If a stimulus (e.g. light) is repeatedly preexposed without consequences, it subsequently develops a weaker association with a reinforcer (e.g. foot shock) than does a non-preexposed stimulus. This retarded conditioning to the preexposed as compared to the non-preexposed stimulus, is latent inhibition (LI). It is well documented that LI is disrupted by low doses of amphetamine and potentiated by neuroleptic drugs, and there is evidence that the action of these agents on LI can be modified by changes in the parameters of preexposure or conditioning. The present experiments tested whether the effects of DA agents on LI are influenced by the nature of the stimulus. In two experiments, LI was assessed using an off-baseline conditioned emotional response (CER) procedure in rats licking for water, consisting of three stages: preexposure, in which the stimulus (a light) to be conditioned, was repeatedly presented without being followed by reinforcement; conditioning, in which the preexposed stimulus was paired with reinforcement (a foot-shock); and test, in which LI was indexed by animals' degree of suppression of licking during stimulus presentation. In both experiments, different groups of animals were preexposed and conditioned with four different preexposed visual stimuli: three steady side-lights, three flashing side-lights, one flashing side-light, and a flashing houselight. Experiment 1 used 40 stimulus preexposures and tested the effects of 1 mg/kg D-amphetamine, whereas experiment 2 used 10 preexposures and tested the effects of 0.1 mg/kg haloperidol. The results showed that of the four stimuli used, both drugs were effective with only one and the same stimulus, namely, flashing houselight. This demonstrates that the disruptive effect of amphetamine and the potentiating effect of haloperidol on LI, are modifiable by manipulating the nature of the preexposed stimulus.