Measurement of continuous reaction times (CRT) was applied to 105 hospitalized controls, 65 patients with brain damage and 47 patients with hepatic encephalopathy. One hundred and fifty reaction times were measured in every patient. The best discrimination between controls and patients with organic brain damage was obtained with the 10 percentile of the reaction time distribution (86% correct classification). The 90 percentile discriminated best between controls and patients with hepatic encephalopathy (86% correct classification). Furthermore, the ratio between the 50 percentile and the difference between the 10 and 90 percentile made it possible to discriminate between the two groups of patients (91% correct classification), which indicates that the CRT test is suitable as a screening device for discrimination between organic brain damage and hepatic encephalopathy.