The aim of this double-blind crossover trial was to compare the objective and subjective effects of flupentixol and lorazepam on human performance, and to reveal possible interactions between flupentixol and diazepam. Twelve healthy students received at 1-week intervals oral single doses of flupentixol 1 mg, flupentixol 2 mg, lorazepam 2.5 mg, placebo, and diazepam 15 mg alone and with flupentixol 1 mg. After the baseline measurements, the drugs were given in capsule form, and the tests were repeated 1.5, 3 and 4.5 h later. Diazepam was given at 1.5 h, to time its peak effect to coincide with that of lorazepam. Drug effects were measured objectively (two tracking tests, digit substitution, letter cancellation, flicker fusion, Maddox wing, tapping, memory) and subjectively (visual analogue scales, questionnaire). Blood samples were taken after each test time. Flupentixol 1 mg did not differ from placebo objectively or subjectively. Flupentixol 2 mg proved nearly inert objectively and on visual analogue scales. Lorazepam impaired objectively measured test performance, the clearest effects occurring at 3 and 4.5 h. It also impaired subjectively assessed performance. Diazepam impaired objective performance less than lorazepam, its effects peaking at 1.5 h after intake. Diazepam caused subjective drowsiness, clumsiness, mental slowness etc. as much as or more than lorazepare. The combination of 1 mg flupentixol and diazepam modified performance as much as diazepam alone. After the administration of 1 mg flupentixol, plasma concentrations were undetectable and levels after 2 mg were hardly detectable. Concentrations of lorazepam exceeded those of diazepam in direct bioassay, but they were much lower when bioassayed after solvent extraction. Flupentixol Img did not modify plasma diazepam levels. The clinical anxiolytic effects of lorazepam and flupentixol have not been compared directly, but we conclude that recommended anxiolytic doses of flupentixol have no or negligible influence on psychomotor performance while lorazepam does impair performance. No relevant interaction between 1 mg flupentixol and diazepam was detected.