Abstract Introduction Cue reactivity (CR) research has reliably demonstrated robust cue-induced responding among smokers exposed to common proximal smoking cues (eg, cigarettes, lighter). More recent work demonstrates that distal stimuli, most notably the actual environments in which smoking previously occurred, can also gain associative control over craving. In the real world, proximal cues always occur within an environment; thus, a more informative test of how cues affect smokers might be to present these two cue types simultaneously. Methods Using a combined-cue counterbalanced CR paradigm, the present study tested the impact of proximal (smoking and neutral) + personal environment (smoking and nonsmoking places) pictorial cues, on smokers’ subjective and behavioral CR; as well as the extent to which cue-induced craving predicts immediate subsequent smoking in a within-subjects design. Results As anticipated, the dual smoking cue combination (ProxS + EnvS) led to the greatest cue-induced craving relative to the other three cue combinations (ProxS + EnvN, ProxN + EnvS, and ProxN ± EnvN), ps < .004. Dual smoking cues also led to significantly shorter post-trial latencies to smoke, ps < .01. Overall CR difference score (post-trial craving minus baseline craving) was predictive of subsequent immediate smoking indexed by: post-trial latency to smoke [B = −2.69, SE = 9.02; t(143) = −2.98, p = .003]; total puff volume [B = 2.99, SE = 1.13; t(143) = 2.65, p = .009]; and total number of puffs [B = .053, SE = .027; t(143) = 1.95, p = .05]. Conclusions The implications of these findings for better understanding the impact of cues on smoking behavior and cessation are discussed. Introduction Stimuli associated with past smoking, or smoking cues, induce robust self-reported cigarette craving in smokers. This effect has been demonstrated across hundreds of cue reactivity (CR) studies using standard proximal smoking cues (eg, cigarettes burning in an ashtray, pictures of lighters).1–8 More recent work has demonstrated that in addition to proximal stimuli, more distal cues, particularly environments in which smoking occurred, also gain associative control over cigarette craving and smoking behavior.9–12 Pictures of environments commonly associated with smoking (eg, bus stop, park bench) but devoid of proximal smoking stimuli can alone induce robust self-report craving.12 Exposure to smoking environment cues has also been shown to decrease latency to smoke and increase the number of puffs smokers take during a simulated lapse scenario.13 Moreover, the magnitude of craving induced by exposure to smoking environments is significantly enhanced when smoking environments are personalized12,14 (ie, using subject-generated pictures of the real-world places where they actually smoke), as is brain reactivity, specifically in the posterior hippocampus and insula.11 The use of personal smoking environment as cues in CR research is a departure from prior studies in which cues were typically experimenter-generated proximal stimuli (cigarettes in ashtrays, packs, lighters).15,16 Standard proximal cues have clear benefits. For example, they are easily developed and they reliably evoke robust craving from smokers regardless of personal smoking experiences. In fact, past research has found that personalizing proximal cues (eg, using one’s own cigarette brand or specific lighter) does not significantly enhance cue saliency.4,17 The opposite appears to be true of environment cues. Although smokers respond with increased craving when exposed to standard smoking environments (eg, a generic outdoor bench) relative to nonsmoking environments (eg, a gymnasium), their cue-induced craving is significantly lower than that evoked by standard proximal smoking and neutral cues.10 However, when environment cues are personalized (eg, photos of a smoker’s actual couch at home, their car), cue-induced craving levels are equivalent to those evoked by standard proximal smoking cues.9,12,13,18 Thus, research aimed at understanding the impact of environments as cues to smoke or refrain from smoking is enhanced by personalization of these contextual cues. Often overlooked across human CR research is the fact that smoking cues of all types do not occur in isolation. That is, proximal cues are always experienced in combination with an environment, albeit one of varying association with past smoking. Yet, most human addiction studies examining combined cues have focused on the impact of combining proximal cues across drug types, for example, proximal smoking cues presented with proximal alcohol or neutral cues,19 or the impact of drinking alcohol on smoking CR,20–22 rather than the effects of combining multiple smoking-related cue types. Examination of urges to smoke and smoking frequency in various environments has been achieved via ecological momentary assessment (EMA). A clear benefit of EMA studies is their naturalism, and access to information about situations that predispose craving, smoking, and relapse.23 For instance, evidence that intermittent smokers are more likely than daily smokers to smoke in bars, when drinking alcohol, and when other people are present.24 However, EMA studies do not allow for tight experimental control of timing and frequency of exposure to specific cues or cue combinations in the absence or presence of other stimuli that may affect various indices of reactivity among smokers. To date, no human laboratory studies have investigated the effect of combining smoking and nonsmoking standard proximal and personal environment cues within a CR paradigm to systematically examine the effects of congruent and incongruent cue combinations on smokers’ subjective and behavioral reactivity. If personalized and combined cues more accurately capture the experience of encountering cues in the real world, they may prove more evocative and thereby enhance smokers’ reactivity. Such enhanced reactivity may increase not only understanding how cues function, but may prove to make CR a better predictor of various clinical measures (eg, likelihood and success in quitting, relapse vulnerability), an area in which CR smoking research has historically fallen short.25–28 Examining actual drug administration as an index of CR in human CR studies has also been largely understudied. Unlike animal research in