The purpose of this study was to examine the effect of the developmental levels of self-reguratory skills and types of instructional control strategies on learning and motivation from computer-based instruction. Subjects participating were 252 sixth graders of elementary school and 252 third graders of middle school. Based on the level of their self-regulatory skills, subjects were randomly assigned to one of the three instructional treatments: Program control, learner control, and learner control with advisement. Self-regulatory skills were measured by 40 items on 14 self-regulatory strategies. Instructional materials were the computer-based verbal information lesson developed for teaching on the ecology of 'Tarantulas' which are spiders that live in the southwestern United States. The learner control group was given control over lesson sequence. review, and practice questions. The learner control with advisement group was presented with the same lesson as the learner control group except that advisement was provided at each decision point in the lesson. A ramdomized block design was employed as the experimental design, and ANOVA, Duncan's posttest, t test, and chi-square test were used to analyze the data for this study. Main findings of this study were as follows. 1. High levels group of self-regulatory skills of 6th grader of elementary school performed better than low levels group in the scores of posttest and program control group of 3rd graders of middle school performed better than learner control with advisement group. But there was no significant interaction effect between self-regulatory skill levels and control strategies on performance. 2. Higher levels group of self-regulatory skills were higher than lower levels group in 6th graders of elementary school and 3rd graders of middle school in the scores of IMMS. But there was no significant interaction effect between self-regulatory skills levels and control strategies on learning motivation. 3. Students under program control took less time to complete their learning tasks than those under learner control with advisement, and students under learner control with advisement took less time than those under learner control in 6th graders of elementary school. But Strudent under program control took less time than those under learner control and learner control with advisement in 3rd graders of middle school.