Information processing was tested in 12 male subjects after smoking marihuana containing 0, 10, or 20 mg. of delta-9-tetrahydrocannabinol (THC) in three consecutive experimental sessions according to a Latin square protocol. Successful dose control was indicated both by the dose-related linear increase observed in heart rate and by preliminary assays of THC metabolites excreted in the urine. During tachistoscopic presentation of varying numbers of circles, statistically significant decrements in information processing occurred as a function of THC dosage. However, adding irrelevant information (triangles) to the display of circles eliminated effects of marihuana on accuracy of counting. Complex reaction times for oddity discrimination increased significantly only after the high dose. Nonetheless, both the social and high doses inappropriately inhibited the general tendency to respond to changing stimuli during oddity discrimination. Marihuana had no effect on field-dependence as measured by the Rod-and-frame test.