Iron deficiency and behavior: criteria for testing causality

Abstract
The association between iron deficiency and poor behavioral-test performance is well established. The question of causality, however, remains unanswered. This paper presents the essential criteria for a valid test of causality. Internal validity is examined in terms of positive and negative findings. Plausibility of the findings requires analyzing the congruency of the results; for example, those who respond to the iron treatment are also those who respond behaviorally. External validity describes the extent to which inferences may be made about other populations or circumstances.