The relationship between drinking motives and alcohol-related interpretation biases

J. Behav. Ther. & Exp. Psychiat. 47 (2015) 102e110

Contents lists available at ScienceDirect

Journal of Behavior Therapy and Experimental Psychiatry journal homepage: www.elsevier.com/locate/jbtep

The relationship between drinking motives and alcohol-related interpretation biases Marcella L. Woud a, *, 1, Eni S. Becker a, Mike Rinck a, Elske Salemink b a b

Behavioural Science Institute, Radboud University Nijmegen, Montessorilaan 3, 6525 HR Nijmegen, The Netherlands ADAPT lab, University of Amsterdam, Weesperplein 4, 1018 XA Amsterdam, The Netherlands

a r t i c l e i n f o

a b s t r a c t

Article history: Received 30 May 2014 Received in revised form 18 November 2014 Accepted 19 November 2014 Available online 26 November 2014

Background and objectives: Numerous studies have investigated drinking motives and alcohol-related interpretation biases (IBs) separately. However, less is known about the relationship between them. Therefore, the present study examined whether coping and enhancement drinking motives were specifically related to negative and positive alcohol-related IBs, respectively. Furthermore, it was investigated whether such biases predict future drinking, especially in individuals with low levels of executive control (EC). Methods: Participants were male and female university students. The Drinking Motives QuestionnaireRevised (DMQ-R; Cooper, 1994) was administered to measure participants' drinking motives. To measure alcohol-related IBs, an adapted version of the Encoding Recognition Task (ERT) was used. During the ERT, participants were asked to read ambiguous alcohol-related scenarios. In a subsequent recognition phase, participants interpreted these scenarios. A classical Stroop was applied to assess levels of EC. Results: Coping motives but not enhancement motives were a unique predictor of the tendency to interpret negatively valenced ambiguous alcohol-relevant situations in an alcohol-related manner. This relationship was significant even when controlling for other relevant predictors. Neither coping nor enhancement motives were predictive of positive alcohol-related IBs. Concerning the prediction of prospective drinking, results showed that particularly the negative alcohol-related IB predicted prospective drinking. However, EC did not moderate the prediction of prospective drinking by either positive or negative interpretation biases. Limitations: The alcohol-ERT might not be the most optimal paradigm for assessing implicit alcoholrelated IBs. Conclusions: The present results emphasize the role of negative affect in the context of drinking motives and alcohol-related IBs. Follow-up studies are needed to test the robustness of these findings, and to further explore the general interplay between drinking motives and alcohol-related IBs. © 2014 Elsevier Ltd. All rights reserved.

Keywords: Drinking motives Alcohol-related interpretation bias Implicit memory associations Executive control Prospective drinking

1. Introduction Research into the contribution of (cognitive) factors related to alcohol misuse and abuse has been characterized by work from various theoretical perspectives. In the area of drinking motives,

* Corresponding author. Present address: Center for the Study and Treatment of Mental Health, Department of Psychology, Ruhr University Bochum, Germany, Massenbergstraße 9-13, 44787 Bochum, Germany. Tel.: þ49 (0)234/32 21502. E-mail address: [email protected] (M.L. Woud). 1 This study was conducted when Marcella L. Woud was still at Behavioural Science Institute, Radboud University Nijmegen, Montessorilaan 3, 6525 HR Nijmegen, The Netherlands. http://dx.doi.org/10.1016/j.jbtep.2014.11.012 0005-7916/© 2014 Elsevier Ltd. All rights reserved.

Cooper and colleagues provided a number of significant contributions, among them the development of the motivational model of alcohol use (Cooper, 1994; Cooper, Frone, Russell, & Mudar, 1995). According to this model, drinking is conceptualized as behavior which is motivated by the need for affect regulation. That is, people drink to reduce negative affective states or enhance positive affective states. As such, drinking motives reflect underlying motivational reasons for alcohol use. Cooper characterizes drinking motives in terms of their reinforcing nature (positive or negative reinforcement) and their source (internal or external), resulting in four drinking motives. The two internal affective motives for alcohol use are coping motives (i.e., drinking to decrease negative affect) and enhancement motives (i.e., drinking to increase positive

M.L. Woud et al. / J. Behav. Ther. & Exp. Psychiat. 47 (2015) 102e110

affect). The two external motives for alcohol use are social and conformity motives (for a review on drinking motives, see Kuntsche, Knibbe, Gmel, & Engels, 2005). A questionnaire suitable for detecting individual differences in drinking motives is the Drinking Motives Questionnaire-Revised (DMQ-R; Cooper, 1994). The DMQ-R has been useful in identifying, for example, students who are at risk to engage in patterns of heavy and/or problematic alcohol use. To illustrate, a study by Magid, MacLean, and Colder (2007) showed that conformity motives were significantly related to alcohol problems, which is line with predictions of Cooper's model. However, most research in the area of drinking motives has been focused on the two internal affective motives, i.e., coping and enhancement. Here, research has repeatedly shown that both coping and enhancement drinking are associated with heavy drinking and alcohol-related problems, although only coping drinking retains its predictive validity for levels of problem drinking after controlling for alcohol consumption (e.g., Cooper, 1994; Grant, Stewart, & Mohr, 2009; Kuntsche, Knibbe, Gmel, & Engels, 2006; Kuntsche et al., 2005; Stewart & Devine, 2000). A rather cognitive perspective on alcohol misuse and abuse is provided by dual process models of addition (e.g., Deutsch & Strack, 2006; Wiers et al., 2007; for critical discussion see Gladwin & Figner, 2014; Gladwin, Figner, Crone, & Wiers, 2011), inspired by dual process models of social behavior (e.g., Evans, 2003; Gawronski & Bodenhausen, 2006; Strack & Deutsch, 2004). According to dual process models of addiction, two types of processes are involved in alcohol use. Please note that specific models differ in their details, but these types of processes are broadly defined as follows. On the one hand, there are fast, impulsive processes that influence alcohol use via associative processes that operate automatically (e.g., via associative networks). These processes can be summarized best as implicit processes. On the other hand, there are slower, more deliberate and reflective processes that influence alcohol use through goal regulation and conscious deliberation. These processes can be summarized best as reflective processes. Given these functional properties, reflective processes but not implicit processes are assumed to depend heavily on the individual's levels of Executive Control (EC) functions (for an in-depth elaboration on the different facets of EC, see Miyake, Friedman, Emerson, Witzki, & Howerter, 2000). According to dual process models, reflective processing can be overruled by implicit processes, implying that (alcohol-related) behavior will be predominantly guided by implicit processes. However, in case an individual has the motivation and cognitive resources, the impact of implicit processes can be regulated. Usually, implicit alcohol-related processes are measured via various subtle biases in behavior rather than via explicit responses on questionnaires (e.g., Deutsch & Strack, 2006; Evans, 2003; Wiers et al., 2007; for review, see Wiers & Stacy, 2006; Stacy & Wiers, 2010). The study of alcohol-related interpretation biases (IBs) has received a great amount of interest in this context. Well-validated paradigms to assess such biases are word association tasks (for a review, see e.g., Stacy, Ames, & Grenard, 2006). During these tasks, participants are asked to give their first, spontaneous association to a variety of ambiguous cues. As these tasks do not directly mention the targeted behavior, such ‘tests are indirect and may have the capability of assessing implicit processes’ (Stacy & Wiers, 2010, p. 7). To illustrate, single ambiguous words, ambiguous phrases containing affective components and open-ended ambiguous scenarios have been used (e.g., Ames & Stacy, 1998; Ames, Sussman, Dent, & Stacy, 2005, 1997; Stacy et al., 2006; Woud, Fitzgerald, Wiers, Rinck, & Becker, 2012; Woud et al., 2014). It has been repeatedly found that individuals who tend to interpret ambiguous information in an alcohol-related manner consume more alcohol. According to associative network accounts (e.g., Anderson, 1983;

