Objectives Examine the association between components of restrained eating, cognitive performance and weight loss maintenance.Methods106 women, all members of a commercial slimming organisation for at least 6months (mean±SD: 15.7±12.4months), were studied who, having lost 10.1±9.7kg of their initial weight, were hoping to sustain their weight loss during the 6month study. Dietary restraint subcomponents flexible and rigid restraint, as well as preoccupying cognitions with food, body-shape and diet were assessed using questionnaires. Attentional bias to food and shape-related stimuli was measured using a modified Stroop test. Working memory performance was assessed using the N-back test. These factors, and participant weight, were measured twice at 6month intervals.ResultsRigid restraint was associated with attentional bias to food and shape-related stimuli (r=0.43, p<0.001 resp. r=0.49, p<0.001) whereas flexible restraint correlated with impaired working memory (r=?0.25, p<0.05). In a multiple regression analyses, flexible restraint was associated with more weight lost and better weight loss maintenance, while rigid restraint was associated with less weight loss.ConclusionsRigid restraint correlates with a range of preoccupying cognitions and attentional bias to food and shape-related stimuli. Flexible restraint, despite the impaired working memory performance, predicts better long-term weight loss. Explicitly encouraging flexible restraint may be important in preventing and treating obesity.