评估基于人类人格驱动的易受虚假信息影响的模拟
Evaluating the Simulation of Human Personality-Driven Susceptibility to Misinformation with LLMs
AI总结：
本文评估了LLM在模拟基于人格特质的易受虚假信息影响方面的有效性，发现某些人格特质与虚假信息的关联可被复制，但其他情况存在系统偏差，揭示了LLM在人格建模中的潜力与局限。
摘要：
大型语言模型（LLMs）使大规模生成合成行为数据成为可能，为人类实验提供了伦理和低成本的替代方案。然而，此类数据是否能够准确捕捉由人格特质驱动的心理差异，仍是一个开放性问题。我们评估了基于大五人格特征的LLM代理在再现基于人格的易受虚假信息影响的变异方面的能力，重点关注新闻辨别能力，即判断真实标题为真实，虚假标题为虚假的能力。利用已发表的数据集，其中人类参与者具有已知的人格特征并对其标题准确性进行了评分，我们创建了匹配的LLM代理，并将它们的响应与原始人类模式进行比较。某些特质-虚假信息关联，特别是涉及宜人性和尽责性的情况，能够被可靠地复制，而其他情况则出现分歧，揭示了LLM在内化和表达人格方面存在系统性偏差。结果既突显了与人格对齐的LLM在行为模拟中的潜力和局限性，也提供了关于人工代理中认知多样性建模的新见解。
英文摘要：
Large language models (LLMs) make it possible to generate synthetic behavioural data at scale, offering an ethical and low-cost alternative to human experiments. Whether such data can faithfully capture psychological differences driven by personality traits, however, remains an open question. We evaluate the capacity of LLM agents, conditioned on Big-Five profiles, to reproduce personality-based variation in susceptibility to misinformation, focusing on news discernment, the ability to judge true headlines as true and false headlines as false. Leveraging published datasets in which human participants with known personality profiles rated headline accuracy, we create matching LLM agents and compare their responses to the original human patterns. Certain trait-misinformation associations, notably those involving Agreeableness and Conscientiousness, are reliably replicated, whereas others diverge, revealing systematic biases in how LLMs internalize and express personality. The results underscore both the promise and the limits of personality-aligned LLMs for behavioral simulation, and offer new insight into modeling cognitive diversity in artificial agents.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