103

Bower, 1981; Collins & Loftus, 1975), this can be explained as follows: During the processing of ambiguous alcohol-related cues, alcohol-related memory schemata that are established during previous alcohol experiences become active. These schemata elicit alcohol-related memory associations, and these are particularly strong, i.e., easily accessible, in individuals who consume a lot of alcohol. As such, individuals with high levels of alcohol consumption generate more alcohol-related interpretations when being confronted with ambiguous alcohol-related cues. This activation is thus rather automatic without conscious deliberation and reflection. In turn, of course, it is also likely that exhibiting strong alcohol-related interpretations could also activate alcohol consumption and/or craving for alcohol. To date, there has been little research investigating the relationship between drinking motives and alcohol-related IBs. The study of Salemink and Wiers (2014) is, to our knowledge, the only study which has addressed this topic so far. They applied a WordeSentence Association Paradigm (WSAP; Beard & Amir, 2009) and investigated whether coping and enhancement drinking motives were specifically related to negative and positive alcoholrelated IBs, respectively. During the WSAP, alcohol-related or alcohol-unrelated words were presented, followed by an ambiguous sentence involving a negative or a positive affect situation, or a neutral control situation. Participants had to indicate whether word and sentence were related, and their choice served as the dependent variable for calculating alcohol-related IBs. Results showed that coping motives (but not enhancement motives) predicted the tendency to associate alcohol-related words with negative affect situations, while enhancement motives (and not coping motives) predicted the tendency to associate alcohol-related words with positive affect situations (for studies that addressed the relation between drinking motives and other implicit alcohol-related cognitive processes, see e.g., Birch et al., 2008; Stewart, Hall, Wilkie, & Birch, 2002; Palfai & Wood, 2001; Ralston, Palfai, & Rinck, 2013). Given the important role of drinking motives and alcoholrelated IBs in the context of alcohol abuse and misuse, the scarcity of research relating them to each other served as the main motivation for the present study. Both concepts are clearly related to heavy and/or problematic levels of alcohol consumption (for recent meta-analyses, see Reich, Below, & Goldman, 2010; Rooke, Hine, & Thorsteinsson, 2008). Hence, studies are needed that further disentangle the functional relationship of and interplay between drinking motives and alcohol-related IBs. Such work could provide an important empirical basis for identifying individuals at risk for excessive drinking. Furthermore, it could inspire the development of prevention programs which would be crucial for adolescents with risk profiles. Here, the work of Conrod, Castellanos, and Mackie (2008) could provide a useful and inspiring base. They developed an intervention program whereby at risk individuals are matched to preventative interventions on the basis of their personality characteristics. This approach could be extend and adapted. That is, programs could be developed that target drinking motives and at risk individuals could be matched on the basis of their enhancement or coping drinking motive. The aim of the present study is twofold. First, the present study sought to further examine the association between and interplay of drinking motives and alcohol-related IBs. Second, the present study sought to further examine the predictive power of alcohol-related IBs on prospective drinking behavior, as well as the moderating role of executive control. As such, the present study can be regarded as a conceptual replication and extension of Salemink and Wiers (2014): We applied different measures but aim to examine the same processes (i.e., drinking motives, alcohol-related IBs and EC). To assess alcohol-related IBs, the present study employed an

104

M.L. Woud et al. / J. Behav. Ther. & Exp. Psychiat. 47 (2015) 102e110

adapted version of the Encoding Recognition Task (ERT) (for a validation study, see Salemink & van den Hout, 2010). During the alcohol-ERT, participants were presented with ambiguous alcoholrelated scenarios and were asked to rate possible interpretations of these scenarios in a second phase. Half of the scenarios described negative situations and the other half described positive ones. Concerning the first aim of the study, we expected that DMQ-R coping motives would specifically predict the tendency to interpret negative ambiguous situations in an alcohol-related manner, while DMQ-R enhancement motives would specifically predict the tendency to interpret positive ambiguous situations in an alcoholrelated manner. The theoretical foundation for this prediction is derived from assumptions put forward by dual process models of addiction (Deutsch & Strack, 2006; Wiers et al., 2007), and in particular from the mood-memory-congruency perspective which is incorporated in associative network accounts (e.g., Anderson, 1983; Bower, 1981; Collins & Loftus, 1975). According to this mood-memory-congruency perspective, not only ambiguous alcohol-related cues have the potential to activate alcohol-related memory schemata, but also emotional cues (for similar reasoning, see e.g., Baker, Morse, & Sherman, 1987; Baker, Piper, McCarthy, Majeske, & Fiore, 2004; Koob & Le Moal, 2001). For example, if an individual consistently consumes alcohol in response to negative affective states, there is a strong association between negative affect and alcohol. Hence, when confronted with negative, ambiguous alcohol-related sentences, alcohol-related memory schemata should become active. Following the mood-memory-congruency perspective, this activation should be stronger than when this individual is confronted with positive, ambiguous alcohol-related sentences. Consequently, one would expect that coping motives are specifically related to negative but not positive alcohol-related IBs. In contrast, enhancement motives should be specifically related to positive but not negative alcohol-related IBs. Concerning the second aim of the present study, we expected that alcohol-related IBs in negative and positive affect situations would be predictive of prospective drinking behavior, particularly among those with lower levels of EC. Again, the theoretical foundation for this prediction is derived from dual process models of addiction (Deutsch & Strack, 2006; Wiers et al., 2007), and in particular their predictions concerning the role of EC. Here, it is assumed that deficits in EC may be a risk factor for, and a consequence of, alcohol abuse and misuse (for more information, see e.g., De Wit, 2009; Koob & Volkow, 2010; Verdejo-García, Lawrence, & Clark, 2008). Indeed, research has shown that for low levels of control, reflective factors become irrelevant and alcohol use is best predicted by impulsive processes (e.g., Grenard et al., 2008; Thush et al., 2008). Moreover, Salemink and Wiers (2014) showed that alcohol-related IBs in positive affect situations were predictive of prospective alcohol use and number of binges, depending on the individuals' level of working memory capacity. To summarize, the present study tested whether coping and enhancement drinking motives were uniquely related to negative and positive alcohol-related IBs, respectively. Furthermore, it was examined whether IBs would be predictive of future drinking, especially in individuals with low levels of EC. In both analyses, we controlled for the effects of other relevant predictors in this context (e.g., social and conformity drinking motives). 2. Methods 2.1. Participants In total, 119 students of Radboud University Nijmegen participated (inclusion criteria: having consumed alcohol in the past week, university student, age 18e28 years). Eleven participants

