9658. doi:10.1890/08-2330.1. URL https://esajournals.onlinelibrary.wiley.com/doi/10.1890/08-2330.1

    work page doi:10.1890/08-2330.1 2009

  21. [21]

    Species co-occurrence networks: can they reveal trophic and non-trophic interactions in ecological communities? Ecology, 99 0 (3): 0 690--699, 2018

    Mara A Freilich, Evie Wieters, Bernardo R Broitman, Pablo A Marquet, and Sergio A Navarrete. Species co-occurrence networks: can they reveal trophic and non-trophic interactions in ecological communities? Ecology, 99 0 (3): 0 690--699, 2018

    work page 2018

  22. [22]

    Melián, and Victor M

    Sandra Hervías-Parejo, Mar Cuevas-Blanco, Lucas Lacasa, Anna Traveset, Isabel Donoso, Ruben Heleno, Manuel Nogales, Susana Rodríguez-Echeverría, Carlos J. Melián, and Victor M. Eguíluz. On the structure of species-function participation in multilayer ecological networks. Nature Communications, 15: 0 8910, 10 2024. ISSN 2041-1723. doi:10.1038/s41467-024-53...

    work page doi:10.1038/s41467-024-53001-1 2024

  23. [23]

    Microbial interactions: from networks to models

    Karoline Faust and Jeroen Raes. Microbial interactions: from networks to models. Nature Reviews Microbiology, 10 0 (8): 0 538--550, 2012

    work page 2012

  24. [24]

    Deciphering microbial interactions in synthetic human gut microbiome communities

    Ophelia S Venturelli, Alex V Carr, Garth Fisher, Ryan H Hsu, Rebecca Lau, Benjamin P Bowen, Susan Hromada, Trent Northen, and Adam P Arkin. Deciphering microbial interactions in synthetic human gut microbiome communities. Molecular Systems Biology, 14, 6 2018. ISSN 1744-4292. doi:10.15252/msb.20178157. URL https://www.embopress.org/doi/10.15252/msb.20178157

    doi:10.15252/msb.20178157 2018

  25. [25]

    Microbial interactions in theory and practice: when are measurements compatible with models? Current Opinion in Microbiology, 75: 0 102354, 10 2023

    Aurore Picot, Shota Shibasaki, Oliver J Meacock, and Sara Mitri. Microbial interactions in theory and practice: when are measurements compatible with models? Current Opinion in Microbiology, 75: 0 102354, 10 2023. ISSN 13695274. doi:10.1016/j.mib.2023.102354. URL https://linkinghub.elsevier.com/retrieve/pii/S1369527423000917

    work page arXiv 2023

  26. [26]

    Microbiomes through the looking glass, 4 2025

    Jacopo Pasqualini, Amos Maritan, Andrea Rinaldo, Sonia Facchin, Edoardo Savarino, Ada Altieri, and Samir Suweis. Microbiomes through the looking glass, 4 2025. URL https://elifesciences.org/reviewed-preprints/105948v1

    work page 2025

  27. [27]

    Bender, Ted J

    Edward A. Bender, Ted J. Case, and Michael E. Gilpin. Perturbation experiments in community ecology: Theory and practice. Ecology, 65: 0 1--13, 2 1984. ISSN 0012-9658. doi:10.2307/1939452. URL https://esajournals.onlinelibrary.wiley.com/doi/10.2307/1939452

    work page doi:10.2307/1939452 1984

  28. [28]

    Laska and J

    Mark S. Laska and J. Timothy Wootton. Theoretical concepts and empirical approaches to measuring interaction strength. Ecology, 79: 0 461--476, 1998. ISSN 00129658. doi:10.1890/0012-9658(1998)079[0461:TCAEAT]2.0.CO;2

    work page doi:10.1890/0012-9658(1998)079 1998

  29. [29]

    José Luiz Attayde and Lars‐Anders A. Hansson. Press perturbation experiments and the indeterminacy of ecological interactions: Effects of taxonomic resolution and experimental duration. Oikos, 92: 0 235--244, 1 2001. ISSN 00301299. doi:10.1034/j.1600-0706.2001.920205.x

    work page internal anchor Pith review Pith/arXiv arXiv 2001

  30. [30]

    Yeakel, Andrew E

    Mark Novak, Justin D. Yeakel, Andrew E. Noble, Daniel F. Doak, Mark Emmerson, James A. Estes, Ute Jacob, M. Timothy Tinker, and J. Timothy Wootton. Characterizing species interactions to understand press perturbations: What is the community matrix? Annual Review of Ecology, Evolution, and Systematics, 47: 0 409--432, 11 2016. ISSN 1543-592X. doi:10.1146/a...

    doi:10.1146/annurev-ecolsys-032416-010215 2016

  31. [31]

