To purpose of the present study is to investigate the feasibility and efficacy of a computerized working memory training in improving cognitive functioning and alcohol use outcomes, in individuals with alcohol use disorders.

This study will employ a randomized, placebo-controlled, double blind design. The intervention group will perform repeated cognitive training (5 sessions à 30-40 min / week) with increased difficulty during 5 consecutive weeks using the Cogmed® software. The non-intervention group however, will perform the same training tasks, but without the increase in difficulty. Thus our study controls for non-specific effects such as regular use of computer, adhering to a training program and expectation effects. In order the complete the study, each participant has to complete minimum 20 training sessions within a maximum time period of 6 weeks. The study will be conducted at the Center for Dependence Disorders in Stockholm, situated within the premises of the Karolinska University Hospital.

The Research Subjects:

They study will include 50 subjects with a diagnosis of alcohol use disorder. Research subjects will be recruited through public advertisement and from within the Center for Dependence Disorders in Stockholm. After an initial brief telephone screening, potential subjects will be invited to the clinic at the Karolinska University Hospital, Stockholm and screened in-person for study participation.

The Study intervention: Cogmed® Training Program Cogmed® working memory training is a computer program that will be administered online. The program consists of 7 verbal and non-verbal working memory tasks and gives immediate performance feedback to the subject undergoing training. Each training session is 90 trials and lasts for approximately 30 minutes per day, 5 days a week and during a total training period of 5 weeks (i.e. total training time approximately 12,5 hours). In order to complete the study each subject must complete 20 - 25 training sessions within maximum 6 weeks. The program adjusts the difficulty of the different tasks trial-by-trial, which allows for each subject to train at a level adjusted to his or her current WM capacity (see (Klingberg et al. 2005) for detailed description). As a control condition, equivalent to the placebo arm in a pharmaceutical trial, the Cogmed® program consists of a Standard Training program that lets the user train the same tasks but without the increase in difficulty. Thus it allows clinical trials of the software to control for non-specific effects such as regular use of computer, adhering to a training program and expectation effects.

The Study Procedures After inclusion-visit, subjects will come back to the clinic for weekly follow-up visits over the span of 5 weeks. The visits can be scheduled ±3 days from the outline below, to accommodate the schedule of each subject.

Telephone screening:

AUDIT questionnaire will be administered through the telephone. If the subject has a total score > 16 and at least 3 points on question number 1 (3-4 drinks per week or more) the subject will called in for a screening visit.

Screening Written and verbal research subject information about the study details from the study physician. Data will be collected on a number of items related to current alcohol use and life time history of dependence. In addition scales related to physical well being and trauma will also be collected.

The following is the procedure for the inclusion and for the treatment period:

Day 1: Inclusion ( week 0)

• Urine dip test and breathalyzer to confirm absence of psychoactive substance and alcohol
• Urine sample for analysis of U-Ethyglucoronide
• Blood samples for bio-bank storage (genetic markers, plasma and serum)
• Assessment of depressive symptoms (MADRS) and craving (Short-DAQ (right now) and OCDS (last 5 days))
• Kirby Monetary Choice Questionnaire
• Neuropsychological Testing using the CANTAB® + Digit Span Task
• Introduction to the Cogmed® software and verbal and written training instructions
• Administration of personal login and password for Cogmed® software and completion of first training session

Randomization procedure Randomization will be done according to a randomised block design. The total sample size is 50 individuals. The personal login for each subject is associated with either active or control training - however neither study coordinator or the subjects know which group they are since this is not evident during the first day of training.

Study period: (Week 1-Week 5)

Week 1: Follow-up visit Day 7

• Breathalyser to confirm sobriety
• TLFB for alcohol and nicotine since last visit
• Urine sample for analysis of U-Ethyglucoronide
• Assessment of depressive symptoms (MADRS) and craving (Short-DAQ (right now) and OCDS (last 5 days)

Week 2, 3 and 4 Follow up visits

• Same procedure as Week 1

Week 5: Test Day (Day 35)

• Urine dip test and breathalyser to confirm absence of psychoactive substance
• Urine sample for analysis of U-Ethyglucoronide
• TLFB for alcohol and nicotine since last visit; Confirm sobriety on day before test day
• Assessment of depressive symptoms (MADRS) and craving (Short-DAQ (right now) and OCDS (last 5 days))
• Kirby Monetary Choice Questionnaire
• Neuropsychological Testing using the CANTAB® + Digit Span Task
• Payment for completed training sessions in food coupons (50 SEK / completed session) + their choice on Kirby Monetary Choice Questionnaire
• Voluntarily enrolment in treatment program at the Center for Dependency Disorders, Stockholm.

Other measures:

Urine Analysis:

Urine sample will be taken at every visit to the clinic in order to analyze presence of psychoactive substances (screening, inclusion and test day) and for analysis of U-ethylglucoronide (inclusion, every week visit and test day).

Breathalyser:

At each visit at the clinic, the research subjects will be tested for presence of alcohol using the alcohol breathalyser (Lion S-D2).

Blood Samples:

Each subject will leave blood samples for the bio-bank for genetic analysis and plasma and serum analysis of inflammatory markers, neuroendocrine hormones etc which will be analyzed after the completion of study. Total blood sampling for the bio-bank will be 28 ml of blood.