which self-administration is the most frequent measure of drug cue salience, human CR addiction studies most commonly assess self-reported craving.2 Relatively few studies have examined the impact of cue exposure on actual smoking behavior (ie, smoking topography),8,11,13,29–31 or the relationship between cue-induced craving and smoking behavior. The absence of such studies was the subject of considerable debate regarding what CR reveals about nicotine dependence, and led to a call for future CR work to assess the impact of cues on actual smoking, as well as to examine the relationship between the commonly assessed subjective craving and more objective measures of actual smoking.25–27 Since then, we recently found that both cue-induced craving and smoking behavior are significantly enhanced during exposure to proximal smoking stimuli compared to neutral stimuli; and, that the magnitude of cue-induced craving predicted immediate subsequent smoking, indexed as latency to smoke, number of puffs, and puff volume.8 Similarly, Hogarth et al.31 found that smoking cues enhanced both craving and smoking topography, and that these two effects were significantly correlated (r = .57). Yet another recent study evaluating exposure to personal smoking-related environments on activation in the hippocampus (HPC) and other brain areas involved in conditioned reward (ie, amygdala, medial prefrontal cortex, insula, striatum), examined correlations between brain reactivity, cue-provoked craving, and smoking behavior. Findings revealed similar effects of heightened craving and smoking behavior in addition to increased HPC and insula reactivity to personal smoking environments.18 On balance, recent CR studies examining both cue-induced craving and actual smoking behavior provide burgeoning evidence that exposure to smoking cues not only increases subjective craving, but actual smoking behavior as well, and has shown, in some cases, that those two indices of reactivity may be highly correlated. To expand CR work incorporating self-report craving and objective measures of smoking, as well as the relationship between these two indices, the present CR aimed to test combinations of standard proximal and personal environment smoking and nonsmoking cues. Although examining compound cues has been rare in human cue research, animal research on stimulus-compounding has tested how two cues that independently signal an outcome might affect reactivity when combined. For example, Panlilio et al.32 found that after independently establishing two discriminative stimuli for cocaine self-administration (a light and a tone), presentation of a combined light + tone cue led to a three-fold increase in cocaine seeking in rats when cocaine was not available and a twofold increase in cocaine consumption when it was available. Likewise, research on conditioned suppression using combined cues has found that two conditioned cues for shock, which independently suppress drug-seeking response, further enhance that suppression by 30% when presented in combination.33 These examples suggest that two cues that signal self-administration enhance that administration when combined, and two cues that suppress it will, when combined, increase suppression. If combined smoking cues in the human lab function similarly, a dual smoking-cue combination might lead to greater increases in craving and smoking behavior compared to a dual nonsmoking cue combination. Anticipating how incongruent cue combinations (a smoking + a nonsmoking cue) might affect craving and smoking behavior is more difficult. Combining a cue that signals nonsmoking might be expected to reduce reactivity to cues associated with smoking. However, a smoker who encounters a pack of cigarettes in a place where smoking is prohibited might still crave, but refrain from smoking, thereby demonstrating a dissociation between craving and behavior. In contrast, human research on the impact of drug availability might predict a reduction in both craving and smoking behavior, as a clear signal that smoking cannot occur (eg, no cigarettes available, an environment that disallows it), should not only suppress the behavior, but should drive craving down as well.3,34,35 No human CR research to date has systematically examined how congruent and incongruent cue combinations affect both craving and actual smoking behavior. The goal of the present study was to test smokers’ self-report and behavioral reactivity to pictorial stimuli of personal smoking and nonsmoking environments (EnvS and EnvN) in combination with standard smoking and neutral proximal cues (ProxS and ProxN). We hypothesized that personal smoking environments (EnvS) combined with proximal smoking cues (ProxS), which combines the two smoking cue categories, should evoke the strongest reactivity; and, EnvN + ProxN, which combines the nonsmoking and neutral cue types, should lead to the lowest. Regarding incongruent cue combinations, EnvS + ProxN and ProxS + EnvN (ie, personal smoking environment combined with a neutral proximal cue and a smoking proximal cue with a nonsmoking environment, respectively) should evoke comparable levels of craving to smoke and smoking behavior given that our past research shows that personalized environment and standard proximal cues evoke comparable levels of craving. However, the present study will be the first to examine if these cue combination lead to differential or congruent effects on craving and smoking behavior. In addition, we anticipate a strong predictive relationship between cue-induced craving and subsequent immediate smoking behavior, specifically that greater cue-induced craving will predict shorter latency to smoke, as well as increased number of puffs and puff volume during immediate subsequent in-session smoking. Method Participants Forty-eight daily smokers (24 men and 24 women) were recruited for the study through newspaper advertisements and flyers inviting, “healthy men and women smokers, ages 18–65 [to participate in] a research study investigating smoking cues.” Participants had to be daily smokers, who were not currently quitting or cutting back, between the ages of 18 and 65 (M = 33.65; SD = 14.2; range = 18–63), who smoked 10 or more cigarettes per day for at least a year (M = 18.1.15; SD = 5.60; range = 10–40), and had a carbon monoxide concentration (CO) greater