发表机构
The Chinese University of Hong Kong; The City University of Hong Kong（香港中文大学; 香港城市大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究通过理论参数化构建4种LLM生成智能体，经1920次模拟验证其可准确复现儿童不同攻击亚型，为心理学相关研究提供了可控框架。
AI 中文摘要
本研究检验基于大语言模型（LLM）的生成智能体在模拟儿童反应性攻击、主动性攻击及共病攻击方面的构念效度。基于社会信息加工模型的理论驱动参数化，构建了4种不同的智能体。在8种社会场景中开展了总计1920次模拟运行，采用混合盲编码流程提取32项量化行为指标。结果显示，相较于非攻击性基线，具有稳健的判别效度，效应量较大；高跨种子可靠性证实，行为差异由潜在心理参数而非模型随机性驱动；质性叙事分析进一步与已有的实证文献一致。总体而言，这些发现表明，经理论参数化的LLM智能体可准确复现不同攻击亚型，为假设生成、干预试点及心理测量工具的优化提供了可扩展、高可控的框架。
英文摘要
This study examines the construct validity of LLM-based generative agents in simulating reactive, proactive, and co-occurring aggression in children. Four distinct agents were instantiated using a theory-driven parameterization grounded in the social information processing model. A total of 1,920 simulation runs were conducted across eight social scenarios, employing a hybrid blind-coding pipeline to extract 32 quantitative behavioral indicators. Results demonstrate robust discriminant validity relative to a non-aggressive baseline, with large effect sizes. High cross-seed reliability confirms that behavioral differentiation is driven by underlying psychological parameters rather than model stochasticity. Qualitative narrative analyses further converged with established empirical literature. Overall, these findings indicate that theory-parameterized LLM agents can accurately reproduce distinct aggression subtypes, offering a scalable, highly controllable framework for hypothesis generation, intervention piloting, and the refinement of psychological measurement tools.
Comments25 pages, 5 tables, 4 figures
