An important limitation in behavioral tests of the hyperalgesic and analgesic effects of drugs is the low sensitivity of the measuring technique associated with inter- and intra-subject variability. Since familiarity with the experimental situation and novelty of the stimulus significantly influences an animal's behavior, we have tested the hypothesis that intensive training of rats, in the Randall-Selitto paw-withdrawal test, will increase the sensitivity of this test for detecting hyperalgesic effects of intradermal injections of the inflammatory mediator, bradykinin (BK). Mean baseline of the mechanical nociceptive threshold decreased between the first and second, and stabilized by the third day of training. Although there was no significant difference between the effects of BK or its vehicle (saline) on nociceptive threshold on day 1, on subsequent days BK produced a dose-dependent decrease in nociceptive threshold that could be distinguished from the effect of vehicle. These data demonstrate that training in the experimental paradigm can increase sensitivity for detecting hyperalgesic effects of inflammatory mediators in behavioral tests.