Approach-avoidance modification aims to target psychopathology’s cognitive-affective (implicit) level of information processing. Especially in Alcohol Use Disorder (AUD) alcohol-avoidance training (AAT) is found to be a successful add-on treatment. Both the desired outcome of the intervention (altering the predisposition to approach alcohol-related cues) and the methodology itself (employing a computerized task with subtle directives) inherently engage implicit cognitive processes. However, the connection between the reduction of the alcohol-approach bias through AAT and explicit (conscious) memory capacity, as well as the impact of the number of training sessions, remains unclear. A better understanding of this relation is relevant for two reasons: First for clinic practice since AAT could be an interesting treatment option for patients with alcohol-related neurocognitive disorders. Second, it would help refine the attributes of the AAT. Therefore the question whether change in alcohol-approach bias can be predicted by the quality of explicit memory encoding was tested.

A non-experimental, explorative, design was used in a clinical setting. 64 participants participated in an AAT with twelve sessions. All participants are diagnosed with AUD either without (N = 25) or with alcohol-related neurocognitive disorders (N = 31) including Korsakoff’s syndrome (N = 5). All participants received neuropsychological assessment. The total encoding error score on the Location Learning Task was used as a measure of explicit memory. Using Mixed Linear Modeling it was tested whether the explicit memory encoding score (EMES) and/ or session number were predictive of gain in the reduction of alcohol-approach bias score.

A better explicit memory encoding capacity predicted reduction of alcohol-approach bias. Neither Session number nor their interaction showed significant results. Moreover, computed regression coefficients showed that over time of training lower explicit memory capacity was conditional for a change from decreased to increased bias reduction which occurred at average on session six.

Change in alcohol-approach bias as a result of AAT, is related with explicit memory capacity: Initially a larger explicit memory capacity is an advantage, however over time (in the second part of training) bias reduction is related to lower explicit memory capacity. This result is consistent with implicit learning theory (the acquisition and consolidation phases of learning) and interaction between implicit and explicit memory processes. Therefore it is important to consider the interrelationship between implicit and explicit memory processes to optimise AAT properties and performance. And importantly, this study supports our previous recommendation: The (prolonged) AAT can (or should) be applied to patients with AUD and (severe) explicit memory disorders as it offers this patient group an additional treatment option.