were excluded from all further analyses: Eight did not consume alcohol and three were tested despite not meeting the age requirement. The final sample (n ¼ 108) consisted of 57 females and 51 males with a mean age of 21.2 years (SD ¼ 2.8). 2.2. Materials 2.2.1. Self-report measures 2.2.1.1. Time-Line-Follow-Back questionnaire (TLFB). During the lab session, alcohol use was measured with an adapted version of the TLFB questionnaire (Sobell & Sobell, 1992; Wiers, Hoogeveen, Sergeant, & Gunning, 1997). Participants summed up for every day of the past week, how many and what kind of standard alcoholic drinks they had consumed, and how many drinks they would typically drink on an average day. These responses were used to calculate a drinking mean score per participant, for both the actual and average number of drinks. Per participant, both mean scores were then used to calculate an overall mean score. In addition, participants indicated the number of occasions on which they had drunk five or more standard glasses of alcohol during the previous 2 weeks. Alcohol use in the week following the lab session was measured with a shorter online version of the TLFB. Here, participants were asked to indicate how many glasses of alcoholic beverages they had consumed during each day of the past week. 2.2.1.2. Drinking Motives Questionnaire e Revised (DMQ-R). The DMQ-R (Cooper, 1994) is a 20-item self-report questionnaire assessing the relative frequency of drinking for each of the four drinking motives, including a scale ranging from 1 (almost never/ never) to 5 (almost always/always). Of primary interest here were the drinking motives ‘enhancement’ and ‘coping’. For example: some coping items were: ‘To forget your worries’, ‘To forget about your problems’. Some enhancement items were: ‘To get high’, ‘Because it gives you a pleasant feeling’. The analyses also included the two other drinking motives, i.e., social and conformity. A sum score for each of the four scales was created per participant. Internal consistency for the present sample was acceptable: coping: Cronbach's a ¼ .75; enhancement: Cronbach's a ¼ .79: social: Cronbach's a ¼ .79; conformity: Cronbach's a ¼ .70. 2.2.1.3. Alcohol Use Disorders Identification Test (AUDIT). The AUDIT (Saunders, Aasland, Babor, De La Fuente, & Grant, 1993) is a selfreport questionnaire measuring alcohol use and levels of hazardous drinking. It comprises a multiple-choice answering format and consists of 10 questions (3 related to alcohol use, 7 to alcoholrelated problems). A sum score of all 10 questions was created per participant. Internal consistency for the present sample was acceptable, Cronbach's a ¼ .71. 2.3. Stroop The classical color-interference Stroop task (Stroop, 1935) was used as an indicator of EC as previous studies successfully used this task as a moderator (e.g., Houben & Wiers, 2009; Wiers, Beckers, Houben, & Hofmann, 2009). During this task, participants categorized stimuli according to their print color. On the compatible card, the words ‘yellow’, ‘green’, ‘red’, ‘blue’ and ‘white’ were shown, and the word's meaning matched its print color. The same words were shown on the incompatible card; however, now the word's meaning did not match its print color (e.g., the word ‘red’ was printed in green). Finally, there was a card that showed meaningless colored strings of XXX. The cards were presented randomly. Each card contained 40 stimuli, i.e., 8 stimuli distributed across 5 columns. Each card appeared on the screen after a mouse click initiated by the experimenter. As soon as the participant had

M.L. Woud et al. / J. Behav. Ther. & Exp. Psychiat. 47 (2015) 102e110

named the last word's print color, the experimenter clicked again and the card disappeared. Reaction times were saved by the computer, for each card separately, and these reaction times were used in the analyses. Participants' errors were recorded by the experimenter who was blind to the type of card which was presented. 2.4. Encoding Recognition Task (ERT) The ERT consisted of two parts. In part one, 24 ambiguous alcohol-related scenarios (12 positive, 12 negative) were presented on the computer screen. Their contents were modeled on items of the coping and enhancement scales of the DMQ-R (Cooper, 1994). Each scenario contained a title and 3 lines. In the final line, a word was missing, and this missing word appeared as soon as the space bar was pressed. However, the missing word was presented as a word fragment which participants were asked to complete. Specifically, participants were instructed to press the space bar again as soon as they knew the complete word. Then, they were supposed to type in the first missing letter of the word fragment. If correct, the complete word appeared on the screen (see Table 1 for examples). If incorrect, an error message appeared on the screen (a red X), and participants had the option to enter another letter. Once the word fragment was completed correctly, the program moved on to a new scenario. To encourage thorough processing of the scenarios' meaning, a comprehension question was included after each scenario which participants had to answer with ‘yes’ or ‘no’. In case participants answered the question incorrectly, an error message appeared on the screen as well. However, here participants were

Table 1 Example of a negative and positive scenario during the Encoding Recognition Task (ERT). Scenario type

Title

“Wanting to forget your problems”

Negative

Encoding sentence

“You are agitatedly walking up and down your room. You are very dissatisfied with your present situation. Your study loan piles up, you failed an exam and now you lost your wallet. You just want to forget all your … (pro_lems/ problems)

Recognition sentences: Alcohol-related target

Alcohol-unrelated target Foil 1 Foil 2

“You want to forget all your problems and decide to have a few extra drinks tonight.” “You want to forget all your problems and decide to go for a long run.” “You want to forget all your problems and grab your jacket.” “You want to forget all your problems and you are going outside.”

Scenario type

Title

“Celebrating your housemate's birthday.”

Positive

Encoding sentence

“Your housemate is celebrating his birthday and invited many people. The party gets going quickly and the atmosphere is great. You want to get in the … even more. (m_od/mood).”

Recognition sentences: Alcohol-related target Alcohol-unrelated target Foil 1 Foil 2

“You want to get in the mood even more, so you get a bottle of beer.” “You want to get in the mood even more, so you turn up the music.” “You want to get in the mood even more, so you stand up from the table.” “You want to get in the mood even more, so you turn towards the person sitting next to you.”

