The main aim of this research is to investigate whether the use of cognitive event-related potentials is an interesting way to identify subgroups of alcoholic patients displaying specific clinical symptoms and cognitive disturbances in order to help clinicians to adapt the pharmaceutical approach to the specific needs of the patient.

Nowadays, a fundamental question remains: How can investigators identify among alcoholic patients who are likely to benefit from the use of naltrexone, acamprosate or baclofen, and those who are not? The goal of this application is to identify subgroups of alcoholic patients displaying specific clinical symptoms and cognitive disturbances linked to consistent biological markers. Investigators propose that this might help clinicians improve their treatment of alcoholic patients by focusing therapy on individual cognitive disturbances, and by adapting pharmaceutical approaches to the identified brain pathophysiology. In other words, investigators suggest that specifying the cognitive profile of each individual patient may help clinicians in their choice of a suitable drug program.

To reach this aim, investigators suggest that a joined investigation of early (P100) and late (P300) brain event-related potentials (ERP) components may help create subgroups of alcoholic patients with homogenous cognitive deficits, and that this ''classification'' might help optimize drug treatment. More precisely, investigators suggest that relapse in chronic alcoholism is partly due to (1) the preferential attentional allocation to alcohol-related information (e.g. the sight of a bottle of wine). As the P100 component has already been shown to be enhanced by motivationally relevant stimuli, investigators think that this component is well-suited for this purpose; and (2) the impairment of the inhibitory control, which is necessary to suppress an inappropriate prepotent response. The Go/No-Go task is a simple procedure, which has already proven to be highly reliable to evidence a deficit in inhibitory control processing in alcoholics, indexed by a No-Go P3 of decreased amplitude and less anterior topography. In summary, investigators have two simple experimental procedures, an oddball task and a Go/No-Go task, which can be easily carried out in clinical settings, and which can provide interesting data concerning, respectively, the existence of an implicit attentional bias towards alcohol-cues and the deficit of inhibitory control towards a prepotent response, through the observation of well-known and well-described cognitive ERP components, i.e. the P100 and P3b components. The main goal of this project will be to test the effect of different drug medications on both attentional (P100) and inhibitory (P300) deficits observable in alcoholic patients.

Inclusion Criteria:

• anti-craving medication free at arrival may be included in the study;
• stable doses of antidepressants for two months prior to screening will be permitted

Exclusion Criteria:

• participants with a history of neurological disorders, or other serious medical conditions (severe heart, lung, kidney or liver disease), assessed during the intake interview, will be excluded, as well as pregnant women; liver tests (aspartate aminotransferase (AST), alanine aminotransferase (ALT) resulting higher than 5 times the usual values are considered as an exclusion factor due to risk linked to naltrexone consumption;
• patients with current addiction to drugs other than nicotine (assessed through urine screening) will be excluded; individuals with positive cannabis screens will be excluded only if they had a history of cannabis dependence

Patients (age-range 18-65 years) undergoing a three-weeks detoxification cure in Alcohol Unit of CHU Brugmann, diagnosed with alcohol dependence, according to the DSM-IV-TR (diagnostic and statistical manual of mental disorders, 4th revision)