A controlled experiment was conducted on the effects of two types of animation - motion and highlighting - on learning. The treatment consisted of a 3.5 minute multimedia presentation that described the workings of a flushing toilet tank. A 2x2 factorial design ({motion, no-motion} x {highlight, no-highlight}) was employed with two dependent measures of learning (retention and transfer). Participants consisted of 65 undergraduates. Highlighting animation had a positive effect on both retention and transfer while motion animation had a negative effect on transfer. No significant interaction was detected between motion and highlighting.

In addition, the experiment tested the illusion of understanding hypothesis as a causal mechanism for the negative influence of motion animation on learning based on three predictions: With motion animation, learners (a) find instructional content less difficult, (b) generate more optimistic self-assessment of learning, and (c) are less able to perform mental visualization of the content. The results of the experiment were consistent with all three predictions. Furthermore, motion animation learners generated less accurate self-assessment than static image learners.

This experiment controlled for confounds found in prior animation effect studies: navigational control, content segmentation, narration modality, and delivery media. The experiment also implemented a double-blind design.