Social psychology is presently dominated by cognitive theories that emphasize the importance of personal beliefs and intellective processes as the immediate determinants of behavior.The present paper explores two areas of.research within this tradition :(1) beliefs about the external world, and (2) beliefs about the self. The paper concludes with a brief critique of the cognitive approach to social psychology. Social psychology, along with many other areas of behavioral research, is presently dominated by the cognitive point of view. In preparing to write this paper it was therefore difficult to determine what aspects of social psychology could legitimately be regarded as &dquo;noncognitive,&dquo; and hence irrelevant to the topic under consideration. I have therefore taken the liberty of covering the field somewhat selectively, so that I might have the opportunity to discuss some continuing themes that strike me as particularly interesting and noteworthy.For the social psychologist, the cognitive approach normally implies an emphasis on personal beliefs and hypotheses as the immediate determinants of behavior.Cognitive social psychologists thus focus on what Lewin called the perceived world, including the inferences and illusions that derive from cognitive activity, as opposed to the unembellished objective reality of the physical and social environment. The dominant tone emphasizes man's rational, intellective processes and downgrades the importance of irrational and self-serving motives, mindless conformity, and mechanistic learning principles. Thus, a number of investigators have recently focussed attention on the number and the quality of the arguments that support a given position, as important determinants of attitude change (Burnstein & Vinokur, 1977; Eagly, 1974;Greenwald, 1968). On the other hand, man's cognitive capacities are finite, and these limits should perhaps be more explicitly recognized by social psychologists (Dawes, 1975). For example, Nisbett and Wilson (1977) have shown that people have a limited capacity to report on their own cognitive processes, and as we shall see below, people's intuitive judgments are often unaffected by information that should (logically) be influential. ' Before getting into the main body of our discussion, we should note that the social psychologist's expression of the cognitive viewpoint is frequently rather different from the cognitive models that have been proposed in experimental psychology. For one thing, the cognitive social psychologist has often been concerned with the substantive inferences and generalizations we derive from our past and present experiences, generalizations that can be expressed in simple verbal terms (e.g., if Bob and .