The impact of digital media-driven affective polarisation on epidemic dynamics

Satoshi Komuro · 2025

Read The impact of digital media-driven affective polarisation on epidemic dynamics by Satoshi Komuro (2025) 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/arxiv-2503-19756.txt.

Introduction

Satoshi Komuro wrote a 2025 math paper on how online-media-driven dislike between groups can shape the course of an epidemic. Past work studied how news flow affects outbreaks, but emotional splits got less attention. Komuro builds a model that links the split level to spread in one frame. Stronger media influence drives a wider split in the model.

The effect on case counts flips with spread speed, though. At low spread rates, a wider split tracks with fewer infections. At high spread rates, the link reverses and split tracks with more infections. The author offers the frame as a basis for judging such risks in health planning.

Catalog record

CREATORSatoshi Komuro
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE36 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-19756

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

The impact of digital media-driven affective polarisation on epidemic dynamics Satoshi Komuro Abstract While prior studies have examined the influence of information diffusion on epidemic dynamics, the role of affective polarisation—driven by digital media usage—remains less understood. This study introduces a mathematical framework to quantify the interplay between affective polarisation and epidemic spread, revealing contrasting effects depending on transmission rates. The model demonstrates that greater digital media influence leads to increased polarisation. Notably, the results reveal opposing trends: a negative correlation between polarisation and the infected population is observed when transmission rates are low, whereas a positive correlation emerges in high-transmission scenarios. These findings provide a quantitative foundation for assessing how digital media-driven polarisation may exacerbate health crises, informing future public health strategies. Preprint: APS/123-QED I Introduction The interplay between information diffusion and epidemic dynamics has been widely studied, with models capturing how health-related behaviours spread through populations [ 1 , 2 , 3 ] . However, these studies often overlook the role of polarisation—a process wherein individuals align strongly with like-minded peers while dismissing opposing views, particularly in digital media environments. Given the increasing political and social polarisation observed in public health debates, understanding its impact on disease spread is essential. Throughout history, humanity has contended with a wide range of infectious diseases, including smallpox, the plague, and the Spanish flu [ 4 ] . More recently, the COVID-19 pandemic has caused substantial loss of life and widespread societal disruption. A key factor influencing the spread of infectious diseases is individual behaviour [ 1 ] . The extent to which populations adopt preventive measures, such as vaccination, mask-wearing, lockdowns, and social distancing, can significantly alter epidemic dynamics. In recognition of this, various mathematical models have been developed to capture the spread of infectious diseases while accounting for the collective behavioural responses of populations [ 2 , 3 , 5 ] . With the rise of digital media, particularly social networking services (SNS), people are now able to access, gather, spread, and create information instantly. Consequently, both the quantity and quality of information circulating in society have evolved, influencing opinion dynamics . During the COVID-19 pandemic, digital platforms became flooded with information, both accurate and inaccurate. Some of this content is considered misinformation (false or misleading information shared without harmful intent) [ 6 ] , while other content is classified as disinformation (false or misleading information shared with harmful intent) [ 7 ] . The sheer volume of information spread at an unprecedented rate, mirroring the rapid transmission of an infectious disease. This phenomenon was widely recognised as an “infodemic,” a term that garnered significant attention [ 8 ] . As WHO Director-General Tedros Adhanom Ghebreyesus stated at the 56th Munich Security Conference: “We’re not just fighting an epidemic; we’re fighting an infodemic. Fake news spreads faster and more easily than this virus, and is just as dangerous” [ 9 ] . The term “infodemic” and the phrase “go viral,” commonly used to describe rapidly spreading content online, highlight the striking similarities between information dissemination in digital media and epidemic spread. Mathematically, these processes can be modelled in similar ways, leading to growing research interest in the coupled dynamics of information diffusion and epidemic transmission. Many studies in this area apply compartmental models , originally developed for epidemic modelling, to information diffusion as well, facilitating parallel modelling for analytical simplicity. However, while these models capture some aspects of information spread, they often fail to account for the role of polarisation , a key social phenomenon observed in public health debates. During the COVID-19 pandemic, societal polarisation emerged between vaccine proponents and opponents [ 10 ] . Additionally, affective polarisation , where individuals show affinity toward those with similar views and hostility toward those with opposing views, played a critical role [ 11 ] . Political identity has also been linked to vaccine attitudes: in the United States, vaccine hesitancy correlated with support for Donald Trump [ 12 ] , whereas in Japan, it was more closely associated with liberalism [ 13 ] . In Europe, research on conspiracy theorists found no direct link between vaccine scepticism and political orientation but identified correlations with educational background and a negative association between religiosity and education level [ 14 ] . These findings suggest that opinions on vaccination extend beyond mere health concerns, serving as markers of broader political and social identities. This raises an important question: How does polarisation, often amplified by digital media, influence epidemic dynamics? To address this, we build on the framework proposed by Törnberg (2022) [ 15 ] , which demonstrates that when homophily (the tendency to interact with like-minded individuals) is strong, affective polarisation tends to intensify as the probability of engaging in random interactions, akin to exchanges on digital media, increases. These results suggest that greater digital media influence drives affective polarisation. In this paper, we investigate how digital media-driven polarisation influences epidemic dynamics across different scenarios. Specifically, we model the interplay between information diffusion and epidemic spread on multiplex networks, adopting Törnberg’s information diffusion model for the information layer and a compartmental model for the epidemic layer. Our findings provide insights into how digital media can shape public health outcomes, with potential implications for policy interventions aimed at mitigating the effects of misinformation and polarisation. II Related Work This section reviews previous work on mathematical models designed to capture information diffusion, epidemic spreading, and the interactions between these processes. II.1 Compartmental Model A foundational mathematical framework for modelling such processes is the compartmental model, which categorises individuals in a population into distinct states (compartments) and describes transitions between these states. One of the earliest and most influential models in this context is the SIR (Susceptible-Infected-Recovered) model [ 16 ] . Proposed in the 1920s, the SIR model was designed to represent the spread of infectious diseases. Its simplicity and effectiveness have led to its widespread application in epidemiology and beyond. In this model, individuals are classified into three compartments: Susceptible (S), Infected (I), and Recovered (R). The transitions between these states are governed by the rates of infection and recovery, providing a framework for understanding the dynamics of disease spread within a population. Infection spreads when susceptible individuals come into contact with infected individuals, with a transition rate proportional to the infection rate \beta . Infected individuals recover at a constant rate, represented by the recovery rate \gamma . The dynamics of these transitions are defined by the following system of ordinary differential equations: \displaystyle\frac{dS}{dt}=-\beta SI, (1a) \displaystyle\frac{dI}{dt}=\beta SI-\gamma I, (1b) \displaystyle\frac{dR}{dt}=\gamma I. (1c) Although this paper does not delve into the detailed analysis of these equations, analysing them provides insights into key epidemiological properties, such as the peak number of infected individuals and the final number of recovered individuals. In particular, setting S(0)=S_{0} , it is well known that the spread and containment of the disease are determined by the threshold \mathcal{R}_{0}\coloneqq\beta S_{0}/\gamma . This threshold, known as the basic reproduction number , indicates that if \mathcal{R}_{0} is greater than 1, the infection spreads, while if it is 1 or lower, it subsides. The basic reproduction number is a crucial metric for controlling and managing disease outbreaks. Moreover, various compartmental models have been proposed

Read the complete text