The Craving to Smoke in Flight Attendants: Relations With Smoking Deprivation, Anticipation of Smoking, and Actual Smoking

Journal of Abnormal Psychology 2010 American Psychological Association 2010, Vol. 119, No. 1, 248 253 0021-843X/10/$12.00 DOI: 10.1037/a0017778 The Craving to Smoke in Flight Attendants: Relations With Smoking Deprivation, Anticipation of Smoking, and Actual Smoking Reuven Dar, Nurit Rosen-Korakin, and Oren Shapira Tel Aviv University Yair Gottlieb and Hanan Frenk The Academic College of Tel Aviv Yafo and Tel Aviv University In the study, the authors examined the effects of smoking deprivation, anticipation of smoking, and actual smoking on the craving to smoke. Flight attendants who were light to heavy smokers rated their craving to smoke at predetermined time points during a 2-way short flight (each leg 3 5.5 hr) and a 1-way long flight (8 13 hr). In both short and long flights, craving increased gradually and peaked as landing approached. Craving levels at the end of the 1st leg of the short flights were equal to those at the end of the long flight and were much higher than those at the parallel time point in the long flight. In the short flight, craving levels at the beginning of the 2nd leg dropped relative to the end of the 1st leg, both for participants who smoked during the intermission and for those who did not, though the drop was steeper for the former. The results provide additional evidence for the role of psychological factors in determining the craving to smoke in a naturalistic setting. Keywords: smoking, craving, flight attendants, smoking deprivation The craving to smoke has been described as the most fundamental and difficult problem for smokers who are trying to quit (Russell, 1988, p. 68). In the past 2 decades, extensive research efforts have been aimed at explicating the nature of the craving to smoke and the factors that determine it. The craving to smoke was listed as a symptom of nicotine withdrawal in the official nomenclature prior to the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; American Psychiatric Association, 1994) and is still conceptualized as a withdrawal symptom in some of the scientific literature (e.g., McClemon, Westman, & Rose, 2004) as well as in commercial and popular publications (e.g., http:// quitsmoking.about.com/cs/cravingsandurges/a/withdrawal.htm). The view of craving as a withdrawal symptom is too narrow, however. Smokers can experience high levels of craving even when they are smoking freely and are clearly not in a state of withdrawal (e.g., Hughes, Gust, Skoog, Keenan, & Fenwick, 1991). Similarly, craving can be higher after short but forced abstinence than after much longer but habitual and expected abstinence (Jacober, Hasenfratz, & Battig, 1994). More generally, a large body of evidence shows that craving is largely determined by smoking-related cues and expectations. Smoking-related cues, including manipulation of cigarettes and Reuven Dar, Nurit Rosen-Korakin, and Oren Shapira, Department of Psychology, Tel Aviv University, Tel Aviv, Israel; Yair Gottlieb and Hanan Frenk, Department of Psychology, The Academic College of Tel Aviv Yafo, Israel, and Department of Psychology, Tel Aviv University. Reuven Dar and Hanan Frenk have received fees for consulting with lawyers working with tobacco companies. However, all of Reuven Dar s and Hanan Frenk s research projects have been supported exclusively by Tel Aviv University funds. Correspondence concerning this article should be addressed to Reuven Dar, Department of Psychology, Tel Aviv University, Tel Aviv 69978, Israel. E-mail: ruvidar@freud.tau.ac.il exposure to smoking-related objects or contexts, can provoke powerful craving responses in smokers (e.g., Juliano & Brandon, 1998; Sayette & Hufford, 1994; Sayette, Martin, Wertz, Shiffman, & Perrott, 2001; Sayette et al., 2003). These cues appear to operate by evoking an expectancy of smoking, which increases craving (Dols, van den Hout, Kindt, & Willems, 2002; Dols, Willems, van den Hout, & Bittoun, 2000; Thewissen, van den Hout, Havermans, & Jansen, 2005; Wertz & Sayette, 2001). A study by our group (Dar, Stronguin, Marouani, Krupsky, & Frenk, 2005) provided additional evidence that the craving to smoke is primarily determined by nonpharmacological factors and can clearly be dissociated from smoking withdrawal. Twenty orthodox Jewish smokers, who are forbidden by their faith from smoking on the Sabbath, were assessed three times: on a workday when