Young children's food preferences: a comparison of three modalities of food stimuli.

Food preferences are widely agreed to be important determinants of eating behaviour in young children. Existing studies of methods of assessing preferences have suggested tasting and ranking foods can generate reliable responses with young children, but there have been few attempts to assess other methods which might provide a more convenient alternative in situations where the use of real foods could be difficult (e.g. outside the laboratory), or tasting could be undesirable (e.g. if there are large numbers of foods, or foods which children would be unwilling to taste). The present study is a comparison of the reliability of preferences measured using: (i) real foods; (ii) food photographs; and (iii) food models, in 3 to 5-year-old children. The results showed that the tasting method produced good results, replicating existing data from U.S. samples. Food photographs came a close second in reliability but food models produced unreliable rankings, especially in the youngest group. Five-year-olds produced significantly more consistent results than the younger children. These results indicate that using real foods as the stimuli produces the most reliable taste preferences with children in this young age range, but photographs may provide a convenient alternative with adequate reliability.