The study tests existing evidence on the paradoxical effects of exposure to a forbidden snack. Sixty-eight females were assigned randomly to one of two conditions: a temptation group, who were given the instruction to abstain from a favourite snack for 24 h while being exposed to it, or a control group, who were given no specific instructions. A further distinction was made between high-restraint/high-disinhibition (n = 21), high-restraint/low-disinhibition (n = 20) and low-restraint participants (n = 27) based on DEBQ subscale scores. After exposure to the foods, all participants were given free access to the food. Participants ate more of the snack after abstinence with exposure. The high-restraint/high-disinhibition group in particular displayed a substantial disinhibition effect. Results indicate that prohibition with exposure may backfire and increase the risk of loss of control over eating behaviour, particularly in at-risk groups of disinhibited restrained eaters. (c) 2008 Elsevier Ltd. All rights reserved.