Adjunct questions were inserted after each of 12 computer-animated graphics sequences shown to 160 eighth-grade science students. These sequences presented two relationships in a purely spatial fashion. Four randomly assigned treatment groups received questions targeting one of the two relationships (temperature or molecular motion) over all or part of the 12 sequences. A control group received placebo postquestions. Significant interactions among treatment groups and posttest items ("low-level" comprehension) were found for each targeted relationship. Furthermore, interactions were stronger for students receiving postquestions covering all sequences. These findings extend previous research to include spatial as well as textual media. (Author)