The purpose of this study was to investigate the impact of nutritional labelling on energy intake, appetite perceptions and attitudes towards food. During a 10-d period, seventy normal-weight (BMI<25 kg/m2) and seventy-one obese women (BMI≥30 kg/m2) were given three meals per d under ad libitum conditions. Participants were randomly assigned to one of three experimental labelling groups in which the only difference was the label posted on lunch meal entrée: (1) low-fat label, (2) energy label (energy content of the entrée and average daily needs) and (3) no label (control). Average energy intake was calculated by weighing all foods before v. after daily consumption. Hunger and fullness perceptions were rated on visual analogue scales immediately before and after each meal. Satiety efficiency was assessed through the calculation of the satiety quotient (SQ). The appreciation and perceived healthiness of the lunch entrées were rated on eight-point Likert scales. There was no difference in energy intake, SQ and attitudes towards food between the three labelling groups. Fasting hunger perception was higher in the low-fat label group compared with the two others groups (P=0·0037). No interactions between labelling groups and BMI categories were observed. In conclusion, although labelling does not seem to influence energy intake, a low-fat label may increase women’s fasting hunger perceptions compared with an energy label or no label.