There is evidence that impairment of impulse regulation is involved in the development and maintenance of eating disorders, especially in Binge Eating Disorder (BED). BED is characterized by recurrent episodes of binge eating with experienced loss of control over eating. Controlling impulsive behaviour, cognitive flexibility, planning and decision making are key abilities of impulse regulation. Some of these impaired cognitive functions are linked to decreased activity of certain brain regions. Transcranial direct current stimulation (tDCS) is a well-established method to alter brain activity. In the current project, we explore if a computer-assisted training programme for patients with BED that is combined with tDCS is feasible and able to ameliorate impulse regulation and impulsive eating behaviour. We hypothesize that the cognitive training programme with additional tDCS will result in a greater decrease of BED symptoms and a stronger increase in impulse regulation skills compared with the cognitive training programme without tDCS by using a placebo stimulation.

Patients with BED form a subgroup of obese patients with a disinhibited eating pattern that is associated with major impairments in cognitive control. The dorsolateral prefrontal cortex (dlPFC) has been identified as a brain region closely tied to cognitive control processes and crucially involved in the control of eating behaviour. This suggests the dlPFC as a target for the modulation of cognitive control processes over eating in BED. This modulation can be achieved by both, a cognitive training task and by noninvasive brain stimulation using tDCS. In studies with healthy normal-weight participants, (1) different cognitive training task enhance control over eating behaviour, (2) tDCS is an effective tool to ameliorate cognitive control processes, and (3) has beneficial effects on motivational aspects of eating behaviour, i.e. food craving. Based on this evidence, we will use a combination of a cognitive control task and tDCS to enhance cognitive control over eating in patients with BED. This is to the best of our knowledge one of the first studies to use tDCS as an intervention to enhance cognitive control over eating in patients with BED.

o Sample size: We will allocate 40 patients in the trial, i.e. 20 patients in each study arm.

o Recruitment: Patients are recruited by announcements, mails to the distributor list of the university hospital, existing databases of patients, and current patients of the university hospital. Patients are screened by a standardized checklist. In- and exclusion criteria are checked during the screening procedure and during the baseline diagnostic before randomisation.

o Standard Operating Procedures: For the recruitment, diagnostic and experimental sessions, Standard Operating Procedures are documented for the experimenter. This includes the order of clinical interviews, questionnaires and operating with the technical measurement instruments. The experimenters are regularly supervised. All adverse events will be listed and severe adverse events will be reported immediately to the PIs.

o Quality assurance plan: Randomized allocation to the stimulation condition (verum vs. sahm) and statistical analyses is done externally by the Institute of Clinical Epidemiology and Applied Biometry, University Tübingen, Germany (ICEAB). Objective technical measurement instruments are mostly used to record data.

o Data checks: Data is recorded mostly by objective technical measurement instruments, so no external monitoring is needed. We will spot check entered questionnaire data, in particular binge eating frequency in the past four weeks will be double checked as this is the primary outcome (PO).

o Source data verification: Data are spot checked by comparing the entries in the source data with the entries in the database. Each PO entry will be double checked. There are pre-defined criteria for entering data into the database.

• Data dictionary:

A data dictionary that contains detailed descriptions of each variable and how to be entered is available.

o Plan for missing data: Missing data and invalid data as well as the reasons will be recorded.

o Statistical analysis plan: A mixed model approach will be used to analyze the PO and secondary outcomes (SOs).

Inclusion Criteria:

• BED according to the Diagnostic and Statistical Manual of Mental Disorders (DSM-5)
• Legal age
• BMI above 20 kg/m2

Exclusion Criteria:

• Insufficient knowledge of German language
• Current pregnancy or lactation period
• Current or lifetime psychotic disorder, bipolar-I disorder, current substance dependence, suicidality
• Past bariatric surgery
• Severe physical disease which influence weight or eating behaviour (e.g. severe diabetes) or neurologic disease
• Non-removable metal parts in the area of the head
• Pacemaker
• Neuroleptics and benzodiazepine
• impaired vision, ametropia, eye diseases

The participants as well as the experimenters are blinded to the treatment group by using a unique 5-digit code for each participant which activates either verum or sham stimulation. The participants and experimenters are blinded to the randomisation condition as current in the sham condition is applied for 40 seconds at the beginning of each training session and during this time, no training of the cognitive control task is done. As participants perceive typical sensations of tDCS (e.g. tingling) and tDCS electrodes are actually mounted, this is considered as a valid placebo control.

The PIs and the biostatistician from the external ICEAB who is doing the statistical analysis are blinded as well until the end of the analysis.