Telling Interactive Player-specific Stories and Planning for It: ASD + PaSSAGE = PAST

Abstract

Artificial Intelligence (AI) techniques have been used to improve overall player satisfaction in video games. A thriving area of such research is interactive AI-assisted story-telling. One such line of research explored gains of automatically fitting the story to each individual player during the game via player modeling. Another research line used AI planning techniques to create contingency stories at design time. In this paper we propose a principled way to combine these two lines of research. We describe a system that uses a player model both at the design time and at the play time to generate and select stories fitting a specific player. We implement an early prototype and present it in the paper.