than 8 ppm (M = 22.76; SD = 12.10). During initial phone screening, potential participants were informed that the study was not a treatment study. Individuals who reporting being interested in treatment were given referral information from a list of ongoing local smoking treatment studies and services. Participants had an average Fagerström Test of Nicotine Dependence (FTND; 33 score of 5.3 (SD = 1.9; range = 2–10) and received $125 for completing the study. Session 1 The first session lasted approximately 90 min. After signing initial consent forms and registering time since last cigarette, a CO measure was taken using a Vitalograph CO monitor (Vitalograph; Lenexa, KS). Participants then completed several forms: Smoking History Form, FTND,36 and the Balanced Inventory of Desired Responding-Impression Management section (BIDR-IM).37 The BIDR is a factor-analytically derived questionnaire of the extent to which an individual engages in conscious dissimilation of responses attempting to create favorable responses).38 Inclusion in the current study allowed for investigation of possible associations between impression management and participants’ answers on self-report measures. Using methods validated in prior research for identifying and taking pictures of personal smoking and nonsmoking environments,11,12 participants then listed and ranked 10 environments in which they most often smoke during a typical week, as well as 10 environments in which they do not. A semi-structured interview was conducted to identify the participant’s top four smoking and four nonsmoking environments. Smoking environments had to be places in which the individual spent time at least once a week and smoked at least 7 out of 10 times when in that environment. Nonsmoking environments also had to be places in which the individual spent time at least once a week and smoked 3 or less times out of 10 times in that place. We allowed places in which limited smoking occurred, rather than no smoking, to ensure that participants could identify four places where they would regularly go without smoking. However, post hoc evaluation revealed that 95.8% of the nonsmoking pictures used in the cue sessions were rated 0 for the number of times the participant smoked out of 10 times being in that environment, suggesting that we captured personal nonsmoking places (either by individual choice or public law) rather than “infrequent smoking” places. Participants took pictures of their top four smoking and four nonsmoking environments, but only three of each were used in the later experimental sessions. This was done to create a buffer if one set of pictures was not clear enough, or could not be taken by the subject. Once the environments were selected, participants went through picture-taking training using the experiment room as a sample environment. They were instructed to take pictures from two angles approaching the environment and two from within the environment. Before leaving, participants received an FujiFilm Finepix J38 digital camera to borrow for picture taking (Fuji Optical Co., Ltd.; Ontario, Canada), as well as a written reminder of the environments of which to take pictures and the time of their next session. Session 2 The second session was scheduled approximately 1 week after the first to give participants adequate time to take pictures. At this brief session, participants supplied a CO sample aimed at capturing typical mid-day smoking exposure and dropped off the borrowed camera with their environment pictures. The experimenter then scheduled the participant’s third session. Sessions 3–6 CR to Combined Cues The four CR sessions each lasted approximately 2 h. The first of these took place approximately 1 week after the second session. This time was needed to edit subjects’ pictures and set up the individual computerized stimuli presentation. All pictures were edited with Adobe Photoshop software to ensure equal pixilation and to remove any proximal smoking cues, people, or alcohol from the pictures. Unwanted stimuli in the photographs were replaced with a continuation of the background to appear as if nothing was ever there. This was achieved using the Adobe Photoshop clone stamping and spot healing functions (see Supplementary Material, for example). Sessions 3–6 occurred within 3 days of each other. Each session included only one type of combined cue (EnvS + ProxS, EnvS + ProxN, EnvN + ProxS, EnvN + ProxN), with the order of sessions equally counterbalanced across subjects. This was done to get a clean assessment of ad lib smoking in response to each cue combination type and to control for session order. The participant gave an arrival CO sample, then smoked one cigarette using the CreSS system (which was used during ad lib smoking later in the session), followed by a second CO sample. Next, participants completed two self-report ratings: the 4-item Questionnaire of Smoking Urges (QSU-4)3 and the Diener and Emmons Mood Form.39 The participant was given an overview of the remainder of the session, including instructions for viewing cues and completing post-trial ratings (see detail below). Proximal cues were displayed on a 22in. monitor (ViewSonic Corporation; Walnut, CA) and environment pictures were presented on a 8 ft × 15ft. projection wall using a InFocus DLP short throw projector (InFocus Corporation; Portland, OR). These two screen types were used to present stimuli near life-size and in proximity to the subject as they would likely appear in real-life. After the instructions, the participant completed a practice trial to ensure that he/she could follow the automated CR procedure. The experimenter then left the room and the 6 CR picture trials began. Each trial followed a standard format: 20 s relaxation, 20 s baseline, 40 s picture viewing (10 s for each of the four angles within a picture set), and post-trial self-report craving rating assessed with the 4-item QSU, as well as a 1-item Likert rating of vividness, and a 2-item Likert-type rating of arousal. The presentation of the pictorial stimuli was controlled by a program written specifically to time the presentation of proximal and environment pictures on the two screens in synch and was run on a Compaq Evo computer (Hewlett Packard Company; Palo Alto, CA). The program also controlled the post-trial ratings, which were