Pith. sign in

REVIEW 3 major objections 5 minor 37 references

Control strategies and trends to equilibrium for kinetic models of opinion dynamics driven by social activity

T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Inactivity drives opinion polarization, and a balanced control strategy steers the population to consensus through a Beta-type equilibrium.

desk verdict A genuinely new kinetic model linking social activity to opinion dynamics, with a clean polarization/consensus dichotomy; the main convergence theorem is real but conditional on an unproved L^q regularity assumption that the paper itself flags. read the letter →

arxiv 2506.07840 v3 pith:UKKAXH3M submitted 2025-06-09 math.AP physics.soc-phq-bio.PE

Andrea Bondesan , Jacopo Borsotti This is my paper
classification math.APphysics.soc-phq-bio.PE MSC 35Q8435B4035K6582C40
keywords opiniondynamicssocialactivitykineticmodelsFokker-Planckequationpolarizationconsensusformationcontrolstrategyrelativeentropy
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 tries to establish a link between social activity and the emergence of extreme opinions, and to prove that a simple control mechanism can break that link. In the kinetic model, each agent has an opinion $w\in[-1,1]$ and an activity level $A$; interaction probability grows with $A$, so active, undecided, and inactive agents follow different opinion dynamics. The model's partial equilibria are Beta distributions whose exponents depend on activity, and the paper argues that this is why inactive agents polarize while active agents reach consensus. The mathematical core is a convergence theorem: with opinion leaders, the marginal opinion distribution of the controlled model converges in $L^1$ to a Beta equilibrium at rate $O(t^{-1/2})$, even though the governing Fokker-Planck equation has time-dependent coefficients. A sympathetic reader would care because the result turns a qualitative story about polarization into a quantitatively checkable relaxation statement.

What carries the argument

The carrying object is the relative-entropy method applied to a time-dependent local equilibrium. For the Fokker-Planck equation (44), the authors define $h_{eq}(t,w)$ as a Beta distribution whose exponents depend on the instantaneous mean opinion $m_w(t)$ (equivalently on parameters $b_1(t),b_2(t)$), so that the flux $Q_p(h)+Q_\ell(h)$ vanishes at $h_{eq}$. The relative entropy $H(h\mid h_{eq})$ has a nonnegative entropy production controlled from below by the squared Hellinger distance, while the explicit time derivative of $h_{eq}$ contributes only an exponentially decaying error because $m_w(t)\to\mu_\ell$ at rate $e^{-\tau t}$. Integrating the differential inequality gives $L^1$ relaxation $O(t^{-1/2})$. A companion $L^q$ estimate, proved under the consensus-formation condition $\tau_q<\tau$, controls the logarithmic boundary terms that otherwise would require the uniform-in-time $L^q$ assumption.

What would settle it

Simulate equation (44) for parameters in the polarization regime (one equilibrium exponent negative) with a compactly supported initial datum, and check whether the solution leaves $L^q(I)$ for every $q>1$ in finite time; if yes, Theorem 2's hypothesis fails. Independently, measure $\|h(t,\cdot)-h_\infty\|_{L^1(I)}$ at long times: a decay slower than $t^{-1/2}$ would refute the stated rate.

Watch

Extended reading notes

Core claim

The central claim is Theorem 2: if $h(t,w)$ solves the Fokker-Planck equation (44) — the reduced opinion equation for the controlled model with both peer interactions and leaders — and if $h\in L^\infty(\mathbb{R}_+,L^q(I))$ for some $q>1$, then $\|h(t,\cdot)-h_\infty\|_{L^1(I)}=O(t^{-1/2})$ as $t\to+\infty$. The limit $h_\infty$ is the Beta distribution (47), whose parameters are fixed by the leaders' mean opinion $\mu_\ell$ and by the balance between compromise and self-thinking. Before this, the paper proves the controlled activity marginal converges to a Dirac delta at $A_c^*>\gamma$, so every agent becomes active in finite time; after that transient the opinion marginal obeys the reduced equation. The authors also show the population's mean opinion converges to $\mu_\ell$ exponentially, and that the local equilibrium $h_{eq}(t,w)$ — a Beta density tracking the current mean opinion — converges to $h_\infty$ exponentially. The stated $L^1$ rate follows by the triangle inequality.

