A path model was used to test the unique and interactive effects of cognitive and motivational variables when learning in a supportive online learning system, Collaborative Inquiry System (CIS). In this student-centered learning environment, students interact with computer simulations and are assisted by online scaffolds intended to help them learn complex scientific concepts. The present study also explored the relationships between students' motivational, cognitive, and metacognitive strategy use and online performance. In total, 178 tenth-grade students participated in the study. The statistical analyses revealed that students' learning goals and cognitive preferences predicted metacognitive strategy use and later influenced their performance. Prior knowledge is a predictor of neither metacognitive strategy use nor learning goals and need for cognition, but is nevertheless an important determinant of online performance. Discussions on how to accommodate the different needs of students with varying levels of prior knowledge, goal orientation, and cognitive preference in a supportive learning system are provided based on the results. (Contains 2 tables and 2 figures.)