Cognitive Bias Modification for Alcohol Trial

A randomised controlled trial of Cognitive Bias Modification training during early recovery from alcohol dependence.


This cognitive bias modification (CBM) research project aimed to tests whether 4 short sessions of “brain-training”, using a laptop and joystick and delivered while clients are in residential withdrawal treatment (i.e., “detoxification”), helps people avoid relapse to alcohol use after discharge from treatment. The majority of people relapse to alcohol use after leaving withdrawal treatment, usually within a few weeks. One potential reason for this is that many people with an alcohol use disorder appear to develop “approach bias” towards alcohol, meaning an automatic tendency to approach alcohol in response to signals (such as images, smells, etc.) that remind them of alcohol. This “approach bias” might not be fully under one’s conscious control.

Our randomised controlled trial (RCT) compared a computerised CBM training task designed to dampen approach bias to another similar task that used the same images and equipment, but which was not specifically designed to change approach bias, to see whether the CBM task leads to higher rates of abstinence from alcohol after people leave residential withdrawal treatment. We recruited 300 participants from 4 residential withdrawal units around Melbourne from 2017-2019. We followed up participants 2 weeks, 3 months, 6 months, and 1 year after they left withdrawal treatment.

We found that 4 sessions of CBM significantly reduced approach bias to alcohol images. Indeed, the change was so large that it was reversed into an “avoidance bias”, where on average, participants’ automatic reaction to alcohol images was to avoid alcohol images rather than approach them. This change was unique to the CBM group, because the control group continued to show an approach bias to alcohol images following training.
Importantly, CBM helped prevent relapse to alcohol use after people left withdrawal treatment. Among those who completed all 4 sessions of training, the proportion who abstained from alcohol during the first 2 weeks after leaving withdrawal treatment was 64% in those who received CBM, compared to 47% in the control group. Even after 1 year, although rates of complete abstinence from alcohol since withdrawal treatment were much lower, they were still significantly higher among those who had received CBM (11%) compared to controls (4%). At the 3-month follow-up we also found a substantial difference in the proportion of participants who had abstained from alcohol within the past month (47% in the CBM group, 26% in controls, among those who completed 4 sessions of training). These differences occurred because participants who received CBM were much slower to return to drinking (average of 53 days between completing withdrawal treatment and first drink) than controls (average of 12 days until first drink).

Project team

Turning Point team

Professor Victoria Manning, Professor Dan Lubman, Dr Joshua Garfield, Mr Hugh Piercy, Ms Katherine Mroz, Mr Sam Campbell, Ms Kristina Vujcic, Dr Paul Sanfilippo, and Ms Nyssa Ferguson.

Project partners

Associate Professor Antonio Verdejo-Garcia (Monash University), Dr Petra Staiger (Deakin University), Dr Kate Hall (Deakin University), Professor Anthony Harris (Monash University), Associate Professor Jarrad Lum (Deakin University), Associate Professor John Reynolds (Monash University), Dr Martyn Lloyd-Jones (St Vincent’s Hospital Melbourne), Prof Reinout Wiers (University of Amsterdam), and Dr Yvonne Bonomo (St Vincent’s Hospital Melbourne).

Our funder

National Health and Medical Research Council (NHMRC: project grant 1124604)

Pilot projects:

  • This project is a follow-up to a pilot study which examined CBM in a smaller group of people undergoing alcohol withdrawal treatment, who were mainly recruited from Turning Point’s withdrawal unit, Wellington House. Read more.

  • An additional related pilot project we conducted at Wellington House tested combining CBM with working memory training in alcohol withdrawal patients. Read more.