Individual variability in preference for energy-dense foods fails to predict child BMI percentile

Many studies show that higher dietary energy density is associated with greater body weight. Here we explored two propositions: i) that child BMI percentile is associated with individual differences in children's relative preference for energy-dense foods, ii) that child BMI percentile is associated with the same individual differences between their parents. Child-parent dyads were recruited from a local interactive science center in Bristol (UK). Using computerized tasks, participants ranked their preference and rated their liking for a range of snack foods that varied in energy density. Children (aged 3–14 years, N = 110) and parents completed the tasks for themselves. Parents also completed two further tasks in which they ranked the foods in the order that they would prioritize for their child, and again, in the order that they thought their child would choose. Children preferred (t(109) = 3.91, p < 0.001) and better liked the taste of (t(109) = 3.28, p = 0.001) higher energy-dense foods, and parents correctly estimated this outcome (t(109) = 7.18, p < 0.001). Conversely, lower energy-dense foods were preferred (t(109) = − 4.63, p < 0.001), better liked (t(109) = − 2.75, p = 0.007) and served (t(109) = − 15.06, p < 0.001) by parents. However, we found no evidence that child BMI percentile was associated with child or parent preference for, or liking of, energy-dense foods. Therefore, we suggest that the observed relationship between dietary energy density and body weight is not explained by individual differences in preference for energy density.