The day-night paradigm, where children respond to a pair of pictures with opposite labels for a series of trials, is a widely used measure of interference control. Recent research has shown that a happy-sad variant of the day-night task was significantly more difficult than the standard day-night task. The present research examined whether the perceptual discriminability of the happy-sad task pictures impacts young children's interference control. When test card pairs were equally distinct between conditions, children performed similarly regardless of whether the happy-sad or day-night response terms were employed (Experiment 1). Two versions of the happy-sad task were administered in Experiment 2, and children experienced significantly more interference when the stimuli were perceptually similar than when they were distinct. Theoretical implications for the role attentional demands play in interference control in Stroop-like tasks for pre-readers are discussed.