Modeling Duelling Contagions of True and False Information in the Face of Inherent Individual biases

This study investigates how competing narratives—specifically true versus manipulative information—spread through social networks. While traditional models often treat information diffusion as a simple contagion, this paper introduces the Competing Information – Social Reinforcement and Willingness (CI-SRW) framework. This model integrates network dynamics with two core psychological theories: complex contagion (where repeated exposure reinforces belief) and the spiral of silence (where individuals withhold opinions if they perceive themselves to be in the minority).

The authors use agent-based modeling to simulate how these cognitive biases, such as optimism bias and confirmation bias, interact with network structures. They validate their model using data from large-scale online behavioral experiments where human participants made decisions under uncertainty, allowing the researchers to test how initial seeding and individual biases shape collective outcomes.

The simulations reveal that the initial distribution of information is a primary determinant of long-term diffusion dynamics. When manipulative narratives are introduced early, they tend to dominate the network, especially in populations with strong optimism bias. The researchers found that individuals are more likely to incorporate favorable information than unfavorable information, which creates a systematic bias that hinders the adoption of accurate, albeit potentially alarming, information.

Crucially, the study demonstrates that the spiral of silence creates a feedback loop where the perceived opinion climate suppresses the expression of accurate information. However, the authors identify a strategic intervention: seeding the network with agents who have lower self-censorship. These individuals act as a counter-force, breaking the spiral of silence and allowing accurate information to gain traction, ultimately leading to a more well-informed populace.

Understanding the interplay between cognitive biases and network diffusion is essential for designing policies that promote healthier public discourse. By identifying that the "who" and "when" of information seeding are critical, this research provides a theoretical basis for interventions that can mitigate the impact of disinformation without relying solely on content moderation. It highlights that fostering environments where individuals feel comfortable expressing accurate but unpopular views is a powerful tool for social cohesion.