We investigated whether seductive details (i.e., interesting but irrelevant adjuncts) are harmful to learning only when students (mistakenly) think that they are relevant. We therefore conducted a study in which participants (N = 86) learned either without seductive details (control condition) or with seductive details--in the latter case with or without being informed about the seductive details' irrelevance. In line with our hypotheses, only participants who were "not" informed about the irrelevance of seductive details revealed worse learning outcomes than those in the control condition, thereby revealing a seductive details effect. Extraneous cognitive load, but not perceived time pressure, mediated the negative effects of being uninformed about the irrelevance of seductive details on learning outcomes. Taken together, our results suggest that the perceived relevance of seductive details is a boundary condition of the seductive details effect.