It has been found that nonreinforeed preexposure to the to-be-conditioned stimulus results in a decrement in performance during the acquisition phase of a classical defensive leg response conditioning procedure. The present study shows that: (a) this decrement is a function of the frequency of preexposure, (b) the size of the decrement remains constant over 4 days of acquisition training, and (c) the decrement as a function of preexposure is not significant during extinction. In a previous experiment Lubow and Moore (1959) demonstrated a phenomenon which they labeled latent inhibition. This term describes the decrement in conditioning performance 'which results from the nonreinforeed preexposure of the to-be-conditioned stimulus. The purpose of the following study was to extend the earlier findings. In that study, experimental and control observations were made from the same &. That is, S was preexposed for 10 trials to either a flashing light or a moving rotor; following this, S was conditioned on alternate trials to the light and to the rotor. Reinforced acquisition training was started immediately after the preexposure trials, and continued until a predetermined criterion of 10 CRs was reached in that session. In the present study we were interested in the effect of frequency of nonreinforeed preexposure of the to-be-conditioned stimulus on acquisition and extinction performance; and in the duration of latent inhibition, both within an acquisition session and between sessions. Finally, we were interested in demonstrating the latent inhibition phenomenon within the context of a different experimental design in which different Ss were used for different groups.