AI 中文总结
该研究通过语言流畅性数据，运用基于轨迹的自然语言处理指标，比较人类与三个大语言模型的语义搜索动态，量化相关维度，发现人类语义搜索在局部利用和全局探索间有独特平衡，当前模型架构无法重现。
AI 中文摘要
语义记忆检索可被概念化为在概念空间中的导航。我们使用语言流畅性数据比较了人类与三个大语言模型（GPT-4o、Gemini-2.5-Pro、Claude-Sonnet-4.5）之间的语义搜索动态。通过将基于轨迹的自然语言处理指标应用于82名人类参与者生成的项目以及大语言模型在八个温度设置下的输出，我们量化了三个互补维度：熵（步长可预测性）、到下一个的距离（连续语义步）和到质心的距离（全局离散度）。人类表现出比所有大语言模型更高的熵、更大的语义步和更广泛的离散度，表明搜索更具变化性和探索性。温度调整仅产生部分对齐，因为在特定设置下人类和大语言模型的个别指标相匹配，但没有一种配置能重现完整的人类特征（在所有维度上）。这些发现表明，人类语义搜索在局部利用和全局探索之间实现了一种独特的平衡，而当前的模型架构无法重现。
英文摘要
Semantic memory retrieval can be conceptualized as navigation through conceptual space. We compared semantic search dynamics between humans and three large language models (GPT-4o, Gemini-2.5-Pro, Claude-Sonnet-4.5) using verbal fluency data. By applying trajectory-based NLP metrics to the items generated by 82 human participants and LLM output across eight temperature settings, we quantified three complementary dimensions: entropy (step size predictability), distance to next (successive semantic steps), and distance to centroid (global dispersion). Humans exhibited higher entropy, larger semantic steps and broader dispersion than all LLMs, indicating more variable and exploratory search. Temperature tuning produced only partial alignments, as individual metrics matched between humans and LLMs at specific settings, but no configuration reproduced the complete human profile (in all dimensions). These findings suggest that human semantic search implements a distinctive balance between local exploitation and global exploration that current model architectures fail to reproduce.
CommentsCogsci paper 2026
Journal refProceedings of the Annual Meeting of the Cognitive Science Society, 48(0), 2026
