نوشیدن برای تقویت منفی: انگیختگی معنایی مفاهیم الکل

Cognitive models of alcohol abuse posit that the context typically associated with alcohol use, such as negative affect, implicitly activates alcohol use cognitions, which in turn leads to alcohol consumption. We selected 40 undergraduate women based upon their alcohol use and reported anxiety sensitivity, and proposed that drinking for the purpose of negative reinforcement would predict increased semantic priming between anxiety and alcohol concepts. A lexical decision task compared the response latencies of alcohol targets preceded by anxiety words to those same targets preceded by neutral words (anxiety–alcohol priming). Level of anxiety sensitivity did not relate to anxiety–alcohol priming, but drinking following social conflict was associated with increased anxiety–alcohol priming. This study specifically suggests that the contextual antecedents to drinking behavior relate to the organization of semantic information about alcohol, and more generally supports cognitive models of substance abuse.

The implicit cognition theory (ICT) of drug urges and drug use behavior assumes that implicit memory associations are central to understanding alcohol use. Based upon neural network or connectionist models of cognition (e.g., Collins & Loftus, 1975), ICT posits that alcohol-related contextual cues are held within automatized action-use schemas (Tiffany, 1990), which develop after multiple pairings between contextual cues and alcohol use. As individuals encounter contextual cues associated with alcohol use, these implicit memory schemas are brought forth to subsequently influence behavior without awareness (Wiers & Stacy, 2006). Thus, ICT argues that cognitions are key to understanding drug use behavior, as cognitions mediate the relationship between affective states and substance use (see review, Birch, Stewart, & Zack, 2006). There has been increased interest in applying ICT and its measurements to alcohol abuse for several reasons. First, evidence suggests that implicit tasks are not as subject to demand characteristics or participant self-presentation bias as are explicit measures (Wiers et al., 2002 and Wiers et al., 2002). Additionally, implicit measures may show incremental validity over explicit measures in situations in which behavior is more difficult to control, or in which motivation to control that behavior is low (e.g., alcohol abuse or dependence; Fazio and Olson, 2003 and Ostafin and Palfai, 2005). 1.1. Means of measuring implicit cognitions Cognitive psychologists have developed specific experimental methods, such as Stroop designs, Implicit Association Tests (IATs), the Extrinsic Affective Priming Task (EAST) and lexical decision tasks to measure implicit cognitions (DeHouwer, 2003). These methods are thought to measure underlying implicit associative memory networks, although there is some speculation that only lexical decision tasks do so directly (e.g., Gawronski & Bodenhause, 2005). Lexical decision tasks utilize a semantic priming paradigm to measure the associative strength between two words. Participants must make a lexical decision (i.e., decide if a word is indeed an English term) for a target word that is preceded by a prime word. The lexical decision response latency is typically shorter (faster) when the target word is preceded by highly associated primes (e.g., cat–dog) than less associated primes (e.g., tree–dog; see review, Neely, 1991). Semantic priming refers to the response latency between the prime–target pair, with shorter latencies reflecting increased priming or a stronger association between two concepts, and longer latencies reflecting decreased semantic priming or a weaker association between two concepts (see review, Neely, 1991). This paradigm has known-groups validity in that it can distinguish between alcohol dependent and non-dependent individuals (Hill & Paynter, 1992). 1.2. Using ICT to explain drinking in response to negative affect While numerous studies have found support for ICT when differentiating problem and non-problem drinkers (e.g., Hill and Paynter, 1992, Palfai and Ostafin, 2003 and Wiers et al., 2002), less research has utilized implicit measures, as described above, to investigate differences between problem drinking individuals who report distinct motives or antecedents for alcohol consumption. For a large subgroup of drinkers the experience of negative affect, including anxiety symptoms, serves as a cue for alcohol consumption (see review, Cooper, Frone, Russell, & Mudar 1995). Individuals who drink in response to negative affect are at an increased risk of developing problematic patterns of alcohol use and alcohol use diagnoses (Kushner, Abrams, & Borchardt, 2000). Individuals who report this style additionally demonstrate increased alcohol consumption levels (Park & Levenson, 2002), binge drinking behavior (Ichiyama and Kruse, 1998 and Park and Levenson, 2002) and drinking for the purpose of intoxication (Tyssen, Vaglum, Aasland, Gronvold, & Ekeberg, 1998). Furthermore, college students who report using alcohol to regulate negative affect may be more likely to maintain or develop alcohol-related problems after leaving school (Park & Levenson, 2002). Thus, understanding factors, such as implicit cognitions, that may relate to drinking in response to negative affect is an issue of extreme importance. According to ICT, drinking in response to negative affect should create an automatic association between negative affect and alcohol-related cognitions, which in turn leads to drinking behavior. To date, two studies have found support for ICT’s predictions when examining individuals who drink during experiences of negative affect (see review, Birch et al., 2006). Zack, Toneatto, and Macleod (1999) examined individuals diagnosed with an alcohol use disorder and found that high psychiatric distress and drinking during negative mood states each predicted faster lexical decision time (increased priming) for negative affect–alcohol pairs in a semantic priming task. Similarly, Stewart, Hall, Wilkie, and Birch (2002) selected individuals who reported an extreme tendency to use alcohol to alleviate negative emotions (coping-motivated drinkers). Utilizing a Stroop task, they found that coping-motivated drinkers showed the predicted longer naming latencies (more interference) for negative affect, alcohol pairs (e.g., sad–beer) than for negative affect, non-alcohol pairs (e.g., sad–tree). In all, these studies provide preliminary evidence that individuals who drink in response to negative moods may implicitly associate negative affect with alcohol, although this effect may be specific to the implicit task used and/or the sample population studied. 