During 12 years of investigation, data were obtained from 2,500 high school students and 10,000 college students concerning the effect of specific types of visualizations in producing optimum levels of achievement. The studies considered such qualities of illustrations as: realism, color, relation to oral presentations, relation to educational objectives, interaction time. Specially designed criterion tests were developed to measure: drawing, identification, terminology, and comprehension. The results showed that visuals do not necessarily enhance learning. When visuals are effective, the amount learned depends on such factors as the medium through which the visual was presented, its relation to the educational objective, and the time of interaction. (EMH)