EVER SINCE De Sanctis1differentiated a group of children designated as dementia praecocissima from among the mentally subnormal, debate has raged as to the validity of the subgrouping and the origins and course of the designated disturbance. Over the years two principal positions have emerged. One view2-6has argued that the development of a psychosis in childhood is underlain by organismic factors particularly reflected in atypical central nervous system (CNS) development and organization. The other7-10views such disorder as stemming primarily from disturbed interpersonal relations, and in particular from seriously disordered relations between the child and his parents. Between these extremes a number of intermediate positions have also been taken.11-14Such intermediate conceptualizations have tended either to see the pattern of behavioral disturbance as the possible end result of independent influences (eg, due in some cases to brain damage, but