The primary purpose of this study was to examine the effects of replacing multiple-choice questions in tutorial software practice interactions with a learning strategy (i.e., learner-generated summaries) on immediate recall. Subjects were 35 students--32 females and 3 males--from an undergraduate computer course in the College of Education at Brigham Young University (Utah). The intervention for all treatment groups consisted of an interactive videodisk in developmental biology which had been developed using the traditional approach of instructional segment followed by multiple choice questions. Students were randomly assigned to three treatment groups and a pretest was administered. Students in Group 1 then answered multiple-choice questions after each of four instructional modules. In Group 2, students were asked to generate a written summary, which they typed in their own words on the computer, about the information presented in the module. Students in Group 3 were also asked to write a summary of the material just presented to them; however, after they wrote their summaries, they were provided feedback in the form of a system-generated summary of all of the main points in the module and told to compare the system-generated summary with their own. A posttest was administered and students were interviewed to determine whether they had used self-initiated learning strategies in addition to those designed in the interventions. It was found that replacing multiple-choice questions with learner-generated summaries not only did not facilitate immediate recall of the information presented, but reduced the efficiency of the learning process, since students who generated their own summaries took significantly longer to complete the instruction. (33 references) (BBM)