A comprehensive battery of neuropsychological tests was administered to 125 adolescents with a history of insulin-dependent diabetes, and to 83 demographically similar nondiabetic control subjects. To test the hypothesis that developing this disease early in life greatly increases the risk of manifesting significant cognitive impairments, diabetic subjects were assigned to an "early-onset" (diagnosis before age 5 years) or a "later-onset" subgroup. Results showed that subjects with early onset of diabetes performed more poorly than either subjects with later onset of diabetes or nondiabetic control subjects on virtually all tests, including measures of intelligence, school achievement, visuospatial ability, memory, motor speed, and eye-hand coordination. Moreover, multiple regression analyses demonstrated that the age at onset and the duration of diabetes seem to affect neuropsychological functioning in very different ways. The duration of the disease best predicted performance on those tests requiring highly overlearned, primarily verbal, skills whereas the age at onset best predicted scores on tests requiring the ability to process relatively unfamiliar, typically nonverbal, information in novel ways. Although the etiology of these deficits remains unclear, there is a possibility that they are secondary to mild brain damage that develops as a consequence of multiple episodes of serious hypoglycemia early in life.