LAUSR

Students experience a variety of emotions following achievement outcomes which stand to influence how they learn and perform in academic settings. However, little is known about the link between student outcome emotions and dimensions of performance feedback in computer-based learning environments (CBLEs). Understanding the dynamics of this relationship is particularly important for high-stakes, competency-based domains such as medical education. In this study, we examined the relationship between medical students’ (N = 30) outcome emotion profiles and their performance on a diagnostic reasoning task in the CBLE, BioWorld. We found that participants could be organized into distinct emotion groups using k-means cluster analyses based on their self-reported outcome emotion profiles: an expected positive emotion cluster and negative emotion cluster and an unexpected low intensity emotion cluster. A clear relationship was found between emotion clusters and diagnostic performance such that participants classified to the positive emotion cluster had the highest performance; those classified to the negative emotion cluster had the lowest performance; and those classified to the low intensity emotion cluster had performance outcomes that fell between the other two. Our discussion focuses on the theoretical implications of emotion classification and design recommendations for learning environments and emotional interventions in computer-based contexts.