PROJECT SUMMARY Anorexia nervosa (AN) is a devastating illness with morbidity and mortality rates among the highest associated with any psychiatric disorder. Treatments for adults with AN have limited efficacy due to an inadequate understanding of the mechanisms underlying AN's core symptoms. The salient feature of AN is extreme restriction of food, particularly dietary fat intake, a behavior both highly resistant to change and central to relapse. There is a pressing need to understand the neural bases of restrictive eating and use this knowledge to develop novel, mechanism-based treatments. Neuroimaging research of food restriction in AN has found that compared to healthy controls, patients with AN exhibit greater choice-related dorsal striatum activation, suggesting that activity in this region may underlie maladaptive eating behavior. To confirm the significance of these brain regions, a critical next step is to extend correlational findings by directly probing this region using repetitive transcranial magnetic stimulation (rTMS) and evaluating its effects on brain and behavior. TMS is a non-invasive form of neuromodulation that delivers an electromagnetic pulse to the cortex and modulates neuronal activity of both targeted cortical regions and subcortical regions within related circuits. Previous research indicates the DLPFC may be a cortical TMS target through which to engage the dorsal striatum. Preliminary data from our group shows that compared to sham, high-frequency rTMS (HF-rTMS) to a specific region of the right DLPFC previously implicated in AN significantly increases selection of high-fat foods among inpatients with AN. The goal of this Mentored Patient-Oriented Research Career Development Award is to examine the neural mechanisms underlying restrictive food choice in AN using HF-rTMS as an experimental manipulation of neural activation. The proposed study will combine neuronavigation-guided rTMS, individualized functional TMS targets, fMRI and a computerized food-choice task to test the effects of HF-rTMS on brain and behavior. The study will utilize a double-blinded randomized parallel design: adult inpatients with AN will receive either HF-rTMS (n=36) or sham (n=36) to the DLPFC. We will compare group changes in brain and behavior to test whether activity in the dorsal striatum underlies restrictive eating behavior in AN (Aim 1) and test the effects of HF-rTMS to the DLPFC on restrictive eating behavior (Aim 2). Our use of individualized TMS targets will enable us to identify whether resting-state functional connectivity predicts neural or behavioral response to HF-rTMS (Exploratory Aim). Data yielded from the proposed study will enable the field to generate causal inferences about the role of brain regions underlying maladaptive eating behavior in AN, and will support the preparation of an R01 application to develop novel treatments for this disorder. Together, the expertise developed through this K23 award will support the appl...