Several theoretical views of automaticity are discussed. Most of these suggest that automaticity should be diagnosed by looking at the presence of features such as unintentional, uncontrolled/uncontrollable, goal independent, autonomous, purely stimulus driven, unconscious, efficient, and fast. Contemporary views further suggest that these features should be investigated separately. The authors examine whether features of automaticity can be disentangled on a conceptual level, because only then is the separate investigation of them worth the effort. They conclude that the conceptual analysis of features is to a large extent feasible. Not all researchers agree with this position, however. The authors show that assumptions of overlap among features are determined by the other researchers' views of automaticity and by the models they endorse for information processing in general. : automatic, unintentional, uncontrolled, autonomous, unconscious Automaticity is a concept with a long-standing history in psychology (e.g., James, 1890;Wundt, 1903). It has been invoked in domains as diverse as perception (MacLeod, 1991), memory (Jacoby, 1991), social cognition (Wegner & Bargh, 1998), learning (Cleeremans & Jiménez, 2002), motivation (Carver & Scheier, 2002), and emotion (Scherer, 1993). Despite its central nature, there is no consensus about what automaticity means. The aim of this article is to provide an in-depth analysis of the concept and, in particular, the features that have been subsumed under the term. We also discuss implications of this analysis for future research.
KeywordsIn the first section, we consider several contrasting views of automaticity. This overview reveals that most scholars analyze automaticity in terms of one or more features. The implication is that automaticity can be diagnosed by looking for the presence of these features in performance or processes. 1 Contemporary approaches to automaticity, moreover, have begun to argue for the separate investigation of automaticity features. In the second section, we discuss the most essential features in greater detail. One purpose is to examine whether they can be separated on a conceptual level. We analyze the features to a point at which the overlap among them is minimized. Another purpose is to show that assumptions of overlap are pervasive in the research literature despite the recent claim that individual features should be investigated separately. These assumptions arise from authors' specific views of automaticity, as well as from the specific underlying information-processing model that they endorse.
Approaches to AutomaticityJames (1890), Jastrow (1906), and Wundt (1896/1897, 1903 offered some of the earliest descriptions of automaticity, and many of their ideas have reemerged in contemporary accounts. The views presented below are also based on early studies of skill development (Bryan & Harter, 1899) and early dual-task studies (Solomons & Stein, 1896; see review by Shiffrin, 1988).
Automaticity as Processing With No or Minimal AttentionC...