105

not required to correct their answer. Finally, participants were asked to indicate on a seven-point Likert scale ranging from 1 (not at all) to 7 (completely) how well they had been able to imagine themselves in the presented scenario (M ¼ 4.88, SD ¼ .71). In part two, the original 24 scenario titles from part one were presented randomly, each followed by a set of 4 sentences. For each sentence, participants had to indicate how close in meaning the sentence was to the original meaning of the scenario, using a fourpoint Likert scale (1 ¼ ‘not at all’ to 4 ¼ ‘very much’). Of the four sentences, one represented a possible alcohol-related interpretation of the original ambiguous scenario (i.e., alcohol-related target sentence), and one represented a possible alcohol-unrelated interpretation (i.e., alcohol-unrelated target sentence/control sentence). In addition, two more sentences were presented. Their contents were broadly similar to the original scenario but did not represent a resolution of the ambiguity (i.e., foil sentences) (see Table 1). Per participant, similarity ratings per sentence type were collapsed into four mean scores, i.e., one mean score for each sentence type. Internal consistency for the target sentences of the present sample was acceptable to good: negative alcohol-related target: Cronbach's a ¼ .83; positive alcohol-related target: Cronbach's a ¼ .83: negative alcohol-unrelated target: Cronbach's a ¼ .72; positive alcohol-unrelated target: Cronbach's a ¼ .75. 2.5. Procedure Upon arrival, participants gave informed consent and answered demographic questions. This was followed by the Stroop task and ERT. Finally, the TLFB, DMQ-R, and AUDIT were administered. One week later, participants received an email containing a link to complete the online version of the TLFB questionnaire. 3. Results 3.1. Correlations self-report measures and alcohol-related IBs We computed correlations between all four drinking motives (DMQ-R scores of coping, enhancement, social and conformity motives) and negative and positive alcohol-related IBs (scores on alcohol-related target sentences), negative and positive control sentences (scores on alcohol-unrelated target sentences), foil sentences, alcohol consumption during the past seven days (TLFB lab version), level of hazardous drinking (AUDIT scores), alcohol consumption during the week following the lab session (TLFB online version), executive control (Stroop index) and gender (see Table 2). Regarding correlations with the TLFB lab version, one participant was excluded due to an extreme value on this measure (3 SDs above the group's mean). Results showed that coping and enhancement drinking motives correlated significantly with the negative alcohol-related IB. Moreover, the negative alcohol-related IB was associated with social drinking motives, levels of alcohol consumption during the past seven days, levels of hazardous drinking, and prospective drinking. That is, individuals who had a stronger tendency to interpret negative affect situations as alcohol-related more often drank in response to negative as well as positive emotions, had consumed more alcohol during the past 7 days, showed higher levels of hazardous drinking and drank more during the seven days following the lab session. As expected, scores on control sentences in negative affect situations did not significantly correlate with any of the selfreport measures. Enhancement motives correlated significantly with the alcohol-related IB in positive affect situations. Moreover, the positive alcohol-related IB was associated with social drinking motives, levels of alcohol consumption during the past 7 days, levels of hazardous drinking, and prospective drinking. That is,

106

M.L. Woud et al. / J. Behav. Ther. & Exp. Psychiat. 47 (2015) 102e110

Table 2 Correlations between alcohol-related interpretation bias in positive and negative affect situations, drinking motives, alcohol use (TLFB lab version), levels of harmful drinking (AUDIT scores), prospective drinking (TLFB online version), executive control (Stroop Index) and gender and means and standard deviations per measure. N ¼ 108. 1. 1. ART negative affect 2. ART positive affect 3. AUT negative affect 4. AUT positive affect 5. FS negative affect 6. FS positive affect 7. DMQ-R coping 8. DMQ-R enhancement 9. DMQ-R social 10. DMQ-R conformity 11. TLFB Ia 12. AUDIT 13. TLFB IIb 14. Stroop index 15. Gender M SD

e .70** .20* .29** .04 .07 .27** .25** .21* .10 .31** .25** .34* .14 .16 2.12 .55

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

e .27** .54** .13 .14 .16y .25* .34** .07 .27** .21* .22* .03 .09 2.72 .58

e .64** .50** .46** .16 .05 .00 .07 .08 .08 .03 .13 .19y 2.77 .48

e .44** .34** .05 .09 .02 .05 .08 .13 .02 .10 .13 3.00 .49

e .73** .04 .04 .05 .02 .08 .09 .10 .17y .10 2.73 .44

e .06 .01 .02 .01 .04 .00 .10 .06 .07 2.80 .41

e .27* .31** .35** .10 .25* .16 .01 .05 8.69 2.91

e .6** .12 .37** .45** .43** .02 .04 13.55 3.89

e .21* .23* .33** .32** .02 .04 15.87 3.67

e .07 .05 .15 .12 .08 6.68 2.19

e .66** .72** .04 .26* 12.86 8.72

e .67** .09 .11 9.99 4.18

e .04 .13 1.83 1.74

e .27* 10.77 5.59

e 57\ 51_

Significances: yp < .1, *p < .05, **p < .001. ART Negative Affect: Alcohol-Related Target Negative Affect Situations; ART Positive Affect: Alcohol-Related Target Positive Affect Situations; AUT Negative Affect: Alcohol-Unrelated Target Negative Affect Situations; AUT Positive Affect: Alcohol-Unrelated Target Positive Affect Situation; FS Negative Affect: Foil Sentences Negative Affect; FS Positive Affect: Foil Sentences Positive Affect; DMQ-R Coping: Coping subscale Drinking Motives Questionnaire Revised; DMQ-R Enhancement: Enhancement subscale Drinking Motives Questionnaire Revised; DMQ-R Social: Social subscale Drinking Motives Questionnaire Revised; DMQ-R Conformity: Conformity subscale Drinking Motives Questionnaire Revised; TLFB I: Time Line Follow Back lab version; AUDIT: Alcohol Use Disorders Identification Test; TLFB II: Time Line Follow Back online version (i.e., one week follow-up); Stroop Index: RT incompatible e RT compatible; Gender: 0 ¼ female and 1 ¼ male. a n ¼ 107 for TLFB I due to one outlier. b n ¼ 92 for prospective drinking.

individuals who had a stronger tendency to interpret positive affect situations as alcohol-related more often drank in response to positive emotions, had consumed more alcohol during the past 7 days, showed higher levels of hazardous drinking and drank more during the seven days following the lab session. Furthermore, there was a marginally significant correlation between coping drinking motives and the alcohol-related IB in positive affect situations. Scores on positive-affect control sentences and foil sentences did not significantly correlate with any of the self-report measures.

3.2. Prediction of alcohol-related interpretation biases via drinking motives To examine whether coping and enhancement drinking motives predicted negative and positive alcohol-related IBs, respectively, two regression analyses were conducted. All variables (i.e., criterion and predictors) were first z-standardized (Aiken & West,

Table 3 Summary of regression analyses for prediction of alcohol-related Interpretation Biases (IBs) in negative and positive affect situations. Outcome

Predictors

B

ART negative affect

ART positive affect DMQ-R coping DMQ-R enhancement DMQ-R social DMQ-R conformity

.69** .17* .12 .15y .01

ART positive affect

ART negative affect DMQ-R coping DMQ-R enhancement DMQ-R social DMQ-R conformity

.68** .08 .05 .25* .02

Note. ART negative affect: Alcohol-Related Target Negative Affect Situations; ART positive affect: Alcohol-Related Target Positive Affect Situations; DMQ-R coping ¼ coping subscale of DMQ-R. DMQ-R enhancement ¼ enhancement subscale of DMQ-R; DMQ-R social: social subscale of DMQ-R; DMQ-R conformity: conformity subscale of DMQ-R. yp ¼/< .10 *p < .05; **p < .001.

