1001 Nights

From Player-Centered to Agent-Centered: A New Paradigm for Interactive Narrative in the Age of Large Language Models

This paper argues that large language model (LLM) driven interactive narrative is shifting the locus of pleasure from player-centred choice toward relationships with semi-autonomous agents that remember, infer, improvise, and sometimes fail.

• The paper argues that LLM-driven interactive narrative shifts pleasure from player-centred choice toward relationships with semi-autonomous agents that remember, infer, improvise, and sometimes fail.
• It proposes the Agent-Centered Narrative Pleasure (ACNP) framework, linking three dimensions: agency distribution, authorship configuration, and pleasure structures.
• Its evidence is exploratory and comparative, based on qualitative coding of five agentic systems (1001 Nights, Generative Agents, AI Town, Instantale, SIMA 2), with no survey data or quantitative sample sizes reported.