Overview

Cognitive science is the study of mental architecture, based on the fundamental assumption that cognition is information processing. In this book we are thinking of mental architectures in terms of three basic questions. Here they are again.

1. In what format does a particular cognitive system carry information?

2. How does that cognitive system transform information?

3. How is the mind organized so that it can function as an information processor?

In Part III we looked in detail at the two most important models of information processing – the physical symbol system hypothesis and the model associated with neurally inspired computing. We turn now to different ways of thinking about the third question.

Our topic in this chapter is the overall organization of the mind. We start thinking about this in section 10.1 by taking a detour through what are known as agent architectures in AI. Agent architectures are blueprints for the design of artificial agents. Artificial agents can be anything from robots to internet bots. Looking at different architectures allows us to see what is distinctive about cognitive systems (as opposed, for example, to reflex systems, or reflex agents). Reflex systems are governed by simple production rules that uniquely determine how the system will behave in a given situation. In contrast, cognitive systems deploy information processing between the input (sensory) systems and the output (effector) systems.