1991). The negative and positive alcohol-related IBs were used as outcomes in the regressions (i.e., the scores on alcohol-related target sentences). In each regression, the DMQ-R enhancement and coping drinking motives were included. Moreover, there were three control variables, i.e., the alcohol-related IB which did not serve as outcome (e.g., when predicting the negative alcoholrelated IB, the positive alcohol-related IB was included as a predictor) and the DMQ-R social and conformity drinking motives (see Table 3). Regarding the prediction of the negative alcohol-related IB, the overall model was significant, R2 ¼ .53 (adjusted R2 ¼ .50), F(5,102) ¼ 22.76, p < .01. In line with our expectations, the DMQ-R coping motive was a significant predictor, B ¼ .17, SE ¼ .08, p < .03, indicating that individuals who more often drank in response to negative emotions had a stronger tendency to interpret negative affect situations as alcohol-related. Moreover, the alcohol-related IB for positive affect situations was a significant predictor, B ¼ .7, SE ¼ .07, p < .001, indicating that individuals who a stronger tendency to interpret positive affect situations as alcohol-related also had a stronger tendency to interpret negative affect situations as alcohol-related. None of the other predictors were significant. Regarding the prediction of the positive alcohol-related IB, the overall model was also significant, R2 ¼ .53 (adjusted R2 ¼ .51), F(5,102) ¼ 23.38, p < .001. Here, the DMQ-R enhancement motive was not a significant predictor, B ¼ .05, SE ¼ .09, p ¼ .55. That is, there was no association between drinking in response to enhancement-related drinking motives and interpretations of ambiguous, alcohol-related positive affect situations. However, two other predictors were significant: The alcohol-related IB in negative affect situations, B ¼ .68, SE ¼ .07, p < .001; and social drinking motives, B ¼ .25, SE ¼ .09, p < .01. Concerning the former, this indicates that individuals who had a stronger tendency to interpret negative affect situations as alcohol-related had a stronger tendency to interpret positive affect situations as alcohol-related. Concerning the latter, this indicates that individuals who more often drank in response to social-related drinking motives had a

M.L. Woud et al. / J. Behav. Ther. & Exp. Psychiat. 47 (2015) 102e110

stronger tendency to interpret positive affect situations as alcoholrelated. The other predictors were not significant.2 3.3. Prediction prospective drinking via alcohol-related interpretation biases and moderation executive control To test whether participants showed the expected Stroop interference effect, RTs of compatible cards were subtracted from RTs of incompatible cards. A positive difference score indicates interference, i.e., slower reaction times on the incompatible versus compatible card. The average difference score was indeed positive, M ¼ 10.77 (SD ¼ 5.59). Furthermore, results of a one sample t-test showed that this positive difference score differed significantly from zero, t(107) ¼ 20.04, p < .001, indicating that participants showed the expected Stroop interference effect. To examine whether alcohol-related IBs predicted prospective drinking, and whether this was further moderated by individual levels of executive control (EC), a moderated multiple regression analysis was performed including z-standardized variables (i.e., criterion and predictors) (Aiken & West, 1991). The quantity of alcohol consumed during the week following the lab session (TLFB online version) was used as outcome. As predictors, we entered negative and positive alcohol-related IBs (Step 1), DMQ-R coping and enhancement motives (Step 2), DMQ-R social and conformity motives (Step 3), executive control (Stroop interference score) (Step 4), and the Stroop  negative IB and Stroop  positive IB interactions (Step 5). In total, 92 individuals completed the online TLFB version. One participant was identified as an outlier (studentized deleted residuals > 3) and was excluded. As Table 4 shows, all five models were significant. Adding the predictors to the model in Step 3 improved the prediction of future drinking significantly. The addition in Step 4 was not significant. However, the final addition of predictors in Step 5 was significant. Regarding the predictors of primary interest, we found the following when looking at the model which includes all relevant predictors (i.e., Step 3). The alcohol-related IB in negative affect situations was a significant predictor, B ¼ .28, SE ¼ .13, p < .05, showing that the higher this bias, the higher levels of prospective drinking. However, the alcohol-related IB in positive affect situations was not significant, B ¼ .03, SE ¼ .12, p ¼ .85. Finally, we did not find a moderation effect of levels of executive control regarding the prediction of prospective drinking via the two types of biases (i.e., Step 5): IB negative  Stroop, B ¼ .20, SE ¼ .15, p ¼ .12, and IB positive  Stroop, B ¼ .01, SE ¼ .13, p ¼ .94. 4. Discussion The present study tested the relationship between drinking motives and alcohol-related IBs by employing a new alcohol Encoding Recognition Task (ERT). The main results are as follows. Correlations revealed that individuals who more often interpreted ambiguous negative affect situations as alcohol-related, drank more often in response to negative and positive emotions, had

2 We also conducted regression analyses with the same predictors and alcoholunrelated target sentences as the outcome. Regarding the regression including alcohol-unrelated target sentences in negative affect situations as criterion: The model was significant (R2 ¼ .44, adjusted R2 ¼ .41, F(5,102) ¼ 16.07, p < .001), and so was the following predictor: alcohol-unrelated target sentences in positive affect situations, B ¼ .64, SE ¼ .07, p < .001 (marginal significant: coping drinking motive, B ¼ .15, SE ¼ .08, p ¼ .072). Regarding the regression including alcohol-unrelated target sentences in positive affect situations as criterion: The model was significant (R2 ¼ .43, adjusted R2 ¼ .40, F(5,102) ¼ 15.17, p < .001), and so was the following predictor: alcohol-unrelated target sentences in negative affect situations, B ¼ .64, SE ¼ .08, p < .001 (marginal significant: enhancement drinking motive, B ¼ .17, SE ¼ .1, p ¼ .076).

107

Table 4 Summary of hierarchical regression analysis for variables predicting alcohol use at one week follow-up. Predictors

Prospective drinking Significance model R2/adjusted R2 R2 change B

STEP 1

F(2,88) ¼ 7.15, p < .01

.14/.12

.14

ART negative affect ART positive affect STEP 2

.36* .03 F(4,86) ¼ 7.36, p < .001

.26/.22

.12*

ART negative affect ART positive affect DMQ-R coping DMQ-R enhancement STEP 3

.27* .01 .04 .37* F(6,84) ¼ 6.23, p < .001

.31/.26

.05

ART negative affect ART positive affect DMQ-R coping DMQ-R enhancement DMQ-R social DMQ-R conformity STEP 4

.30* .03 .02 .32* .11 .24* F(7,83) ¼ 5.68, p < .001

.32/.27

.02

ART negative affect ART positive affect DMQ-R coping

.28* .05 .00 .01 .34* .10 .26* .13

DMQ-R enhancement DMQ-R social DMQ-R conformity Stroop index STEP 5 ART negative affect ART positive affect DMQ-R coping DMQ-R enhancement DMQ-R social DMQ-R conformity Stroop index Stroop  ART negative affect Stroop  ART positive affect

