introduction about different types of AI:

artificial intelligence - GOFAI - good old fashioned AI or symbolic AI or traditional AI
In traditional AI the modules that are defined are information processing, or functional. Typically these modules might be a perception module, a planner, a world modeler, a learner etc. The components directly participate in functions such as perceiving, planning, modeling, learning, etc. Intelligent behaviour of the system, such as avoiding obstacles, standing up, controlling gaze, etc., emerges from the interaction of the components

artificial intelligence - connectionism
The central connectionist principle is that mental phenomena can be described by interconnected networks of simple units (wikipedia: connectionism). Good discussion in wikipedia on the connectionism versus computationalism debate


artificial intelligence - behaviour based (Rodney Brooks)
In behaviour based AI the modules that are defined are behaviour producing. Typically these modules might be an obstacle avoidance behaviour, a standing up behaviour, a gaze control behaviour etc. The components directly participate in producing behaviours such as avoding obstacles, standing up, controlling gaze, etc. Intelligent functionality of the system, such as perception, planning, modeling, learning etc. emerges from the interaction of the components.