smoking as usual, on a Sabbath when they never smoked, and on a forced abstinence workday. Craving, irritability, and other commonly reported smoking withdrawal symptoms were assessed retrospectively at several time points during the preceding 24 hr. As predicted, craving, irritability, and self-reported difficulty in abstaining were much lower during the Sabbath than during the two other test days. Notably, craving during the smoking-as-usual workday was just as high as during the forced abstinence workday, demonstrating that craving had no relation to abstinence in this setting. In the same vein, craving in the evening before the test day was higher than in the next morning, despite the overnight abstinence that preceded the morning assessment. Although the results of Dar et al. (2005) corroborate the hypothesis that the craving to smoke is largely determined by smoking-related cues and expectations, those results have several limitations. Most important, because of religious laws prohibiting the participants from working (including writing) on the Sabbath, craving and withdrawal were assessed retrospectively rather than in real time. In addition, as smoking has been associated with stress (see Kassel, Stroud, & Paronis, 2003 for a review), the 248

BRIEF REPORTS 249 reduction of craving on the Sabbath, a day of rest, may have been partially a result of lowered stress. The goal in the present study was to replicate the findings of Dar et al. (2005) and extend them to another naturalistic context, while addressing the limitations noted above. In the present study, we examined craving levels among flight attendants who are not permitted to smoke during flights due to airline regulations. Each participant was assessed twice: during a two-way flight comprising two short legs (each leg 3 5.5 hr long) with a 1-hr intermission and during a long, one-way flight (8 13 hr long). Craving was assessed in real time at several points during the flights. We also assessed self-reported stress, expectations of smoking, and actual smoking during the intermission in the two-legged flight. We assumed that if the craving to smoke is mainly determined by smoking deprivation, craving levels should be related primarily to the time that has elapsed since the beginning of the flight. Alternatively, if craving is mainly determined by smoking-related cues and expectations, as our previous findings suggest, then craving should be predicted by the time remaining to the end of the flight, which signals the opportunity to smoke. Specifically, we hypothesized that the craving to smoke would rise faster in short, as opposed to long, flights. We also predicted that participants would report similar craving levels toward the end of the flight, regardless of its length. Finally, we predicted that craving before the intermission between legs would be higher among participants who expected to smoke during the intermission than among participants who did not expect to smoke during the intermission. Method Participants Participants were 53 flight attendants (38 women and 15 men) in El-Al Israeli Airlines. They were personally approached by Nurit Rosen-Korakin, who was herself a flight attendant at the time. Participants were asked to volunteer to take part in a study about smoking but were not told about its specific purpose. Those agreeing to participate signed an informed consent. Demographic and smoking data of the participants are provided in Table 1. Procedure Upon agreeing to participate in the study, each participant was handed two questionnaires, one for the short flight and one for the long flight, which also included the Fagerström Test for Nicotine Table 1 Characteristics of Participants Characteristic M SD Range Age (years) 28.50 3.71 22 41 Education (years) 14.47 1.90 12 18 Cigarettes per day 12.09 6.21 2 30 Smoking duration (years) 11.00 3.84 3 24 Number of family members who smoke 1.19 1.37 0 6 Number of quitting attempts 1.75 3.32 0 20 FTND score 2.41 2.26 0 7 Note. FTND Fagerström Test for Nicotine Dependence. Dependence (FTND; Heatherton, Kozlowski, Frecker, & Fagerström, 1991, see the Measures subsection, below) and a detailed smoking history inventory. The questionnaires were to be completed independently by each participant, according to his or her individual work schedule. Participants were to complete the first part of each questionnaire, which assessed personal and flight data, on the way to the aircraft, when they had no further opportunity to