Thinking humanly: The cognitive modeling approach

If we are going to say that a given program thinks like  a  human, we must have some way of determining how humans think. We need to get  inside  the actual workings of human minds. There are two ways to do this: through introspection trying to catch our own thoughts as they go by and through psychological experiments. Once we have a sufficiently precise theory of the mind, it becomes possible to express the theory as a computer program. If the program's input/output and timing behaviors match corresponding human behaviors, that is evidence that some of the program's mechanisms could also be operating in humans. For example, Allen Newell and Herbert Simon, who developed  GPS,  the  " General Problem Solver "

(Newell and Simon, 1961), were not content to have their program solve problems correctly. They were more concerned with comparing the trace of its reasoning steps to traces of human subjects solving the same problems. The interdisciplinary field of  cognitive science  brings together computer models from AI and experimental techniques from psychology to try to construct precise and testable theories of the workings of the human mind. 

Cognitive science is a fascinating field, worthy of an encyclopedia in itself (Wilson and Keil, 1999). We will not attempt to describe what is known of human cognition in this book. We will occasionally comment on similarities or differences between AI techniques and human cognition. Real cognitive science, however, is necessarily based on experimental investigation of actual humans or animals, and we assume that the reader has access only to a computer for experimentation.

In the early days of AI there was often confusion between the approaches: an author would argue that an algorithm performs well on a task and that it is  therefore  a good model of human performance, or vice versa. Modern authors separate the two kinds of claims; this distinction has allowed both AI and cognitive science to develop more rapidly. The two fields continue to fertilize each other, especially in the areas of vision and natural language. Vision in particular has recently made advances via an integrated approach that considers neurophysiological evidence and computational models.

Thinking rationally: The  " laws of thought "  approach

The Greek philosopher Aristotle was one of the first to attempt to codify  " right thinking, "  that is, irrefutable reasoning processes. His  syllogisms  provided patterns for argument structures that always yielded correct conclusions when given correct premises - for example,  " Socrates is a man; all men are mortal; therefore, Socrates is mortal. "  These laws of thought were  supposed to govern the operation of the mind; their study initiated the field called  logic.

Logicians in the 19th century developed a precise notation for statements about all kinds of things in the world and about the relations among them. (Contrast this with ordinary arithmetic notation, which provides mainly for equality and inequality statements about numbers.) By 1965, programs existed that could, in principle, solve  any  solvable problem described in logical notation. The so - called  logicist  tradition within artificial intelligence hopes to build on such programs to create intelligent systems.

There are two main obstacles to this approach. First, it is not easy to take informal knowledge and state it in the formal terms required by logical notation, particularly when the knowledge is less than  100%  certain. Second, there is a big difference between being able to solve a problem  " in principle "  and doing so in practice. Even problems with just a few dozen facts can exhaust the computational resources of any computer unless it has some guidance as to which reasoning steps to try first. Although both of these obstacles apply to  any  attempt to build computational reasoning systems, they appeared first in the logicist tradition.

Acting rationally: The rational agent approach

An  agent  is just something that acts  (agent  comes from the Latin  agere,  to do). But computer

agents are expected to have other attributes that distinguish them from mere  " programs, "

such as operating under autonomous control, perceiving their environment, persisting over a

prolonged time period, adapting to change, and being capable of taking on another's goals.

RATIONAL AGENT: 

A rational agent  is one that acts so as to achieve the best outcome or, when there is uncertainty, the best expected outcome.

In the  " laws of thought "  approach to  AI,  the emphasis was on correct inferences. Making correct inferences is sometimes  part  of being a rational agent, because one way to act rationally is to reason logically to the conclusion that a given action will achieve ones goals and then to act on that conclusion. On the other hand, correct inference is not  all  of rationality, because there are often situations where there is no provably correct thing to do, yet something must still be done. There are also ways of acting rationally that cannot be said to involve inference. For example, recoiling from a hot stove is a reflex action that is usually more successful than a slower action taken after careful deliberation.

All the skills needed for the Turing Test are there to allow rational actions. Thus, we need the ability to represent knowledge and reason \with it because this enables us to reach good decisions in a wide variety of situations. We need to be able to generate comprehensible sentences in natural language because saying those sentences helps us get by in a complex society. We need learning not just for erudition, but because having a better idea of how the world works enables us to generate more effective strategies for dealing with it. We need visual perception not just because seeing is fun, but  to  get a better idea of what an action might achieve for example, being able to see a tasty morsel helps one to move toward it.

For these reasons, the study of AI as rational agent design has at least two advantages. First, it is more general than the  " laws of thought "  approach, because correct inference is just one of several possible mechanisms for achieving rationality. Second, it is more amenable to scientific development than are approaches based on human behavior or human thought because the standard of rationality is clearly defined and completely general. Human behavior, on the other hand, is well - adapted for one specific environment and is the product, in  part, of a complicated and largely unknown evolutionary process that still is far from producing perfection. This book will therefore concentrate on general principles  of  rational agents and on components for constructing them. We will see that despite the apparent simplicity with which the problem can be stated, an enormous variety of issues come up when we try to solve it.