pith. sign in

arxiv: 2506.20603 · v2 · pith:A6QXX4VYnew · submitted 2025-06-25 · ⚛️ physics.soc-ph

Endogenous Feedback in Coevolutionary Games Reshapes the Stability of Cooperation

Federico Maria Quetti , Andrea Civilini , Giacomo Frigerio , Silvia Figini , Giacomo Livan , Vito Latora This is my paper

Pith reviewed 2026-05-22 01:06 UTC · model grok-4.3

classification ⚛️ physics.soc-ph
keywords evolutionary game theoryendogenous feedbackcoevolutioncooperationchimera gamespayoff dynamicsdelayed feedbackreplicator dynamics
0
0 comments X

The pith

Endogenous feedback in games creates stable cooperation even where fixed payoffs predict collapse.

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

The paper models evolutionary games in which the payoff matrix itself changes as a direct function of the current population cooperation level. This endogenous link produces new stable states, called chimera games, that support cooperation in settings where standard fixed-game models predict defection or collapse. The same mechanism also shows that time delays in the feedback can replace stability with sustained oscillations, and that nonlinear forms of the feedback can create multiple equilibria whose selection depends on initial conditions. A reader would care because the incentives for cooperation or defection are no longer imposed from outside but generated by the population's own past behavior.

Core claim

By letting the payoff matrix coevolve directly with the instantaneous level of cooperation, the model generates feedback-induced regimes termed chimera games in which stable cooperation arises despite being incompatible with the predictions of standard fixed-game dynamics. Delayed feedback destabilizes these equilibria and produces sustained oscillations, while nonlinear feedback reshapes equilibrium structure and introduces path dependence. The results demonstrate that cooperation can be promoted, suppressed, or destabilized by incentives generated endogenously by the collective behavior itself.

What carries the argument

The endogenous feedback rule that makes the payoff matrix a direct, time-dependent function of the instantaneous population cooperation level.

If this is right

  • Cooperation can persist in games that are dominated by defection when payoffs adjust with the population state.
  • Time delays in feedback convert stable cooperation points into limit-cycle oscillations.
  • Nonlinear feedback creates multiple coexisting equilibria whose selection depends on initial conditions.
  • The same population can generate either pro-cooperation or anti-cooperation incentives depending on the form of its own feedback.

Where Pith is reading between the lines

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

  • The framework could be tested by letting experimental subjects' payoffs update in real time according to observed cooperation rates.
  • Similar feedback structures appear in opinion dynamics and norm evolution where collective behavior alters individual incentives.
  • The model suggests that policy interventions might work by modulating how strongly payoffs respond to observed behavior rather than setting fixed rules.

Load-bearing premise

The payoff matrix changes directly and instantaneously as a function of the current cooperation level without any external variables or delays.

What would settle it

Numerical integration or stability analysis of the replicator dynamics with linear feedback on a prisoner's dilemma payoff matrix that yields a stable interior equilibrium with positive cooperation instead of convergence to zero.

Figures

Figures reproduced from arXiv: 2506.20603 by Andrea Civilini, Federico Maria Quetti, Giacomo Frigerio, Giacomo Livan, Silvia Figini, Vito Latora.

Figure 1
Figure 1. Figure 1: FIG. 1. Stylized representation of the [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. Game dynamics for different [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. Levels of cooperation in time in the PD target scenario, for different [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
read the original abstract

In Evolutionary game theory the payoffs are typically fixed or shaped by external environmental variables. Here, we introduce an endogenous-feedback model in which the game played coevolves directly with the population state: the payoff matrix is a time-dependent function of the level of cooperation. This allows strategic incentives to be continuously modified by the collective behavior they generate. Even in the simplest case of linear and instantaneous feedback, the model reveals feedback-induced regimes, termed chimera games, in which stable cooperation arises despite being incompatible with the predictions of standard fixed-game dynamics. We further show that delayed feedback can destabilize these equilibria and generate sustained oscillations, while nonlinear feedback reshapes equilibrium structure and introduces path dependence. Our results show how cooperation can be promoted, suppressed, or destabilized by incentives generated endogenously by the very same population's collective behavior. We conclude by outlining how our framework connects to real-world systems shaped by endogenous feedback.