smoke prior to the flight. They were asked to complete the remaining part of the questionnaire, including craving ratings, at predetermined time points during the flight, as detailed below. Finally, they were asked to return the completed questionnaires to the experimenter s mailbox and were thanked for their cooperation. Measures The short flight questionnaire. This questionnaire was to be completed during a two-way flight from Tel Aviv to relatively proximal destinations, with each leg lasting 3 5.5 hr. In the first section participants were asked to provide demographic details (age, sex, education, and number of months working in El-Al) and flight details (flight number, date, boarding time, planned departure time, actual departure time, and length of each leg). They were also asked to indicate when they expected to smoke their next cigarette. In the second section, participants were asked to rate their current level of craving for a cigarette on a 1 7 scale (1 not at all,7 very much) at three time points during each leg: right after serving the meal (T1, T5; 1.5 2 hr after departure), 1 1.5 hr later (T2, T6), and right after the passengers have left the plane (T3, T7). They were also asked to rate their craving just before the passengers boarded the plane on the way back (T4). As the nature of the participants work did not allow for constant intervals between measurements, they were asked to document the exact timing of each rating. Between legs, participants were also asked to indicate whether they smoked during the intermission and, if so, how many cigarettes. In addition, at the end of each leg, participants were asked to rate how stressful the leg had been on a 1 7 scale (1 not at all, 7 very much). Finally, at the end of the second leg, participants were asked to indicate at which moment during the flight they had felt the strongest craving to smoke. The long flight questionnaire. This questionnaire was to be completed during a one-way flight that lasted 8 13 hr. It was similar to the short flight questionnaire, with the following exceptions: First, participants were asked to rate their level of craving for a cigarette at five time points rather than at three time points during the flight. Second, on long flights, one meal is served after departure and another before landing. In between, the flight attendants rest in two shifts (see Figure 1). As a result, three time points were shared by all participants, whereas the remaining two differed between the two shifts. All participants rated their craving right after the first meal (T1; 1.5 2 hr after departure), right after the second meal (T4), and right after the passengers have left the aircraft at the end of the flight (T5). In addition, those in the first rest shift (N 28) rated their craving right after returning from rest (T2) and approximately 1.5 hr later (T3), whereas those in the second rest (N 24) shift rated their craving approximately 1.5 hr after the first meal (T2) and approximately1.5 hr later, right before taking the rest (T3). Participants were asked to indicate to which

250 BRIEF REPORTS 7 6 5.11 5 first shift second shift Craving 4 3 2 1 return from rest return from rest 2.64 2.04 2.17 2.12 1.82 1.75 3.31 3.70 2.50 T1 T2 T3 T4 T5 Time Figure 1. Craving levels by time and rest shift for long flights. For the first shift, all differences between consecutive measurements are significant between the time after the return from rest (T2) and the time after the passengers have left the aircraft at the end of the flight (T5). For the second shift, the only significant difference between consecutive measurements is between the time after the return from rest (T4) and T5. All reported significant differences are at the.01 level. T1 right after the first meal, T4 right after the second meal, T5 right after the passengers have left the aircraft at the end of the flight, T2 right after returning from rest or approximately 1.5 hr after the first meal, and T3 approximately 1.5 hr after returning from rest or right before taking the rest. Error bars represent standard errors of the mean. rest shift they were assigned and to document the exact time of each rating. The Hebrew version of the FTND (Heatherton, Kozlowski, Frecker, & Fagerström, 1991). The FTND is the most commonly used scale for nicotine and smoking dependence and includes six items, with a total score ranging between 0 and 10. Detailed smoking history questionnaire. A detailed smoking history questionnaire was