Response to: Beyond Accuracy: Interpreting Artificial Intelligence Performance in Surgical Decision-Making

We thank Daneshi et al for their interest in our recent publication, “Human vs Machine: The Future of Decision-making in Plastic and Reconstructive Surgery.”1,2 We appreciate their engagement with the topic, but we would like to take this opportunity to clarify the rationale behind the design of our study and to address several of the points raised in their letter. We believe that ongoing academic discourse on the evolving role of artificial intelligence (AI) in surgical education is both necessary and constructive, and we are pleased to contribute to the conversation through both our current and future research. We acknowledge the authors’ concerns regarding the limitations of multiple-choice questions (MCQs) in fully capturing clinical reasoning, but we would like to emphasize that the primary aim of our study was to assess knowledge acquisition following exposure to continuing medical education (CME) material. MCQs remain a valuable and standardized tool in medical education, particularly for ensuring consistency and comparability across learners and systems. Although they may not reflect the full depth of individualized decision-making, making a selection among multiple options is in itself a form of decision-making. Particularly when MCQs are designed to reflect clinically relevant scenarios, they challenge the test-taker to apply reasoning, eliminate incorrect choices, and justify a preferred answer, which are processes that engage cognitive processes that are similar to those used in clinical environments.