To test the hypothesis that endogenous opiate peptides selectively influence hedonic response to sweet and high-fat foods, the opiate antagonist naloxone, opiate agonist butorphanol, and a saline placebo were administered by intravenous infusion to 16 obese and 25 normal-weight women. Twenty of the women (10 obese, 10 lean) fulfilled DSM-III-R diagnostic criteria for bulimia nervosa, as determined by psychiatric interview. During drug infusion the women tasted and rated 20 sweetened dairy products and were presented with eight snack foods of varying sugar and fat content. Naloxone suppressed hedonic responses in all subject groups and suppressed the consumption of sweet and high-fat foods in binge eaters, but not in nonbingers. Food intakes of obese women were not affected by naloxone. Butorphanol had no effect on either hedonic response or on food consumption in any group. Although opiate blockade is not a viable strategy for weight reduction in the treatment of obesity, it may be useful in the clinical management of the binge-eating disorder.