Han Jin, Battu Balaraju, Romić Ivan, Rahwan Talal, Holme Petter
Department of Computer Science, Aalto University, Espoo, Finland.
New York University Abu Dhabi, Abu Dhabi, United Arab Emirates.
PLoS One. 2025 May 22;20(5):e0320094. doi: 10.1371/journal.pone.0320094. eCollection 2025.
Large language models (LLMs) are increasingly used to model human social behavior, with recent research exploring their ability to simulate social dynamics. Here, we test whether LLMs mirror human behavior in social dilemmas, where individual and collective interests conflict. Humans generally cooperate more than expected in laboratory settings, showing less cooperation in well-mixed populations but more in fixed networks. In contrast, LLMs tend to exhibit greater cooperation in well-mixed settings. This raises a key question: Are LLMs about to emulate human behavior in cooperative dilemmas on networks? In this study, we examine networked interactions where agents repeatedly engage in the Prisoner's Dilemma within both well-mixed and structured network configurations, aiming to identify parallels in cooperative behavior between LLMs and humans. Our findings indicate critical distinctions: while humans tend to cooperate more within structured networks, LLMs display increased cooperation mainly in well-mixed environments, with limited adjustment to networked contexts. Notably, LLM cooperation also varies across model types, illustrating the complexities of replicating human-like social adaptability in artificial agents. These results highlight a crucial gap: LLMs struggle to emulate the nuanced, adaptive social strategies humans deploy in fixed networks. Unlike human participants, LLMs do not alter their cooperative behavior in response to network structures or evolving social contexts, missing the reciprocity norms that humans adaptively employ. This limitation points to a fundamental need in future LLM design-to integrate a deeper comprehension of social norms, enabling more authentic modeling of human-like cooperation and adaptability in networked environments.
大语言模型（LLMs）越来越多地被用于对人类社会行为进行建模，最近的研究探索了它们模拟社会动态的能力。在此，我们测试大语言模型在个体利益与集体利益冲突的社会困境中是否反映人类行为。在实验室环境中，人类通常比预期更愿意合作，在混合良好的群体中合作较少，但在固定网络中合作较多。相比之下，大语言模型在混合良好的环境中往往表现出更多的合作。这就引出了一个关键问题：大语言模型是否即将在网络中的合作困境中模仿人类行为？在本研究中，我们研究了在混合良好和结构化网络配置中，智能体反复进行囚徒困境博弈的网络交互，旨在识别大语言模型和人类在合作行为上的相似之处。我们的研究结果表明了关键差异：虽然人类在结构化网络中往往更愿意合作，但大语言模型主要在混合良好的环境中表现出更多合作，对网络环境的适应性有限。值得注意的是，大语言模型的合作也因模型类型而异，这说明了在人工智能体中复制类人社会适应性的复杂性。这些结果凸显了一个关键差距：大语言模型难以模仿人类在固定网络中部署的细微、适应性强的社会策略。与人类参与者不同，大语言模型不会根据网络结构或不断演变的社会环境改变其合作行为，缺乏人类适应性采用的互惠规范。这一局限性指出了未来大语言模型设计中的一个基本需求——整合对社会规范的更深入理解，以便在网络环境中更真实地模拟类人合作和适应性。
