Depression is a serious mental health problem that affects millions. Depression is usually treated using drugs and/or psychotherapy, but neither approach is successful for everyone, and some people do not respond to either. Therefore it is crucial that we continue to seek new methods for treating depression, and develop enhancements to existing treatments. In recent years, trials have documented improvements in depressive symptoms using noninvasive brain stimulation techniques, such as transcranial direct current stimulation, or tDCS. Our aim in this research is to investigate the effects of brain stimulation combined with psychological therapy in depression, an area that remains largely unexplored. Specifically, stimulation of the dorsolateral prefrontal cortex (DLPFC), a brain region known to work inefficiently in depression, has been shown to result in an improvement of depressive symptoms, as well as in the patient's 'cognitive control' abilities. Because 'cognitive control' processes, such as concentrating and ignoring distracting thoughts, are engaged during psychological therapies for depression, we predict that DLPFC stimulation should improve how patients respond to psychological therapy. This study has considerable implications as it will potentially benefit a large number of patients for which current treatments are ineffective.

We propose to investigate the effect of applying transcranial direct current stimulation (tDCS) over the left dorsolateral prefrontal cortex (LDLPFC) immediately prior to each of 8 sessions of cognitive behavioural therapy (CBT), a type of psychological therapy. In addition we will investigate whether tDCS effects on CBT are due to changes in cognitive control using both behavioural and neuroimaging measures.

We hypothesize that tDCS, thanks to its enduring effects, should improve the benefits of CBT, through the enhancement of cognitive control in the patient.

Null hypothesis: mood and cognitive performance will be equivalent in depressed individuals who undergo tDCS and those who undergo sham stimulation.

Experimental hypothesis: mood and cognitive performance will be improved in depressed individuals who undergo tDCS relative to those who undergo sham stimulation.

Sixty patients suffering from depression will be recruited, and a double-blind, sham-controlled, randomised design will be used. This design means that neither participants nor researchers are aware of the conditions they have been assigned to, and has been chosen to eliminate participant expectancy ("placebo") effects and researcher bias. To ensure that 30 participants are included in each group, we will use a balanced assignment algorithm at entry into the study, which will maximise statistical power. Since the results of a study of this type have never previously been reported (to our knowledge), clinical equipoise exists, mandating the use of a sham stimulation arm. Patients with depression will be randomly assigned to one of two groups: tDCS or sham stimulation. To ensure that 30 participants are included in each group, we will use a balanced assignment algorithm at entry into the study. The whole study will be spread over 16 weeks for each participant with the following time course:

• start: baseline MRI brain scan while completing a cognitive control task; measures of depressive symptoms
• weeks 1 to 8: one session per week of tDCS or sham stimulation (20 min, while completing a cognitive control task) immediately followed by CBT (1 hour); measures of depressive symptoms
• end of week 8: post-treatment MRI scan while completing a cognitive control task and measures of depressive symptoms
• week 9 to 16: up to one session per week of CBT as usual (1 hour) without stimulation
• end of week 16, or end of CBT (whichever is sooner): measures of depressive symptoms

MRI scans will have two purposes, (1) localising the area of the DLPFC for stimulation, and (2) comparing brain responses to a cognitive control task before and after treatment.

The tDCS will be delivered using neuroConn stimulators. A 1 milliamp (mA) current will be delivered for 20 minutes through damp sponges soaked in saline solution and attached to the head of the patient using a cap. One electrode will be placed on the forehead over the LDLPFC, and the other on the left shoulder blade or left cheek to record baseline electrical signal. In the sham condition, there will be a brief current change at the beginning and end of each stimulation session, in order to mimic the effect of an actual stimulation, but no current will be delivered in between.

CBT sessions will be delivered weekly, immediately after the tDCS or sham stimulation for the first 8 weeks, and provided by qualified psychological therapists as part of standard National Health Services (NHS) care.

Mood will be assessed before the start of the study (first MRI scan), each week for the 8 week stimulation phase, at the second MRI scan and after 16 weeks, using the Montgomery and Asberg Depression Rating Scale (MADRS).

Patients will perform a task to assess their cognitive control abilities during the MRI scans, as well as during each of the 8 stimulation sessions. During this task, visual stimuli will be presented to the subject, who will have to make different responses to these stimuli by pressing buttons on a response box.

Inclusion Criteria:

• Patients suffering from unipolar major depressive disorder
• First depressive episode onset before 40 years old
• Right-handedness
• English as first language
• Intention to commence a course of cognitive behavioural therapy

Exclusion Criteria:

• Antidepressant or other psychotropic medication at any time during the study or within previous 4 weeks (8 for fluoxetine)
• Recent illicit drug use
• Prior mixed, manic, or psychotic symptoms or other psychiatric or neurological illness
• Standard exclusion criteria for MRI scanning: pregnancy, breast feeding, any immovable metal in the body, weight above 250 lbs, claustrophobia
• tDCS safety criteria: skin disease or skin treatment that could potentially cause irritation with electrical stimulation