Comprehension of complex science topics occurs from the creation of new understanding of the information by the learner. However, learners are not very successful generating their own meaning, especially in computer based learning environments in which learners are required to make decisions about their learning process, since they rarely regulate their own learning process cognitively or metacognitively. This study examined the instructional effects of generative learning strategy and metacognitive feedback on learners' comprehension and self-regulation while learning a complex science topic in a computer-based learning environment. The 36 participants were assigned to either static visualized instructional material dealing with the human heart system, the same material with generative strategy, or the same material with generative strategy and metacognitive feedback. The results of this study revealed that the generative strategy with metacognitive feedback group scored significantly higher on comprehension and self-regulation measures. The results also found a significant positive relationship between self-regulation and comprehension. (Contains 4 tables and 5 figures.)