presented on the small computer after each of the six picture trials. Instructions to participants read, in part, “A prompt will appear on the screen instructing you to sit back in the chair and relax. Following that, pictures of items will appear on the computer screen in front of you, while pictures of environments are displayed on the wall in front of you. You are to focus on having those items right now while thinking about actually being in the environments. Keep focusing on the items and environments that are presented until the computer screen changes and prompts you to answer questions based on how you felt while focusing on those objects in those places”. No pictures were present during the actual ratings. After the final of the six picture trials, a screen appeared informing the participant that, “When the pictures reappear you may smoke as much or as little as you like. You do not have to smoke if you choose not to. If you smoke, you must light the cigarette and use the cigarette holder the same way you did earlier.” A tray with covered cigarette paraphernalia (the participant’s cigarettes, CreSS cigarette holder, a lighter, and an ashtray) was on a side table next to, but out of the line of sight of the participants. The next screen instructed the participant to move the tray forward, uncover it, put a cigarette in the holder, set it back down, and advance to the next screen. This started the smoking latency timer. Pictures then appeared and repeated randomly over a 12-min ad lib period during which the participant was free to smoke if he/she chose to. Behavioral smoking indices of latency to light, puff volume, and number of puffs were collected during this time. The experimenter then returned to the room and scheduled the next session, or if it was the last session, debriefed the participant and paid him/her. Analytic Strategy The analyses for the current study focused on the impact of four proximal and environmental cue combinations on self-reported craving and behavioral indices of CR. The primary analyses reported in the current study were evaluated via mixed model regressions using SAS PROC MIXED (SAS Institute, Inc, 2015) with Bonferroni post hoc comparisons to ascertain significant differences in cue combination effects on craving and smoking topography variables while controlling for Type 1 error rates. Mixed model regression features advantages over traditional forms of regression, due to the ability to handle repeated assessments of both predictor variables (eg, proximal and environmental stimulus combination) and outcome variables (eg, smoking topography measures of latency to smoke, number of puffs, and total puff volume) in combination with between-subject predictors (eg, trait scores of impression management and nicotine dependence). Analyses of mixed model regressions tested the main effects of cue combination session (1 = EnvN + ProxN, 2 = EnvN + ProxS, 3 = EnvS + ProxN, and 4 = EnvS + ProxS) on craving and smoking topography variables. Separate models were evaluated for the following outcome variables: craving, latency to smoke, total number of puffs, and total puff volume. If a participant refrained from smoking during the ad lib period (EnvS + ProxS = 11, EnvN + ProxS = 26, EnvS + ProxN = 17, EnvN + ProxN = 26) latency was entered a 12 min, and puff volume and number of puffs were entered as 0. Regarding the effects of cue combination session on self-reported craving, baseline craving levels were entered as a covariate, to control for their predictive influence on this outcome variable. Secondary analyses examined main effects of whether magnitude of cue-induced craving (post-trial minus session baseline) predicted smoking topography variables, with separate models evaluating the dependent variables of total puff volume, total number of puffs, and latency to smoke. For each of these models, cue-induced craving × trial interaction effects were also examined, to ascertain, for example, whether latency to smoke was quickest amid those who displayed elevated craving exclusive to smoking cues. Results Descriptive statistics pertaining to primary sample characteristics are featured in Table 1. Preliminary analyses tested for pertinent demographic and trait predictors of the primary dependent variables via mixed model regressions. Evidence indicated participant age (B = 4.23, SE = 2.15; t(144) = 1.96, p = .05) and education (B = 88.6, SE = 30.4; t(144) = 2.92, p = .004) to predict longer post-trial latencies to smoke. Further, age was inversely predictive of post-trial craving, after controlling for baseline craving: B = −.4, SE = .2; t(144) = −1.95, p = .05. Finally, FTND score was significantly predictive of shorter post-trial latencies to smoke (B = −32.4, SE = 15.3; t(144) = −2.12, p = .036). Consequently, age was entered as a covariate for all subsequent analyses examining participant craving, whereas age, education, and nicotine dependence were entered as covariates when examining post-trial latencies to smoke. Table 1. Descriptive Statistics of Sample Characteristics Variables Freq (%) M (SD) Age 33.7 (14.3) Number cigarettes smoked per day 16.2 (5.7) Fagerström Test of Nicotine Dependence 4.42 (2.16) Gender Male 24 (50%) Female 24 (50%) Ethnicity African American 8 (17%) Asian/Pacific Islander 1 (2%) Biracial 2 (4%) Caucasian 37 (77%) Education Some high school 2 (4%) High school diploma 10 (21%) Some college 27 (56%) Baccalaureate degree 5 (10%) Some graduate school 3 (6%) Graduate degree 1 (2%) Variables Freq (%) M (SD) Age 33.7 (14.3) Number cigarettes smoked per day 16.2 (5.7) Fagerström Test of Nicotine Dependence 4.42 (2.16) Gender Male 24 (50%) Female 24 (50%) Ethnicity African American 8 (17%) Asian/Pacific Islander 1 (2%) Biracial 2 (4%) Caucasian 37 (77%) Education Some high school 2 (4%) High school diploma 10 (21%) Some college 27 (56%) Baccalaureate degree 5 (10%) Some graduate school 3 (6%) Graduate degree 1 (2%) View Large Table 1. Descriptive Statistics of Sample Characteristics Variables Freq (%) M (SD) Age 33.7 (14.3) Number cigarettes smoked per day 16.2 (5.7) Fagerström Test of Nicotine Dependence 4.42 (2.16) Gender Male 24 (50%) Female 24 (50%) Ethnicity African American 8 (17%) Asian/Pacific Islander 1 (2%) Biracial 2 (4%) Caucasian 37 (77%) Education Some high school 2 (4%) High school diploma 10 (21%) Some