Learning group behavior in games using Cultural Algorithms and the land bridge simulation example

The L-CAPS is a functional interface for learning group behavior using Cultural Algorithms in the land bridge simulation example. It enables the Cultural Algorithms process to implicitly communicate with, modify and evaluate, autonomous game agents restricted to an external virtual world. Such a mechanism has not been explored to the same extent that it is implemented in L-CAPS. The L-CAPS succeeds in extending the work presented by Kinnaird-Heether, and is able to examine the viability of using CA in learning group behavior, as well as the ability for CA to be extended to more general game designs.;The results can be equated to an inaugural phase for future game designs that might rely on the CA process for agent learning and AI. However, to do so, the L-CAPS would need to produce a framework relatable to modern game designs. Using the XNA API, it has done so, and therefore, has opened the way for future explorations in using Cultural Algorithms to manipulate mass amounts of external game agents. In future work, CA could become the entire CA subsystem that is manipulating world agents from their expressive interactions that define their social fabric.