We should be teaching students how to think; instead we are primarily teaching them what to think. This misdirection of effort in education is the inevitable consequence of an overemphasis on objectively measurable outcomes. In brief, we are more concerned with what answers are given than with how they are produced. But recent developments in cognitive studies have made it possible to conduct systematic scientific research on high-order human cognitive processes. Cognitive process instruction applies these research techniques to problems in education in order to develop instructional programs which teach students how to think. The long-range goal is to improve the level of reasoning typically used by students; some even predict that it will raise the level of human intelligence in general. At the very least, research on human thought processes should be able to help us eliminate some of the more stultifying aspects of education and bring thinking back into the classroom without sacrificing content or substance.
Interest in cognitive process instruction has grown steadily during the past decade. To a certain extent this reflects nothing more than a reaction to the behaviorist excesses with which psychology has burdened both itself and education over the past 70 years. But there are also substantive reasons for the renewed interest in the human thought process. Major developments in the fields of artificial intelligence, cognitive psychology and genetic epistemology now make it possible to study mental processes in a more systematic and sophisticated manner than was possible in the time of William James and John Dewey. Computers have provided a mechanism for simulating thought and a framework for conceptualizing mental processes. A new approach to Cognitive Psychology has overcome old taboos against the analysis of verbalized thought processes and created a wider recognition of the importance of Jean Piaget's work. Although they are very different in style, computer simulation and clinical interview techniques have produced strikingly similar conclusions on the nature of intelligence. One is the important role existing knowledge plays in determining how experience is perceived, and hence in how new knowledge is constructed. This constructivist epistemology suggests that much of our effort in education has been seriously misguided.
One hundred years ago schools strove to develop good habits of mind. Since little was known concerning the detailed structure of these habits they could only be taught indirectly. It was widely believed that subjects such as Latin and mathematics required, and therefore developed, both discipline and logical analysis. Although this model of implicit instruction lost favor after it was "discredited" by Thorndike's transfer experiments, it is still used to justify many practices in higher education. However the impact of this approach has been weakened because students are no longer asked to consider how they think or to