A Planning-Based Approach to Define and Analyze Inter-Player Dependence in Cooperative Game Playthroughs

Abstract

In cooperative game levels, instances of players depending on each other are frequently an integral and intentional part of the game's design. Particularly in nonlinear games where players can take many approaches to their objectives, understanding how players might rely on each other while playing the game can be challenging, especially when each player may have different preferences about what they want to do. Though many people have an intuitive sense of inter-player dependence, formal definitions to identify dependence in a planning context are not well established. In this paper, I take a planning-based approach to define three types of inter-player dependence (weak dependence, strong dependence, and preferential dependence) to quantify dependence within playthroughs represented by solution plans. I then illustrate the use of an implementation to quantify dependence for potential use by human game designers and automated procedural content generation systems through worked examples. My intent with this work is to provide a method to quantify various dependence types for game levels as well as a strong formal foundation for future work analyzing inter-player dependence in cooperative games.