Childhood obesity has multiple causes, most of them capable of explaining onlyone part of the problem. The population-wide impact of sedentary lifestyles andavailability of energy-dense food is undeniable, but substantial individualdifferences in body weight persist, suggesting that individuals responddifferently to the ‘obesogenic’ environment. One plausiblemechanism for this variation is the early expression of appetitive traits,including low responsiveness to internal satiety signals, high responsiveness toexternal food cues, high subjective reward experienced when eating liked foodsand preferences for energy-dense foods. Case–control studies supportthe existence of abnormalities in these traits among obese children comparedwith normal-weight children, and correlations between psychometric measures ofchild appetite and child weight suggest that appetitive trait profiles may notonly promote obesity but also protect against it. The origins of appetitivetraits are as yet uncharted, but will include both genetic and environmentalinfluences. Parental feeding style may affect the development of appetite butthe exact nature of the relevant behaviours is unclear and many studies arecross-sectional or begin late in childhood, obscuring causal relationships.Future research should explore determinants and biological mechanisms by usingprospective designs beginning early in life, measuring relevant biomarkers suchas gut hormones and incorporating neuroimaging and genotyping technologies.Potential clinical applications include the identification of ‘atrisk’ children early in life and interventions to modify appetitivetraits or ameliorate their impact on intake and weight.