Load-bearing premise

Theorem 2 assumes the opinion distribution never becomes too concentrated — it must stay in some $L^q$ space uniformly for all time — and the paper proves this only under extra consensus-formation constraints, so the convergence result stands or falls on that unproved-in-general bound.

Editorial extensions

If this is right

  • With leaders present, the final opinion distribution depends on the leaders' mean opinion and on the compromise/self-thinking balance, while the initial population opinion influences only the transient.
  • Under the controlled dynamics all agents become active in finite time, so the reduced opinion Fokker-Planck model is valid after a finite transient.
  • If the control is misbalanced ($a_p$ too large), the strategy can decrease the number of active agents instead of increasing it.
  • For the uncontrolled model no global equilibrium exists once the activity marginal can drift, so polarization among inactive agents is a persistent phenomenon.
  • Under the theorem's hypothesis, the time-dependent coefficient does not slow relaxation: the $O(t^{-1/2})$ rate matches the constant-coefficient Wright-Fisher opinion equation.

Reading between the lines

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

  • A testable sociological extension: in online discussion data, sort users by interaction frequency and compare empirical opinion distributions per activity band with the model's Beta-type partial equilibria; the model predicts a sharp threshold activity below which polarization appears.
  • The local-equilibrium technique should transfer to other opinion models in which a slowly drifting moment (such as a moving average opinion) enters the drift of a Fokker-Planck equation, giving explicit rates without requiring constant coefficients.
  • Because agents become active in finite time, an intervention that temporarily raises interaction propensity among inactive agents could permanently move the population into the consensus regime; the control need not be maintained forever.
  • Rerunning the same entropy decomposition on a graphon-structured population, with the leaders' fixed mean replaced by a local mean field, would likely produce analogous local-equilibrium Beta distributions with graphon-weighted exponents.
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

3 major / 5 minor

Summary. The paper introduces a kinetic model for opinion dynamics in which each agent carries an opinion w in I=[-1,1] and a social activity level A in R. Through a quasi-invariant limit of binary interactions it derives Fokker-Planck-type equations (13) for the full distribution, with agent-agent and agent-leader contributions. The authors study macroscopic observables, derive Beta-type partial equilibria whose exponents are used to classify polarization versus consensus, propose a control strategy on the activity variable, and then analyze the large-time behavior of the marginals g(t,A) and h(t,w). The main results are: Proposition 3 and Corollary 2 on the activity marginal converging to a Dirac delta under the control; Proposition 5 proving convergence of the mean opinion to the leaders' mean; Theorem 1 on L^1 convergence of h(t,w) to a Beta equilibrium in the controlled all-active case; Theorem 2, the headline result, giving O(t^{-1/2}) convergence to the equilibrium (47) for the time-dependent Fokker-Planck equation (44) under an a priori L^∞(R+,L^q) assumption; and Propositions 6, 7 and Lemma 2 on nonnegativity, uniqueness and a related analyticity condition.

Significance. The topic is timely and the model is a natural kinetic generalization of Moussaïd's activity-based opinion model. The paper contains genuinely useful elements: the control conditions (29)-(32) are derived, not fitted; the transient activity dynamics are solved explicitly in formulas (39)-(41); and the relative-entropy strategy for a time-dependent-coefficient Wright-Fisher-type equation with Beta equilibria is a plausible extension of [2,19]. The polarization-versus-consensus interpretation based on the exponents of partial equilibria (22)-(24) is also conceptually attractive. However, the central convergence theorem is conditional on an unproved a priori regularity assumption, and the main applied claim about polarization emerging among inactive agents rests on partial equilibria whose attractiveness is not proved. If the regularity issue is repaired and the claims are realigned with what is proved, the paper would be a solid contribution; in its present form the most visible claims outrun the proofs.