F(9,81) ¼ 5.09, p < .001

.36/.29

.04y .38* .07 .01 .31* .13 .30* .16y .20 .01

Note. N ¼ 91 ART Negative Affect: Alcohol-Related Target Negative Affect Situations; ART Positive Affect: Alcohol-Related Target Positive Affect Situations; DMQ-R coping ¼ coping subscale of DMQ-R; DMQ-R enhancement ¼ enhancement subscale of DMQ-R; DMQ-R social ¼ social subscale of DMQ-R; DMQ-R conformity ¼ conformity subscale of DMQ-R; Stroop Index: RT incompatible e RT compatible. yp < .10; *p < .05; **p < .001. Deviations in adjusted R2 are due to rounding.

consumed more alcohol during the past week, showed higher levels of hazardous drinking, and drank more during the week following the lab session. Similarly, individuals who more often interpreted ambiguous positive affect situations as alcohol-related, more often drank in response to positive emotions, had consumed more alcohol during the past week, showed higher levels of hazardous drinking, and drank more during the week following the lab session. Importantly, scores on control sentences in both negative and positive affect situations did not correlate with any of the alcohol-related self-report measures. Regarding the regression analyses testing the valence-specific association between drinking motives and alcohol-related IBs, the following results were obtained: As predicted, DMQ-R coping motives (but not enhancement motives) were a unique predictor of the tendency to interpret negative ambiguous alcohol-relevant situations in an alcoholrelated manner. Importantly, the predictive validity of coping

108

M.L. Woud et al. / J. Behav. Ther. & Exp. Psychiat. 47 (2015) 102e110

drinking motives was significant although we controlled for other relevant predictors. However, in contrast to our prediction, DMQ-R enhancement motives were not a unique predictor of the tendency to interpret positive ambiguous alcohol-relevant situations in this manner. Regarding the regression analyses testing the predictive validity of alcohol-related IBs for prospective drinking, the alcoholrelated IB in negative affect situations was a significant predictor (Step 3), showing that the more often participants interpreted negative affect situations as alcohol-related, the more they drank during the week after the study. However, individual levels of EC did not moderate this relation (Step 5). Moreover, prospective drinking was not predicted by the alcohol-related IB in positive affect situations, nor was there a moderation by individual levels of EC. To summarize, we found a valence-congruent association between coping drinking motives and negative alcohol-related IBs. Following assumptions of associative network accounts and the mood-memory-congruency perspective (Anderson, 1983; Bower, 1981; Collins & Loftus, 1975), this can be explained as follows: Individuals who mainly drink alcohol to cope, associate alcohol with something that allays and decreases their negative affect. Hence, when presented with negative ambiguous alcohol-related sentences, alcohol-related memory schemata become activated. Given this match in valence, this activation should be strong, or at least stronger than when being presented with positive ambiguous alcohol-related sentences. As such, individuals who drink to cope are biased towards these scenarios, i.e., have a greater tendency to interpret negative ambiguous alcohol-related sentences in an alcohol-related manner, compared to positive ambiguous alcoholrelated sentences. Regarding the association between enhancement drinking motives and positive alcohol-related IBs, we did not find such a valence-specific association. To conclude, the present findings are only partly in line with findings by Salemink and Wiers (2014). By means of an alcohol Word Sentence Association Paradigm (WSAP; Beard & Amir, 2009), they could show that coping motives (but not enhancement motives) predicted the tendency to associate alcohol-related words with negative affect situations, while enhancement motives (and not coping motives) predicted the tendency to associate alcoholrelated words with positive affect situations. Moreover, they found that alcohol-related IBs in positive affect situations predicted prospective alcohol use and number of binges, depending on the individuals' level of working memory capacity. In the context of drinking motives, the study by Salemink and Wiers was the first to examine the predictive validity of alcohol-related IBs, and this motivated us to conceptually replicate and extend these findings. However, there are a number of unexpected findings. First, social drinking motives (but not enhancement motives) were a significant predictor of positive alcohol-related IBs. The most parsimonious explanation of this result is the construct validity of the positive, ambiguous alcohol-related scenarios. The contents of these scenarios was supposed to be in line with cognitions and affective states relevant in enhancement-related drinking. However, given the situations described in the scenarios (e.g., being with friends, joining a party), there might have been a closer and/or stronger conceptual overlap with social drinking motives. The correlational analysis supports this assumption, as the correlation between social drinking motives and positive alcohol-related IBs was stronger than the correlation between enhancement drinking motives and positive alcohol-related IBs (r ¼ .34 versus r ¼ .25). Salemink and Wiers only included coping and enhancement motives in their analyses. However, given the present results, it is possible that the addition of the social and conformity motives would have affected their results too. A second unexpected finding was that in both regression analyses, the affectively opposite alcohol-related IB was

a significant predictor (i.e., positive alcohol-related IBs predicted negative alcohol-related IBs and vice versa). So, on the one hand, results show predictive validity of a valence congruent predictor (i.e., coping drinking predicted negative alcohol-related IBs), which is generally in line with findings of Salemink and Wiers as well as research putting forward that the activation of alcohol-related schemata also depends on contextual cues (e.g., Krank & Wall, 2006). On the other hand, results also show a predictive validity of a valence incongruent predictor. However, also other studies reported such valence-incongruent effects. To illustrate, Stewart et al. (2002) applied an affective priming task to study the activation of alcohol-related schemata in coping and enhancement motivated drinkers. In addition to the expected valence-congruent priming effects, they found that coping drinkers were primed by both positive and negative stimuli. Stewart et al. (2002) put forward that thus both types of stimuli can activate alcohol-related cognitions. Moreover, previous research has shown that situations involving pleasant times with others as well as unpleasant emotions predicted DMQ-R coping scores in students (Carrigan, Samoluk, & Stewart, 1998). Hence, one could tentatively argue that individuals could also be classified as ‘emotional drinker’ versus ‘non emotional drinker’. Regarding our hypothesis that the two alcohol-related IBs would be predictive of prospective drinking, we found the negative alcohol-related IB to be a significant predictor of prospective drinking (Step 3). This confirms the important role of negative affect in the context of alcohol (ab)use. Results also showed that enhancement and conformity motives were predictive of prospective drinking. Particularly the latter finding highlights the need to also conduct more research on the relationship between alcohol use and external drinking motives (Cooper, 1994). In contrast to our expectations, the positive alcohol-related IB did not predict prospective drinking. Moreover, we did not find any moderating effects of executive control at all. Also these findings deviate from those obtained by Salemink and Wiers (2014), and they are not in line with predictions of dual process models of addiction (e.g., Deutsch & Strack, 2006; Wiers et al., 2007). Regarding the latter finding, there are at least three potential explanations. First, the alcohol-ERT might not be adequate for assessing implicit alcohol-related cognitive processes, i.e., spontaneous, unintentional and automatic processes (Moors & De Houwer, 2006). Participants read the ambiguous scenarios and interpreted them later. This ‘off-line’ measurement does not assess participants' interpretations immediately like ‘on-line’ measurements such as the alcohol-WSAP does (Salemink & Wiers). Hence, the latter might be more sensitive to individual differences in executive control. A second explanation for not finding a moderating role of executive functioning is the fairly homogenous sample we tested. Since they were all university students, it seems likely that the variance in Stroop scores was relatively low. However, as there are studies that applied the Stroop in university samples and did find a moderating effect (e.g., Houben & Wiers, 2009; Wiers et al., 2009), this explanation should be treated with caution. Finally, according to Miyake et al. (2000), EC should not be regarded as a unitary concept and can be divided into three distinctive abilities (i.e., response inhibition, working memory updating, and mental set shifting) (for similar reasoning, see Hofmann, Friese, & Roefs, 2009). Hence, though highly speculative, finding EC to be a moderator in this context may have required another facet of EC. The present study is not without limitations. First, scenarios were not balanced for social contents. To illustrate, the negative scenarios included more scenarios during which the protagonist was alone. This might have confounded the valence of the scenarios. Second, our sample involved students. Hence, follow-up studies are needed that involve community and clinical samples.

