Benzodiazepines are generally reported to be active in tests based on punished responding and in procedures involving exploratory behaviour, but the effects of 5-HT drugs thus far reported are inconsistent. The effects of the two 5-HT(1A) agonists 8-OH-DPAT and buspirone were studied in a passive avoidance test and in an elevated plus-maze test. In the passive avoidance test 8-OH-DPAT and buspirone, as well as diazepam and chlordiazepoxide, were effective, while, in the elevated plus-maze test, the two benzodiazepines were active whereas buspirone and 8-OH-DPAT were not. Comparing the effects of the 5-HT(1A) agonists with the two benzodiazepines in the passive avoidance test it is suggested that this test can be predictive for drugs influencing anxiety. The elevated plus-maze test has many advantages, such as the absence of noxious stimuli, compared to punishment procedures, but since the 5-HT(1A) agonists do not act as anxiolytic compounds in this test, it is suggested that the test does not provide a suitable model of anxiety.