Physiological Measurements:

Heart rate and blood pressure will be measured by study nurse, on each visit to the clinic.

Participant Information and Consent:

In accordance with the Declaration of Helsinki, all potential research subjects will be competent and adequately informed by the investigator of every relevant aspect of the study before consenting to participate. Both oral and written information will be given, addressing the aims, methods, sources of funding, institutional affiliations of the researchers, the benefits and potential risks of the study and the discomfort it may entail.

The potential research subject will also be informed of the right to refuse to participate in the study or to withdraw consent to participate at any time without reprisal. After ensuring that the potential subject has understood the information, the investigator will seek the potential subject's freely-given written informed consent.

Ethical Considerations:

Risk/Benefit Ratio of the Study The risks involved in this study are justified by the anticipated benefits for the participants. Besides the potential benefit of the intervention, the patients establish contact with addiction healthcare professionals, which hopefully will encourage them to enroll in treatment programs. Thus, many patients with AUD that never had any treatment will be referred to treatment after the study. Furthermore, research subjects will also be compensated with 50 SEK / completed training session (maximum 25 x 50 = 1250) and additional 100-200 SEK depending on their choice on the Kirby Monetary Choice Questionnaire.

Potential risks for the individual subject are discussed below:

Potential Risks:

It is not anticipated that subject enrollment in the present study will cause any significant health related risks. However, the following potential risks are recognized:

1. Inconvenience of repeated visits and possible loss of confidentiality (related to clinical information, psychiatric assessment scales, and urine screens).
2. Risks associated with blood drawing include minor pain of venipuncture and some soreness after removal of the needle.
3. The Monetary Choice Questionnaire involves "fooling" the subject since we tell them that one of their choices will be selected at random - while in fact we have decided beforehand which of the different options they will receive.

Procedures for Minimizing Potential Risks:

1. As a guard against the loss of confidentiality, all information will be stored on locked files which can be accessed only by members of the research staff for this project. No names or other identifying information will be used in publications that stem from this research.
2. The blood sampling will be performed by an experienced research nurse, and venipuncture will be performed maximum one time per visit.
3. Even though we may "fool" our subjects, this is necessary in order to make sure that no one gets a lot more compensation than any other subject. Furthermore, this procedure will not result in any discomfort for the individual, only a small increase in compensation for their study participation. This procedure is necessary in order to get the subject to answer truthfully which option they would choose, since any offer could potentially be the real one.

Power Estimation:

We base our power estimation on effect sizes reported by Klingberg (2010) in a review of previous studies on computerized cognitive training. Difference between active and control groups regarding pre/post-test performance on working memory tasks have been around 40 % for trained tasks and 15 % for untrained tasks, with an effect size of approximately 1.0 (Cohens d). However, effect sizes for improvement in other cognitive or reasoning tasks have been weaker (around 0.4).

Given an effect size of 1.0, alpha level of 0.05 and power of 0.80 a sample size of 36 is sufficient in order to detect a difference between the groups regarding working memory improvement (calculated using STATA software power program). Thus, our sample size of 50 is adequate.

Inclusion Criteria:

• Male or a non-pregnant/non-nursing female
• Minimum of nine years of education
• AUDIT > 11 (performed at tele-screening) and scores 3 or 4 points on first AUDIT question (drinks 2-3 times per week or more)
• Fulfils the DSM-5 criteria for Alcohol Use Disorder
• Minimum 3 and maximum 14 alcohol-free days before day of inclusion (including day of inclusion)
• Access to a home computer with internet connection during the study
• Be willing to give informed consent and comply with study procedures

Exclusion Criteria:

• Fulfils current DSM IV diagnosis of any other substance dependence disorder (except nicotine)
• Fulfils current DSM IV diagnosis of the any major psychiatric disorder such as schizophrenia, bipolar disorder or severe major depression
• Suicidal ideation at screening
• Homelessness
• Acute withdrawal in need of pharmacological treatment (Ciwa score > 14) on day of inclusion
• Previous withdrawal-induced delirium tremens or seizures
• Regular intake of psychotropic medication (including acamprosate, naltrexone, disulfiram, antipsychotics, mood stabilizers, antiepileptic medication, benzodiazepines) the last 6 months before screening (exceptions include SSRI against anxiety or depressive illness currently in remission + temporary use of benzodiazepiner against withdrawal and non addictive substances e.g. prometazin (Lergigan®), propiomazin (Propavan®), alimemazin (Theralen®) or hydroxizin (Atarax®)
• Presence of non-stabilized severe medical illness e.g. liver cirrhosis, untreated severe essential hypertension (>160/100 mm Hg), uncontrolled diabetes mellitus, ischaemic heart disease, epilepsy
• History of stroke, intracranial hemorrhage or severe head trauma/traumatic brain injury
• Impaired sense of smell
• Use of any illegal drugs for the last 30 days
• Traces of alcohol as measured by breathalyzer on the Inclusion or Test day
• Traces of any other psychoactive substance (e.g. central stimulant amines, THC, benzodiazepines, opiates, cocaine) in urine sample on inclusion day or during the course of the study