used to assess how long participants have smoked, the number and brand of cigarettes habitually smoked, the number of family members and coworkers who smoked, the history of any previous quit attempt, any current use of nicotine replacement devices (none reported), and whether participants smoked more than usual during the hour before boarding the plane. Results Craving Patterns To examine the pattern of the craving to smoke during the long flight (see Figure 1), we conducted a 5 (time: T1 T5) 2 (rest shift: first vs. second) mixed-model analysis of variance (ANOVA), with time as the repeated measures factor and rest shift as the between-subjects factor. There was a significant effect of time on craving, F(4, 192) 53.05, p.001, and a significant Time Rest Shift interaction, F(4, 192) 5.78, p.001. To examine the nature of this interaction, we analyzed craving patterns separately for each shift with univariate repeated measures analyses. There was a significant effect of time in both the first shift, F(4, 100) 53.30, p.001, and the second shift, F(4, 92) 13.92, p.001. However, pairwise contrasts having an alpha of.01 for control of familywise inflation of Type I error rate revealed a different pattern of craving in the two shifts. In the first rest shift, craving did not change significantly between T1 and T2 but then did rise significantly with each consecutive measurement until T5. In the second rest shift, in contrast, there were no significant differences between consecutive measurement until T4, and craving rose significantly only between T4 and T5 (see Figure 1). Thus, in both shifts, craving did not change substantially until participants returned from their rest and then started ascending gradually until the end of the flight. The patterns of craving in the short flight were analyzed with two univariate repeated measures analyses, one for the first leg (T1 T3) and one for the second leg (T4 T7). The fourth measurement in the short flights (T4) was analyzed along with the second leg because it was taken after the intermission, when participants no longer had the opportunity to smoke. There was a significant effect of time on craving in both the first leg, F(2, 104) 98.46, p.001, and the second leg, F(3, 156) 37.99, p.001. Pairwise contrasts showed that in the first leg, craving rose significantly with each measurement. In the second leg craving rose significantly with each measurement between T5 and T7, but did not differ between T4 and T5 (Figure 2; all reported significant differences are at the.01 level). Next, we examined whether craving levels rose at different rates during short and long flights. We compared craving levels between the first half of the long flights (T1 T3) and the first leg of the short flights (T1 T3), separately at each time point. Paired samples t tests did not show significant differences in craving levels between flights at T1 and T2, t(51) 0.11, p.91; t(52) 1.93, p.18, respectively. However, consistent with our prediction,

BRIEF REPORTS 251 Craving 7 6 5 4 3 2 1 1.87 2.23 4.55 craving was significantly higher at the third measurement of the short flights (M 4.55, SD 1.93) than in the parallel measurement of the long flight (M 2.36, SD 1.40), t(52) 8.76, p.001. 1 To examine whether craving levels at the end of the flight differed as a function of the flight s length, we compared the last craving measurements of the three segments (long flight, first leg, second leg). Consistent with our prediction, a univariate repeated measures ANOVA showed no significant differences between the measurements, F(2, 100) 0.24, p.78. In the same vein, the actual length of the flight was not correlated with peak craving levels in any of the three segments (rs.22,.09, and.10 in the long flight, the first leg of the short flight, and the second leg of the short flight, respectively; ps.1). Craving levels at all three segments peaked at approximately 4.5. Effects of Expectation of Smoking and Actual Smoking During the Intermission Between legs, flight attendants are sometimes allowed to take a smoking break at the destination terminal. To test the role of explicit expectations of smoking, we compared craving levels at the end of the first legs of the short flights (T3) between participants who expected to smoke during the between-legs intermission (N 33) and those who did not (N 19). In contrast to our prediction, an independent samples t test did not show a significant difference between the two groups, t(50) 0.44, p.66. Next, we