A bio-inspired evolutionary approach to model fake news diffusion

Opinion dynamics and information integrity are seriously threatened by the proliferation of fake news and disinformation on social media. Like diseases in biological systems, misinformation may spread quickly and mutate when it comes into contact with new hosts. Specifically, complex contagion models such as the Watts threshold model have been used to capture the social reinforcement mechanisms that drive information spread and opinion dynamics. However, these models typically assume static content, overlooking the dynamic nature of fake news that can mutate as individuals engage with and reinterpret the message. Recent studies have highlighted the role of individual thresholds and peer influence in information cascades. However, there is currently no model describing how misinformation evolves in content while it spreads, particularly under varying social pressures and network topologies. Here we propose a bio-inspired evolutionary framework based on the Watts threshold model, integrating mutation dynamics to simulate how misinformation changes as it cascades through a social network. Our agent-based Monte Carlo simulations incorporate a peer pressure coefficient that modulates node activation, potentially pushing individuals above their threshold to adopt the message. This coevolutionary model provides a new perspective on the evolutionary dynamics of misinformation by combining complex contagion and content evolution. It allows for the exploration of parameter spaces and cascade regimes, providing a foundation for detecting critical thresholds and early-warning signals in misinformation diffusion.