Effects of learning content in context on knowledge acquisition and recall: a pretest-posttest control group design

BACKGROUND: It is generally assumed that learning in context increases performance. This study investigates the relationship between the characteristics of a paper-patient context (relevance and familiarity), the mechanisms through which the cognitive dimension of context could improve learning (activation of prior knowledge, elaboration and increasing retrieval cues), and test performance. METHODS: A total of 145 medical students completed a pretest of 40 questions, of which half were with a patient vignette. One week later, they studied musculoskeletal anatomy in the dissection room without a paper-patient context (control group) or with (ir)relevant-(un)familiar context (experimental groups), and completed a cognitive load scale. Following a short delay, the students completed a posttest. RESULTS: Surprisingly, our results show that students who studied in context did not perform better than students who studied without context. This finding may be explained by an interaction of the participants’ expertise level, the nature of anatomical knowledge and students’ approaches to learning. A relevant-familiar context only reduced the negative effect of learning the content in context. Our results suggest discouraging the introduction of an uncommon disease to illustrate a basic science concept. Higher self-perceived learning scores predict higher performance. Interestingly, students performed significantly better on the questions with context in both tests, possibly due to a ‘framing effect’. CONCLUSIONS: Since studies focusing on the physical and affective dimensions of context have also failed to find a positive influence of learning in a clinically relevant context, further research seems necessary to refine our theories around the role of context in learning. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12909-015-0416-0) contains supplementary material, which is available to authorized users.