Major depressive disorder (MDD) is the most prevalent psychiatric disorder, affecting between 10 and 20 % of the world population at some point in their lives. MDD is characterized by a high risk for relapse after recovery (40% within 2 years). Therefore, understanding and changing the highly recurrent course of MDD is of high clinical and societal importance. In this talk, I will review results from the Netherlands Study of Depression and Anxiety (NESDA) Neuroimaging study (n=301), in which we studied risk factors, associations, and consequences of an unfavorable course of major depressive disorders. We investigated functional Magnetic Resonance Imaging (fMRI) characteristics associated with emotional processing, executive functioning, brain connectivity, in addition to structural brain characteristics. Results indicate differential predictors and consequences of an unfavorable course. Implications of these results for neurocognitive models of depression will be discussed. Additionally, a neurocognitive model of relapse and relapse prevention will be presented, which formed the basis of the NEW-PRIDE (Neurocognitive Working Mechanisms of Preventing Relapse in Depression) study. In this Randomized Controlled Trial , fMRI, pupillometry and neuropsychological assessments are employed to 1) understand the working mechanisms of preventive cognitive therapy for preventing relapse and 2) to develop neurocognitive predictors of individual treatment success. This study aims to contribute to effective preventive-treatment allocation, lower relapse-rates, and ultimately lower conversion into chronic-MDD by taking a neurocognitive approach. Acknowledgement: Many thanks to André Aleman, Rozemarijn van Kleef, Dick J. Veltman, Nic J.A. van der Wee, Jan-Bernard Marsman, Claudi Bockting, Esther Opmeer, and Hui Ai for collaborations on the work presented during this talk.