Large Language Models (LLMs) acting as artificial agents offer the potential for scalable behavioral research, yet their validity depends on whether LLMs can maintain stable personas across extended conversations. We address this point using a dual-assessment framework measuring both self-reported characteristics and observer-rated persona expression. Across two experiments testing four persona conditions (default, high, moderate, and low ADHD presentations), seven LLMs, and three semantically equivalent persona prompts, we examine between-conversation stability (3,473 conversations) and within-conversation stability (1,370 conversations and 18 turns). Self-reports remain highly stable both between and within conversations. However, observer ratings reveal a tendency for persona expressions to decline during extended conversations. These findings suggest that persona-instructed LLMs produce stable, persona-aligned self-reports, an important prerequisite for behavioral research, while identifying this regression tendency as a boundary condition for multi-agent social simulation.

翻译：作为人工代理的大型语言模型（LLMs）为可扩展的行为研究提供了潜力，但其有效性取决于LLMs能否在长时间对话中保持稳定的人格表征。我们通过一个双重评估框架来解决这一问题，该框架同时测量自我报告特征和观察者评定的人格表达。通过两项实验——测试四种人格条件（默认、高、中、低ADHD表现）、七个LLMs模型和三种语义等效的人格提示——我们检验了对话间稳定性（3,473次对话）和对话内稳定性（1,370次对话，18轮对话）。自我报告在对话间和对话内均保持高度稳定。然而，观察者评分显示，在长时间对话中人格表达存在衰减趋势。这些发现表明，经人格指令调整的LLMs能够产生稳定且与人格一致的自述报告，这是行为研究的重要前提；同时，这种回归趋势的识别为多智能体社会模拟划定了边界条件。