compared the change in craving during the betweenlegs intermission between participants who actually smoked during the intermission (N 29, of whom 25 expected to smoke) and those who did not (N 24, of whom 15 did not expect to smoke). We conducted a 2 (time: T3 vs. T4) 2 (smoking in the intermission: yes vs. no) mixed-model ANOVA, with time as the repeated measures factor and smoking as the between-subjects factor. We found a main effect for time, F(1, 51) 50.35, p.001, such that craving was significantly lower after the intermission (M 2.72, SD 1.76) than preceding it (M 4.55, SD 2.72 2.30 2.85 T1 T2 T3 T4 T5 T6 T7 1 st leg intermission 2 nd leg Time Figure 2. Craving levels by time for short flights. All differences between consecutive measurements are significant ( p.01), except for the difference between the time before the passengers boarded the plane on the way back (T4) and 1.5 2 hr after departure (T5). T1 and T5 right after serving the meal, T2 and T6 1 1.5 hr later, T3 and T7 right after the passengers have left the plane, and T4 just before the passengers boarded the plane on the way back. Error bars represent standard errors of the mean. 4.58 1.93). Simple effects analyses showed that craving dropped during the intermission not only among participants who smoked, t(28) 6.67, p.001, but also among participants who did not smoke, t(23) 3.39, p.003. In addition, there was a significant Time Smoking interaction, F(1, 51) 21.79, p.001, indicating that the drop in craving was larger among flight attendants who smoked during the intermission than among those who did not. Stress, FTND Scores and Peak Craving Levels We computed the correlations between stress ratings, FTND scores, and craving ratings at the last measurement in each segment. Participants were generally not stressed during the flights (long flight: M 2.14, SD 1.36; first leg: M 2.13, SD 1.43; second leg: M 2.40, SD 1.32). Correlations between stress ratings and craving ratings were weak and insignificant (long flight: r.22, p.11; first leg: r.13, p.36; second leg: r.23, p.10). Correlations between the FTND and craving ratings were modest but significant (long flight: r.33, p.02; first leg: r.29, p.03; second leg: r.31, p.02). 2 No other smoking-related variable was correlated with craving levels. Participants retrospective reports about the moment during the flight when their craving to smoke peaked were consistent with real time ratings. Referring to the long flights, 38% answered, right after landing or at the (destination) terminal, 28% answered, just before landing and only 20% answered, during the flight. Referring to the short flights, 42% reported their craving peaked just before or after landing back in Tel Aviv (end of second leg), 25% answered, when approaching landing at destination (end of first leg) or during the between-leg intermission, 7.7% answered, during both landings (at destination and back in Tel Aviv), 5.8% answered, after the meal (during the first leg), and 5.8% reported that craving had peaked during the second leg. Finally, 45 participants (86.5%) reported that they smoked more than usual during the hour before boarding the plane. Discussion In this study of flight attendants, the craving to smoke was related to the time remaining to the end of the flight rather than to the time that elapsed since the last cigarette or to the total length of the flight (i.e., duration of abstinence). In both short and long flights, craving increased gradually and peaked toward the end of the flight, but craving levels at the final assessment point in the first leg of the short flights were much higher than were craving levels at the parallel time point in the long flight. In addition, 1 Due to differences in the schedule of tasks between short flights and long flights, the third measurement (T3) of the long flights was on average farther away in time from departure (M 7.09 hr) than the T3 of the short flights (M 5.54 hr). 2 We also tested whether smoking dependence moderated the other findings reported in this article. This was tested by dividing participants by a median split on the FTND to low-dependency participants (FTND: 0 2, N 30) versus high-dependency participants (FTND: 3 7, N 23). We entered this variable as an additional factor in all the reported analyses. None of the findings reported in the article was significantly moderated by this dependency factor (Fs for the interaction term ranged from 0.005 to 2.74, ps ranged from.95 to.10).

