Project Abstract Background/Description. Given our mutual interest in direct brain stimulation as an effective treatment for non-adherent eating disorders associated with refractory obesity, our multidisciplinary team at Stanford University has developed a collaboration with NeuroPace, Inc, a company that recently received FDA approval for a responsive neurostimulator. We previously found that electrically stimulating the nucleus accumbens (NAc) of mice attenuates binge-like eating. In addition, increased power in low frequency oscillations appears to temporally correlate with anticipation of food reward and predict the onset of a binge. There is increasing awareness that obese individuals frequently lose control over food, which leads to binge-like eating. We hypothesize that a responsive neurostimulator could be used to identify low frequency oscillations that represent loss of control over eating and deliver responsive stimulation to the NAc to prevent a binge. Objective. To test stimulation parameters and detection algorithms for responsive neurostimulation in humans in an Early Feasibility Study. Methods. All needed regulatory avenues will be pursued, and an Early Feasibility Study will be performed in human subjects with refractory obesity due to loss of control eating. We will primarily assess device function and safety, but will utilize multimodal controlled and ambulatory measures to test the potential of this clinical program for LOC eating in obesity.