Symbolically Scaffolded Play: Designing Role-Sensitive Prompts for Generative NPC Dialogue

TL;DR

This paper explores how symbolic scaffolding in prompts affects NPC dialogue in games, revealing role-dependent effects and proposing a new framework to balance coherence and improvisation in LLM-driven interactions.

Contribution

It introduces Symbolically Scaffolded Play, a novel framework that uses fuzzy symbolic structures to improve dialogue stability without sacrificing improvisation.

Findings

Scaffolded prompts did not significantly improve player experience.

Role-dependent effects observed: NPC interviewer gained stability, suspects lost believability.

Fuzzy-symbolic scaffolding balances coherence and improvisation.

Abstract

Large Language Models (LLMs) promise to transform interactive games by enabling non-player characters (NPCs) to sustain unscripted dialogue. Yet it remains unclear whether constrained prompts actually improve player experience. We investigate this question through The Interview, a voice-based detective game powered by GPT-4o. A within-subjects usability study ($N=10$) compared high-constraint (HCP) and low-constraint (LCP) prompts, revealing no reliable experiential differences beyond sensitivity to technical breakdowns. Guided by these findings, we redesigned the HCP into a hybrid JSON+RAG scaffold and conducted a synthetic evaluation with an LLM judge, positioned as an early-stage complement to usability testing. Results uncovered a novel pattern: scaffolding effects were role-dependent: the Interviewer (quest-giver NPC) gained stability, while suspect NPCs lost improvisational believability. These findings overturn the assumption that tighter constraints inherently enhance play. Extending fuzzy-symbolic scaffolding, we introduce \textit{Symbolically Scaffolded Play}, a framework in which symbolic structures are expressed as fuzzy, numerical boundaries that stabilize coherence where needed while preserving improvisation where surprise sustains engagement.