college 27 (56%) Baccalaureate degree 5 (10%) Some graduate school 3 (6%) Graduate degree 1 (2%) Variables Freq (%) M (SD) Age 33.7 (14.3) Number cigarettes smoked per day 16.2 (5.7) Fagerström Test of Nicotine Dependence 4.42 (2.16) Gender Male 24 (50%) Female 24 (50%) Ethnicity African American 8 (17%) Asian/Pacific Islander 1 (2%) Biracial 2 (4%) Caucasian 37 (77%) Education Some high school 2 (4%) High school diploma 10 (21%) Some college 27 (56%) Baccalaureate degree 5 (10%) Some graduate school 3 (6%) Graduate degree 1 (2%) View Large Craving and Behavioral Responses to Experimental Cues In accord with expectation, mixed model regression yielded a significant main effect for cue combination on post-trial craving, after controlling for baseline craving and participant age, B = 8.14, SE = 1.02; t(142) = 7.95, p < .001. Bonferroni post hoc comparisons revealed post-trial craving to be significantly elevated following dual smoking cue combination relative to the other three cue combination stimulus exposures, ps ≤ .004 (Figure 1). Similarly, main effect analyses of cue combination exposure on smoking topography variables yielded significant effects for post-trial latency to smoke (B = −80.95, SE = 13.09; t(143) = −6.18, p < .001) after controlling for participant age, education, and nicotine dependence; total puff volume (B = 71, SE = 16.79; t(143) = 4.23, p < .001); and total number of puffs (B = 1.6, SE = .398; t(143) = 4.03, p < .001) (Figures 2–4). Specific comparisons using Bonferonni post hoc adjustment indicated the dual smoking cue combination to predict significantly shorter post-trial latencies to smoke relative to EnvN + ProxN and EnvN + ProxS, ps < .001, and marginally shorter post-trial latencies to smoke relative to EnvS + ProxN, p = .06. Further, Bonferroni comparisons showed that the dual absence of smoking cues predicted significantly lower post-trial total puff volume and total number of puffs compared to EnvS + ProxN and dual smoking cue combinations, ps ≤ .05. Taken together, these findings indicate that exposure to dual smoking cue combinations reliably predicts increases in self-reported craving and shorter post-trial latencies to smoke above and beyond exposure to all other cue combinations. Likewise, the dual absence of smoking cues predicted the least post-trial total puff volume and fewest number of puffs relative to EnvS + ProxS, whereas the mixed cue combinations (EnvN + ProxS and EnvS + ProxN) were not significantly different from one another in their relationships to craving and smoking topography variables. Figure 1. View largeDownload slide Mixed model regression on craving displaying means with standard error bars across experimental stimuli cue combinations. Bonferroni post hoc comparisons indicated dual smoking cue exposure to predict significantly higher post-trial craving than any other combination of experimental stimuli, ps ≤ .004. Figure 1. View largeDownload slide Mixed model regression on craving displaying means with standard error bars across experimental stimuli cue combinations. Bonferroni post hoc comparisons indicated dual smoking cue exposure to predict significantly higher post-trial craving than any other combination of experimental stimuli, ps ≤ .004. Figure 2. View largeDownload slide Mixed model regression on post-trial latency to smoke in seconds with standard error bars across cue combinations. Bonferroni post hoc comparisons revealed dual combination of experimental smoking cue stimulus exposure (EnvS + ProxS) to predict significantly shorter latencies to smoke than EnvN + ProxN and EnvN + ProxS, ps < .001, and marginally significant shorter latencies than EnvS + ProxN, p = .06. Figure 2. View largeDownload slide Mixed model regression on post-trial latency to smoke in seconds with standard error bars across cue combinations. Bonferroni post hoc comparisons revealed dual combination of experimental smoking cue stimulus exposure (EnvS + ProxS) to predict significantly shorter latencies to smoke than EnvN + ProxN and EnvN + ProxS, ps < .001, and marginally significant shorter latencies than EnvS + ProxN, p = .06. Figure 3. View largeDownload slide Mixed model regression on total puff volume (mL) with standard error bars across cue combinations. Bonferroni post hoc comparisons demonstrated dual absence of smoking cue stimulus exposure (EnvN + ProxN) to be linked to the lowest volume puff inhalation compared to EnvS + ProxN and EnvS + ProxS stimulus combinations, ps ≤ .05. Figure 3. View largeDownload slide Mixed model regression on total puff volume (mL) with standard error bars across cue combinations. Bonferroni post hoc comparisons demonstrated dual absence of smoking cue stimulus exposure (EnvN + ProxN) to be linked to the lowest volume puff inhalation compared to EnvS + ProxN and EnvS + ProxS stimulus combinations, ps ≤ .05. Figure 4. View largeDownload slide Mixed model regression on total number of puffs, displaying means with standard error bars across experimental stimuli cue combinations. Bonferroni post hoc comparisons demonstrated dual absence of smoking cue stimulus exposure to be linked to significantly fewer post-trial number of puffs compared to EnvS + ProxN and EnvS + ProxS stimulus combinations, ps ≤ .05. Figure 4. View largeDownload slide Mixed model regression on total number of puffs, displaying means with standard error bars across experimental stimuli cue combinations. Bonferroni post hoc comparisons demonstrated dual absence of smoking cue stimulus exposure to be linked to significantly fewer post-trial number of puffs compared to EnvS + ProxN and EnvS + ProxS stimulus combinations, ps ≤ .05. Craving Reactivity as a Predictor of Smoking Behavior Main effect analyses of craving difference score (post-trial craving minus baseline craving) on smoking topography variables indicated significant effects for post-trial latency to smoke after controlling for participant age, education, and nicotine dependence (B = −2.69, SE = 9.02; t(143) = −2.98, p = .003); total puff volume (B = 2.99, SE = 1.13; t(143) = 2.65, p = .009); and total number of puffs (B = .053, SE = .027; t(143) = 1.95, p = .05). These findings indicate magnitude of post-trial craving to predict shorter latencies to smoke, deeper puff inhalations, and higher number of puffs. Mixed model regressions of interactions between