major comments (3)
  1. [§4.2, Theorem 2 and inequality (52)] Theorem 2 is stated for any solution h of (44) satisfying h∈L∞(R+,Lq(I)) for some q>1, but this hypothesis is load-bearing and is not established for the parameter range covered by the statement. It is used exactly at inequality (52) to control the time-dependent term T2 in the relative entropy inequality (53). Lemma 1 provides only the exponentially growing bound (62), and Remark 5 merely says the proof can be adapted when condition (63) holds, without stating or proving the adapted theorem. In the polarization regime the equilibrium (47) lies in Lq only for q<q* in (48), and no uniform-in-time Lq bound for the solution is available. Thus Theorem 2 as written does not deliver the advertised strong convergence for the interesting polarization parameters; the authors should either prove the required regularity in a stated parameter regime or reformulate Theorem 2 explicitly as a conditional statement about a class of solutions with that a priori regularity, and move the consensus-regime result proved in Remarks 5 and 6 into a theorem with its hypotheses clearly listed.
  2. [§2.3, equations (22)-(24) and Abstract] The paper's main applied claim, that polarization arises among inactive agents while active agents reach consensus, is based on the signs of the exponents of the partial equilibria (22)-(24). No theorem shows that solutions of the full model approach these partial equilibria, and the later rigorous convergence results concern the marginal h under the all-active reduction, with Theorem 2 itself conditional. Consequently the abstract's statement that the paper shows polarization can arise among low-activity agents is stronger than what is proved. The authors should either prove a statement about the large-time behavior of the full model or explicitly qualify the polarization/consensus discussion as an analysis of formal partial equilibria, with the rigorous results limited to the reduced settings of Section 4.
  3. [§4.2, proof of Theorem 2, equations (60) and (63)] Even apart from the unproved Lq hypothesis, the convergence argument has a gap in the control of the difference (C∞-C(t))/C(t). After equation (60) the authors need to bound (1-w)^{b1(t)-b1∞}-1, and in the delicate case ξ<0 they assert that for t sufficiently large a function of the form log(1-w)(1-w)^{sξe^{-τt}} belongs to Lq′(I). The existence of a uniform threshold T is not proved, nor is it shown that the exponential rate is preserved uniformly in the parameter s involved in the integral representation. Moreover, this part of the proof also uses heq∈L∞(R+,Lq(I)), which again depends on the same unproved regularity assumption or on the additional restrictions (63). The final O(t^{-1/2}) conclusion therefore needs a completed estimate before Theorem 2 can be considered proved.
minor comments (5)
  1. [§3.2 and §4.1, displayed formulas] Several displayed equations contain corrupted or unreadable symbols, in particular the formulas for dρi/dt and dρa/dt in Section 3.2 and the explicit characteristic solutions in the proof of Proposition 4. The manuscript must be typeset cleanly so that the conditions (29)-(32) and the characteristic calculations can be verified by a reader.
  2. [§2.1, equation (11) onward] The quasi-invariant limit is presented as formal, which is acceptable if stated clearly, but the phrase 'additional terms cancel out assuming ⟨|η|^3⟩<∞' leaves several second-order terms and the remainder without a displayed estimate; the authors should either give the details in an appendix or state explicitly that the Fokker-Planck equations are the formal mean-field limits used as the model from that point onward.
  3. [§2, notation for a_•] The notation a_• (or a●) is used before it is defined; the authors should introduce it as a placeholder for a_p and a_ℓ explicitly at its first occurrence in Section 2.
  4. [§3, equation (26)] The control interaction uses λ^c_A in the microscopic update but λ_c in the Fokker-Planck operator (28); the notation should be made consistent.
  5. [§5, Conclusions] There is a grammatical error at 'The proof of our this result involved the use of the relative entropy method'; it should read 'The proof of this result involved...'.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the model, control conditions, and convergence theorems are derived from the stated microscopic rules, and the acknowledged L^q gap is a conditional limitation, not a circularity.

