Behavioral effects of lysin-vasopressin (LVP) were investigated applying two paradigms from human experimental psychology. The first task was designed to simulate amnesic symptoms in normals. The second task addressed the emotional value of the items to be processed. Additionally, EEG recordings were used as indicators of the central nervous system effectiveness of LVP. Blood pressure and heart rate measured peripheral arousal. The co-twin control method was employed to increase experimental power. Contrary to the prediction of the vasopressin memory hypothesis none of the specific memory parameters was improved by LVP treatment. Changes in the electrical activity of the brain, but not in blood pressure and heart rate indicated central nervous system actions of LVP. However, interpretation of LVP effects in terms of memory processing seems not to be justified.