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.

Referee Report

0 major / 3 minor

Summary. The paper introduces an endogenous-feedback model in evolutionary game theory where the payoff matrix coevolves directly with the population state x (cooperation level). For the linear instantaneous case this produces the autonomous ODE dx/dt = x(1-x)(a + b x). Equilibria and stability are obtained by standard phase-line analysis, revealing 'chimera games' in which stable high-cooperation states exist that have no counterpart under any fixed-matrix replicator dynamics with the same parameters. The analysis is extended to delayed feedback (which can destabilize equilibria and produce oscillations) and nonlinear feedback (which introduces path dependence). The framework is offered as a way to understand how population-generated incentives can promote, suppress, or destabilize cooperation.

Significance. If the central derivations hold, the work supplies a mathematically tractable extension of replicator dynamics that generates qualitatively new stability regimes for cooperation. The explicit construction of the autonomous ODE, the direct contrast with fixed-game predictions, and the clean treatment of delay and nonlinearity are strengths that could stimulate both theoretical follow-up and empirical tests in systems where behavior modifies the rules of interaction.

minor comments (3)
  1. The abstract introduces the term 'chimera games' without a one-sentence definition; adding a brief parenthetical characterization would improve immediate accessibility for readers outside the subfield.
  2. The transition from the general time-dependent payoff matrix to the specific linear ODE form would benefit from an explicit intermediate equation showing how the payoff difference becomes the affine function a + b x.
  3. A small comparison table (or figure panel) juxtaposing the stability regions of the endogenous model against the corresponding fixed-matrix replicator dynamics would make the central 'incompatible' claim visually immediate.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the careful and accurate summary of our manuscript, for recognizing its potential significance, and for recommending minor revision. We appreciate the positive evaluation of the endogenous-feedback framework, the chimera-game regimes, and the extensions to delay and nonlinearity.

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The manuscript defines an endogenous linear feedback making the payoff difference an affine function of instantaneous cooperation x, which directly yields the autonomous ODE dx/dt = x(1-x)(a + b x) by standard substitution into the replicator equation. Equilibria and stability follow from ordinary phase-line analysis of this ODE; the comparison to fixed-matrix replicator dynamics is constructed explicitly as a contrast rather than a fit. Extensions to delayed and nonlinear feedback are obtained by the same direct substitution without parameter tuning or self-referential closure. No load-bearing step reduces to a fitted input renamed as prediction, a self-citation chain, or an ansatz smuggled via prior work by the same authors. The derivation chain is self-contained against external benchmarks and uses only standard dynamical-systems methods.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The model relies on assumptions about the form of feedback (linear, delayed, nonlinear) and introduces parameters for these effects.

free parameters (1)
  • feedback strength parameter
    Controls how strongly the payoff matrix depends on cooperation level.
axioms (1)
  • domain assumption Payoff matrix is a time-dependent function of cooperation level
    Core modeling choice for endogenous feedback.

pith-pipeline@v0.9.0 · 5700 in / 1049 out tokens · 57645 ms · 2026-05-22T01:06:31.360227+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

52 extracted references · 52 canonical work pages

  1. [1]

    Axelrod and W

    R. Axelrod and W. D. Hamilton, science 211, 1390 (1981)

    work page 1981

  2. [2]

    Maynard Smith, American scientist 64, 41 (1976)

    J. Maynard Smith, American scientist 64, 41 (1976)

    work page 1976

  3. [3]

    P. D. Taylor and L. B. Jonker, Mathematical biosciences 40, 145 (1978)

    work page 1978

  4. [4]

    M. A. Nowak, Evolutionary dynamics: exploring the equations of life (Harvard university press, 2006)

    work page 2006

  5. [5]

    J. M. Smith and G. R. Price, Nature 246, 15 (1973)

    work page 1973

  6. [6]

    J. M. Smith, Evolution and the Theory of Games (Cambridge University Press, 1982)

    work page 1982

  7. [7]

    M. A. Nowak and R. M. May, nature 359, 826 (1992)

    work page 1992

  8. [8]

