How Media Competition Fuels the Spread of Misinformation

Abstract

Competition among news sources may encourage some sources to share fake news and misinformation to influence the public. While sharing misinformation may lead to a short-term gain in audience engagement, it may damage the reputation of these sources, resulting in a loss of audience. To understand the rationale behind sharing misinformation, we model the competition as a zero-sum sequential game, where each news source influences individuals based on its credibility-how trustworthy the public perceives it-and the individual's opinion and susceptibility. In this game, news sources can decide whether to share factual information to enhance their credibility or disseminate misinformation for greater immediate attention at the cost of losing credibility. We employ the quantal response equilibrium concept, which accounts for the bounded rationality of human decision-making, allowing for imperfect or probabilistic choices. Our analysis shows that the resulting equilibria for this game reproduce the credibility-bias distribution observed in real-world news sources, with hyper-partisan sources more likely to spread misinformation than centrist ones. It further illustrates that disseminating misinformation can polarize the public. Notably, our model reveals that when one player increases misinformation dissemination, the other player is likely to follow, exacerbating the spread of misinformation. We conclude by discussing potential strategies to mitigate the spread of fake news and promote a more factual and reliable information landscape.

Figures (8)

• Figure 1: a. The credibility with respect to political bias for 223 news sources from Ad-Font websites. The hyper-partisan sources are less credible compared to the centrist ones. b. Normalized media influence intensity with respect to news source credibility and individual susceptibility. Credibility becomes irrelevant for susceptible individuals. c-f. Visualization of how credibility and disseminating misinformation affect individuals with different levels of susceptibility. Green users are resistant to misinformation, while yellow users are susceptible. The opacity of the arrows increases with increased influence. c,e. Sources with high credibility (blue) significantly influence all users, even when disseminating misinformation. d,f. Sources with low credibility (red) have almost no influence over less susceptible individuals; however, disseminating misinformation substantially increases its influence over susceptible individuals.
• Figure 2: Opinion dynamics become polarized when news source strategies for disseminating misinformation reflect real-world policies. a,e. Evolution of opinions among $N=500$ individuals for two different strategies c,g. with $\eta=1,~\xi=2,~\beta_1=3,~\beta_2=2$. b,f. Snapshots of susceptibility and opinion distribution over time steps $t=\{0,100,200\}$. Initially, hyper-partisan sources start with high credibility while frequently disseminating misinformation, thereby gaining significant influence over the public. However, as their integrity wanes, individuals resilient to misinformation shift towards the center. d,h. Final misinformation exposure, $\Gamma$, with respect to opinion bias. Unbiased individuals are exposed to considerably less misinformation in comparison to others.
• Figure 3: Different candidates for the reward function $r(x)$. The players seek to radicalize public opinion in their favor and minimize the influence of the opponent. The exponent of the reward function controls players' preference for radicalization. Larger exponents (red, green) indicate a stronger preference for radicalized public opinion.
• Figure 4: The equilibrium for $M^2$ actions a,b. and nine selected actions c,d.. Both cases entirely polarize public opinion, but the complete action achieves greater separation between peaks. (b) The evolution of opinion when each source can choose any action at each time for the equilibrium. c The evolution of opinion distributions when news source actions are limited to strategies depicted in e. a,d. The final susceptibility-opinion distribution for complete action and limited strategies, respectively. e. The 9 action profiles that players can choose from in each round.
• Figure 5: Whenever a player increases misinformation dissemination below equilibrium, the other player's optimal response causes it to share more misinformation. a. The left player changes its policy to share more misinformation by deviating from equilibrium, while the right player responds optimally. b. The left player avoids disseminating misinformation, allowing the right player to radicalize the community in its favor. c. Both players take action based on equilibrium d. The left player disseminates more misinformation, forcing the other player to lower its credibility accordingly.