Experiments were designed to determine the role of associative conditioning in reflex modification of the acoustic startle response using gaps in background noise. Experiments were conducted with independent, naive groups of adult Long Evans hooded rats tested using 20-msec gaps in white noise (ON 80dB/OFF 35dB) as the prestimulus (S1, ISI = 190 msec) and a 120-dB, 40-msec 13-kHz pure tone as the eliciting stimulus (S2). The first experiment characterized the effects of repeated testing for 9 days. The second experiment was a test of associative conditioning. Three groups of rats were tested daily for 6 days under one of the following conditions: S1 and S2 paired in a contingent manner, S2 only, or S1 only. All groups then received the contingent pairing of S1 and S2 for an additional 9 days of testing. In the third experiment, a separate group of rats was tested using either contingent or non-contingent presentation of stimuli in a contingent fashion. Results indicate that the amount of inhibition increases with repeated, daily testing, and achieve asymptotic levels of inhibition 5-6 daily sessions. The paired presentation of S1 and S2 is a necessary, but not a solely sufficient condition for normal development of inhibition.