Studies of nutritional effects on behavior of children present many of the same methodological problems as studies of drugs on behavior of children. Dependent measures must be made across several classes of response at different time scales and levels of precision. A schema is presented for an integrated series of nutritional studies. The effects of missing breakfast and of simple sugars are used to illustrate several applications from this schema, including autonomic (evoked cardiac response), central (event-related potentials), and behavioral measures. Each class of measures is found to be sensitive to these nutritional manipulations. Attention and motor behavior appear to be quite sensitive to subtle nutritional effects.