发表机构
School of Electrical and Computer Engineering, College of Engineering, University of Tehran; Sharif University of Technology; Missouri University of Science and Technology; Tehran Institute for Advanced Studies, Khatam University（德黑兰大学工程学院电气与计算机工程学院; 谢里夫理工大学; 密苏里科技大学; 哈塔姆大学德黑兰高等研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究提出PsychoAgent架构，分离事实与情感记忆并通过冲突感知控制器整合，在三类冲突场景中其冲突关键记忆检索率优于相关基线，为LLM智能体建模类人冲突效应提供了可检查机制。
AI 中文摘要
类人认知并非仅通过主题相似性选择过往经验：情感重要性与未解决的冲突同样会影响可提取的内容。我们提出PsychoAgent，一种面向大语言模型（LLM）智能体的认知架构，它将事实记忆与情感记忆分离，并通过冲突感知执行控制器整合两者。情感记忆首先通过语义相关性过滤，再通过显著性重新排序，在保持主题适配性的同时，允许情感重要的痕迹进入提示词。在三个受控冲突场景中，完整架构检索到的冲突关键记忆多于语义-情感RAG基线和单一记忆RAG基线（0.933对0.500和0.667），仅存在微小的语义相似性成本。五名盲评者评估了27个输出，经评者内标准化后，完整架构的总体均值最高（+0.22标准差），但校正后的成对差异不显著。一段为期三天的示例轨迹进一步显示了持续的情感、离线记忆重组和选择性记忆加权。这些发现支持情感敏感检索作为一种可检查的机制，用于在LLM智能体中建模类人冲突效应。
英文摘要
Human-like cognition does not select past experience by topical similarity alone: affective significance and unresolved conflict also shape what becomes accessible. We present PsychoAgent, a cognitive architecture for LLM agents that separates factual and affective memory and integrates both through a conflict-aware executive controller. Affective memories are first filtered by semantic relevance and then re-ranked by salience, preserving topical fit while allowing emotionally important traces to enter the prompt. Across three controlled conflict scenarios, the full architecture retrieved more conflict-critical memories than semantic-affective and single-memory RAG baselines (0.933 vs. 0.500 and 0.667), with a small semantic-similarity cost. Five blinded raters evaluated 27 outputs. After within-rater standardization, the full architecture had the highest overall mean (+0.22 SD), but corrected pairwise differences were not significant. A three-day illustrative trace further shows persistent affect, offline memory recombination, and selective memory reweighting. The findings support affect-sensitive retrieval as an inspectable mechanism for modeling human-like conflict effects in LLM agents.
Comments22 pages total (12-page main paper + 10-page supplementary material). Revised after peer review. Accepted for oral presentation and publication in the BICA 2026 proceedings, Springer Lecture Notes in Electrical Engineering (LNEE)
