Do worked examples foster medical students' diagnostic knowledge of hyperthyroidism?

Objectives  In an initial experimental study in the domain of learning about hypertension, a case-based, worked example approach was found to be most effective when erroneous examples and elaborated feedback were provided. However, combining erroneous examples with knowledge of correct result (KCR) feedback impaired learning. This study was designed to establish whether these findings could be replicated in the domain of learning about hyperthyroidism. Methods  A total of 124 medical students were randomly assigned to the four conditions of a 2 × 2 design (with errors versus without errors; elaborated feedback versus KCR feedback). Diagnostic knowledge was operationalised by a multiple-choice test, key feature problems and problem-solving tasks. Acceptance and subjective learning outcomes were assessed on three rating scales. Results  The combination of erroneous examples and elaborated feedback was the most effective learning condition, whereas erroneous examples with KCR feedback impaired knowledge acquisition. These effects were independent of differences in prior knowledge and time on task and replicated key findings of the study on hypertension diagnostics. Additionally, results showed that students in conditions with elaborated feedback assessed their learning outcomes as significantly higher than students receiving KCR feedback only. Conclusions  By providing erroneous examples in combination with elaborated feedback in a computer-based learning environment, diagnostic knowledge was fostered. The approach of ‘learning from worked examples’ was successfully adapted to a complex domain and was found to support the acquisition of complex competencies.