Childhood obesity may be responsible for a major alteration of life quality in children and in future adults. The causes of obesity are multiple and obesity results from a chronic imbalance between food intake and energy expenditure. The challenge is to understand the cause of this imbalance. In this context, this study aimed at evaluating the contribution of the attraction toward salty, sweet and fatty foods. Thus, the first aim of this study was to establish a tool to assess children’s attraction toward sweet, salty and fatty foods that could be used later in large-scale cohorts. The second objective was to compare this questionnaire to sensory tests measuring the overall liking for sweet, salt and fat applied to the same subjects. Both tools have been initially developed in adults and were then adapted to the child population (age 7-10 y). The third objective was to compare results between children and adults. The initial questionnaire was revised by replacing some foods not adapted for children by others, and by adding pictures of foods and detailed instructions. The questionnaire included 83 items (sweet, 21; salt, 11; fat-sweet, 20; fat-salt, 31) scattered into: liking towards foods, preferred level of seasoning, preferred drinks in a menu and behavior-related questions. The sensory test included 14 sets of food models (sweet, 4; salt, 4; fat, 6), each with each featuring 5 grading levels of sweetness, saltiness or fattiness. Within each set, the samples were tasted in a random order and rated on a hedonic scale. The study included 123 children. The sensory test was divided in 3 sessions. Then, children filled in the questionnaire at home. For the questionnaire, scores of attraction toward sweet, salty or fatty foods were computed. For the sensory tests, an optimum level of sweetness, saltiness or fattiness was calculated for each food model and then average for each sensation. Children’s optima of liking (sensory) for sweet and salt were higher than for adults. Attraction (questionnaire) for fat and to a lesser extent for sweet was also higher for children than for adults. Scores from the questionnaire and from the sensory tests were correlated for salt (r=0.33, P=0.001) and fat (r=0.34, P=0.0005). For sweet, these two scores were only moderately correlated (r=0.21, P=0.05) compared to the scores obtained with adults. In conclusion, the questionnaire and the sensory tests help to highlight some common variance but also refer to different aspects of liking.