According to the World Health Organization, MDD is attributed as the leading cause of disability worldwide, leaving 300 million individuals affected. Despite the efficacy of pharmacotherapy, a subset of MDD patients, classified as TRD, exhibit suboptimal response and thus require alternative treatment options such as rTMS. Emotional-laden "hot"and Neutral "cold" cognitions are shown to be dysfunctional in depression. Potential pro-cognitive effects remain inconclusive. In this study the investigators seek to investigate whether visual scanning patterns of emotionally laden images may be a biological marker and predictor of rTMS antidepressant efficacy. If so, then changes in visual scanning patterns are expected to precede clinical symptom improvement. Furthermore, changes in visual scanning patterns (which characterizes the state of hot cognition) are compared simultaneously to changes in cold cognition in order to elucidate the neural mechanisms underlying rTMS-induced changes in cognition. It is hypothesized that participants who are responders to rTMS will exhibit a decrease in the amount of time spent looking at dysphoric images will precede clinically detectable changes in mood as measured by a reduction in the scores on the 17-item Hamilton Depression Rating Scale (HDRS-17). The hypothesis for this study corresponds to the alleviation of the dysfunction within the hot cognitive system as a result of rTMS and a potential compensatory effect of cold cognition as a natural reaction of resetting the allocation of cognitive resources.

Repetitive transcranial magnetic stimulation (rTMS) is prescribed as first-line treatment for patients with Major Depressive Disorder. rTMS is safe, tolerable and non-invasive neurostimulation procedure. rTMS is characterized by an advantage in comparison to traditional pharmacotherapy in the sense that it exerts higher spatial precision in anatomical targeting of the specifically intended brain region compared to medications. The brain region of interest targeted with deep TMS in this study is the left-dorsolateral prefrontal cortex (L-DLPFC) which has been implicated in pathophysiology of depression.

Emotionally-laden functions within the brain are classified as "hot cognition" and contrastingly, affective-neutral areas regulating domains such as concentration, attention, learning and memory and executive function correspond to the characterization of "cold cognition" system. Both ''hot'' (emotion-laden) and ''cold'' (emotion-independent) cognition are known to take a role in the pathophysiology of MDD.

It is known that individuals suffering from depression experience impairments in hot cognition and as a result dedicate excessive focus to negative information, thereby producing negativity bias. A dysfunctional cold cognition system in depression also persists due to hypoactivity and abnormal functional connectivity of its underlying networks. The potential effects of rTMS on hot and cold cognition are currently uncertain. This study seeks to simultaneously evaluate hot and cold cognitive processes in response to deep TMS in order to elucidate the underlying cognitive mechanisms involved. More specifically, the investigators seek to explore the interaction between the two in terms of the time course and magnitude of the improvements in each system in response to TMS, and their relationship to mood improvement.

Methodology The subjects in the TRD group will undergo High Frequency deep TMS (HF-dTMS) stimulation over L-DLPFC, at the frequency of 18Hz, at 120% motor threshold value, for 5 daily sessions per week, over the course of 6 weeks. After the completion of treatment course at the conclusion of the 6 week mark, one post-treatment follow-up visit will take place at 1 month in order to assess any longer-term effects on cognition and depressive symptoms.

• Visual scanning The visual scanning test will consist of the presentation of 81 slides: 32 test slides (consisting of emotional images taken from Karolinska Directed Emotional Faces (KDEF) standardized library of facial expressions as well as a set of images with suicidal valence) and 49 filler slides. Each of the test and filler slides will contain four images at each of the four quadrants of the slide. The subjects' visual scanning behavior of the test slides will be analyzed while the scanning behavior of the filler slides will not be analyzed. Filler slides will change from session to session while the images on the test slides will remain the same. The position as well as the order of presentation of the four images on each test slide will randomly change between sessions. By using a large number of filler slides (different filler slides for each session), a randomized order for the presentation of the test slides and a randomized order for the position of the four images on each test slide, the investigators are aiming to minimize the possibility that patients' scanning behavior will be affected by recall from previous presentations. Each slide will be presented for 11 seconds for a total presentation time of 15 minutes.
• CANTAB measures For measuring the domains of cold cognition, the CANTAB measure which is a series of cognitive tests will be utilized. This battery is comprised of tests in the following cognitive domains: Attention, Executive Function, Memory and Social-Emotional Cognition. For the purposes of this study, the following cognitive tests will be implemented: For the domain of Attention, Motor screening task (MOT) [2 minutes] which evaluates potential sensorimotor or comprehension deficits, and Rapid Visual Information Processing (RVP) [7 minutes] which measures sustained attention are administered. For the domain of Executive Function, One Touch Stockings of Cambridge (OTS) [10 minutes] test that is built upon the foundation of "Tower of Hanoi" test and evaluates working memory and spatial planning subdomains of executive function will be implemented. For the domain of Memory, Verbal Recognition Memory (VRM) [10 minutes] will be used. VRM is a measure of new learning and verbal memory through the ability of collecting verbal information and its retrieval. For the domain of Social-Emotional Cognition, Emotion Recognition Task (ERT) [6-10 minutes] will be utilized which assesses the subject's ability in identification of six basic human emotions in facial expression along that expression's magnitude spectrum. Overall time allocated to conducting the aforementioned CANTAB measures will be approximately 40 minutes. As the subjects are engaging with these tests, the eye-tracking apparatus will be set up in the meantime [15-20 minutes] to enhance efficiency in time management and optimize the use of time. Following the completion of CANTAB tests, the subjects will promptly begin and complete the visual eye-tracking exercise [15 minutes]. Total time allocated to these measures of hot and cold cognition will approximate 1 hour.
• Bloodwork Past research indicates elevated levels of plasma inflammatory markers in depressive patients compared to non-depressed individuals among which markers such as tumor necrosis factor ⍺ (TNF-⍺) and interleukin-6 (IL-6) have been most reliably seen to exhibit heightened concentration levels in plasma. It has also been shown that depressed individuals had a higher likelihood of portraying non-response towards antidepressant medication carry a higher level of inflammatory markers at baseline and follow-up than those who responded to antidepressant therapy.

