Children were included in this challenge study if according to parental report (1) the child's hyperactive behavior had been noticeably improved for at least 3 mo. as a result of adherence to the Feingold diet and (2) dietary violations such as those used in the study were reported to have a noticeable negative effect. Evaluations of 14 objective measures in a double-blind, cross-over design yielded no significant differences between diet infraction and noninfraction conditions. In addition to questioning the stated efficacy of the diet, findings suggest that one should not depend solely on parental report when evaluating a dietary effect and should also weight potential negative effect of adherence to the diet.