There is a problem in writing a book such as this on a still young and controversial field. One wants to present the body of knowledge and theory concerning which there is a high degree of consensus in order to represent most accurately the state of the art. But, when there is much disagreement even about what are the important facts, that is difficult to do. Moreover, a theoretically neutral book presenting either no speculation at all or all sides of every issue can make dull reading. I have thus tried to steer a middle course: Describe the perceptual phenomena that most investigators consider important, discuss the major theories that have been proposed to explain them, and give my own point of view and the evidence supporting it. When I advance a somewhat idiosyncratic theory or speculation, I say so.