Designing artificial agents to detect the motive profile of users in virtual worlds and games

摘要

Understanding the motive profile of users in virtual worlds, computer games or interactive simulation environments promises to complement research on game flow, player experience modelling and game data mining. Traditionally work in these areas examines statistics from users interacting with existing games. This paper takes an alternative approach and explores the design of artificial agents that can aid with detection of a user or player's motive profile. We demonstrate these agents in a decision-making scenario that requires the opponent to trade off social and risk-taking goals. We conclude by identifying a group of four agents that we hypothesize will be able to distinguish between users or players with different motive profiles.