Elsevier

Addictive Behaviors

Volume 40, January 2015, Pages 21-26
Addictive Behaviors

Alcohol Cognitive Bias Modification training for problem drinkers over the web

https://doi.org/10.1016/j.addbeh.2014.08.010Get rights and content

Highlights

  • First web-based test of different varieties of cognitive training in problem drinkers

  • As usual in internet research large dropout (half after inclusion, another half during training)

  • Drinking went down in all conditions, including sham-training control

  • Combinations of motivational intervention and cognitive training should be studied

Abstract

Following successful outcomes of cognitive bias modification (CBM) programs for alcoholism in clinical and community samples, the present study investigated whether different varieties of CBM (attention control training and approach-bias re-training) could be delivered successfully in a fully automated web-based way and whether these interventions would help self-selected problem drinkers to reduce their drinking. Participants were recruited through online advertising, which resulted in 697 interested participants, of whom 615 were screened in. Of the 314 who initiated training, 136 completed a pretest, four sessions of computerized training and a posttest. Participants were randomly assigned to one of four experimental conditions (attention control or one of three varieties of approach-bias re-training) or a sham-training control condition. The general pattern of findings was that participants in all conditions (including participants in the control-training condition) reduced their drinking. It is suggested that integrating CBM with online cognitive and motivational interventions could improve results.

Introduction

Dual-process models emphasize the importance of both impulsive and reflective processes in many behaviors (e.g., Kahneman, 2003, Strack and Deutsch, 2004), including addiction (Bechara, 2005, Stacy and Wiers, 2010, Wiers et al., 2007). From this perspective, in addiction impulsive processes (e.g., attentional and approach biases to drug-related stimuli) gain control over behavior, even when the long-term perils of continued drug use are reflectively known. Cognitive Bias Modification (CBM) programs have been developed to target these cognitive processes in addiction, either through strengthening control processes (e.g., Houben, Wiers, & Jansen, 2011), or through changing biases in attention and action tendencies (review: Wiers, Gladwin, Hofmann, Salemink, & Ridderinkhof, 2013).

Problem drinkers have an attentional bias for alcohol-related stimuli, often assessed with varieties of the visual probe or alcohol-Stroop tasks (reviews: Cox et al., 2006, Field and Cox, 2008). Following the seminal study on attentional re-training in anxiety by MacLeod, Rutherford, Campbell, Ebsworthy, and Holker (2002), researchers in addiction began to investigate the effects of a single session of attentional re-training using similarly adapted visual probe tasks in heavy drinkers. Results showed that an attentional bias for alcohol could be changed, but after a single session no generalization was found to untrained stimuli or to behavior (Field et al., 2007, Schoenmakers et al., 2007). Two studies examining the effects of repeated attentional re-training found more encouraging results.

Fadardi and Cox (2009) developed the Alcohol Attention Control Training Program (AACTP), aimed at increasing control over distraction by alcohol. The AACTP uses a variety of pictorial alcohol-Stroop tasks to train participants’ attention away from alcohol-stimuli, with increasing levels of difficulty. The AACTP was tested with hazardous and harmful drinkers, using a within-subjects, waiting-list design. This design was chosen in part because the AACTP uses explicit goal-setting, which makes it difficult to create a sham-control condition. The AACTP significantly reduced both participants’ attentional bias for alcohol and alcohol consumption, in the absence of changes during the initial one-month waiting period. Changes in bias and consumption were maintained at a three-month follow-up.

Schoenmakers et al. (2010) performed the first small RCT of repeated attentional re-training in alcohol-dependent patients. Patients were randomly assigned to re-training based on the visual probe task (5 sessions) or sham training, which included the same pictures and motivating feedback as the re-training. Results showed an increase in attentional bias in the control group and a significant decrease in attentional bias for alcohol stimuli in the experimental group, with generalization to untrained pictures. The CBM group was also discharged from treatment sooner and relapsed later than the control group.

Problem drinkers show an approach-bias for alcohol-related stimuli (Field et al., 2008, Wiers et al., 2009). In the alcohol Approach-Avoidance Task (Wiers et al., 2009), participants are instructed to pull or push in response to a stimulus-feature unrelated to the contents of the picture (e.g. pull the joystick in response to pictures in landscape format, push pictures in portrait format). With a re-training variety of this task, it was demonstrated that students’ alcohol approach-bias could be successfully modified (Wiers, Rinck, Kordts, Houben, & Strack, 2010). In a subsequent RCT with alcohol-dependent patients (Wiers, Eberl, Rinck, Becker, & Lindenmeyer, 2011), patients were randomly assigned to one of two experimental conditions or one of two control conditions. Both experimental conditions involved four sessions of approach-bias re-training with a modified AAT, one with explicit instruction (push alcohol pictures), the other with implicit instructions (push pictures in one format; during training all alcohol pictures came in the push-format). One control condition consisted of continued assessment (alcohol pictures pushed and pulled equally often); the other consisted of pre- and posttest assessment only. Results indicated reduced approach bias in both experimental groups, both on the alcohol AAT (generalization to untrained pictures) and on an Implicit Association Test (Ostafin & Palfai, 2006). Patients in the experimental conditions relapsed 13% less often in the year to follow, compared with patients in the control-groups. In a replication-study (Eberl et al., 2013), 509 alcohol-dependent patients received either approach-bias re-training or no training. Training resulted in 10% less relapse after a year, mediated by the change in alcohol-approach action-tendencies.