full rationale

The paper is self-contained at the level of its claimed derivations. The controlled Fokker–Planck equation (27) and the marginal equation (44) are obtained by explicit quasi-invariant limits from the binary interactions (2), (3), and the control-strategy conditions (29)–(32) are computed from the signs of the fluxes (C_i, C_a). The convergence results are not fitted: Theorem 1 imports the known O(t^{-1/2}) Wright–Fisher relaxation from external sources [19,20], while Theorem 2 is proved internally via relative entropy estimates, with the local equilibrium h_eq(t,w) constructed from the same coefficients that define (44) and with h_eq→h∞ following from the exponential convergence of m_w(t) to μ_ℓ. The paper explicitly flags that the a priori assumption h∈L∞(R+,L^q(I)) in Theorem 2 is not generally proved, that Lemma 1 only gives an exponentially growing bound, and that recovering the needed integrability requires the extra constraint (63); this is an honest conditional gap, not a reduction of the theorem to its assumptions by construction. The self-citation [4] is used for nonnegativity preservation and as a template for Lemma 1, but the relevant arguments are restated or adapted in the paper and are not load-bearing for the central convergence theorem. No parameter is fitted to a target output, no known result is merely renamed, and no uniqueness claim is imported from the authors' prior work.

Assumptions & free parameters 4 free parameters · 7 assumptions · 0 invented entities

The model introduces several hand-chosen parameters (a_p, λ_c, θ, ν_p, ν_ℓ) that set the regimes for polarization vs consensus and for control effectiveness. None are fitted to data. The main theorems rest on standard domain assumptions plus two ad hoc conditions: the a priori L^q regularity of the solution (Theorem 2) and the extra no-flux conditions (61) used in Lemma 1.

free parameters (4)
  • a_p = a_p = ω_p/2 + ε (Eq. 4)
    Hand-chosen to make the mean activity drift vanish under symmetric initial data (∫_{-M}^M (Āω_p + ε - a_p) dA = 0). This selects the realistic case (ii).
  • λ_c = λ_c (often set to λ_A)
    Control strength in the controlled interaction (26); must satisfy (29) for the strategy to reduce inactive and increase active agents. Chosen by hand, not fitted.
  • θ = θ ∈ (0,1) satisfying (31) when λ_c = λ_A
    Fraction of agents following the controlled interaction. The effectiveness conditions are derived, but the actual value is a free modeling choice.
  • ν_p and ν_ℓ (ratios σ^2/λ) = unspecified positive constants
    These ratios determine whether the Beta equilibrium exponents are negative (polarization) or positive (consensus). No empirical calibration is provided.
assumptions (7)
  • domain assumption Binary interaction rules (2)-(3) with interaction probability P(Ã=1)=Āω+ε define the microscopic dynamics.
    These are postulates of the model; the probability of interaction depends on the activity level via the piecewise-linear function Ā.
  • domain assumption The quasi-invariant limit produces the Fokker-Planck equation (13) via Taylor expansion and vanishing remainder.
    The derivation is formal and relies on standard kinetic theory arguments, assuming finite third moments of the noise and no-flux boundary conditions.
  • domain assumption Restriction to G≡1 and D(w)=√(1-w²) for the partial equilibrium analysis and for the convergence theorems.
    These choices simplify the drift and diffusion; the authors note other choices are possible but do not prove the results for them.
  • domain assumption Initial distribution has compact support in A and w.
    Used in Corollary 1 and Theorem 1 to guarantee equi-bounded support of the activity marginals.
  • ad hoc to paper The marginal solution h(t,w) belongs to L^∞(R+, L^q(I)) for some q>1 (Theorem 2).
    This is the key unproven regularity assumption, used in the relative entropy estimate; the authors discuss it and only prove it in special consensus cases.
  • ad hoc to paper Additional no-flux boundary conditions (61): h(t,±1)=0 and (1-w²)h^{q-1}∂_w h=0 at the boundaries (Lemma 1).
    These boundary conditions are imposed to derive the L^q bound; they are not established for the original equation.
  • standard math m_A(t) is analytical in Propositions 1 and 2.
    Analyticity is assumed to show all derivatives vanish and hence m_A ≡ 0 under the symmetric initial condition.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Control strategies and trends to equilibrium for kinetic models of opinion dynamics driven by social activity." pith.science (2026). https://pith.science/paper/UKKAXH3M

@misc{pith2026250607840,
  author       = {Pith},
  title        = {Pith review of: Control strategies and trends to equilibrium for kinetic models of opinion dynamics driven by social activity},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UKKAXH3M}},
  note         = {Machine review of arXiv:2506.07840}
}
read the original abstract

We introduce new kinetic equations modeling opinion dynamics inside a population of individuals, whose propensity to interact with each other is described by their level of social activity. We show that opinion polarization can arise among agents with a low activity level, while active ones develop a consensus, highlighting the importance of social interactions to prevent the formation of extreme opinions. Moreover, we present a realistic control strategy aimed at reducing the number of inactive agents and increasing the number of socially active ones. At last, we prove several (weak and strong) convergence to equilibrium results for such controlled model. In particular, by considering additional interactions between individuals and opinion leaders capable of steering the average opinion of the population, we use entropy method-like techniques to estimate the relaxation toward equilibrium of solutions to a Fokker-Planck equation with time-dependent coefficients.

