واکنش پذیری به نشانه های نیکوتین در طول جلسات واکنش پذیری نشانه تکرار شده

Abstract The present study investigated whether reactivity to nicotine-related cues would attenuate across four experimental sessions held 1 week apart. Participants were nineteen non-treatment seeking, nicotine-dependent males. Cue reactivity sessions were performed in an outpatient research center using in vivo cues consisting of standardized smoking-related paraphernalia (e.g., cigarettes) and neutral comparison paraphernalia (e.g., pencils). Craving ratings were collected before and after both cue presentations while physiological measures (heart rate, skin conductance) were collected before and during the cue presentations. Although craving levels decreased across sessions, smoking-related cues consistently evoked significantly greater increases in craving relative to neutral cues over all four experimental sessions. Skin conductance was higher in response to smoking cues, though this effect was not as robust as that observed for craving. Results suggest that, under the described experimental parameters, craving can be reliably elicited over repeated cue reactivity sessions.

. Introduction While craving occupies a central conceptual role in addictive behaviors research, no consensus has been reached regarding its definition (Pickens & Johanson, 1992). Despite the lack of consensus, one common feature of many definitions of craving is that it is a subjective state of desire with motivational properties that have an important role in the procurement and consumption of drugs (cf., Sayette et al., 2000) as well as relapse to drug use (Childress et al., 1988 and Wise, 1988). The cue reactivity paradigm has been used to systematically study craving (Drobes et al., 2001 and Drummond et al., 1995). In a typical cue reactivity study, cues such as drug-related pictures, movies, and/or paraphernalia are presented to participants in a controlled laboratory setting. While numerous response systems have been monitored during cue presentations, the most conceptually important of these responses is subjective craving. Generally, drug-dependent individuals report greater craving in response to cues associated with their drug of choice as compared to neutral cues (Drobes et al., 2001 and Drummond et al., 1995). Measures of craving and cue reactivity have been used within the field of addictions research to predict relapse in drug-dependent individuals as well as assess the efficacy of novel pharmacological agents for the treatment of addictions (Carter and Tiffany, 1999 and Modesto-Lowe and Kranzler, 1999). Although numerous theoretical accounts of craving and cue reactivity have been posited, most models postulate a central role for associative learning mechanisms (Childress, Ehrman, Rohsenow, Robbins, & O'Brien, 1992). These models maintain that cues that have been reliably paired with previous drug administration or withdrawal will elicit Pavlovian conditioned reactions when the cues are subsequently encountered. While conditioning models have many similar features, they differ with regard to the nature and form of the conditioned reactions. Appetitively based models assume that the conditioned reactions are consistent with the rewarding or excitatory effects of the drugs (Stewart et al., 1984 and Wise, 1988) whereas aversive or withdrawal-based accounts assume the reactions take the form of withdrawal-like responses (Poulos et al., 1981, Siegel, 1983 and Wikler, 1980). Other models suggest that cues may activate both appetitively based and withdrawal-based mechanisms that in turn contribute to the generation of cravings (Baker et al., 1987 and Niaura et al., 1988). Despite the differences between these theoretical accounts, all of the models argue that cue-elicited craving and reactivity are important motivators of drug seeking and consummatory behaviors. If cue-elicited craving and reactivity have a causal role in the maintenance of substance use (among non-treatment seeking individuals), then it should be expected that the response-eliciting properties of substance-paired cues should be maintained over time or across multiple cue reactivity assessments (i.e., if cues don’t maintain their potency over time, they aren’t likely to have an important role in addictive behavior). This expectation would also be consistent with the aforementioned learning models since active use should sustain the associative learning mechanisms that presumably underlie cue-elicited craving and reactivity. Despite decades of research, relatively few studies have examined cue reactivity over repeated laboratory sessions. A number of 2-session studies have been conducted to assess initial efficacy of various medications for the treatment of cocaine dependence. These studies have generally found evidence for a reduction in reactivity from the first to the second session (Hersh et al., 1995, Modesto-Lowe et al., 1997 and Reid et al., 1999), regardless of whether the medication under study produced any significant effects. However, with only two sessions of cue reactivity, it is difficult to know whether the reduction was the product of acclimation to the novel experimental setting and assessment procedures, or rather a genuine reduction in craving and reactivity in response to the substance cues. Moreover, since these studies had treatment components (i.e. experimental medication), it is possible that the perceived demand characteristics of the study may have led to a reduction in self-reported cravings. Within the nicotine literature, there have been relatively few studies that have investigated the cue reactivity over repeated sessions. In one such study, Drobes and Tiffany (Drobes & Tiffany, 1997) presented smoking imagery cues and in vivo cues (a confederate smoking) in one session, after which participants either continued to smoke as usual or abstained for 6 h, then completed a second cue reactivity session. Results suggested that individuals experienced greater cue-induced craving in the second session, but since the study had some participants abstain from smoking, it is unclear whether the greater second session craving was a function of this experimental manipulation. Tiffany et al. ( Tiffany, Cox, & Elash, 2000) presented in vivo cues (also involving a confederate smoking) in one session, after which participants wore either a nicotine or placebo patch and abstained for 6 h, then completed a second cue reactivity session. Results indicated increased baseline craving in the second session, but decreased cue-induced craving. Cooney et al. ( Cooney, Cooney, Pilkey, Kranzler, & Oncken, 2003) examined alcohol-dependent smokers that participated in two cue reactivity sessions held 48–72 h apart. All participants were assessed, in a counterbalanced fashion, under conditions of nicotine deprivation (approximately 36 h of abstinence) and non-deprivation (ad libitum smoking). During both sessions, participants rated their alcohol and nicotine craving in response to alcohol and alcohol-associated imagery. Results indicated that, regardless of nicotine deprivation and cue condition, there was a general reduction in self-reported craving from the first to the second session. Recently, a large 4-session cue reactivity study was completed in which slides depicting nicotine-related cues were presented ( Carter et al., 2006). Although slides were rated at the end of the study after participants had already viewed them 4 times, the smoking-related pictures evoked robust cravings. It is important to note, however, that over the course of this earlier multi-session study, participants were either deprived of nicotine for 12 h, and/or received nicotine nasal spray prior to cue presentation. Thus, it is unclear whether or not these manipulations influenced the maintenance of craving over the 4 sessions. Taken together, the long-term nature of cue-induced reactivity is unclear. As noted above, a number of cocaine studies have noted first to second session decrements. There are many potential causes for these decrements, including extinction to the cues themselves, a reduction in novelty from the first to second session, or even demand characteristics, since all of the above-cited studies had some form of treatment component. Within the nicotine literature, some multi-session studies exist, but there has typically been some form of between-session manipulation such as nicotine deprivation and/or some form of nicotine replacement during cue reactivity sessions. In contrast to these earlier studies, the present study assessed reactivity to smoking-related cues over four assessment sessions in nicotine-dependent men who where neither treatment seekers nor deprived of nicotine. This controlled, multi-session methodology not only permitted the assessment of a potential novelty-related decrement in craving and reactivity, but also allowed for evaluation of the potency of smoking related cues over a longer period of time and across a greater number of assessments under non-deprived conditions. It was expected that smoking-related cues would evoke greater cue reactivity (i.e. greater craving and physiological reactivity) relative to neutral cues across all 4 experimental sessions.