cue combination and craving differences score indicated nonsignificant interactions for all three smoking topography variables of post-trial latency to smoke, total puff volume, and total number of puffs, ps > .4. Therefore, magnitude of self-reported craving did not interact with cue combination condition to predict smoking topography variables. Discussion Past research has demonstrated that standard proximal smoking cues and personal smoking environments both independently evoke robust craving and changes in smoking behavior when presented to current smokers.9,11,12 The results of the present study offer evidence that combining those two cue types leads to greater subjective and behavior reactivity from smokers above that evoked by cue combinations incorporating only one smoking cue. As hypothesized, exposure to combined smoking cues led to significantly enhanced cue-induced craving over all other cue combinations. The dual smoking cue combination also led to the greatest increase in smoking behavior indexed as latency to smoke, number of puffs, and puff volume. Further, the dual absence of smoking cues led to the lowest self-report craving and behavioral reactivity. In addition to finding that combining cues enhances cue-induced craving and smoking, the present study found that cue-induced craving was predictive of immediate subsequent smoking. Specifically, greater cue-induced craving was found to be a significant predictor of subsequent immediate smoking as indexed by shorter latencies to smoke, deeper puff inhalations, and greater number of puffs. As noted, examining smoking topography in response to cues has been a largely understudied area of CR, a field in which self-report craving has been the most commonly included measure, thus disallowing elucidation of the link between cue-induced craving and actual smoking behavior. The present study adds to an important growing body of research revealing a positive relationship between cue-induced craving and the speed and amount of immediate subsequent smoking in which one engages.8,31 What the present work cannot address is the extent to which CR relates to clinical outcome. As often as the magnitude of CR has been shown to predict such variables as likelihood of quit initiation40 and successful abstinence,41–44 it has failed to show any association with treatment outcome.1,17,45,46 Thus, in developing more evocative cues, like those used in the present study (cue of various types, combinations, and personalization), researchers may be better able to capture the key situations that provoke an individual’s strongest craving and thereby potentially render CR a more effective predictor of smoking and clinical outcomes in the future. The magnitude of self-reported craving did not interact with experimental cue combination to predict smoking topography variables. That is, regardless of specific stimuli, only amount of cue-induced craving predicted immediate subsequent smoking as no cue combination uniquely predicted how quickly or how much someone subsequently smoked. As noted above, intuitively, it seems likely that exposure to a proximal smoking cue while in an environment where smoking is prohibited might have a differential effect on craving and actual smoking behavior. For example, craving might be elevated upon exposure to cigarettes, but no actual smoking would occur if the environment is not permissive (eg, seeing cigarettes in one’s purse while at church). However, the results are more consistent with an availability explanation of conditioning and CR. Research on the impact of drug availability has shown that a non-permissive environment should signal a lack of opportunity to engage in drug use, which should keep both craving and behavior low, as should a permissive environment combined with a lack of drugs. That is, if an individual knows that self-administration cannot occur, craving and drug use behavior should be low.3,34,35,47 This may explain why exposure to one smoking cue combined with one nonsmoking cue led to moderate craving and smoking behavior compared to reactivity evoked by dual smoking or dual nonsmoking cue combinations in the present study. However, the dual smoking cue combination led to only marginally greater smoking behavior compared to the EnvS + ProxN combination. This may be because a ProxN cue is simply the absence of the drug stimulus (eg, a toothbrush instead of a cigarette), which functions as a neutral cue more than a nonsmoking or inhibitory cue. By comparison, a nonsmoking environment represents a conditioned stimulus, an environment in which smoking is not allowed to occur (either by personal choice or law), which is a clearer negative predictor of smoking. Therefore, the EnvS + ProxN may function less to inhibit smoking compared to the reverse, EnvN + ProxS. Although these two cue conditions did not differ in cue-induced craving, fewer participants chose to smoke during the EnvS + ProxN (n = 17) ad lib period compared to the EnvN + ProxS (n = 26). A few limitations are worth noting. This was not a treatment seeking sample, and quit interest was not assessed. Thus, the results add to overall understanding of subjective and behavioral cue responding among daily smokers exposed to various cues and cue combinations, but may not generalize to samples of smokers trying to quit. However, past research has shown that deprivation during CR versus satiety has an additive, not interactive, effect on craving, suggesting that abstinence may simply enhance the magnitude of reactivity.47,48 However, future studies might determine if the cognitive impact of motivation to quit leads to more interactive effects than were found here. It is also not feasible to examine behavioral responding to cues within a study of quitting smokers. However, past studies have examined prequit levels of self-report craving40 and brain reactivity43 to smoking cues as a predictor of quit initiation and success. Thus, future work could look at prequit behavioral cue responding to combined cues as a predictor of outcome, as well as examine self-report reactivity to combined cues throughout, and or following long-term abstinence. Similarly, future studies could examine the impact of various cues combined with internal states, such as negative affect induction, which has been shown to affect craving and smoking