Corpus Analysis of Interactional Response Words as Indicators of Shared Understanding in Medical Problem-Based Learning Tutorials

Author(s): Olukayode Matthew Tokode, Reg Dennick

Background: Shared understanding is essential to effective collaborative learning. Interactive processes occurring in problem- based learning (PBL) tutorials have been explored to determine their cognitive and social advantages, but shared understanding is a relatively under-researched social process of PBL. The objective of this study is to describe how medical students share understanding in medical problem-based learning tutorials. Method: We recruited participants from first-year medical students in a single institution’s problem-based learning graduate entry curriculum. Transcripts from full cycles of eight tutorial groups were compiled to form the study corpus. Small interactional response words as indicators of shared understanding were measured using the Wmatrix 3 programme, and concordance lines were analysed manually to determine word functions. Results: Interactional response words were most prevalent in session 1 and least prevalent in session 2 of the PBL cycle. Interactional response words were used to mark unexpanded and simple and complex content expansion functions. While affirmation content expansion functions and reactive content expansion functions were more prevalent in sessions 1 and 3, negation content expansion functions were more frequent in session2. The frequency of interactional response words and their functions seem to align with the focus of each PBL tutorial session. Conclusions: Demonstrating the feasibility of corpus linguistics methodology for PBL concept analysis, this study showed that students in PBL tutorials attained sophisticated levels of shared understanding. We discussed the implication of the results for interprofessional teamworking and patient-doctor communication.