252 BRIEF REPORTS craving levels at the last measurement point were equal in short and long flights. These findings suggest that expectations and cues for an approaching opportunity to smoke after landing affected craving more than did smoking deprivation. Our findings show that craving levels are partially reset by taking a break or a rest, demonstrating again that craving does not simply accumulate as a function of deprivation. In the long flight, craving levels began to rise following the rest, creating a distinctly different pattern for those who rested earlier than for those who rested later in the flight. This finding is consistent with the studies by Dar et al. (2005) and Jacober et al. (1994), in which craving markedly decreased following a night s sleep but increased substantially during a shorter abstinence period upon waking. In the two-legged flights, craving levels at the beginning of the second leg were lower than those at the end of the first leg, even for participants who did not smoke during the intermission between legs, though the drop was steeper for participants who did smoke during the intermission. This pattern again suggests a partial reset effect and also shows that actual smoking has an additional craving reduction effect. As we did not a priori predict the reset effect, further research is needed to substantiate and account for this finding. Our findings are consistent with those of our previous study (Dar et al., 2005), in which orthodox Jews reported low levels of craving and other smoking withdrawal symptoms during the Sabbath. The present study addressed some of the limitations of the previous study. First, we measured craving in real time rather than retrospectively. Second, we assessed the role of stress, which was a potential confound in our previous study. The finding that stress during the flight was only weakly and insignificantly related to craving suggests that our previous findings were not produced by a difference in stress between the Sabbath and the weekdays. Yet, it is possible that had we also measured stress outside the flight, we would have found higher correlations between stress and craving. Consistent with earlier findings (reviewed by Donny, Griffin, Shiffman, & Sayette, 2008) craving was modestly correlated with dependence as measured by the FTND but was unrelated to any other individual differences, including the duration of smoking and the mean number of cigarettes per day. More generally, our findings accord with experimental evidence that smoking urges are primarily determined by smoking-related cues and expectations, including an upcoming opportunity to smoke (e.g., Jacober et al., 1994; Juliano & Brandon, 1998; Sayette & Hufford, 1994; Sayette et al., 2001, 2003). Particularly relevant are the findings that smokers experience higher levels of the urge to smoke when exposed to cues predicting the occurrence of smoking than when exposed to cues predicting abstinence from smoking (Dols et al., 2002, 2000; Thewissen et al., 2005). When smoking flight attendants board a flight, they enter a familiar context that signals that smoking is not to be expected soon. This context acts to suppress craving, even when several hours have elapsed since the last cigarette, as when our participants boarded the return leg of the two-way flight. As cues for an approaching landing multiply (e.g., the seatbelts signs are turned-on, passengers are seated), so rises the expectancy of smoking soon and the craving for a cigarette. Together with our previous study, the present findings demonstrate that smoking restrictions create contexts that not only reduce smoking but that also reduce the craving to smoke. Craving levels at the final assessment point in the short flights were not significantly higher among participants who expected to smoke in the between-legs intermission than among participants who did not. This finding suggests that craving may be triggered not only by explicit expectations but also by cues that have been conditioned to smoking opportunities. It should, however, be kept in mind that the explicit expectations of our participants were not always accurate, which constrains any inferences from this finding. Studying the role of conscious and unconscious factors in evoking smoking as a habituated behavior may be an important direction for further research. A limitation of the present study is that intervals between measurements were not constant or strictly fixed, which did not allow for a more precise analysis of the craving patterns. Another limitation is that we used a one-item craving measure, as we could not consume too much of the flight attendants time during work. Finally, our participants were mostly young and were not very heavy smokers (only 20.8% of the participants reported smoking at least one pack of cigarettes per day), so our results may not generalize to older and heavier smokers. We partially ruled out this possibility by verifying that our findings were not moderated by nicotine dependency level, but this null result does not provide strong evidence for a lack of such a relation. A stronger test of the impact of deprivation on craving would require a sample that includes heavier smokers than those recruited in the current study. In light of these limitations and previous data that do indicate an effects of deprivation on craving (e.g., Hendricks, Ditre, Drobes, & Brandon, 2006), the question of the relative contribution of deprivation versus psychological factors to craving should be considered very much open. In future studies, heavier and older smokers should be included, and designs should allow constant assessment intervals and assessment of craving with multiple-item measures. In summary, our results provide converging evidence from a naturalistic setting for the importance of psychological factors such as habits, cues, and expectations in determining the craving to smoke. Craving is the principal problem for smokers trying to quit (e.g., Russell, 1988; West & Schnieder, 1987), and understanding the determinants of craving is of utmost importance not only in relation to smoking but also in relation to other drug and nondrug habits. Indeed, models of craving (e.g., Tiffany, 1990) are relevant not only to drug urges but also to craving in relation to strong appetitive habits, such as binge eating, or to impulse control disorders like trichotillomania and pathological gambling. Explicating the determinants of craving in this larger context is essential to the understanding, treatment, and relapse prevention of a wide range of unwanted and harmful behaviors. References American Psychiatric Association. (1994). Diagnostic and statistical manual of the mental disorders (4th ed.). Washington, DC: American Psychiatric Association. Dar, R., Stronguin, F., Marouani, R., Krupsky, M., & Frenk, H. (2005). Craving to smoke in orthodox Jewish smokers who abstain on the Sabbath: A comparison to a baseline and a forced abstinence workday. Psychopharmacology, 183, 294 299. Dols, M., van den Hout, M., Kindt, M., & Willems, B. (2002). The urge to smoke depends on the expectation of smoking. Addiction, 97, 87 93. Dols, M., Willems, B. W., van den Hout, M., & Bittoun, R. (2000).

