Relevant information as a formalised approach to evaluate game mechanics
We present a new approach to use adaptive AI in the game design process to aid evaluating of game mechanics. During production, this is a crucial task to improve the player satisfaction with a game title. The problem with automated game evaluation via AI is the measurement of values that indicate the quality of the game mechanics. We apply the Information Theory based concept of “Relevant Information” to this problem and argue that there is a relation between enjoyment related gameplay properties and Relevant Information. We also demonstrate, with a simple game implementation, how an adaptive AI can be used to approximate the Relevant Information, and how those measurable numerical values related to certain game design flaws.