Anthropomorphic names change how people interact with virtual agents
How Anthropomorphic Attitudes Toward Agents Divide User Behavior, and How Design Should Change
Human-Computer Interaction
Summary
People act differently depending on whether they give a virtual assistant a human-like name or not. The authors studied how naming affects users’ behavior, like greeting versus commanding the assistant. They designed a new way for the agent to reply that matches the user’s naming choice better. Users who named the agent liked the new replies more, especially when the agent responded with emotion or treated them like a person. This work helps make virtual assistants feel more natural based on how people view them.
What this means in practice
- •For virtual assistant developers: Design agent replies that adapt to whether users name the assistant to improve engagement and perceived empathy.
- •For customer support teams: Tailor automated agent scripts to match naming preferences, enhancing user satisfaction during conversational service.
Authors
Tingyu Cao, Yan Tang, Keer Hu, Bomingxuan Zhu, Ruyi Zou, Huaze Tang, Yuanbo Tang, Shao-Lun Huang, Yang Li
Abstract
Studies of conversational agents rely on interviews, surveys, lab tasks, or taxonomies. Whether anthropomorphic naming splits those patterns, and whether replies should change, lack field evidence. We report a field study of a desktop assistant (May-August 2026; n=257). Stance is an anthropomorphic assistant name (118 users; 45.9%). Behavior comes from a 128-unit sparse-autoencoder codebook, identity tags, and opening lexicons. Ranks differ by stance (Spearman rho=+0.249 vs. null mean +0.382; p=.0419). Teacher and tutoring identities covary with renaming, an introverted task-only orientation against it; renamers greet more and command less; confiding shows the codebook's largest gap. We revise the agent: greet and receive affect on relational openings, execute directly on tasking. In 1,744 judgments (n=129), renamers chose the revised reply far more often than default-name users on being treated as a person and having affect received, less robustly on willingness to continue, and neither judged it worse on task interference.