The new potential anxiolytic ritanserin was studied in a double-blind manner vs. alprazolam, diazepam and placebo in 23 healthy subjects. The subjects belonged either to a high anxiety level group or a low anxiety level group, in order to study the effect of the anxiety level on the pharmacodynamic responses. The assessments included cognitive function (memory tests), psychomotor performance [Critical Flicker Fusion (CFF), Choice Reaction Time (CRT)], subjective ratings of alertness and overnight sleep and stress paradigm. Ritanserin (10 mg), alprazolam (0.75 mg), diazepam (10 mg) and placebo were given as single oral doses following a latin square design. Groups were well contrasted on the Cattell anxiety scale and were not overlapping. On no psychometric variable have there been any interactions between the anxiety level and the drug factor. At baseline an anxiety-related difference between the two groups was observed: lower CFF value in the high anxiety group (-1.4 Hz). Both benzodiazepines impaired psychomotor assessment and memory function and increased sleepiness. Ritanserin decreased CFF values without significantly affecting CRT on which nevertheless a trend to impairment was observed. Memory tests, and subjective ratings of alertness were unaffected by ritanserin. A trend to an antistress effect was observed on electrodermogram after ritanserin. Both benzodiazepines decreased central nervous system arousal and memory while ritanserin was inactive except on CFF. Recent data support the hypothesis that 5-HT2 blockers decrease pupil diameter which is a well known covariate of flicker frequency.