behavior,49–53 or social cues such as people.54 Further, more nuanced cue paradigms might be examined for their efficacy in assessing relapse vulnerability among smokers, or determining who may benefit most from interventions that target reactivity to different types of salient smoking cues and cue combinations. In sum, combined cue paradigms enhance subjective and behavior reactivity among daily smokers. This new methodology should be exploited and further evaluated as a means of better assessing cue sensitivity in smokers, as well as for its potential to predict measures of clinical interest. Supplementary Material Supplementary data are available at Nicotine and Tobacco Research online. Funding This work was supported by a National Institutes of Health grant (R01DA023646) awarded to CAC. Declaration of Interests None declared. References 1. Abrams DB, Monti PM, Carey KB, Pinto RP, Jacobus SI. Reactivity to smoking cues and relapse: two studies of discriminant validity. Behav Res Ther . 1988; 26( 3): 225– 233. Google Scholar CrossRef Search ADS PubMed 2. Carter BL, Tiffany ST. Meta-analysis of cue-reactivity in addiction research. Addiction . 1999; 94( 3): 327– 340. Google Scholar CrossRef Search ADS PubMed 3. Carter BL, Tiffany ST. The cue-availability paradigm: the effects of cigarette availability on cue reactivity in smokers. Exp Clin Psychopharmacol . 2001; 9( 2): 183– 190. Google Scholar CrossRef Search ADS PubMed 4. Conklin CA, Tiffany ST. The impact of imagining personalized versus standardized urge scenarios on cigarette craving and autonomic reactivity. Exp Clin Psychopharmacol . 2001; 9( 4): 399– 408. Google Scholar CrossRef Search ADS PubMed 5. Conklin CA, Tiffany ST, Vrana SR. The impact of imagining completed versus interrupted smoking on cigarette craving. Exp Clin Psychopharmacol . 2000; 8( 1): 68– 74. Google Scholar CrossRef Search ADS PubMed 6. Niaura RS, Rohsenow DJ, Binkoff JA, Monti PM, Pedraza M, Abrams DB. Relevance of cue reactivity to understanding alcohol and smoking relapse. J Abnorm Psychol . 1988; 97( 2): 133– 152. Google Scholar CrossRef Search ADS PubMed 7. Shiffman S, Dunbar M, Kirchner Tet al. Smoker reactivity to cues: effects on craving and on smoking behavior. J Abnorm Psychol . 2013; 122( 1): 264– 280. Google Scholar CrossRef Search ADS PubMed 8. Conklin CA, Vella EJ, Joyce CJ, Salkeld RP, Perkins KA, Parzynski CS. Examining the relationship between cue-induced craving and actual smoking. Exp Clin Psychopharmacol . 2015; 23( 2): 90– 96. Google Scholar CrossRef Search ADS PubMed 9. Conklin CA. Environments as cues to smoke: implications for human extinction-based research and treatment. Exp Clin Psychopharmacol . 2006; 14( 1): 12– 19. Google Scholar CrossRef Search ADS PubMed 10. Conklin CA, Robin N, Perkins KA, Salkeld RP, McClernon FJ. Proximal versus distal cues to smoke: the effects of environments on smokers’ cue-reactivity. Exp Clin Psychopharmacol . 2008; 16( 3): 207– 214. Google Scholar CrossRef Search ADS PubMed 11. McClernon FJ, Conklin CA, Kozink RVet al. Hippocampal and insular response to smoking-related environments: neuroimaging evidence for drug-context effects in nicotine dependence. Neuropsychopharmacology . 2016; 41( 3): 877– 885. Google Scholar CrossRef Search ADS PubMed 12. Conklin CA, Perkins KA, Robin N, McClernon FJ, Salkeld RP. Bringing the real world into the laboratory: personal smoking and nonsmoking environments. Drug Alcohol Depend . 2010; 111( 1–2): 58– 63. Google Scholar CrossRef Search ADS PubMed 13. Stevenson JG, Oliver JA, Hallyburton MB, Sweitzer MM, Conklin CA, McClernon FJ. Smoking environment cues reduce ability to resist smoking as measured by a delay to smoking task. Addict Behav . 2017; 67: 49– 52. Google Scholar CrossRef Search ADS PubMed 14. Ahnallen CG, Tidey JW. Personalized smoking environment cue reactivity in smokers with schizophrenia and controls: a pilot study. Psychiatry Res . 2011; 188( 2): 286– 288. Google Scholar CrossRef Search ADS PubMed 15. Conklin CA, Tiffany ST. Applying extinction research and theory to cue-exposure addiction treatments. Addiction . 2002; 97( 2): 155– 167. Google Scholar CrossRef Search ADS PubMed 16. Conklin CA, Tiffany ST. Cue-exposure treatment: time for change. Addiction . 2002; 97( 9): 1219– 1221. Google Scholar CrossRef Search ADS PubMed 17. Shadel WG, Niaura R, Abrams DBet al. Scripted imagery manipulations and smoking cue reactivity in a clinical sample of self-quitters. Exp Clin Psychopharmacol . 1998; 6( 2): 179– 186. Google Scholar CrossRef Search ADS PubMed 18. McClernon FJ, Conklin CA, Kozink RVet al. Hippocampal and insular response to smoking-related environments: Neuroimaging evidence for drug context effects in nicotine dependence. Neuropsychopharmacology . 2015. www.nature.com/npp/journal/vaop/naam/abs/npp2015214a.html. Accessed September 14, 2015. 19. Burton SM, Tiffany ST. The effect of alcohol consumption on craving to smoke. Addiction . 1997; 92( 1): 15– 26. Google Scholar CrossRef Search ADS PubMed 20. Sayette MA, Martin CS, Wertz JM, Perrott MA, Peters AR. The effects of alcohol on cigarette craving in heavy smokers and tobacco chippers. Psychol Addict Behav . 2005; 19( 3): 263– 270. Google Scholar CrossRef Search ADS PubMed 21. Gulliver SB, Rohsenow DJ, Colby SMet al. Interrelationship of smoking and alcohol dependence, use and urges to use. J Stud Alcohol . 1995; 56( 2): 202– 206. Google Scholar CrossRef Search ADS PubMed 22. Rohsenow DJ, Monti PM, Colby SMet al. Effects of alcohol cues on smoking urges and topography among alcoholic men. Alcohol Clin Exp Res . 1997; 21( 1): 101– 107. Google Scholar CrossRef Search ADS PubMed 23. Shiffman S. Ecological momentary assessment (EMA) in studies of substance use. Psychol Assess . 2009; 21( 4): 486– 497. Google Scholar CrossRef Search ADS PubMed 24. Shiffman S, Terhorst L. Intermittent and daily smokers’ subjective responses to smoking. Psychopharmacology . 2017; 234( 19): 2911– 2917. Google Scholar CrossRef Search ADS PubMed 25. Perkins KA. Does smoking cue-induced craving tell us anything important about nicotine dependence? Addiction . 2009; 104( 10): 1610– 1616. Google Scholar CrossRef Search ADS PubMed 26. Tiffany ST, Wray J. The continuing conundrum of craving. Addiction . 2009; 104( 10): 1618– 1619. Google Scholar CrossRef Search ADS PubMed 27. Shiffman S. Responses to smoking cues are relevant to smoking and relapse. Addiction . 