In the present web-based study, attention control training (AACTP) and approach-bias re-training (varieties of the AAT) were directly compared. We hypothesized that all varieties of training would result in a greater reduction in drinking than sham-training. The study was different from earlier studies with clinical samples in which the goal of treatment was abstinence from alcohol; in the present study, participants’ goal was to reduce their drinking. Also, the study was entirely web-based, which has strong potential advantages, when effective (great potential impact at low costs, cf. Riper et al., 2008, Riper et al., 2011).

To summarize, we tested the efficacy in reducing problem drinking with internet-delivered training interventions, all of which have produced promising outcomes in stand-alone versions. We also included a number of psychological measures on which the AACTP had previously shown positive results (e.g., self-efficacy). The present design had three additional advantages: The large majority of participants would receive a training from which we expected a positive effect (attractive for participants). Second, the AACTP intervention could be compared with a sham-training control group. Third, the different types of alcohol-CBM for which previous research had reported positive results could be directly compared.

Section snippets

Participants

Participants were recruited via newspaper and online advertisements and a website. A total of 697 participants completed screening, 7 of whom were excluded for age (younger than 18), 42 for scoring lower than eight on the AUDIT (Alcohol Use Disorder Identification Test; Saunders, Aasland, Babor, de la Fuente, & Grant, 1993), 13 for not completing the AUDIT, and 20 for dyslexia. Color-blind participants were not excluded, but could not be assigned to the AACTP experimental condition, which

Results

In addition to alcohol, the large majority of participants were current or past smokers (77%), without much lifetime use of other substances (< 30% reported marijuana on 20 or more occasions, < 20% medication on more than 20 occasions, and less than 10% other drugs on 10 or more occasions, in the absence of differences between the groups, F40,1172 = 1.07, p = .35. Dropout was not significantly associated with condition (p > .30). Two participants in the AACTP condition did not finish any training

Discussion

The effects of web-based CBM was tested in self-selected problem drinkers. The AACTP program aimed at increased attentional control was compared with four versions of approach-bias re-training, including a sham-control condition. The main finding was that drinking was reduced across all conditions at posttest. In complete case analyses, at one and three-months follow-up approach-bias re-training had a stronger effect than AACTP. However, these findings should be interpreted with caution for two

Role of funding sources

The present study was supported by STW-Cognition (applied technical sciences), grant MSW.7353, awarded to the first author. RWW is supported by a grant from the Dutch National Science Foundation, N.W.O. (VICI 453-08-001). WMC and JSF have been supported by Economic and Social Research Council (ESRC) Grant RES-000-22-0314 and by Economic and Social Research Council Grant RES-000-23-1269.

The funding agencies had no involvement in design, data-collection, analysis, interpretation of the data or

Contributions

RWW and KH designed the study. WMC and JSF contributed to the translation and adaptation for the web of the AACTP intervention. PvB helped in data collection and description of the study. RWW wrote the ms., which all authors contributed to and approved for publication.

Conflict of interest

RWW gave a paid talk for Lundbeck pharmaceutical company.

RWW was co-applicant in two awarded grants from ERAB (the European Foundation for Alcohol Research), and KH was PI of one awarded grant for ERAB. ERAB is an independent foundation that awards alcohol-related research after an independent scientific evaluation (peer reviewed), with guarantee of completely independent scientific expression (in accordance with the Dublin principles), //www.api.or.at/sp/alcoholpolicy%20dokumente/dublinpriciples.pdf

Acknowledgements

The present study was supported by STW-Cognition (applied technical sciences), grant MSW.7353, awarded to the first author. RWW is supported by a grant from the Dutch National Science Foundation, N.W.O. (VICI 453-08-001). WMC and JSF have been supported by Economic and Social Research Council (ESRC) Grant RES-000-22-0314 and by Economic and Social Research Council Grant RES-000-23-1269. The authors thank Michiel Vestjens, Charlie Bonnemeijer and Thomas Pronk for technical support, and Thomas E.

References (40)

  • W.M. Cox et al.

    The addiction-stroop test: Theoretical considerations and procedural recommendations

    Psychological Bulletin

    (2006)
  • S. Dovis et al.

    Can motivation normalize working memory and task persistence in children with attention-deficit/hyperactivity disorder? The effects of money and computer-gaming

    Journal of Abnormal Child Psychology

    (2012)
  • A. Dijkstra et al.

    Self-efficacy expectations with regard to different tasks in smoking cessation

    Psychology & Health

    (2000)
  • G. Eysenbach

    The law of attrition

    Journal of Medical Internet Research

    (2005)
  • A.P. Field

    Discovering Statistics Using IBM SPSS Statistics. And Sex and Drugs and Rock ‘n’ Roll

    (2013)
  • M. Field et al.

    Experimental manipulation of attentional biases in heavy drinkers: Do the effects generalise?

    Psychopharmacology

    (2007)
  • L. Forsberg et al.

    The Readiness to Change Questionnaire: Reliability and validity of a Swedish version and a comparison of scoring methods

    British Journal of Health Psychology

    (2004)
  • K. Houben et al.

    Getting a grip on drinking behavior: Training working memory to reduce alcohol abuse

    Psychological Science

    (2011)
  • D. Kahneman

    A perspective on judgment and choice: Mapping bounded rationality

    American Psychologist

    (2003)
  • C. MacLeod et al.

    Selective attention and emotional vulnerability: assessing the causal basis of their association through the experimental manipulation of attentional bias

    Journal of Abnormal Psychology

    (2002)
  • Cited by (0)

    View full text