M.L. Woud et al. / J. Behav. Ther. & Exp. Psychiat. 47 (2015) 102e110

Another interesting line for follow-up research could concern the causal relationship between drinking motives and alcohol-related IBs. To illustrate, alcohol-related IBs could be induced via an experimental design, and the effect of this induction on actual drinking behavior could be tested. Furthermore, mood induction studies may further advance our understanding of drinking motives and alcohol-related IBs. To illustrate, Grant, Stewart, and Birch (2007) showed mood-congruent effects on implicit alcohol-related concepts after a mood manipulation (see also Ralston et al., 2013), and such a relation can also be expected in the present context. Finally, investigating whether alcohol-related IBs are associated with other cognitive biases (e.g., attention biases) could be very informative. To conclude, we found a unique and valence-specific association between coping drinking motives and negative alcohol-related IBs. However, such an association was absent for the relation between enhancement drinking motives and positive alcohol-related IBs. Concerning the prediction of prospective drinking, results showed that particularly the negative alcohol-related IB was predictive of prospective drinking. However, EC did not moderate the prediction of prospective drinking by the two types of biases. These findings highlight the role of negative affect in the context of drinking motives and alcohol-related IBs. Hence, future research should focus on testing the robustness of these findings, and should further examine the general interplay between drinking motives and alcohol-related IBs. Role of funding organizations This research was funded by the Behavioural Science Institute (BSI), Radboud University Nijmegen, The Netherlands. The BSI had no role in study design; in the collection, analysis and interpretation of data; in the writing of the paper; and in the decision to submit the article for publication. Acknowledgments We are grateful to the students who helped with the data collection of this study. Moreover, we would like to thank Thomas Gladwin for his help with the preparations of the manuscript. References Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Thousand Oaks, CA: Sage. Ames, S. L., & Stacy, A. W. (1998). Implicit cognition in the prediction of substance use among drug offenders. Psychology of Addictive Behaviors, 12, 272e281. Ames, S. L., Sussman, S., Dent, C. W., & Stacy, A. W. (2005). Implicit cognition and dissociative experiences as predictors of adolescent substance use. American Journal of Drug and Alcohol Abuse, 31, 129e162. Anderson, J. R. (1983). The architecture of cognition. Cambridge, MA: Harvard, University Press. Baker, T. B., Morse, E., & Sherman, J. F. (1987). The motivation to use drugs: a psychological analysis of urges. In P. C. Rivers (Ed.), The Nebraska Symposium on Motivation: Alcohol use and abuse (pp. 257e323). Lincoln: University of Nebraska Press. Baker, T. B., Piper, M. E., Fiore, M. C., McCarthy, D. E., & Majeskie, M. R. (2004). Addiction motivation reformulated: an affective processing model of negative reinforcement. Psychological Review, 111, 33e51. Beard, C., & Amir, N. (2009). Interpretation in social anxiety: when meaning precedes ambiguity. Cognitive Therapy and Research, 33, 406e415. Birch, C. D., Stewart, S. H., Wiers, R. W., Klein, R. M., MacLean, A. D., & Berish, M. J. (2008). The mood-induced activation of implicit alcohol cognition in enhancement and coping motivated drinkers. Addictive Behaviors, 33, 565e581. Bower, G. H. (1981). Mood and memory. American Psychologist, 36, 129e148. Carrigan, G., Samoluk, S. B., & Stewart, S. H. (1998). Examination of the short form of the inventory of drinking situations (IDS-42) in a young adult university student sample. Behaviour Research and Therapy, 36(7e8), 789e807. Collins, A. M., & Loftus, E. (1975). A spreading activation theory of semantic processing. Psychological Review, 82, 407e428.

