Large Language Models (LLMs) have demonstrated remarkable capabilities in reasoning and generation, serving as the foundation for advanced persona simulation and Role-Playing Language Agents (RPLAs). However, achieving authentic alignment with human cognitive and behavioral patterns remains a critical challenge for these agents. We present HUMANLLM, a framework treating psychological patterns as interacting causal forces. We construct 244 patterns from ~12,000 academic papers and synthesize 11,359 scenarios where 2-5 patterns reinforce, conflict, or modulate each other, with multi-turn conversations expressing inner thoughts, actions, and dialogue. Our dual-level checklists evaluate both individual pattern fidelity and emergent multi-pattern dynamics, achieving strong human alignment (r=0.91) while revealing that holistic metrics conflate simulation accuracy with social desirability. HUMANLLM-8B outperforms Qwen3-32B on multi-pattern dynamics despite 4x fewer parameters, demonstrating that authentic anthropomorphism requires cognitive modeling--simulating not just what humans do, but the psychological processes generating those behaviors.

翻译：大语言模型（LLMs）在推理与生成方面展现出卓越能力，为高级人物模拟与角色扮演语言智能体（RPLAs）奠定了基础。然而，如何使这些智能体真正与人类认知及行为模式对齐，仍是一个关键挑战。我们提出HUMANLLM框架，将心理模式视为相互作用的因果力。我们从约12,000篇学术论文中构建了244种模式，并合成了11,359个场景，其中2-5种模式相互增强、冲突或调节，通过多轮对话表达内心想法、行动与言语。我们的双层检查表同时评估了单一模式的保真度与多模式交互的涌现动态，实现了高度的人类对齐（r=0.91），同时揭示整体性指标容易混淆模拟准确性与社会期望度。尽管参数规模仅为后者的四分之一，HUMANLLM-8B在多模式动态模拟上仍超越了Qwen3-32B，这表明真实的拟人化需要认知建模——不仅要模拟人类的行为，更要模拟产生这些行为的心理过程。