Mediation of cognitive bias modification for alcohol addiction via stimulus-specific alcohol avoidance association

Background Cognitive bias modification (CBM) studies have provided evidence that cognitive biases play a causal role in alcohol use disorders. In this study, data from a CBM experiment in alcoholic patients were re-analyzed. In the original study, no mediation by associations measured with an Implicit Association Test (IAT) was found. In this study, we explored the possibility that relevant alcohol-related automatic processes may be cue-specific. Methods Data from a previous clinical study in a sample of 214 alcohol-addicted patients were re-analyzed. Patients were assigned to a CBM intervention or control condition, performed an alcohol-approach IAT, and were followed up for relapse data a year after training. In this study, bias scores measured via the IAT were calculated and analyzed separately for different stimulus categories: Alcohol, Soft drink, Approach, and Avoid. Results Training reversed the alcohol-approach bias for all categories. This reversal of bias also predicted reduced relapse, but involved a complex stimulus category-dependent pattern in which an avoidance bias for Alcohol stimuli was most predictive of reduced relapse. Conclusions The results contribute to evidence that CBM indeed affects relapse probability via changes in automatic processes, although future study is needed to determine the precise nature of mediating processes. Automatic processes underlying alcohol-related associations may be stimulus-specific, which may be important for the methods of future studies involving implicit measures