The learning styles hypothesis--and particularly the meshing hypothesis--state that learners' preferences about their preferred modality of learning (i.e., visual, aural, or kinesthetic) predict learning gains on academic tasks. Despite the fact that this hypothesis is not borne out by the scientific evidence available to us, it still remains in widespread classroom use. This article begins by discussing the evidence against learning styles. Second, the article discusses why teachers might continue to believe in and use learning styles in their classroom as well as why essentialist beliefs about learning are not helpful. Finally, 3 variables that do impact student learning--knowledge, strategies, and interest are discussed. Each is defined, their development and measurements are discussed, and finally some instructional examples are given. Replacing the use of learning styles in the classroom with instructional decisions based on the development of knowledge, strategies, and interest can improve student learning outcomes across a wide range of subjects and grade levels.