    J. W. Weibull, Evolutionary game theory , 1st ed. (MIT Press, 2004)

    work page 2004

  9. [9]

    Von Neumann and O

    J. Von Neumann and O. Morgenstern, in Theory of games and economic behavior (Princeton university press, 2007)

    work page 2007

  10. [10]

    R. B. Myerson, Game theory (Harvard university press, 2013)

    work page 2013

  11. [11]

    Livan, Royal society open science 6, 191255 (2019)

    G. Livan, Royal society open science 6, 191255 (2019)

    work page 2019

  12. [12]

    De Domenico, F

    F. De Domenico, F. Caccioli, G. Livan, G. Montagna, and O. Nicrosini, Royal Society Open Science 11, 240177 (2024)

    work page 2024

  13. [13]

    Axelrod and D

    R. Axelrod and D. Dion, Science 242, 1385 (1988)

    work page 1988

  14. [14]

    Doebeli, C

    M. Doebeli, C. Hauert, and T. Killingback, Science 306, 859 (2004)

    work page 2004

  15. [15]

    M. A. Nowak, science 314, 1560 (2006)

    work page 2006

  16. [16]

    M. Perc, J. J. Jordan, D. G. Rand, Z. Wang, S. Boccaletti, and A. Szolnoki, Physics Reports 687, 1 (2017)

    work page 2017

  17. [17]

    Levin, T

    S. Levin, T. Xepapadeas, A.-S. Cr´ epin, J. Norberg, A. De Zeeuw, C. Folke, T. Hughes, K. Arrow, S. Barrett, G. Daily, et al. , Environment and development economics 18, 111 (2013)

    work page 2013

  18. [18]

    Ak¸ cay, The American Naturalist195, 315 (2020)

    E. Ak¸ cay, The American Naturalist195, 315 (2020)

    work page 2020

  19. [19]

    L. S. Shapley, Proceedings of the national academy of sciences 39, 1095 (1953)

    work page 1953

  20. [20]

    Z. Wang, A. Szolnoki, and M. Perc, Physical Review E 90, 032813 (2014)

    work page 2014

  21. [21]

    Hilbe, ˇS

    C. Hilbe, ˇS. ˇSimsa, K. Chatterjee, and M. A. Nowak, Nature 559, 246 (2018). 11

    work page 2018

  22. [22]

    Kleshnina, C

    M. Kleshnina, C. Hilbe, ˇS. ˇSimsa, K. Chatterjee, and M. A. Nowak, Nature Communications 14, 4153 (2023)

    work page 2023

  23. [23]

    G. Wang, Q. Su, L. Wang, and J. B. Plotkin, Proceedings of the National Academy of Sciences 121, e2406993121 (2024)

    work page 2024

  24. [24]

    J. R. Peck and M. W. Feldman, The American Naturalist 127, 209 (1986)

    work page 1986

  25. [25]

    Bowles, J.-K

    S. Bowles, J.-K. Choi, and A. Hopfensitz, Journal of theoretical biology 223, 135 (2003)

    work page 2003

  26. [26]

    A. R. Tilman, S. Levin, and J. R. Watson, Journal of theoretical biology 454, 205 (2018)

    work page 2018

  27. [27]

    Estrela, E

    S. Estrela, E. Libby, J. Van Cleve, F. D´ ebarre, M. Deforet, W. R. Harcombe, J. Pe˜ na, S. P. Brown, and M. E. Hochberg, Trends in ecology & evolution 34, 6 (2019)

    work page 2019

  28. [28]

    M. H. Cortez, S. Patel, and S. J. Schreiber, Proceedings of the Royal Society B 287, 20192298 (2020)

    work page 2020

  29. [29]

    J. S. Weitz, C. Eksin, K. Paarporn, S. P. Brown, and W. C. Ratcliff, Proceedings of the National Academy of Sciences 113, E7518 (2016)

    work page 2016

  30. [30]

    A. R. Tilman, J. B. Plotkin, and E. Ak¸ cay, Nature communications 11, 915 (2020)

    work page 2020

  31. [31]

    Ito and M

    H. Ito and M. Yamamichi, PNAS nexus 3, pgae455 (2024)

    work page 2024

  32. [32]