1.3. ICT and anxiety sensitivity Anxiety sensitivity, or the propensity to attribute a negative interpretation to feelings of anxiety, appears to be related to specific drinking motives and may be an especially important predictor of drinking in response to negative affect (Stewart, Zvolensky, & Eifert, 2001). Individuals with high anxiety sensitivity are more likely to report drinking during negative affective states than individuals with low anxiety sensitivity (Reyno et al., 2006 and Samoluk and Stewart, 1998). As noted above, ICT would predict that the tendency of anxiety sensitive drinkers to consume alcohol in response to negative affect should cause anxiety-related concepts to automatically activate alcohol-related cognitions (Zack, Poulos, Fragopolous, & Macleod, 2003). That is, individuals high in anxiety sensitivity should demonstrate a stronger linkage between anxiety concepts and alcohol concepts than between neutral concepts and alcohol concepts. Thus far, only one study has examined anxiety–alcohol associations in an anxiety sensitive population. In an innovative design, Zack et al. (2003) had college-age drinkers complete a phrase priming task, where negative and positive mood-related phrases served as primes for alcohol target words. Although this study found that self-reported tendency to drink in “bad” moods predicted negative mood phrase priming, the study did not find any association between negative mood phrase priming and anxiety sensitivity level. However, there were a few limitations to this study. First, a “drinker” in this study was an individual who had consumed one or more alcoholic drinks in the last week. It is unclear whether underclassmen in the high anxiety sensitivity condition had sufficient drinking experience to build an association between negative mood and alcohol. Secondly, individuals in the high anxiety subject group only needed to be at or above the 70th percentile for their anxiety sensitivity score. Recent data indicates that anxiety sensitivity may be taxonic in nature and may only affect a small proportion of the population (Bernstein et al., 2006), and thus it is unclear whether this study utilized a truly anxiety sensitive sample given their more liberal group criterion scores. 1.4. The relationship between gender, anxiety sensitivity, and alcohol use As noted, anxiety sensitivity is a risk factor for drinking in response to negative affect, and drinking in response to negative affect is associated with subsequent alcohol use disorder development (Kushner et al., 2000 and Novak et al., 2000). However, there is some evidence that this relationship may be gender specific. Stewart et al. (2001) found that women with high anxiety sensitivity levels were more likely to report drinking in an effort to cope than were their less anxiety sensitive counterparts, but comparable results were not found for men. Likewise, Zack et al. (2003) found that female drinkers with high anxiety sensitivity were more likely to drink in negatively reinforcing situations than females with low anxiety sensitivity. Again, this pattern of results did not hold for male participants. Given the unique role that anxiety sensitivity plays in drinking in response to negative affect, additional research should examine these issues within a female sample. More generally, there is comparatively less research on factors that may cause or maintain alcohol use in women than in men, despite the evidence that women may be more vulnerable to the effects of alcohol (National Institute on Alcohol Abuse and Alcoholism; NIAAA, 2004). Female drinkers are at a greater risk for sexual trauma, organ damage, suicide, motor vehicle accidents and interpersonal and legal problems than male drinkers. The study of factors which maintain drinking in young women, who ideally have not yet developed these severe consequences, is an issue of extreme clinical significance (NIAAA, 2004). Due to the paucity of research on correlates of alcohol use in young women, the current study chose to utilize a within-gender approach. The current approach allows for an understanding of factors within a female sample that may predict specific drinking behaviors, and thus acknowledges within-sex variability in drinking correlates and behaviors. 1.5. The current study The current study sought to answer three questions. First, it examined whether alcohol use and level of anxiety sensitivity would predict priming between anxiety and alcohol concepts (anxiety–alcohol priming). Secondly, this study sought to examine whether drinking context (drinking in negatively reinforcing situations) would influence priming between anxiety and alcohol concepts. Finally, this study examined whether a specific drinking context (i.e., drinking in negatively reinforcing situations) would mediate the relationship between anxiety sensitivity and anxiety–alcohol priming. To examine these three questions the authors utilized an auditory priming modality, rather than a visual priming modality. Evidence suggests that auditory semantic priming effects are more robust (as measured by response latency) than visual priming effects (Holcomb & Neville, 1990), thereby allowing for the detection of more subtle effects. The predictions were as follows: 1) The magnitude of negative–alcohol priming would be greater for problem drinking females with high anxiety sensitivity (hi drink/hi anx) than for problem drinking females with low anxiety sensitivity (hi drink/lo anx); 2) Drinking in negatively reinforcing situations would be associated with increased anxiety–alcohol priming (shorter response latencies); 3) Drinking in negatively reinforcing situations would mediate the relationship between anxiety sensitivity and anxiety–alcohol priming for hi drink/hi anx participants.