    Pugnaire

    Cristina Armas and Susana Rodríguez-Echeverría and Francisco I. Pugnaire . A field test of the stress-gradient hypothesis along an aridity gradient . Journal of Vegetation Science, 22: 0 818--827, 2011. ISSN 16541103. doi:10.1111/j.1654-1103.2011.01301.x

    Pith/arXiv arXiv 2011

  32. [32]

    Toxicity drives facilitation between 4 bacterial species

    Philippe Piccardi, Björn Vessman, and Sara Mitri. Toxicity drives facilitation between 4 bacterial species. Proceedings of the National Academy of Sciences of the United States of America, 116: 0 15979--15984, 2019. ISSN 10916490. doi:10.1073/pnas.1906172116

    doi:10.1073/pnas.1906172116 2019

  33. [33]

    Oxidative stress changes interactions between 2 bacterial species from competitive to facilitative

    Rita Di Martino, Aurore Picot, and Sara Mitri. Oxidative stress changes interactions between 2 bacterial species from competitive to facilitative. PLoS Biology, 22: 0 1--21, 2024. ISSN 15457885. doi:10.1371/journal.pbio.3002482. URL http://dx.doi.org/10.1371/journal.pbio.3002482

    doi:10.1371/journal.pbio.3002482 2024

  34. [34]

    Bruce A. Menge. Dection of direct versus indirect effects: Were experiments long enough? The American Naturalist, 149: 0 801--823, 5 1997. ISSN 0003-0147. doi:10.1086/286025. URL https://www.journals.uchicago.edu/doi/10.1086/286025

    work page doi:10.1086/286025 1997

  35. [35]

    Network structure beyond food webs: mapping non‐trophic and trophic interactions on chilean rocky shores

    Sonia K \'e fi, Eric L Berlow, Evie A Wieters, Lucas N Joppa, Spencer A Wood, Ulrich Brose, and Sergio A Navarrete. Network structure beyond food webs: mapping non‐trophic and trophic interactions on chilean rocky shores. Ecology, 96: 0 291--303, 1 2015. ISSN 0012-9658. doi:10.1890/13-1424.1. URL https://esajournals.onlinelibrary.wiley.com/doi/10.1890/13-1424.1

    work page doi:10.1890/13-1424.1 2015

  36. [36]

    Meacock and Sara Mitri

    Oliver J. Meacock and Sara Mitri. Environment‐organism feedbacks drive changes in ecological interactions. Ecology Letters, 28, 1 2025. ISSN 1461-023X. doi:10.1111/ele.70027. URL https://onlinelibrary.wiley.com/doi/10.1111/ele.70027

    work page doi:10.1111/ele.70027 2025

  37. [37]

    Dynamic metabolic adaptation can promote species coexistence in competitive microbial communities

    Leonardo Pacciani-Mori, Andrea Giometto, Samir Suweis, and Amos Maritan. Dynamic metabolic adaptation can promote species coexistence in competitive microbial communities. PLoS Computational Biology, 16: 0 1--18, 2020. ISSN 15537358. doi:10.1371/journal.pcbi.1007896. URL http://dx.doi.org/10.1371/journal.pcbi.1007896

    work page doi:10.1371/journal.pcbi.1007896 2020

  38. [38]

    Ritchie and David Tilman

    Mark E. Ritchie and David Tilman. Predictions of species interactions from consumer-resource theory: experimental tests with grasshoppers and plants. Oecologia, 94: 0 516--527, 7 1993. ISSN 0029-8549. doi:10.1007/BF00566967. URL http://link.springer.com/10.1007/BF00566967

    work page doi:10.1007/bf00566967 1993

  39. [40]

    Macarthur's consumer-resource model

    Peter Chesson. Macarthur's consumer-resource model. Theoretical Population Biology, 37: 0 26--38, 2 1990. ISSN 00405809. doi:10.1016/0040-5809(90)90025-Q. URL https://linkinghub.elsevier.com/retrieve/pii/004058099090025Q

    work page arXiv 1990

  40. [41]

    Capitán, and David Alonso

    Gian Marco Palamara, José A. Capitán, and David Alonso. The stochastic nature of functional responses. Entropy, 23: 0 575, 5 2021. ISSN 1099-4300. doi:10.3390/e23050575. URL https://www.mdpi.com/1099-4300/23/5/575

    work page doi:10.3390/e23050575 2021

  41. [42]

    Strogatz

    Steven H. Strogatz. Nonlinear Dynamics and Chaos. CRC Press, 5 2018. ISBN 9780429961113. doi:10.1201/9780429492563. URL https://www.taylorfrancis.com/books/9780429961113

    arXiv 2018

  42. [43]

    Chemical and Biological Processes in Fluid Flows

    Zoltán Neufeld and Emilio Hernández-García. Chemical and Biological Processes in Fluid Flows. PUBLISHED BY IMPERIAL COLLEGE PRESS AND DISTRIBUTED BY WORLD SCIENTIFIC PUBLISHING CO., 2009. doi:10.1142/p471. URL https://www.worldscientific.com/doi/abs/10.1142/p471

    work page doi:10.1142/p471 2009

  43. [44]

