Introduction. Research on the effectiveness of approaches for the teaching of clinical reasoning is scarce. A recent study showed hypothetico-deduction to be slightly more beneficial than self-explanation for students’ diagnostic performance. An account for this difference was unclear. This study investigated whether hypothetico-deduction leads to consideration of more alternative diagnoses while practicing with cases, and whether its advantage over self-explanation remains when diseases slightly different from the ones previously studied are tested.

Methods. One-hundred thirty-nine 2nd-year students from a six-year medical school participated in a two-phase experiment. In the learning phase, they worked in small groups on five clinical vignettes of cardiovascular diseases by following different approaches depending on their experimental condition. Students under the self-explanation condition provided the most likely diagnosis and pathophysiological explanation for the clinical findings. Students under the hypothetico-deduction condition hypothesized about plausible diagnoses for clinical findings presented sequentially. In a one-week-later test, all students diagnosed eight cases of cardiovascular diseases with clinical presentations similar to the ones previously studied but different diagnoses.

Results. The hypothetico-deduction condition generated more alternative diagnoses in the learning phase than the self-explanation condition, F(1,177) = 199.51, p = .001, h2 = .53; the effect size was large. A small difference in favour of hypothetico-deduction was observed in the proportion of accurate diagnoses: F(1,138) = 4.08, p = .05, h2 = .03.

Discussion. Relative to self-explanation, hypothetico-deduction induced consideration of more alternative diagnoses during practice with cases. This may explain the slight benefit of hypothetico-deduction over self-explanation regarding students’ diagnostic performance.