The present experiment demonstrated a “perceptual learning” effect found in the animal literaturewith human participants. The common finding in animal work is that intermixed exposures to twostimuli prior to conditioning facilitates their subsequent discrimination on a generalization test morethan the same amount of exposure to the stimuli in a blocked arrangement. The method was asuppression task implemented in a video game. Participants learned to suppress a baseline response(mouse clicking) when a colored sensor (i.e., CS) predicted an attack (i.e., US). First, prior toconditioning, they received either intermixed pre-exposures to two sensor CSs, blocked preexposures,no pre-exposures, or pre-exposure to the individual visual elements of the CSs. Second, inconditioning, one of the sensor CSs was paired with an attack US. Finally, generalization ofsuppression to the other sensor CS was assessed. Pre-exposures to the sensor CSs reducedgeneralization relative to no-exposure at all, with intermixed pre-exposures producing the greatestreduction in generalization. The importance of the present work is that it reduces the possibleidiosyncrasy of existing results with humans that used evaluative-conditioning methods bydemonstrating the effect with a method that has been used to reproduce a variety of associativelearningphenomena and is easily amenable to associative-learning explanations