    A. J. McKane and T. J. Newman. Predator-prey cycles from resonant amplification of demographic stochasticity. Physical Review Letters, 94: 0 218102, 6 2005. ISSN 0031-9007. doi:10.1103/PhysRevLett.94.218102. URL https://link.aps.org/doi/10.1103/PhysRevLett.94.218102

    doi:10.1103/physrevlett.94.218102 2005

  44. [45]

    R. T. Paine. Intertidal community structure. Oecologia, 15: 0 93--120, 1974. ISSN 0029-8549. doi:10.1007/BF00345739. URL http://link.springer.com/10.1007/BF00345739

    work page doi:10.1007/bf00345739 1974

  45. [46]

    Goldford, Nanxi Lu, Djordje Bajić, Sylvie Estrela, Mikhail Tikhonov, Alicia Sanchez-Gorostiaga, Daniel Segrè, Pankaj Mehta, and Alvaro Sanchez

    Joshua E. Goldford, Nanxi Lu, Djordje Bajić, Sylvie Estrela, Mikhail Tikhonov, Alicia Sanchez-Gorostiaga, Daniel Segrè, Pankaj Mehta, and Alvaro Sanchez. Emergent simplicity in microbial community assembly. Science, 361: 0 469--474, 8 2018. ISSN 0036-8075. doi:10.1126/science.aat1168. URL https://www.science.org/doi/10.1126/science.aat1168

    doi:10.1126/science.aat1168 2018

  46. [47]

    Treloar, Alex J.H

    Neythen J. Treloar, Alex J.H. Fedorec, Brian Ingalls, and Chris P. Barnes. Deep reinforcement learning for the control of microbial co-cultures in bioreactors. PLoS Computational Biology, 16, 4 2020. ISSN 15537358. doi:10.1371/journal.pcbi.1007783

    work page doi:10.1371/journal.pcbi.1007783 2020

  47. [48]

    Structure and evolution of streptomyces interaction networks in soil and in silico

    Kalin Vetsigian, Rishi Jajoo, and Roy Kishony. Structure and evolution of streptomyces interaction networks in soil and in silico. PLoS Biology, 9, 10 2011. ISSN 15449173. doi:10.1371/journal.pbio.1001184

    work page doi:10.1371/journal.pbio.1001184 2011

  48. [49]

    Berlow, Evie A

    Sonia Kéfi, Eric L. Berlow, Evie A. Wieters, Sergio A. Navarrete, Owen L. Petchey, Spencer A. Wood, Alice Boit, Lucas N. Joppa, Kevin D. Lafferty, Richard J. Williams, Neo D. Martinez, Bruce A. Menge, Carol A. Blanchette, Alison C. Iles, and Ulrich Brose. More than a meal… integrating non‐feeding interactions into food webs. Ecology Letters, 15: 0 291--30...

    arXiv 2012

  49. [50]

    Robert T. Paine. Food web complexity and species diversity. The American Naturalist, 100: 0 65--75, 1 1966. ISSN 0003-0147. doi:10.1086/282400. URL https://www.journals.uchicago.edu/doi/10.1086/282400

    work page doi:10.1086/282400 1966

  50. [51]

    Timothy Wootton

    J. Timothy Wootton. Predicting direct and indirect effects: An integrated approach using experiments and path analysis. Ecology, 75: 0 151--165, 1 1994. ISSN 0012-9658. doi:10.2307/1939391. URL https://esajournals.onlinelibrary.wiley.com/doi/10.2307/1939391

    work page doi:10.2307/1939391 1994

  51. [52]

    https://github.com/jvrglr/Unraveling-the-temporal-dependence-of-ecological-interaction-measures

    Codes and datasets. https://github.com/jvrglr/Unraveling-the-temporal-dependence-of-ecological-interaction-measures. Accessed: 2025-06-04

    work page 2025

  52. [53]

    Immanuel M. Bomze. Lotka-volterra equation and replicator dynamics: A two-dimensional classification. Biological Cybernetics, 48: 0 201--211, 10 1983. ISSN 0340-1200. doi:10.1007/BF00318088. URL http://link.springer.com/10.1007/BF00318088

    doi:10.1007/bf00318088 1983

  53. [54]

    Immanuel M. Bomze. Lotka-volterra equation and replicator dynamics: new issues in classification. Biological Cybernetics, 72: 0 447--453, 4 1995. ISSN 0340-1200. doi:10.1007/BF00201420. URL http://link.springer.com/10.1007/BF00201420

    work page doi:10.1007/bf00201420 1995

This paper was first reviewed by deepseek-v4-flash on August 5, 2026.