发表机构
Graduate School of Informatics, Kyoto University; Equmenopolis, Inc.; Waseda University; School of Informatics, Kyoto University（京都大学信息学研究科; Equmenopolis公司; 早稻田大学; 京都大学信息学系）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究通过纵向多模态研究，探讨对话式人工智能系统中交互如何发展为关系。核心方法是让参与者对五个关系构建要素评分，发现对话质量影响当下愉悦感，感知记忆受关系制约，关系有崩溃和激增等转折点，揭示了人机关系建立的方式。
AI 中文摘要
随着对话式人工智能系统被设计用于重复使用，一个核心问题是一系列交互如何发展成为一种关系。我们展示了一项针对记忆增强对话代理的纵向多模态研究（24名参与者，每人进行10次会话），参与者在每次会话后对五个关系构建要素——熟悉度、自我表露、感知记忆、对话质量和愉悦感进行评分。出现了两种互补动态。首先，对话质量强烈影响当下会话的愉悦感，但不会跨会话延续，而感知记忆受到关系的制约——由先前的关系状态预测，而非仅反映系统能力——并通过后续的自我表露间接影响后来的愉悦感。其次，关系由离散的转折点——崩溃和激增——所标志，这些在多模态行为中部分可追溯，并开启不同的干预窗口：激增在当下行为上更易被检测到，愉悦感激增比愉悦感崩溃恢复得更可靠，一些崩溃从特定个体的行为漂移中能更好地被预测，而非在其发生后才被检测到。这些发现共同表明，纵向人机关系是通过缓慢积累和突然的转折点建立起来的。
英文摘要
As conversational AI systems are designed for repeated use, a central question is how a series of interactions becomes a relationship. We present a longitudinal multimodal study of a memory-augmented conversational agent (24 participants x 10 sessions), in which participants rated five relational constructs -- familiarity, self-disclosure, perceived memory, conversational quality, and enjoyment -- after each session. Two complementary dynamics emerge. First, conversational quality strongly shapes how enjoyable a session feels in the moment but does not carry forward across sessions, whereas perceived memory is relationally conditioned -- predicted by prior relational state rather than reflecting system capability alone -- and it shapes later enjoyment indirectly, via subsequent self-disclosure. Second, relationships are punctuated by discrete turning points -- crashes and surges -- that are partially traceable in multimodal behavior and open different intervention windows: surges are more behaviorally detectable in the moment, enjoyment surges persist more reliably than enjoyment crashes recover, and some crashes are better forecast from person-specific behavioral drift than detected after they have already occurred. Together, the findings suggest that longitudinal human-AI relationships are built through both slow accumulation and abrupt turning points.
Comments15 pages, 3 figures. Accepted to ICMI 2026 (International Conference on Multimodal Interaction), October 5-9, 2026, Napoli, Italy
