The present experiment tested the effects of cueing (spotlights) and of verbal instructions in narrated animations on learning outcomes as well as potential moderator effects of working memory capacity (WMC). We found that the spotlight group dwelled longer on parts that were in the focus of the narration than a no-support group (control group). Despite higher levels of experienced difficulties, the no-support group had higher learning outcomes. More importantly, we found an aptitude–treatment interaction: Students with low WMC were impeded by verbal instruction, whereas students with high WMC were impeded by spotlights. Mediation analyses showed that spotlights increased attention to the focus of the narration that fostered learning outcomes but did not lead to better learning outcomes. Our findings suggest that further research on support procedures for animations should (i) take WMC into account, (ii) consider possible negative side effects, and (iii) address processes that mediate the effects. Copyright © 2012 John Wiley & Sons, Ltd.