Results 3.1. Craving Means and standard errors for baseline and post cue cravings for each session are presented in Table 1. Pre-cue baseline craving was stable, ICC = 0.91, as were the post-cue scores for the neutral cues, ICC = 0.79, and smoking cues, ICC = 0.88. Table 1. Baseline and post-cue means (M) and standard errors (SE) for QSU-B ratings and range-corrected skin conductance values Session Neutral Smoking Baseline Post cue Baseline Post cue M SE M SE M SE M SE QSU-B 1 3.9 0.3 4.1 0.3 3.9 0.3 4.8 0.3 2 3.3 0.3 3.3 0.4 3.0 0.4 3.7 0.5 3 2.8 0.3 3.1 0.3 3.0 0.3 3.9 0.3 4 3.4 0.4 3.8 0.4 3.2 0.4 4.1 0.4 HRa 1 78.1 2.5 78.8 2.66 76.2 2.4 78.2 2.4 2 78.4 2.1 78.5 1.78 76.3 2.3 80.1 1.8 3 75.7 2.3 79.1 2.11 76.1 2.4 78.4 1.97 4 78.4 2.3 80.7 2.08 78.3 2.5 80.8 1.9 SCb 1 0.67 0.09 0.81 0.09 0.80 0.07 0.94 0.09 2 0.72 0.10 0.86 0.10 0.67 0.07 0.90 0.08 3 0.67 0.08 0.83 0.08 0.64 0.07 0.88 0.07 4 0.67 0.08 0.79 0.08 0.65 0.09 0.83 0.08 a Maximum HR response within 90-second segment of data collection, units are beats per minute (bpm). b Maximum SC response within 90-second segment of data collection, units are log(μSiemens + 1). Table options ANOVA results revealed that pre-cue baseline values did not differ with respect to Cue Type (p = 0.37). There was an effect for Session, F(3,54) = 5.25, p < 0.01, with Bonferroni-corrected post-hoc tests indicating that pre-cue baseline craving in Session 1, M = 3.9 (SE = 0.1), was greater than that for Sessions 2, M = 3.1 (SE = 0.1), Session 3 = 2.9 (SE = 0.1), and Session 4, M = 3.4 (SE = 0.1), t’s = 4.15, 5.14, 2.83, respectively, all p's < .05. For baseline-corrected cue-induced (post cue) craving, ANOVA results revealed both a main effect for Session, F(3,60) = 3.35, p < 0.05, and for Cue Type, F(1,18) = 9.58, p < 0.01 but no Cue Type × Session interaction (p = 0.56). Post-hoc tests revealed that the mean Session 1 craving, M = 4.2 (SE = 0.1), was greater than Session 2, M = 3.6 (SE = 0.1), and Session 3, M = 3.7 (SE = 0.1), t's = 4.14 and 3.45, p's = .01. The mean of Session 4, M = 3.9 (SE = 0.1) did not differ from that of any other session. Since this session effect was corrected for baseline responding, it follows that the observed decrement in craving to the cues (from session 1 to sessions 2 and 3) was unrelated to (or independent of) variation in baseline responding. Collapsed across all 4 sessions, mean craving in response to smoking cues, M = 4.14 (SE = 0.1) was greater than craving in response to neutral cues, M = 3.56 (SE = 0.1). When QSU-B scores for the neutral and smoking cues were compared within each session, results revealed significant differences within each session (one-tailed, Table 2), with effect sizes (partial η2) ranging between 0.15 and 0.37. Taken together, these results suggested that smoking cues evoked greater craving than neutral cues across all 4 experimental sessions. Table 2. Results of within-session comparisons of baseline-corrected means for craving and skin conductance Session Neutral Smoking F b < p ηp2 M a SE M SE QSU-B 1 4.11 0.16 4.82 0.16 10.09 0.01 0.37 2 3.27 0.13 3.77 0.13 6.90 0.05 0.29 3 3.13 0.21 3.83 0.21 5.51 0.05 0.25 4 3.71 0.16 4.12 0.16 3.10 0.05 0.15 SC 1 0.86 0.04 0.94 0.04 2.05 0.09 0.12 2 0.83 0.03 0.92 0.03 3.08 0.06 0.18 3 0.81 0.03 0.86 0.03 1.73 0.11 0.11 4 0.76 0.02 0.80 0.02 0.86 0.19 0.06 a Estimated means, SEs, and partial η2 produced by each respective within-session analyses. b Degrees of freedom are 1,18 for QSU and 1,14 for SC. p values are one-tailed. Table options It should be noted that the forgoing data analytic strategy was used to separately analyze the two factors of the QSU. Since the analyses yielded results that paralleled those described above for the QSU total scores, the findings are not presented. 3.2. Heart rate results One participant had missing data for one session and was excluded from the analysis. Means and standard errors are presented in Table 1. Pre-cue baseline HR was stable, ICC = 0.84, as were the post scores for the neutral cues, ICC = 0.78, and smoking cues, ICC = 0.85. Pre-cue baseline values showed no effects for Session (p = 0.74) or Cue Type (p = 0.34). For baseline corrected cue-induced HR reactivity, no effects for Session (p = 0.83) or Cue Type (p = 0.39) were noted. 3.3. Skin conductance results Due to equipment errors, 3 participants had missing data from at least one session and were excluded from the analysis. For the remaining participants, means and standard errors of the log-transformed SC scores are presented in Table 1. Pre-cue baseline SC was stable, ICC = 0.87, as were the post-cue scores for the neutral cues, ICC = 0.81, and smoking cues, ICC = 0.78. ANOVA findings for pre-cue baseline SC revealed no effect for Session (p = 0.68) or Cue Type (p = 0.63). For baseline-corrected cue-induced SC, there was no effect for Session (p = 0.22), nor was the Session by Cue Type interaction significant (p = 0.74). A main effect for Cue Type was observed, F(1,15) = 9.68, p = 0.01. The mean baseline-corrected score for smoking cues, M = 0.88, SE = 0.02, was greater than that for neutral cues, M = 0.82, SE = 0.02. In contrast to the findings for craving ratings, however, the main effect for Cue Type was only apparent when taking into account the data from all four sessions; significant differences were not found when the data from each session were analyzed separately (one-tailed, Table 2).