Get access

The influence of medical students’ self-explanations on diagnostic performance


Dr Martine Chamberland, Department of Medicine, Faculty of Medicine and Health Sciences, University of Sherbrooke, Sherbrooke, Quebec J1H 5N4, Canada. Tel: 00 1 819 346 1110, ext. 15385; Fax: 00 1 819 820 6815; E-mail:


Medical Education 2011: 45: 688–695

Context  Skill in clinical reasoning is a highly valued attribute of doctors, but instructional approaches to foster medical students’ clinical reasoning skills remain scarce. Self-explanation is an instructional procedure, the positive effects of which on learning have been demonstrated in a variety of domains, but which remain largely unexplored in medical education.

Objectives  The purpose of this study was to investigate the effects of self-explanation on students’ learning of clinical reasoning during clerkships and to examine whether these effects are affected by topic familiarity.

Methods  An experimental study with a training phase and an assessment phase was conducted with 36 Year 3 medical students, randomly assigned to one of two groups. In the training phase, students solved 12 clinical cases (four cases on a less familiar topic; four on a more familiar topic; four on filler topics), either generating self-explanations (n = 18) or not (n = 18). The self-explanations were generated after minimal instructions and no feedback was provided to students. One week later, in the assessment phase, students were requested to diagnose 12 different, more difficult cases, similarly distributed among the same more familiar topic, less familiar topic and filler topics, and their diagnostic performance was assessed.

Results  In the training phase the performance of the two groups did not differ. However, in the assessment phase 1 week later, a significant interaction was found between self-explanation and case topic familiarity (F1,34 = 6.18, p < 0.05). Students in the self-explanation condition, compared with those in the control condition, demonstrated better diagnostic performance on subsequent clinical cases, but this effect emerged only for cases concerning the less familiar topic.

Conclusions  The present study shows the beneficial influence of generating self-explanations when dealing with less familiar clinical contexts. Generating self-explanations without feedback resulted in better diagnostic performance than in the control group at 1 week after the intervention.

Get access to the full text of this article