This study was conducted to determine whether learner control of lesson strategy is superior to programmed control in computer-based instruction (CBI), and, if so, whether learner control is more effective when guidance is provided. Subjects were 164 trainees assigned to the Basic Electricity/Electronics School in San Diego. They were randomly assigned to three experimental treatment groups (unguided learner control, guided learner control and programmed control) and a control group. Lesson materials and objectives were the same for all groups; however, experimental groups used PLATO IV student terminals, while the control group used lesson booklets. Within-lesson performance of and lesson strategies used by the experimental groups were compared as well as the performance of the control and experimental groups on the post lesson performance tests and module examinations. Results indicate that learner control, with or without guidance, is not superior to programmed control as a CBI mode. However, the use of the learner control lesson structure may be desirable because of the possible economies to be effected. Also, since it appears that the rule-example-practice lesson strategy may prove to be optimum for all students, consideration may be given to delivering this strategy in an adaptive programmed control mode. A list of 17 references is provided. (Author/CHC)