Computer-based interactive video (CBIV), a hypermedia system that uses the power of the computer to support student interaction with various audio and visual media, can pose such problems for students as the potential for learner disorientation and cognitive overload. Intended to explore the applicability of Wittrock's generative learning hypothesis in the use of pictorial graphic organizers with CBIV instruction, this study tested three hypotheses: (1) that graphical views generated by the students themselves would have a greater effect on tests of learning and retention than would a final form version; (2) that both a participatory graphic organizer (PGO) and a final form organizer (FGO) would have a greater effect on recall than on retention; and (3) that the posttest scores for the two groups using the different types of organizer should correlate positively with measures of analytic reasoning and holistic processing. Subjects for the study were 32 nursing students using a CBIV program. One group of 16 completed the program using a participatory form of graphic organizer which required them to fill in blanks on the organizer. The other group of 13 completed the CBIV program using a final form version of the organizer which required that the students only read and remember the information presented there. No significant difference was found between the two groups on the mean scores for posttest 1; while the PGO group did score slightly higher on posttest 1, the FGO group scored significantly higher on posttest 2; and correlations between the posttest mean scores and tests of analytic reasoning and holistic processing were generally small. It was concluded that although some support was found for the efficacy of generative organizers with CBIV, this study should be replicated with larger treatment groups, and that a "true" advance organizer should be included as one of the treatments. (29 references) (DB)