Skip to content
Book Open access

Typing vs. Dialogue Trees: Do Players Actually Want Open-Ended LLM Conversations in Games?

Nov 2026 · Companion Proceedings of the Annual Symposium on Computer-Human Interaction in Play · 0 citations · 23 references

Abstract

Large Language Models (LLMs) offer the potential to replace static dialogue trees with open-ended, generative interactions in video games. However, the user experience of this transition remains under-researched. We present Project Bard, a custom 2D RPG that directly compares player interactions with LLM-based NPCs (free-text typing) against traditional pre-authored NPCs (dialogue trees) in a preliminary within-subjects study (N = 32). While telemetry shows players spent three times as long interacting with generative NPCs, immersion metrics show no significant difference, and qualitative analysis reveals that this extended time was often spent boundary-testing the system as a mechanical sandbox. Furthermore, players were evenly split in their system preference, highlighting a trade-off between the appeal of conversational agency and the usability friction of switching between gameplay and typing, as well as issues of AI hallucinations. These findings suggest that generative dialogue is not necessarily a universal upgrade, requiring careful alignment of AI mechanics with specific player preferences.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.