It has also been suggested that the number of unsuccessful antidepressant treatment trials indicating resistance and non-responsiveness in treatment resistant depressed individuals is associated with elevated plasma concentration of inflammatory markers of TNF-⍺ and IL-6 compared to responsive depressed individuals who had undergone one adequate antidepressant trial successfully and those with no previous history of antidepressant treatment. As a biological marker of rTMS antidepressant efficacy, bloodwork will be obtained a total of 4 times throughout the course of the study in order to monitor for genotype of serotonin transporter polymorphism (5-HTTLPR), brain-derived neurotrophic factor (BDNF) and potential alterations in present levels of the inflammatory markers of TNF-⍺ and IL-6 involved in the pathophysiological inflammation profile of depression.

● Psychiatric scales The severity of depressive symptoms in subjects will be assessed using the physician-rated HDRS-17 and self-report QIDS-SR16 psychiatric scales.

Analysis Following the end of the last visit at Week 6, patients will be grouped by response status: responder, and non-responder. Responder is defined as a patient with a change of 50% or greater on the HDRS-17 measure from baseline (V2) to the last visit (V8).

For each stage of testing, following the start of treatment (i.e. V2 to V8), differences from baseline (V2) of average visual fixation time, average fixation frequency, HDRS-17 and QIDS-SR16 scores will be calculated. The primary outcome measure, i.e., the difference from baseline of the average fixation time measured on different visits (V2 to V8), will be analyzed using linear mixed effect model with fixed effects and a random subject effect. The investigators will initially fit a mixed model with fixed effects of time and group and their interaction, and random intercepts and slopes. Similar models will be built for the secondary outcome variables comprising of average difference from baseline of fixation frequency, HDRS-17 and QIDS-SR16 scores.

The investigators will use the group main effect and the group-time interaction to test the hypothesis that changes in differences from baseline of fixation time differ between responders and non-responders. The investigators expect the differences from baseline of average fixation time will become more negative (decrease) in the responder group but will remain unchanged in the non-responder group. It is also expected that this reduction in the differences from baseline of average fixation time will precede any significant changes in depressive symptom scores on measures of HDRS-17 or QIDS-SR16.

Similarly, changes in the secondary outcome measures (difference from baseline of average fixation frequency, HDRS-17 and QIDS-SR16 scores) during the study course will also be analyzed in order to demonstrate an expected difference between responders and non-responders. The investigators expect the difference from baseline of fixation frequency will increase and that HDRS-17 and QIDS-SR16 will reduce by the conclusion of this study in the responder but not in the non-responder group. The investigators predict, however that any change in the HDRS-17 or QIDS-SR16 scores will succeed visual scanning parameters (i.e., difference from baseline of average fixation time or average fixation frequency) and only be observed at later time points.

Inclusion Criteria:

• For the patient group, subjects are required to meet DSM-5 criteria for a diagnosis of Major Depressive Disorder.
• For the healthy control group, subjects should be matched to the patient group in domains of age, sex and education; as well as being devoid of any current or previous history of neuropsychiatric illness.
• Patient group subjects must have a score of at least 20 on the HDRS-17 at time of assessment.
• All subjects (patients and healthy controls) should be between the ages of 18 and 65.
• All subjects (patients and healthy controls) must be capable of providing their informed written consent.
• For the patient group, it is essential that subjects who are taking antidepressant medications maintain a stable dose of medication for at least 4 weeks prior to start of treatment and also continue maintaining this stability throughout the course of treatment.

Exclusion Criteria:

• Subjects suffering from unstable medical conditions
• Subjects with current and past history of epilepsy
• Recent history of substance abuse/dependence (excluding nicotine and caffeine) within the past 6 months as defined by DSM-5
• History of suicide attempts or self-harm within the past 12 months
• Current or previous diagnosis of Bipolar Disorder, Schizophrenia or other psychotic disorders (including psychotic disorder due to general medical condition, substance-induced psychotic, psychotic disorder not otherwise specified) as defined by the MINI
• Previous lack of response to rTMS
• Electroconvulsive therapy (ECT) within the 3 months prior to beginning of study
• Inability to communicate in English language

Potential TRD participants in this study will be individuals referred by their family doctor or community psychiatrist for diagnostic clarification and/or the provision of treatment options for their mood symptoms. Additionally, individuals who respond to the study advertisement and based on the telephone screening performed by the study physicians, are deemed to meet the study's inclusion and exclusion criteria will be the provision of treatment options for their mood symptoms as well as the opportunity to participate in this study. Healthy Control participants will be age-,sex- and education- matched to the TRD participants and will be representative of the community sample, consisting of healthy individuals with no psychiatric illness.

Due to confidentiality, IPD will not be shared with other researchers