The pharmacological, stimulus expectancy, and response expectancy effects of light cigarettes (0.60 mg yield of nicotine) compared with virtually nicotine-free cigarettes (less than 0.05 mg yield of nicotine) were examined using a between-subjects design. A total of 103 college-student smokers completed tests of cognition before and after smoking one type of cigarette, which was evaluated on a number of dimensions. Cigarettes with nicotine were preferred on most dimensions, and stimulus expectancy partially mediated the relationship between nicotine and subjective effects of the cigarettes. Stimulus expectancy also mediated the effect of nicotine on tension reduction immediately after smoking, but not near the end of the experiment. Response expectancy effects of nicotine were related to predicted and actual recall performance, such that participants who performed well tended to attribute the effect to the cigarette they smoked. Implications for smoking cessation and research studies using non-nicotine cigarettes are discussed.There is growing consensus that changes in cognitive processing contribute to the reinforcing effects of cigarettes (e.g., Brandon, Herzog, Irvin, & Gwaltney, 2004;Kassel, 1997;Levin, 1992). In college students, for example, smoking behavior increases substantially during periods of intense study (Wesnes, Revell, & Warburton, 1983), and students report increased smoking specifically to increase their level of cognitive arousal (West & Lennox, 1992). Whereas a large, established body of research has focused on the cognitive effects of nicotine (for reviews, see Heishman, Taylor, & Henningfield, 1994;Sherwood, 1993), there is now increasing recognition that numerous factors, including non-pharmacological factors, can play important roles in maintaining smokers' addiction to cigarettes (Johnson, Bickel, & Kirshenbaum, 2004;Perkins, Sayette, Conklin, & Caggiula, 2003;Rose, 2006). The focus of the present study was on the role of expectancy in nicotine's cognitive effects.The term expectancy denotes two different phenomena in nicotine research. First, participants may hold certain beliefs about the effects of a drug on their own internal state, which can be based on past use of the drug or information received from others. This type of expectancy is more precisely termed "response expectancy," and it has been shown to influence a variety of outcomes, including mood states, memory, and drug use (Kirsch, 1999). Response expectancies related to cigarettes have been shown to be a causal factor in smoking motivation (e.g., Copeland & Brandon, 2000). In addition, Hendricks and Brandon (2008) showed that smokers had different expectancies for cigarettes, which were primarily associated with negative health consequences, compared with nicotine, which was more closely associated with addition. Both cigarettes and nicotine were equally associated with arousal. Expectancy can also refer to whether participants expect an active or inactive substance in a study. This variable is more precisely terme...