Using Decision Theory for Player Analysis in Pacman

Benjamin Cowley, Darryl Charles, Michaela M Black, Ray J Hickey

    Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review


    Computer and videogames have been described using several formal systems – in this paper we consider them as Information Systems. In particular, we use a Decision Theoretic approach to model players dynamically in real- time Pacman (Namco 1980). The method described provides low-level in-game data capture which can provide a simple metric of challenge and player skill, which are key components in measuring the optimality of player experience based on Flow theory. Our approach is based on the calculation of optimal choices available to a player based on key utilities for a given game state. Our hypothesis is that observing a player’s deviation from an expected path can reveal their play preferences and skill, and help enhance our player models. Improved models will then enable in-game adaptation, to better suit individual players. In this paper we outline the basic principle of this approach and discuss the results of our first experiment. Keywords:
    Original languageEnglish
    Title of host publicationProceedings of Simulation of Adaptive Behavior 2006 : Workshop on Adaptive Approaches to Optimizing Player Satisfaction in Computer and Physical Games
    EditorsGeorgios Yannakakis, John Hallam
    Number of pages10
    Place of PublicationRome
    Publication date2006
    Publication statusPublished - 2006
    MoE publication typeA4 Article in conference proceedings

    Cite this