109

Conrod, P. J., Castellanos, N., & Mackie, C. (2008). Personality-targeted interventions delay the growth of adolescent drinking and binge drinking. Journal of Child Psychology and Psychiatry, 49, 181e190. Cooper, M. L. (1994). Motivations for alcohol use among adolescents: development and validation of a four-factor model. Psychological Assessment, 6, 117e128. Cooper, M. L., Frone, M. R., Russell, M., & Mudar, P. (1995). Drinking to regulate positive and negative emotions: a motivational model of alcohol use. Journal of Personality and Social Psychology, 69, 990. De Wit, H. (2009). Impulsivity as a determinant and consequence of drug use: a review of underlying processes. Addictive Biology, 14, 22e31. Deutsch, R., & Strack, F. (2006). Reflective and impulsive determinants of addictive behavior. In R. W. Wiers, & A. W. Stacy (Eds.), Handbook of implicit cognition and addiction (pp. 45e57). Thousand Oaks, CA: Sage Publishers. Evans, J. S. B. T. (2003). In two minds: dual-process accounts of reasoning. Trends in Cognitive Science, 7, 454e459. Gawronski, B., & Bodenhausen, G. V. (2006). Associative and propositional processes in evaluation: an integrative review of implicit and explicit attitude change. Psychological Bulletin, 132, 692. Gladwin, T. E., & Figner, B. (2014). “Hot” cognition and dual systems: introduction, criticisms, and ways forward (accepted). In E. Wilhelms, & V. F. Reyna (Eds.), Frontiers of cognitive psychology series: Neuroeconomics, judgment and decision making. Gladwin, T. E., Figner, B. F., Crone, E. A., & Wiers, R. W. (2011). Addiction, adolescence, and the integration of control and motivation. Developmental Cognitive Neuroscience, 1, 364e376. Grant, V. V., Stewart, S. H., & Birch, C. D. (2007). Impact of positive and anxious mood on implicit alcohol-related cognitions in internally motivated undergraduate drinkers. Addictive Behaviors, 32, 2226e2237. Grant, V. V., Stewart, S. H., & Mohr, C. D. (2009). Coping-anxiety and copingdepression motives predict different daily mood-drinking relationships. Psychology of Addictive Behaviors, 23, 226. Grenard, J. L., Ames, S. L., Wiers, R. W., Thush, C., Sussman, S., & Stacy, A. W. (2008). Working memory capacity moderates the predictive effects of drug-related associations on substance use. Psychology of Addictive Behaviors, 22, 426e432. Hofmann, W., Friese, M., & Roefs, A. (2009). Three ways to resist temptation: the independent contributions of executive attention, inhibitory control, and affect regulation to the impulse control of eating behavior. Journal of Experimental Social Psychology, 45, 431e435. Houben, K., & Wiers, R. W. (2009). Response inhibition moderates the relationship between implicit associations and drinking behavior. Alcoholism: Clinical and Experimental Research, 33, 626e633. Koob, G. F., & Le Moal, M. (2001). Drug addiction, dysregulation of reward, and allostasis. Neuropsychopharmacology, 24, 97e129. Koob, G. F., & Volkow, N. D. (2010). Neurocircuitry of addiction. Neuropsychopharmacology Reviews, 35, 217e238. Krank, M. D., & Wall, A.-M. (2006). Context and retrieval effects on implicit cognition for substance abuse. In R. W. Wiers, & A. W. Stacy (Eds.), Handbook of implicit cognition and addiction (pp. 281e292). Thousand Oaks, CA: Sage Publishers. Kuntsche, E. N., Knibbe, R. A., Gmel, G., & Engels, R. C. M. E. (2005). Why do young people drink? A review of drinking motives. Clinical Psychology Review, 25, 841e861. Kuntsche, E., Knibbe, R., Gmel, G., & Engels, R. C. M. E. (2006). Replication and validation of the Drinking Motive Questionnaire Revised (DMQ-R, Cooper, 1994) among adolescents in Switzerland. European Addiction Research, 12, 161e168. Magid, V., MacLean, M. G., & Colder, C. R. (2007). Differentiating between sensation seeking and impulsivity through their mediated relations with alcohol use and problems. Addictive Behaviors, 32, 2046e2061. Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., & Howerter, A. (2000). The unity and diversity of executive functions and their contributions to complex “frontal lobe” tasks: a latent variable analysis. Cognitive Psychology, 41, 49e100. Moors, A., & De Houwer, J. (2006). Automaticity: a theoretical and conceptual analysis. Psychological Bulletin, 132, 297e326. Palfai, T., & Wood, M. D. (2001). Positive alcohol expectancies and drinking behavior: the influence of expectancy strength and memory accessibility. Psychology of Addictive Behaviors, 15, 60e67. Ralston, T. E., Palfai, T. P., & Rinck, M. (2013). The influence of depressed mood on action tendencies towards alcohol: the moderational role of drinking motives. Addictive Behaviors, 38, 2810e2816. Reich, R. R., Below, M. C., & Goldman, M. S. (2010). Explicit and implicit measures of expectancy and related alcohol cognitions: a meta-analytic comparison. Psychology of Addictive Behaviors, 24, 13e25. Rooke, S. E., Hine, D. W., & Thorsteinsson, E. B. (2008). Implicit cognition and substance use: a meta-analysis. Addictive Behaviors, 33, 1314e1328. Salemink, E., & van den Hout, M. (2010). Validation of the “recognition task” used in the training of interpretation biases. Journal of Behavior Therapy and Experimental Psychiatry, 41, 140e144. Salemink, E., & Wiers, R. W. (2014). Alcohol-related memory associations in positive and negative affect situations: drinking motives, working memory capacity, and prospective drinking. Psychology of Addictive Behaviors, 28, 105e113. Saunders, J. B., Aasland, O. G., Babor, T. F., De La Fuente, J. R., & Grant, M. (1993). Development of the alcohol use disorders identification test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption-II. Addiction, 88, 791e804. Sobell, L. C., & Sobell, M. B. (1992). Timeline follow-back: a technique for assessing self- reported alcohol consumption. In R. Z. Litten, & J. P. Allen (Eds.), Measuring

110

M.L. Woud et al. / J. Behav. Ther. & Exp. Psychiat. 47 (2015) 102e110

alcohol consumption: Psychosocial and biochemical methods (pp. 41e72). Totowa, NJ: Humana Press. Stacy, A. W. (1997). Memory activation and expectancy as prospective predictors of alcohol and marijuana use. Journal of Abnormal Psychology, 106, 61e73. Stacy, A. W., Ames, S. L., & Grenard, J. L. (2006). Word association tests of associative memory and implicit processes: theoretical and assessment issues. In R. W. Wiers, & A. W. Stacy (Eds.), Handbook of implicit cognition and addiction (pp. 75e90). Thousand Oaks, CA: Sage Publishers. Stacy, A. W., & Wiers, R. W. (2010). Implicit cognition and addiction: a tool for explaining paradoxical behavior. Annual Review of Clinical Psychology, 6, 551e575. Stewart, S. H., & Devine, H. (2000). Relations between personality and drinking motives in young adults. Personality and Individual Differences, 29, 495e511. Stewart, S. H., Hall, E., Wilkie, H., & Birch, C. (2002). Affective priming of alcohol schema in coping and enhancement motivated drinkers. Cognitive Behaviour Therapy, 31, 68e80. Strack, F., & Deutsch, R. (2004). Reflective and impulsive determinants of social behavior. Personality and Social Psychology Review, 8, 220e247. Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18, 643e661. Thush, C., Wiers, R. W., Ames, S. L., Grenard, J. L., Sussman, S., & Stacy, A. W. (2008). Interactions between implicit and explicit cognition and working memory capacity in the prediction of alcohol use in at-risk adolescents. Drug and Alcohol Dependence, 94, 116e124.

Verdejo-García, A., Lawrence, A. J., & Clark, L. (2008). Impulsivity as a vulnerability marker for substance-use disorders: review of findings from high-risk research, problem gamblers and genetic association studies. Neuroscience Biobehavioral Review, 32, 777e810. Wiers, R. W., Bartholow, B. D., van den Wildenberg, E., Thush, C., Engels, R. C. M. E., Sher, K. J., et al. (2007). Automatic and controlled processes and the development of addictive behaviors in adolescents: a review and a model. Pharmacology, Biochemistry and Behavior, 86, 263e283. Wiers, R. W., Beckers, L., Houben, K., & Hofmann, W. (2009). A short fuse after alcohol: implicit power associations predict aggressiveness after alcohol consumption in young heavy drinkers with limited executive control. Pharmacology Biochemistry and Behavior, 93, 300e305. Wiers, R. W., Hoogeveen, K. J., Sergeant, J. A., & Gunning, W. B. (1997). High and low dose expectancies and the differential associations with drinking in male and female adolescents and young adults. Addiction, 92, 871e888. Wiers, R. W., & Stacy, A. W. (2006). Handbook of implicit cognition and addiction. Thousand Oaks, CA: Sage Publications. Woud, M. L., Fitzgerald, D. A., Wiers, R. W., Rinck, M., & Becker, E. S. (2012). Getting into the spirit: alcohol-related interpretation bias in heavy drinking students. Psychology of Addictive Behaviors, 26, 627e632. Woud, M. L., Pawelczak, S., Rinck, M., Lindenmeyer, J., Souren, P., Wiers, R. W., et al. (2014). Alcohol-related interpretation bias in alcohol-dependent patients. Alcoholism: Clinical and Experimental Research, 38, 1151e1159.