The SIS Competition Model for Conflicting Rumors
Yu Takiguchi, Koji Nemoto · 2026
Read The SIS Competition Model for Conflicting Rumors by Yu Takiguchi, Koji Nemoto (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/openalex-w7162264588.txt.
Introduction
Yu Takiguchi and Koji Nemoto published a 2026 mathematical study of two conflicting rumors spreading through the same population, such as fake news and its correction. Their model borrows from epidemic theory: people can catch and spread either rumor, lose interest, and become susceptible again. Unusually, each person's opinion feeds back on how easily each rumor infects them.
Working out every steady state and its stability, the authors find a surprising coexistence mechanism. Belief in one rumor can paradoxically help the opposing rumor spread, and a rumor with a lower infection rate can sometimes spread further. A majority advantage also emerges on its own, and a rumor dies out unless its starting share passes a critical threshold, which the authors derive.
Catalog record
| CREATOR | Yu Takiguchi, Koji Nemoto |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 45 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | OpenAlex / Journal of the Physical Society of Japan |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
arXiv:2603.03949v1 [physics.soc-ph] 4 Mar 2026 The SIS Competition Model for Conflicting Rumors Yu Takiguchi and Koji Nemoto March 5, 2026 Abstract We propose an SIS competition model describing the propagation of conflicting rumors, such as fake news and its corrections. This simple model captures the interaction between rumor propagation and opinion dynamics, where rumors drive opinion changes and, conversely, individuals’ opinions determine the infection rates of rumors. We analytically derive all steady states and their stability. These results uncover a novel coexistence mechanism. This coexistence corresponds to a scenario where belief in one rumor (e.g., fake news) paradoxically aids the spread of the opposing rumor (e.g., corrective information). Due to this mechanism, a nontrivial but realistic phenomenon occurs where a lower infection rate actually enhances the spread of a rumor. Furthermore, although the model does not explicitly incorporate majority conformity, a phenomenon where the majority gains an advantage emerges spontaneously. Consequently, even if one rumor has a higher infection rate, it may be eliminated by the other if its initial share fails to exceed a critical threshold. We analytically derive this threshold using the singular perturbation method. 1 Introduction The phenomenon of conflicting rumors spreading simultaneously, such as fake news versus its corrections, support versus criticism of a specific politician, or conspiracy theories versus anti-conspiracy arguments, is frequently observed. Under what conditions can fake news be eradicated by corrective information? What mechanisms allow conflicting rumors to coexist? Addressing these questions is crucial in today’s society characterized by developed social media. Theoretical studies on rumor propagation have been advanced using epidemic models[1, 2, 3]. As early as 1964, Daley and Kendall proposed the SIR model (DK model), consisting of individuals who are ignorant of the rumor (susceptible, S), spreading the rumor (infected, I), or have ceased spreading it (recovered/removed, R)[4]. As a model describing multiple rumors, the 2SI2R model has been proposed[5, 6]. This model assumes a situation where an old rumor A and a new rumor B propagate. Individuals can take five states: ignorant of both rumors (S), spreading either rumor (IA , IB ), or knowing but not spreading either rumor (RA , RB ). Those who know the old rumor A (IA , RA ) can transition to state IB , but transitions from the new rumor to the old one are 1 not considered. In this model, the threshold for global rumor diffusion (i.e., a pandemic) has been derived. Karrer and Newman[7] also studied an SIR model for two interacting diseases, showing that even diseases with slow spreading speeds can cause a pandemic if their transmission probability is high. Trpevski et al.[8] extended the SIS model to investigate the diffusion of two rumors by considering three states: S, IA , and IB . In their model, assuming one rumor is more credible or reputable, they assigned asymmetric transition rules for A and B. They revealed through simulation that two rumors can coexist if the social network meets certain conditions regarding properties such as clustering and degree. The situation where two types of diseases spread on respective transmission networks has been investigated by Sanz et al.[9]. Additionally, Chen et al.[10] have comprehensively reviewed epidemic models describing multiple rumors. Which rumor spreads more easily depends on people’s values and opinions. For example, ardent supporters of a political party will actively spread positive rumors about that party, while ignoring or attacking negative ones. Recently, the ”echo chamber phenomenon”, where people with similar values connect densely, amplifying rumors aligned with their values while excluding critical ones, has become a growing concern[11]. However, previous studies using epidemic models[5, 6, 7, 8, 9, 10] have not represented the aspect where rumor propagation is influenced by each individual’s values. The dynamics of how opinions change through communication have been studied using opinion models[2]. In the voter model[12, 13, 14], the simplest opinion model, a randomly selected individual chooses a neighbor at random and adopts the neighbor’s opinion as their own. Conditions under which only one opinion eventually survives, that is, reaching a consensus, have been actively studied. Recently, more concrete opinion models motivated by social media have been proposed[15, 16, 17]. Epidemic models describe the propagation of rumors, while opinion models describe the evolution of beliefs. In real society, people change their opinions under the influence of rumors they hear, and the transmissibility of rumors changes under the influence of the opinions people support. Therefore, to analyze the competitive diffusion of conflicting rumors, it is necessary to integrate these two models. Thus, we propose an SIS competition model that describes both rumor propagation and opinion dynamics. Using this model, we analytically clarify the conditions under which only one rumor survives. Furthermore, we uncover a previously unreported mechanism that enables the coexistence of rumors. 2 Model We construct a model describing the propagation of rumors regarding two exclusive opinions A and B by extending the SIS model. Let ρr be the fraction of individuals who support opinion r ∈ {A, B}. For simplicity, we assume that every individual supports one of the opinions, i.e., ρA + ρB = 1 holds. Each individual takes either state Sr (r ∈ {A, B}), representing those who support opinion r but do not spread the rumor, or state Ir , representing those who support opinion r and are spreading the rumor. Denoting the fractions of individuals in these states as Sr and Ir , the relation ρr = Sr + Ir holds. An individual in state Sr′ hears the rumor promoting r from an individual in state Ir 2 Figure 1: Schematic of transitions in the SIS competition model. and starts spreading r at a transmission rate βr′ →r . The case r ̸= r′ corresponds to a change of opinion triggered by hearing the rumor. An individual spreading a rumor r transitions to state Sr at a recovery rate γr . This implies that they lose interest and cease spreading the rumor. Fig. 1 summarizes these transitions. This model, which we call the SIS competition model, is described by the following ordinary differential equations: ṠA = −βA→A SA IA − βA→B SA IB + γA IA , I˙A = +βA→A SA IA + βB→A SB IA − γA IA , (1) ṠB = −βB→B SB IB − βB→A SB IA + γB IB , I˙B = +βB→B SB IB + βA→B SA IB − γB IB . (3) (2) (4) The transition from S to I represents the propagation of a rumor, while the transition from r to r′ represents a change in an individual’s opinion. Thus, our model describes both rumor propagation and opinion dynamics. Note that in this model, we assume that a change of opinion results in a transition from Sr to Ir′ . This assumption is based on the idea that if one hears a rumor impactful enough to alter their opinion, they will likely be motivated to spread that rumor to others. Although a term for the transition from Sr to Sr′ could be added, we have confirmed that it does not lead to qualitative differences in the results. B B→A We define the basic reproduction numbers RA , and RA→B of 0 , R0 , R0 0 the SIS competition model as