distinguished no longer seem principled enough in processing terms. Finer processing mechanisms are the focus of much current social cognition work. Whether a broadly conceived dual process model can be built from them is unclear. I suspect that no single bifurcation will do justice to the complex processing picture that is currently emerging.
The ambitious early attempt to map broad categories of information (or input) onto broad types of processing also foreshadowed a large theme of contemporary social cognition research. The task of mapping types of social information onto the cognitive processes that utilize them is currently underway in a great deal of research. For example, current research on stereotypes asks which features spontaneously evoke person categories, and when—and which—stereotypic information is processed in more implicit rather than explicit fashion. The emerging story is both more subtle and more complex than would have been predicted even in 1990.
The functional arguments of dual-process model, I have argued, were more implicit. One attraction of these models was their implicit extension of rationality to a variety of social influences and biases. In addition, these models generated an empirical research agenda exploring these functional implications. That research suggests that a variety of social influences on judgment—including consensus, communicator likeability, and some stereotypes—may result at times from perceivers' strategic attempts to use these factors to acquire accurate knowledge, rather than simply reflecting the intrusion of other social motives on judgment and belief. This functional message of early dual-process model was left implicit in much of the early research inspired by those models. To appreciate these implications, therefore, requires reconstructing those dual-process model in functional terms, redefining basic terms, and reinterpreting empirical investigations.
The issue of rationality is just one of the functional issues that has emerged again and again, in different guises, throughout the development of social cognition. Once functional issues are seen as broadly relevant and not restricted to individual differences, one begins to see functional issues lurking behind a great deal of social cognition. On the surface, the 1980s were largely marked by a focus on cognition, with little reference to motivation or functions, but I have attempted to show that functional issues remained implicit in some of that work. By 1990, the interrelationship of motivation and cognition moved to center stage (e.g., Higgins & Sorrentino, 1990; Kruglanski, 1996), and concern with attitude function has reemerged (e.g., Gastil, 1992; Maio & Olson, 1995; Murray, Haddock, & Zanna, 1996; Pratkanis, Breckler, & Greenwald, 1989; Shavitt et al., 1994). As we enter the second century of social psychology, concern with motivation and function seem poised to reemerge as a dominant concern of social cognition.
The views expressed here are solely those of the author and do not represent the National Institute of Justice. The extensive comments of Len Newman, Doug Hazlewood, and Jeff Sherman are gratefully acknowledged.