Figures

Figures reproduced from arXiv: 2506.07840 by the authors.

Figure 1
Figure 1. Graph of the function A¯ defined by (1). The three different parts of the population are highlighted. where (see [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Partial equilibrium distributions for a fixed time and different values of the activity level. Inactive agents are characterized by opinion polarization, while active ones by consensus formation. In (A) mA¯ = 0 while in (B) mA¯ = −0.2, the other values are ρA¯ = 0.65, ωp = 0.8, ε = 0.05, and νp = 0.04. The function c was so chosen so that the area subtended by the graphs was equal to 1, but other choices would have … view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

37 extracted references · 17 canonical work pages

  1. [1]

    G. Albi, M. Bongini, F. Rossi, and F. Solombrino. Leader formation with mean-field birth and death models. Math. Models Methods Appl. Sci., 29(4):633–679, 2019.https://doi.org/10.1142/S0218202519400025

    work page doi:10.1142/s0218202519400025 2019

  2. [2]

    Auricchio, G

    F. Auricchio, G. Toscani, and M. Zanella. Trends to equilibrium for a nonlocal Fokker–Planck equation.Appl. Math. Lett., 145:108746, 2023.https://doi.org/10.1016/j.aml.2023.108746

    work page arXiv 2023

  3. [3]

    M. Bisi, G. Spiga, and G. Toscani. Kinetic models of conservative economies with wealth redistribution.Commun. Math. Sci., 7(4):901–916, 2009.https://doi.org/10.4310/cms.2009.v7.n4.a5

    work page doi:10.4310/cms.2009.v7.n4.a5 2009

  4. [4]

    Bondesan, G

    A. Bondesan, G. Toscani, and M. Zanella. Kinetic compartmental models driven by opinion dynamics: Vaccine hesitancy and social influence.Math. Mod. Meth. Appl. S., 34(6):1043–1076, 2024.https://doi.org/10.1142/ S0218202524400062

    2024

  5. [5]

    Borgs, J.T

    C. Borgs, J.T. Chayes, H. Cohn, and Y. Zhao. AnL p theory of sparse graph convergence II: LD convergence, quotients and right convergence.Ann. Probab., 46(1):337–396, 2018.https://doi.org/10.1214/17-AOP1187

    doi:10.1214/17-aop1187 2018

  6. [6]

    Borgs, J.T

    C. Borgs, J.T. Chayes, H. Cohn, and Y. Zhao. AnL p theory of sparse graph convergence I: Limits, sparse random graph models, and power law distributions.Trans. Amer. Math. Soc., 372(5):3019–3062, 2019.https: //doi.org/10.1090/tran/7543

    doi:10.1090/tran/7543 2019

  7. [7]

    Boudin and F

    L. Boudin and F. Salvarani. A kinetic approach to the study of opinion formation.M2AN Math. Model. Numer. Anal., 43(3):507–522, 2009.https://doi.org/10.1051/m2an/2009004

    work page arXiv 2009

  8. [8]

    Cercignani.The Boltzmann equation and its applications, volume 67 ofApplied Mathematical Sciences

    C. Cercignani.The Boltzmann equation and its applications, volume 67 ofApplied Mathematical Sciences. Springer-Verlag, New York, 1988.https://doi.org/10.1007/978-1-4612-1039-9

    doi:10.1007/978-1-4612-1039-9 1988

Show all 37 references
  1. [9]

    Chen and D.W

    L. Chen and D.W. Stroock. The fundamental solution to the Wright–Fisher equation.SIAM J. Math. Anal., 42(2):539–567, 2010.https://doi.org/10.1137/090764207

    2010 doi

  2. [10]

