Simulated Misinformation Susceptibility (SMISTS): Enhancing Misinformation Research with Large Language Model Simulations

Psychological inoculation, a strategy to build resistance against persuasive misinformation, has been shown to reduce its spread and adverse effects. Although these inoculations are effective, the design and optimization of them typically require substantial financial and human resources. To address these challenges, this work introduces Simulated Misinformation Susceptibility Test (SMIST), leveraging Large Language Models (LLMs) to simulate participant responses in misinformation studies. SMIST employs a life experience-driven simulation methodology, which accounts for various aspects of participants' backgrounds, to mitigate common issues of caricatures and stereotypes in LLM simulations and enhance response diversity. Our extensive experimentation demonstrates that SMIST, utilizing GPT-4 as the backend model, yields results that align closely with those obtained from humansubject studies in misinformation susceptibility. This alignment suggests that LLMs can effectively serve as proxies in evaluating the impact of psychological inoculations. Further, SMIST can be applied to emerging and anticipated misinformation scenarios without harming human participants, thereby expanding the scope of misinformation research.