    Gonz´ alez-Bail´ on, J

    S. Gonz´ alez-Bail´ on, J. Borge-Holthoefer, A. Rivero, and Y. Moreno, Scientific reports1, 1 (2011)

    work page 2011

  33. [33]

    A. R. Kang, J. Park, J. Lee, and H. K. Kim, in Proceedings of the 24th international conference on world wide web (2015) pp. 1079–1084

    work page 2015

  34. [34]

    Kaizoji, Physica A: Statistical Mechanics and its Applications 287, 493 (2000)

    T. Kaizoji, Physica A: Statistical Mechanics and its Applications 287, 493 (2000)

    work page 2000

  35. [35]

    Tanimoto and H

    J. Tanimoto and H. Sagara, BioSystems 90, 105 (2007)

    work page 2007

  36. [36]

    Z. Wang, S. Kokubo, M. Jusup, and J. Tanimoto, Physics of life reviews 14, 1 (2015)

    work page 2015

  37. [37]

    Ito and J

    H. Ito and J. Tanimoto, Royal Society open science 5, 181085 (2018)

    work page 2018

  38. [38]

    Pe˜ na, B

    J. Pe˜ na, B. Wu, and A. Traulsen, J. R. Soc. Interface 13, 20150881 (2016)

    work page 2016

  39. [39]

    Sadekar, A

    O. Sadekar, A. Civilini, J. G´ omez-Garde˜ nes, V. Latora, and F. Battiston, Physical Review E 110, 014306 (2024)

    work page 2024

  40. [40]

    Hauert, F

    C. Hauert, F. Michor, M. A. Nowak, and M. Doebeli, Journal of theoretical biology 239, 195 (2006)

    work page 2006

  41. [41]

    Ak¸ cay and J

    E. Ak¸ cay and J. Roughgarden, Proceedings of the Royal Society B: Biological Sciences 278, 2198 (2011)

    work page 2011

  42. [42]

    Hofbauer and K

    J. Hofbauer and K. Sigmund, Bulletin of the American mathematical society 40, 479 (2003). 12

    work page 2003

  43. [43]

    Hofbauer and K

    J. Hofbauer and K. Sigmund, Evolutionary games and population dynamics (Cambridge university press, 1998)

    work page 1998

  44. [44]

    Szab´ o and C

    G. Szab´ o and C. T˝ oke, Physical Review E58, 69 (1998)

    work page 1998

  45. [45]

    Traulsen, M

    A. Traulsen, M. A. Nowak, and J. M. Pacheco, Physical Review E—Statistical, Nonlinear, and Soft Matter Physics 74, 011909 (2006)

    work page 2006

  46. [46]

    K. Betz, F. Fu, and N. Masuda, Bulletin of Mathematical Biology 86, 84 (2024)

    work page 2024

  47. [47]

    Perc and A

    M. Perc and A. Szolnoki, BioSystems 99, 109 (2010)

    work page 2010

  48. [48]

    Gross and B

    T. Gross and B. Blasius, Journal of the Royal Society Interface 5, 259 (2008)

    work page 2008

  49. [49]

    A. W. Delton, M. M. Krasnow, L. Cosmides, and J. Tooby, Proceedings of the National Academy of Sciences 108, 13335 (2011)

    work page 2011

  50. [50]

    Civilini, O

    A. Civilini, O. Sadekar, F. Battiston, J. G´ omez-Garde˜ nes, and V. Latora, Physical Review Letters 132, 167401 (2024)

    work page 2024

  51. [51]

    Civilini, N

    A. Civilini, N. Anbarci, and V. Latora, Physical Review Letters 127, 268301 (2021)

    work page 2021

  52. [52]

    Sadekar, A

    O. Sadekar, A. Civilini, V. Latora, and F. Battiston, Journal of the Royal Society Interface 22, 20250134 (2025). 13 SUPPLEMENTAL MATERIAL I. SIMULATIONS Here we detail the Monte Carlo simulation protocol performed to validate the analytical results. The standard evolutionary framework adopted for simulations relies on the well mixed assumption and draws ...

    work page 2025