Context: Human brown adipose tissue (BAT) is activated with cold exposure, but it is unknown whether overfeeding activates BAT. Objective: We determined BAT activation with cold, fasting, and overfeeding and the relationship of BAT activation with future weight change. Design, Setting, Participants, and Interventions: Sixteen healthy adults were evaluated during energy balance, fasting, and 24 hours of 200% overfeeding. All subjects had a fluorodeoxyglucose-positron emission tomography (PET) scan after exposure to 16°C to determine cold-induced BAT activity (CIBA). The first six subjects had a second PET scan after 36 hours of fasting to establish the lack of BAT activation at 22°C. The other subjects' second PET scan occurred after 24 hours of overfeeding at 22°C but only if they demonstrated CIBA. Twelve subjects returned at 6 months for reassessment of body composition. Main Outcome Measures: BAT was defined in cool scans as voxels with a standardized uptake value (SUV) of 2.0 or greater and Hounsfield units between −250 and −10. Body composition was assessed by dual-energy x-ray absorptiometry. Results: Although 75% of the subjects demonstrated visible CIBA, none had visual BAT activity after overfeeding. CIBA was greater than that observed in the same defined BAT voxels after fasting (n = 6; 2.9 ± 0.5 vs 1.2 ± 0.2; Δ = −1.7; 95% confidence interval −2.4, −1.0 SUV; P < .01). In the second cohort, CIBA was also higher than observed BAT voxel activity after 24 hours overfeeding (n = 8; 3.5 ± 0.7 vs 0.9 ± 0.2; Δ = −2.6; 95% confidence interval −3.2, −1.9 SUV; P < .01). Baseline CIBA negatively correlated with changes in fat mass after 6 months (r = −0.72, P = .009). Conclusions: BAT may be important in weight regulation unrelated to the response to overeating.