2009; 104( 10): 1617– 1618. Google Scholar CrossRef Search ADS PubMed 28. Perkins KA. Subjective reactivity to smoking cues as a predictor of quitting success. Nicotine Tob Res . 2012; 14( 4): 383– 387. Google Scholar CrossRef Search ADS PubMed 29. Payne TJ, Schare ML, Levis DJ, Colletti G. Exposure to smoking-relevant cues: effects on desire to smoke and topographical components of smoking behavior. Addict Behav . 1991; 16( 6): 467– 479. Google Scholar CrossRef Search ADS PubMed 30. Rickard-Figueroa K, Zeichner A. Assessment of smoking urge and its concomitants under an environmental smoking cue manipulation. Addict Behav . 1985; 10( 3): 249– 256. Google Scholar CrossRef Search ADS PubMed 31. Hogarth L, Dickinson A, Duka T. The associative basis of cue-elicited drug taking in humans. Psychopharmacology (Berl) . 2010; 208( 3): 337– 351. Google Scholar CrossRef Search ADS PubMed 32. Panlilio LV, Weiss SJ, Schindler CW. Cocaine self-administration increased by compounding discriminative stimuli. Psychopharmacology (Berl) . 1996; 125( 3): 202– 208. Google Scholar CrossRef Search ADS PubMed 33. Van Houten R, O’Leary KD, Weiss SJ. Summation of conditioned suppression. J Exp Anal Behav . 1970; 13( 1): 75– 81. Google Scholar CrossRef Search ADS PubMed 34. Droungas A, Ehrman RN, Childress AR, O’Brien CP. Effect of smoking cues and cigarette availability on craving and smoking behavior. Addict Behav . 1995; 20( 5): 657– 673. Google Scholar CrossRef Search ADS PubMed 35. Juliano LM, Brandon TH. Reactivity to instructed smoking availability and environmental cues: evidence with urge and reaction time. Exp Clin Psychopharmacol . 1998; 6( 1): 45– 53. Google Scholar CrossRef Search ADS PubMed 36. Heatherton TF, Kozlowski LT, Frecker RC, Fagerström KO. The Fagerström Test for Nicotine Dependence: a revision of the Fagerström Tolerance Questionnaire. Br J Addict . 1991; 86( 9): 1119– 1127. Google Scholar CrossRef Search ADS PubMed 37. Paulhus DL. Balanced inventory of desirable responding (BIDR). Accept Commit Ther Meas Package . 1988; 41. www.mindfulness-extended.nl/content3/wp-content/uploads/2013/07/Measures-Package.pdf#page=41 . Accessed September 2008. 38. Paulhus DL. Measurement and control of response bias. In: Robinson JP, Shaver PR, Wrightsman LS, eds. Measures of Social Psychological Attitudes, Vol. 1. Measures of Personality and Social Psychological Attitudes. 1991: 17– 59. 39. Diener E, Emmons RA. The independence of positive and negative affect. J Pers Soc Psychol . 1984; 47( 5): 1105– 1117. Google Scholar CrossRef Search ADS PubMed 40. Conklin CA, Parzynski CS, Salkeld RP, Perkins KA, Fonte CA. Cue reactivity as a predictor of successful abstinence initiation among adult smokers. Exp Clin Psychopharmacol . 2012; 20( 6): 473– 478. Google Scholar CrossRef Search ADS PubMed 41. Waters AJ, Shiffman S, Sayette MA, Paty JA, Gwaltney CJ, Balabanis MH. Cue-provoked craving and nicotine replacement therapy in smoking cessation. J Consult Clin Psychol . 2004; 72( 6): 1136– 1143. Google Scholar CrossRef Search ADS PubMed 42. Powell JD, West L, Powell R, Pickering JA. Erratum to: Relapse to smoking during unaided cessation: clinical, cognitive and motivational predictors. Psychopharmacology (Berl) . 2011; 215( 607): 10– 1007. 43. Janes AC, Pizzagalli DA, Richardt Set al. Brain reactivity to smoking cues prior to smoking cessation predicts ability to maintain tobacco abstinence. Biol Psychiatry . 2010; 67( 8): 722– 729. Google Scholar CrossRef Search ADS PubMed 44. Janes AC, Gilman JM, Radoman M, Pachas G, Fava M, Evins AE. Revisiting the role of the insula and smoking cue-reactivity in relapse: a replication and extension of neuroimaging findings. Drug Alcohol Depend . 2017; 179: 8– 12. Google Scholar CrossRef Search ADS PubMed 45. Niaura R, Abrams D, Demuth B, Pinto R, Monti P. Responses to smoking-related stimuli and early relapse to smoking. Addict Behav . 1989; 14( 4): 419– 428. Google Scholar CrossRef Search ADS PubMed 46. Payne TJ, Smith PO, Adams SG, Diefenbach L. Pretreatment cue reactivity predicts end-of-treatment smoking. Addict Behav . 2006; 31( 4): 702– 710. Google Scholar CrossRef Search ADS PubMed 47. Bailey SR, Goedeker KC, Tiffany ST. The impact of cigarette deprivation and cigarette availability on cue-reactivity in smokers. Addiction . 2010; 105( 2): 364– 372. Google Scholar CrossRef Search ADS PubMed 48. Carter BL, Robinson JD, Lam CYet al. A psychometric evaluation of cigarette stimuli used in a cue reactivity study. Nicotine Tob Res . 2006; 8( 3): 361– 369. Google Scholar CrossRef Search ADS PubMed 49. Perkins KA, Karelitz JL, Giedgowd GE, Conklin CA, Sayette MA. Differences in negative mood-induced smoking reinforcement due to distress tolerance, anxiety sensitivity, and depression history. Psychopharmacology (Berl) . 2010; 210( 1): 25– 34. Google Scholar CrossRef Search ADS PubMed 50. Perkins KA, Karelitz JL, Giedgowd GE, Conklin CA. Negative mood effects on craving to smoke in women versus men. Addict Behav . 2013; 38( 2): 1527– 1531. Google Scholar CrossRef Search ADS PubMed 51. Perkins KA, Giedgowd GE, Karelitz JL, Conklin CA, Lerman C. Smoking in response to negative mood in men versus women as a function of distress tolerance. Nicotine Tob Res . 2012; 14( 12): 1418– 1425. Google Scholar CrossRef Search ADS PubMed 52. Conklin CA, Perkins KA. Subjective and reinforcing effects of smoking during negative mood induction. J Abnorm Psychol . 2005; 114( 1): 153– 164. Google Scholar CrossRef Search ADS PubMed 53. Hogarth L, Mathew AR, Hitsman B. Current major depression is associated with greater sensitivity to the motivational effect of both negative mood induction and abstinence on tobacco-seeking behavior. Drug Alcohol Depend . 2017; 176( suppl C): 1– 6. Google Scholar CrossRef Search ADS PubMed 54. Conklin CA, Salkeld RP, Perkins KA, Robin N. Do people serve as cues to smoke? Nicotine Tob Res . 2013; 15( 12): 2081– 2087. Google Scholar CrossRef Search ADS PubMed Published by Oxford University Press on behalf of the Society for Research on Nicotine and Tobacco 2018. This work is written by (a) US Government employee(s) and is in the public domain in the US.
Nicotine and Tobacco Research – Oxford University Press
Published: Jan 23, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera