Abstract

This chapter operationalises the conceptual framework of Computational Infodemiology through a suite of ten structured simulations designed to model how belief systems emerge, evolve, and destabilise within digital ecosystems. Moving from theory to experimentation, each simulation represents a distinct configuration of agent behaviour, network topology, and informational pressure, allowing for the empirical investigation of belief dynamics under variable conditions. Key model parameters include agent population size, misinformation spread probability, fact-checking success rate, belief instability, exploration–exploitation behaviour (epsilon), and temporal evolution across simulation steps. These parameters enable the simulation of diverse environments ranging from stable networks with limited falsehood diffusion to adversarial systems marked by chaotic misinformation cascades. Scenarios such as the Baseline, Deterministic Behaviour, High Misinformation Pressure, Sceptic Drop-Off, and Mass Chaos offer insight into belief convergence, cognitive polarisation, and epistemic resilience. Others like Tug-of-War and Community Defence test competitive equilibrium and cooperative resistance to misinformation. All simulations employ a reinforcement learning schema based on the Upper Confidence Bound (UCB) algorithm, ensuring agents dynamically balance exploitative stability with adaptive learning. This modelling suite offers a powerful testbed for infodemiological interventions, policy experimentation, and the analysis of digital belief ecologies.