Science Fiction in Medical Education: The Case of Learning Styles

Research in medical education has resulted in a wide variety of useful guidelines for how to design instruction such that learning is facilitated. Unfortunately, there are concepts that continue to have a role in the field despite a lack of empirical evidence in favor of them. This letter addresses 1 of these concepts, namely, that of learning styles, and more specifically the hypothesis that tailoring instruction to learning styles improves learning outcomes (ie, henceforth learning styles hypothesis).

One would think that a straightforward scientific way of testing the learning styles hypothesis is to randomly assign learners to a condition in which instruction is tailored to their learning style, or to a condition where that does not happen. Pashler and colleagues¹ provided an excellent review that illustrates studies on learning styles that have adopted this approach, and several found results that contradict the learning styles hypothesis.

There are at least 3 explanations for the lack of evidence in favor of the learning styles hypothesis, in the aforementioned review and ever since. First, learners are categorized in terms of learning styles almost exclusively by means of learning style questionnaires, the usefulness of which is questionable at best, because there is no solid empirical evidence to suggest that learners are aware of their learning styles, let alone that they can reliably and validly report on them.² Second, the fact that at least 71 different learning styles have been identified, and learners can allegedly have all kinds of combinations of learning styles,³ implies that there are more combinations of learning styles than there are people on the planet.⁴ In the light of the learning styles hypothesis, this leaves us with the following take-home message for the design of education: everyone is different. Third, the notion that learners know best what is good for them flies in the face of decades of well-designed instructional design research that clearly demonstrates it is the complexity and the degree of instructional support provided in learning material that need to be accounted for, and that what learners think is good for them may not facilitate their learning at all.⁴

In the light of the previous, it is not surprising that a randomized controlled experiment published recently (which examined whether tailoring instructional style to a medical student's self-perceived learning style improves performance when teaching intravenous catheter placement) found no evidence in favor of the learning styles hypothesis.⁵ Given the massive evidence in favor of instructional design models, such as 4-component instructional design⁴ and the lack of evidence in favor of learning styles,^1,2,5 the field of medical education should focus on evidence-based instructional design models, and place the learning styles hypothesis in the genre of science fiction.