    Cordier, L

    S. Cordier, L. Pareschi, and G. Toscani. On a kinetic model for a simple market economy.J. Stat. Phys., 120:253–277, 2005.https://doi.org/10.1007/s10955-005-5456-0

    2005 doi

  3. [11]

    Della Marca, N

    R. Della Marca, N. Loy, and A. Tosin. An SIR-like kinetic model tracking individuals’ viral load.Netw. Heterog. Media, 17(3):467–494, 2022.https://doi.org/10.3934/nhm.2022017

    2022 doi

  4. [12]

    Dimarco, B

    G. Dimarco, B. Perthame, G. Toscani, and M. Zanella. Kinetic models for epidemic dynamics with social heterogeneity.J. Math. Biol., 83(1):Paper No. 4, 32, 2021.https://doi.org/10.1007/s00285-021-01630-1

    2021 doi

  5. [13]

    D¨ uring, J

    B. D¨ uring, J. Franceschi, M.-T. Wolfram, and M. Zanella. Breaking consensus in kinetic opinion formation models on graphons.J. Nonlinear Sci., 34(79), 2024.https://doi.org/10.1007/s00332-024-10060-4

    2024 doi

  6. [14]

    D¨ uring and M.-T

    B. D¨ uring and M.-T. Wolfram. Opinion dynamics: inhomogeneous Boltzmann-type equations modelling opinion leadership and political segregation.Proc. A., 471(2182):20150345, 21, 2015.https://doi.org/10.1098/rspa. 2015.0345

    2015

  7. [15]

    Epstein and R

    C.L. Epstein and R. Mazzeo. Wright–Fisher diffusion in one dimension.SIAM J. Math. Anal., 42(2):568–608, 2010.https://doi.org/10.1137/090766152

    2010 doi

  8. [16]

    Epstein and R

    C.L. Epstein and R. Mazzeo.Degenerate diffusion operators arising in population biology, volume 185 ofAn- nals of Mathematics Studies. Princeton University Press, Princeton, NJ, 2013.https://doi.org/10.1515/ 9781400846108

    2013

  9. [17]

    T. J. Feddersen and W. Pesendorfer. Abstention in elections with asymmetric information and diverse preferences. Am. Polit. Sci. Rev., 93(2):381–398, 1999.https://doi.org/10.2307/2585402

    1999 doi

  10. [18]

    Feltrin and G

    P. Feltrin and G. Ieraci. Facilitating voting and electoral participation in Italy. On some possible measures to contrast involuntary and imposed abstentionism.Rivista di Digital Politics, 1:29–48, 2023.https://www. rivisteweb.it/doi/10.53227/107475

    2023 doi

  11. [19]

    Furioli, A

    G. Furioli, A. Pulvirenti, E. Terraneo, and G. Toscani. Fokker-Planck equations in the modelling of socio- economic phenomena.Math. Models Methods Appl. Sci., 27(1):115–158, 2017.https://doi.org/10.1142/ S0218202517400048

    2017

  12. [20]

    Furioli, A

    G. Furioli, A. Pulvirenti, E. Terraneo, and G. Toscani. Wright–Fisher-type equations for opinion formation, large time behavior and weighted logarithmic-Sobolev inequalities.Ann. Inst. H. Poincar´ e C Anal. Non Lin´ eaire, 36(7):2065–2082, 2019.https://doi.org/10.1016/j.anihpc...

    2019 doi

  13. [21]

    Furioli, A

    G. Furioli, A. Pulvirenti, E. Terraneo, and G. Toscani. Non-Maxwellian kinetic equations modeling the dynam- ics of wealth distribution.Math. Models Methods Appl. Sci., 30(4):685–725, 2020.https://doi.org/10.1142/ S0218202520400023

    2020

  14. [22]

    Galam, Y

    S. Galam, Y. Gefen (Feigenblat), and Y. Shapir. Sociophysics: A new approach of sociological collective behaviour: I. Mean-behaviour description of a strike.Journal of Mathematical Sociology, 9(1):1–13, 1982. https://doi.org/10.1080/0022250X.1982.9989929

    1982

  15. [23]

