The purpose of this study was to examine whether matching navigation mode of a learning environment with learners' preferred navigation mode would facilitate their learning in a web environment. Sixty-eight undergraduate students were randomly assigned to treatments (linear vs. nonlinear navigation mode) and received four criterion tests designed to measure different educational objectives immediately after interacting with the instructional material about human heart. The results suggested that matching or mismatching navigation mode of the learning environment with learners' preference did not lead to significant differences in learning outcomes. However, there were significant differences in achievement between groups with different navigation mode preferences. Our findings indicated that students preferring nonlinear navigation had significantly higher achievement scores on higher levels of learning outcomes. Based on the results of the study, we discussed specific recommendations for instructional designers and the issue concerning matching/mismatching navigation modes with learning preferences. (Contains 3 tables and 7 figures.)