There is growing evidence that type 1 diabetes leads to an increased prevalence of depressive disorders and preliminary data suggesting that the metabolic disturbances associated with diabetes, both severe hypoglycemia and persistent hyperglycemia, lead to changes in brain structure and cognition. These findings, together with research on structural changes in the brain among depressed patients without diabetes, suggest that diabetes could cause structural changes in the brain that lead to depression. No studies have evaluated mechanisms underlying the etiology of depression among patients with type 1 diabetes. Using a cross sectional research design, we propose to study six groups of subjects: All subjects (N = 180) will be ages 30-40, right-handed and matched according to age, gender and SES. Diabetic patients will have between a 15-25 year history of types of diabetes. There will be three groups of diabetic subjects without psychiatric history: 1. Well controlled (0-1 episodes of severe hypoglycemia; mean HbA1c over history of diabetes equal to or < 8.0%). 2. Recurrent hypoglycemia ( equal to or >3 episodes of severe hypoglycemia; mean HbA1c over time equal to or <8.0%). 3. Persistent hyperglycemia (mean HbA1c over time equal to or >9.0%; 0-1 hypoglycemic episodes). A fourth group of diabetic subjects with a history of unipolar major depression and who equally represent the glycemic control characteristics of the other three diabetic groups will also be studied. In addition, two non-diabetic control groups (history of depression; no psychiatric history) will also be studied. We will assess these patients as to brain structure, using Magnetic Resonance Imaging (MRI); depression, using the Structured Clinical Interview for DMS IV (SCID); and cognition using the Wechsler Adult Intelligence Scale III (WAIS III) and other neuropsychological tests of memory, psychomotor speed and mental efficiency. We will also evaluate medical factors (e.g., glycemic control-HbA1c; and history of severe hypoglycemia). We will examine whether: 1) structural abnormalities are more common in diabetic subjects compared to the matched community controls; 2) there is a relationship between diabetes specific medical variables, such as long-term glycemic control, and brain structure abnormalities; 3) structural abnormalities are more common in diabetic patients with a history of unipolar major depression than diabetic patients without a history of depression; and 4) the frequency of structural abnormalities in the brain among diabetic patients with a lifetime history of unipolar depression differs from the depression control group. The proposed research will, for the first time, provide evidence regarding the linkage between structural changes in the brain and depressive disorder in diabetes, and evidence about the relationship of type I diabetes and its attendant metabolic disturbances to structural changes in the brain.