["Journal of Child Psychology and Psychiatry, EarlyView. ", "\n\nBackground\nFamily‐Based Treatment (FBT) is evidence based for patients younger than 18 with anorexia nervosa; however, continued refinement of FBT is needed to increase response and remission rates.\n\n\nAims\nThis study aimed to compare traditional FBT with an interoceptive exposure (IE) intervention that targets visceral sensitivity and autonomous eating in a family context. We hypothesized that IE would reduce food avoidance to a greater degree than FBT, as measured by laboratory feeding independent of weight gain.\n\n\nMethods\nAdolescents diagnosed with a low‐weight eating disorder (LWED) were randomized to six sessions of FBT or IE. Before and after the intervention, patients completed interviews and self‐report questionnaires, laboratory test meals (single and multi‐item), and functional magnetic resonance imaging. Healthy controls provided a comparison group for these assessments. The study was registered as follows: clinicaltrials.gov NCT02795455 (https://clinicaltrials.gov/study/NCT02795455?term=NCT02795455&rank=1), GCO 15‐0939 Reward Systems and Food Avoidance in Eating Disorders.\n\n\nResults\nAdolescents receiving IE (n = 30) consumed significantly more energy (kcal) during the post‐treatment single item (Mdiff = 98.12 kcal, SDpooled = 119.84, d = 0.82) and multi‐item meals (Mdiff = 137.11 kcal, SDpooled = 301.85, d = 0.45) than the FBT group (n = 30). No significant differences were found for age and sex adjusted % expected body weight between groups. Adolescents with LWED reported significantly more symptoms on all clinical measures than controls (n = 27).\n\n\nConclusions\nSix sessions of a novel form of IE yielded significant changes in eating behavior, namely an approximately two‐fold increase in energy consumed in both multi‐item and single‐item meals among those randomized to IE. Additional research using a fully expanded version of these treatments is needed.\n\n"]