Experimental and Clinical Psychopharmacology Copyright 2000 by the American Psychological Association, Inc. 2000, Vol. 8, No. 4, 518-523 1064-1297/00/$5.00 DOI: 10.1037//1064-1297.8.4.518 How Persistent Is the Effect of Smoking Urges on Cognitive Performance? Rolf A. Zwaan, Robert A. Stanfield, and Carol J. Madden Florida State University Several studies have provided empirical support for S. T. Tiffany's (1990) hypothesis that drug urges interfere with cognitive performance. The authors examined the persistence of this effect. Results from an experiment involving 48 smokers and 46 nonsmokers, using a paradigm developed by R. A. Zwaan and T. P. Truitt (1998), suggest that the effect of smoking urges in cognitive performance dissipates over time. The implications of this finding for cognitive theories of drug urges and for future research on the effects of smoking urges are discussed. Smokers are often confined to smoke-free environments, for example, classrooms and office buildings, in which they are to perform complex cognitive tasks. Given that smokers are unable to respond to their smoking urges in such envi- ronments, there is the potential of a conflict between the urge to smoke and the task at hand. Indeed, researchers have hypothesized that smoking urges have a detrimental effect on cognitive performance (see, e.g., Tiffany, 1990). There is initial evidence for this hypothesis in the literature. For example, both Sayette and Hufford (1994) and Cepeda- Benito and Tiffany (1996) have shown that smoking urges bring about increases in simple tone-reaction times relative to a no-urge condition. More recently, Zwaan and Truitt (1998) showed that smoking urges interfere with higher cognitive processes, in this case, language comprehension. In Tiffany's theory, findings such as these are explained by assuming that smokers have smoking-related action sche- mata stored in long-term memory. These scripts can be activated by external cues, such as the smell of smoke, or by internal cues, such as thoughts about smoking. Urges and cravings are thought to occur when the drug-use action schema is activated but its implementation is impeded by external conditions, for instance, when the smoker is con- fined to a smoke-free environment. More specifically, ad- dicts are hypothesized to use conscious processing re- sources to inhibit the drug-use action plan. As a conse- quence, they should have fewer conscious resources available for other tasks they are engaged in at the time. Although the cited studies have shown that smoking urges elicited by in vivo cues or by imagery scripts may have deleterious effects on cognitive performance, little is known about the time course of these effects. Three patterns Rolf A. Zwaan, Robert A. Stanfield, and Carol J. Madden, Department of Psychology, Florida State University. This research was supported by National Institute on Drug Abuse Grant DA 11678. Correspondence concerning this article should be addressed to Rolf A. Zwaan, Department of Psychology, Florida State Univer- sity, Tallahassee, Florida 32306-1270. Electronic mail may be sent to zwaan@psy.fsu.edu. are theoretically possible: The urge effect may (a) remain constant, (b) dissipate, or (c) increase over time. The pre- diction that smoking urges have a continuous effect on cognitive performance can be derived from Tiffany's model, in which cognitive performance in the presence of a smoking urge is likened to dual-task performance (Tiffany, 1990, p. 160). That is, the smoking urge functions like a secondary task, drawing on resources needed for the main task. The prediction that the effect of smoking urges dissipates over time would be borne out either by decay of the urge itself (e.g., if it is not supported by a continuously present urge cue) or by a coping mechanism that allows the smokers to minimize or eliminate the interference from the urge by adapting their cognitive strategies for performing the pri- mary task. The prediction that the urge effect increases over time can be derived from Wegner's (1994) theory of ironic memory processes. Wegner and colleagues have shown that the active suppression of information can have ironic effects such that the information that is actively being suppressed actually gains more in activation (Wegner, 1989, 1992; Wegner, Quillian, & Houston, 1996; Wegner, Schneider, Carter, & White, 1987). Thus, according to this prediction, the cognitive performance of smokers exposed to the urge script should become increasingly worse. This effect might be enhanced if the inability to implement the smoking- action plan renders smokers increasingly frustrated. This would presumably generate more task-irrelevant thoughts, thereby adding to the deleterious effect of the urge itself. Thus, there are three classes of hypotheses that make different predictions about the time course of the urge effect. According to the first class, the effect remains con- stant. According to the second class, it dissipates. And according to the third class, it increases. The main goal of this study was to distinguish empirically among the three classes of predictions by examining the time course of the urge effect on cognitive performance. A second goal of this study was to provide a conceptual replication of Zwaan and Truitt (1998). They obtained an urge effect in smokers (but not nonsmokers) on sentence 518