Alcohol Use Disorder (AUD) is a major public health problem, characterized by a high rate of relapse. Chronic and excessive alcohol consumption notably induces frontal brain alterations and cognitive impairments such as executive dysfunction and an attentional bias for alcohol, participating to the risk of relapse. In effect, AUD patients preferentially process alcohol-related cues, which could reflect a reorganization of the patients' semantic network. The investigators hypothesize that in AUD patients, semantic associations in memory are reorganized with a higher centrality of alcohol-related elements. To the investigators knowledge, no studies have explored semantic associations and/or semantic networks in AUD.

A study, conducted in patients with neurological damage, showed that frontal lesions are associated with excessive strength in semantic associations, and difficulties to generate remote associations. This excessive strength in semantic associations could reduce the ability to inhibit automatisms and to adapt to new context.

Objective: The objective of this study is to explore whether and how AUD patients have a different organization of semantic associations than healthy controls, and whether this reorganization influences the alcohol consumption over the months following the withdrawal. The investigators will also explore how it relates to neuropsychological assessment of flexibility, executive functions, and impulsivity. To these purposes, the investigators will use two original verbal tasks (Free Generation of Associates Task, FGAT and Associative Judgment Task, AJT) assessing word associations and allowing the estimation of semantic networks using graph theory, in combination with neuropsychological testing, in AUD patients and in healthy controls.

Methods: This study will include a group of 30 AUD patients and a group of 30 healthy controls. Both groups will be assessed twice, at baseline (T1; early in abstinence for AUD patients) and after a three-month period (T3). For the two groups, T1 and T3 assessments will include the two semantic association tasks (FGAT and AJT). For AUD patients, assessments will also involve neuropsychological testing of impulsivity, flexibility, and attentional bias. Besides, in AUD patients, data about alcohol consumptions will be collected six weeks (T2) and three months (T3) following the baseline assessment to classify patients as relapsers or abstainers.

Inclusion Criteria:

For alcohol use disorder patients

• Severe alcohol use disorder
• French as mother tongue
• Right-hander
• Patient abstinent from alcohol since 15 to 30 days at the inclusion
• Patient free from benzodiazepine since at least 48hours at the inclusion
• Patient who gave his informed written consent
• Currently in outpatient or inpatient care

For healthy controls

• French as mother tongue
• Right-hander
• Participant who gave his informed written consent

Exclusion Criteria:

For alcohol use disorder patients

• Patient under guardianship or under justice safeguard measures
• Patient under measure of therapeutic injunction
• Pregnancy or breastfeeding declared
• Meeting Diagnostic and Statistical Manual 5 (DSM) criteria for substance use disorder other than Tobacco
• Meeting DSM-5 criteria for non substance use disorder
• Patient presenting severe or progressive disease that interfere with experimental tasks, such as neurological diseases (TBI, epilepsy, stoke) , hepatic diseases, cancer, HIV, Hepatitis C Virus (HCV), and unstable psychiatric comorbidities.

For healthy controls

• Participant under guardianship or under justice safeguard measures
• Participant under measure of therapeutic injunction
• Pregnancy or breastfeeding declared
• Meeting DSM-5 criteria for alcohol use disorder
• Meeting DSM-5 criteria for substance use disorder other than Tobacco
• Meeting DSM-5 criteria for non substance use disorder
• Currently under benzodiazepine
• Participant presenting severe or progressive disease that interfere with experimental tasks, such as neurological diseases (TBI, epilepsy, stoke) , hepatic diseases, cancer, HIV, HCV, and unstable psychiatric comorbidities.

Two groups of participants will be included in this study. The first one consists of a pool of alcohol use disorder, the second one consists of a pool of healthy controls ; they will be matched as much as possible on age, gender, and educational level.