Social Influence and the Allocation of Scientific Attention in AI Populations

社会影响与人工智能群体中科学注意力的分配

Abstract: AI systems are becoming participants in the evaluation and use of scientific research. They encounter citation counts, download statistics and lists of popular articles developed around human readers, but the collective consequences of these signals for artificial readers remain uncertain.

摘要： 人工智能系统正逐渐成为科学研究评估与应用过程中的参与者。它们会接触到围绕人类读者开发的引用计数、下载统计数据以及热门文章列表，但这些信号对人工智能读者产生的集体影响尚不明确。

This paper adapts the Music Lab design to a market for academic attention. In the first experiment, 1,000 AI agents choose papers from the titles and abstracts of all 114 regular research articles published in the American Economic Review in 2025. The experiment has five independent-choice communities and five social-influence communities, each with 100 sequential agents. Only agents in the social-influence condition observe earlier selections within their community. Agents may select any number of papers.

本文将“音乐实验室”（Music Lab）的设计应用于学术注意力市场。在第一个实验中，1,000 个 AI 智能体从 2025 年发表在《美国经济评论》（American Economic Review）上的全部 114 篇常规研究论文的标题和摘要中进行选择。实验设置了五个独立选择社区和五个社会影响社区，每个社区包含 100 个按顺序操作的智能体。只有处于社会影响条件下的智能体能够观察到其社区内先前的选择。智能体可以选择任意数量的论文。

Social-information communities select 17.2 percent fewer papers per agent, concentrate their choices more heavily, and collectively cover 73 papers, compared with 90 independently. Between-community variation is greater under social information.

与独立选择社区（覆盖 90 篇论文）相比，社会信息社区每个智能体选择的论文数量减少了 17.2%，选择更加集中，且集体覆盖的论文总数为 73 篇。在社会信息条件下，社区间的差异性更大。

In a second experiment with 200 agents across twenty social communities, randomly assigning papers five initial selections raises their subsequent selection rate by 45.55 percentage points (95% CI: 41.20 to 49.90). Choices have modest correspondence with external citations and little correspondence with download counts. The results show how a simple information rule shapes the volume, breadth and distribution of scientific attention in an artificial population.

在第二个实验中，研究人员在 20 个社会社区中部署了 200 个智能体，通过随机为论文分配 5 次初始选择，使其后续被选择率提高了 45.55 个百分点（95% 置信区间：41.20 至 49.90）。实验结果显示，智能体的选择与外部引用计数存在适度相关性，但与下载计数几乎没有相关性。这些结果揭示了一个简单的信息规则是如何塑造人工智能群体中科学注意力的总量、广度及分布的。