Daniel Jeffery Dobson and Kenneth D. Forbus
Conventional simulators lack explanatory power. Traditional simulators provide the dynamic behavior needed for a gaming world, but do not provide conceptual understandings of the world to players or game A.Is. By creating game engines that have a qualitative understanding of the system they describe, authors can more easily build richer and smarter game agents. This paper outlines why game engines should include more conceptual understandings, suggests some relevant AI technologies, and speculates on how this might be accomplished.