• Cocaine hydrochloride was administered to experienced users as an intravenous (IV) loading dose of 40 to 80 mg, followed by four-hour continuous IV infusions of either cocaine or placebo. Rates of cocaine infusion were individualized to maintain steady-state cocaine concentrations for the duration of the infusion. During the infusions, subjects rated themselves on questions that assessed their suspiciousness and paranoia, and nurse-observers took descriptive notes on the subjects' behavior; these notes were later scored on a scale for guarded, suspicious, and paranoid behavior. Nurses observed and rated moderately suspicious behavior when cocaine IV bolus loading doses were followed by cocaine infusions, but not when loading doses were followed by saline solution infusions; subjects did not rate themselves as suspicious during any of the study conditions. Suspiciousness during low-dose cocaine infusions significantly correlated with the amount of cocaine previously administered to the subjects. Suspiciousness during infusions was not related to plasma cocaine concentrations, preadmission drug use, or psychiatric symptoms and history. Cocaine infusions may be a useful tool to pursue the biology of stimulant psychoses.