SummaryA variety of dietary constituents, both naturally occurring and added, may directly or indirectly affect the levels of hyperactivity in children. Studies designed to evaluate the magnitude of this effect on the behaviour of children have revealed the monumental difficulties encountered by the many investigators who have ventured into this area of research. Some of these difficulties have been recognised to be due to the poor and variable definitions of hyperactivity, and the consistently small size of sample investigated in each of the studies. Sample size is greatly influenced by the high drop‐out rates, often reducing the percentage of good responders to a treatment to insignificant numbers. Given these known factors that can influence the outcome of a study of this kind, one can design experiments to surmount these‐difficulties. For example, the high drop‐out rates observed in the majority of the studies dictate that one must start with a large enough number of well‐defined hyperactive children to compensate for the expected losses. The initial diagnosis of hyperactivity also should be the result of a careful diagnostic evaluation including psychiatric and psychological workups as well as birth records and a thorough paediatric allergy screening. The evaluation of behavioural changes following the ingestion of food constituents is equally important. This evaluation should include not only subjective ratings, i.e. the use of parents' and teachers' (Conners) scales, but quantitative measures such as laboratory tests designed to measure attention span, coordination, motor activity, and inability to concentrate, etc. Finally, in view of the fact that hyperactivity in children is an aetiologically complex syndrome, a multidisciplinary approach in the search for treatment and/or prevention of this syndrome is r