CONTEXT-We teach judgement in pieces; that is, we talk about each aspect separately (patient, plan, resources, technique, etc.). We also let trainees figure out how to put the pieces together. In complex situations, this might be problematic. Using data from a drawing-based study on surgeons' experiences with complex situations, we explore the notion of 'problem definition' in real-world clinical judgement using the theoretical lens of systems engineering.METHODS-'Emergence', the sensitising concept for analysis, is rooted in two key systems premises: that person and context are inseparable and that what emerges is an act of choice. Via a 'gallery walk' we used these premises to perform analysis on individual drawings as well as crosscomparisons of multiple drawings. Our focus was to understand similarities and differences among the vantage points used by multiple surgeons.RESULTS-In this paper we challenge two assumptions from current models of clinical judgement: that experts hold a fixed and static definition of the problem and that consequently the focus of the expert's work is on solving the problem. Each situation described by our participants revealed different but complementary perspectives of what a surgical problem might come to be: from concerns about ensuring standard of care, to balancing personal emotions versus care choices, to coordinating resources, and to maintaining control while in the midst of personality clashes.CONCLUSION-We suggest that it is only at the situation and system level, not at the individual level, that we are able to appreciate the nuances of defining the problem when experts make judgements during real-world complex situations.Correspondence: Sayra Cristancho, Centre for Education Research and Innovation, University of Western Ontario, Health Sciences Addition, Room 114 London, Ontario, N6A 5C1, Canada. Tel: 519 661 2111 (ext. 89253); sayra.cristancho@schulich.uwo.ca. Contributors: SC and LL were responsible for the conceptualisation of the project and the analysis and interpretation of the data, made substantial contributions to the drafting and reviewing of the manuscript, and have approved the submitted manuscript and agree to be accountable for its content. MO, RN and TF were responsible for contributing substantive analysis, and drafting and revising the manuscript, and have approved the submitted manuscript and agree to be accountable for its content.