BRIEF REPORTS 253 Smokers can learn to influence their urge to smoke. Addictive Behaviors, 25, 103 108. Donny, E. C., Griffin, K. M., Shiffman, S., & Sayette, M. A. (2008). The relationship between cigarette use, nicotine dependence, and craving in laboratory volunteers. Nicotine & Tobacco Research, 10, 447 455. Heatherton, T. F., Kozlowski, L. T., Frecker, R. C., & Fagerström, K. O. (1991). The Fagerström Test for Nicotine Dependence: A revision of the Fagerström Tolerance Questionnaire. British Journal of Addiction, 86, 1119 1127. Hendricks, P. S., Ditre, J. W., Drobes, D. G., & Brandon, T. H. (2006). The early time course of smoking withdrawal effects. Psychopharmacology, 187, 385 396. Hughes, J. R., Gust, S. W., Skoog, K., Keenan, R. M., & Fenwick, J. W. (1991). Symptoms of tobacco withdrawal. A replication and extension. Archives of General Psychiatry, 48, 52 59. Jacober, A., Hasenfratz, M., & Battig, K. (1994). Cigarette smoking: Habit or nicotine maintenance? Effect of short-term smoking abstinence and oversmoking. Human Psychopharmacology, 9, 117 123. Juliano, L. M., & Brandon, T. H. (1998). Reactivity to instructed smoking availability and environmental cues: Evidence with urge and reaction time. Experimental and Clinical Psychopharmacology, 6, 45 53. Kassel, J. D., Stroud, L. R., & Paronis, C. A. (2003). Smoking, stress, and negative affect: Correlation, causation, and context across stages of smoking. Psychological Bulletin, 129, 270 304. McClemon, F. J., Westman, E. C., & Rose, J. E. (2004). The effects of controlled deep breathing on smoking withdrawal symptoms in dependent smokers. Addictive Behaviors, 29, 765 772. Russell, M. A. H. (1988). Nicotine replacement: The role of blood nicotine levels, their rate of change, and nicotine tolerance. In O. F. Pomerleau & C. S. Pomerleau (Eds.), Nicotine replacement: A critical evaluation (pp. 63 94). New York, NY: Liss. Sayette, M., & Hufford, M. (1994). Effects of cue exposure and deprivation on cognitive resources in smokers. Journal of Abnormal Psychology, 103, 812 818. Sayette, M. A., Martin, C. S., Wertz, J. M., Shiffman, S., & Perrott, M. A. (2001). A multidimensional analysis of cue-elicited craving in heavy smokers and tobacco chippers. Addiction, 96, 1419 1432. Sayette, M. A., Wertz, J. M., Martin, C. S., Cohn, J. F., Perrott, M. A., & Hobel, J. (2003). Effects of smoking opportunity on cue-elicited urge: A facial coding analysis. Experimental Clinical Psychopharmacology, 11, 218 227. Thewissen, R., van den Hout, M., Havermans, R. C., & Jansen, A. (2005). Context dependency of cue-elicited urge to smoke. Addiction, 100, 387 396. Tiffany, S. T. (1990). A cognitive model of drug urges and drug-use behavior: Role of automatic and nonautomatic processes. Psychological Review, 97, 147 168. Wertz, J. M., & Sayette, M. A. (2001). A review of the effects of perceived drug use opportunity on self-reported urge. Experimental and Clinical Psychopharmacology, 9, 3 13. West, R. J., & Schneider, N. (1987). Craving for cigarettes. British Journal of Addiction, 82, 407 415. Received November 21, 2008 Revision received May 14, 2009 Accepted May 15, 2009