Electroencephalograms and behavior ratings of 13 behavior problem children were studied during a six-week interval involving control periods without medication and periods of medication including benzedrine, phenobarbital and dilantin. The results show that behavior improved strikingly under benzedrine medication. Marked improvement of behavior also resulted from dilantin medication, but the drug was less effective than benzedrine. Phenobarbital, following a period of improved behavior under benzedrine, caused an exacerbation of symptoms, although, when compared to the behavior ratings during an initial control period, it was found to improve slightly the behavior of some of the children and make it slightly worse in others.
Although a number of statistically significant changes occurred in the electroencephalograms, they were not marked changes and were not in proportion to the changes in behavior produced by the various drugs. Also the EEG changes were not always consistent from one subject to another, from one head area to another, or from one drug to another when similar changes in behavior occurred. Therefore it must be concluded that behavior, which in problem children presumably has some relationship to the amount of abnormality found in their EEGs, may be improved without modifying essentially the electro-cortical abnormalities.