    Glasscock

    D. Glasscock. What is . . . a graphon?Notices Amer. Math. Soc., 62(1):46–48, 2015.http://dx.doi.org/10. 1090/noti1204

    2015

  16. [24]

    Gualandi and G

    S. Gualandi and G. Toscani. Human behavior and lognormal distribution. A kinetic description.Math. Models Methods Appl. Sci., 29(4):717–753, 2019.https://doi.org/10.1142/S0218202519400049. 36 A. BONDESAN AND J. BORSOTTI

    2019 doi

  17. [25]

    Legnante and P

    G. Legnante and P. Segatti. Intermittent abstentionism and multi-level mobilisation in Italy.Modern Italy, 14(2):167–181, 2009.https://doi.org/10.1080/13532940902871802

    2009 doi

  18. [26]

    Moussa ¨ ıd

    M. Moussa ¨ ıd. Opinion formation and the collective dynamics of risk perception.PLOS ONE, 8(12):e84592, 2013. https://doi.org/10.1371/journal.pone.0084592

    2013 doi

  19. [27]

    Ochrombel

    R. Ochrombel. Simulation of Sznajd sociophysics model with convincing single opinions.Int. J. Mod. Physics. C, 12(7):1091, 2001.https://doi.org/10.1142/S0129183101002346

    2001 doi

  20. [28]

    Pareschi and G

    L. Pareschi and G. Toscani.Interacting Multiagent Systems: Kinetic Equations and Monte Carlo Methods. Oxford University Press, 2013.https://global.oup.com/academic/product/ interacting-multiagent-systems-9780199655465?cc=it&lang=en&

    2013

  21. [29]

    Pareschi and G

    L. Pareschi and G. Toscani. Wealth distribution and collective knowledge: a Boltzmann approach.Philos. Trans. R. Soc. Lond. Ser. A Math. Phys. Eng. Sci., 372:20130396, 2014.https://doi.org/10.1098/rsta.2013.0396

    2014

  22. [30]

    Pareschi, G

    L. Pareschi, G. Toscani, A. Tosin, and M. Zanella. Hydrodynamic models of preference formation in multi-agent societies.J. Nonlinear Sci., 29(6):2761–2796, 2019.https://doi.org/10.1007/s00332-019-09558-z

    2019 doi

  23. [31]

    Stauffer

    D. Stauffer. Percolation and galam theory of minority opinion spreading.Int. J. Mod. Physics. C, 13(7):975–977, 2002.https://doi.org/10.1142/S0129183102003735

    2002 doi

  24. [32]

    Stauffer and P

    D. Stauffer and P. Oliveira. Persistence of opinion in the Sznajd consensus model: Computer simulation.Eur. Phys. J. B, 30:587–592, 2002.https://doi.org/10.1140/epjb/e2002-00418-0

    2002 doi

  25. [33]

    Sznajd-Weron and J

    K. Sznajd-Weron and J. Sznajd. Opinion evolution in closed community.Int. J. Mod. Physics. C, 11(6):1157– 1165, 2000.https://doi.org/10.1142/S0129183100000936

    2000 doi

  26. [34]

    G. Toscani. Kinetic models of opinion formation.Commun. Math. Sci., 4(3):481–496, 2006.https://doi.org/ 10.4310/cms.2006.v4.n3.a1

    2006 doi

  27. [35]

    Ulsh¨ ofer, F

    C. Ulsh¨ ofer, F. Renkewitz, and T. Betsch. The influence of narrative v. statistical information on perceiving vaccination risks.Med. Decis. Making, 31(5):742–753, 2011.https://doi.org/10.1177/0272989X11400419

    2011 doi

  28. [36]

    Weidlich

    W. Weidlich. Sociodynamics––a systematic approach to mathematical modelling in the social sciences.Chaos Solitons Fract., 18(3):431–437, 2003.https://doi.org/10.1016/S0960-0779(02)00666-5

    2003 doi

  29. [37]

    M. Zanella. Kinetic models for epidemic dynamics in the presence of opinion polarization.Bull. Math. Biol., 85(5):1–29, 2023.https://doi.org/10.1007/s11538-023-01147-2. Andrea Bondesan Department of Mathematical, Physical and Computer Sciences University of Parma Parco Area de...

    2023 doi

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.