People often believe they understand familiar systems in greater detail than they can explain. Rozenblit and Keil called this the "illusion of explanatory depth" after participants rated their understanding of everyday mechanisms highly, then struggled to describe them step by step.
Orthopaedic judgment combines pattern recognition, memory, interpretation, risk estimation, and the capacity to revisit an initial impression. Repetition makes predictions fast and accurate before a causal account has been assembled. This efficiency supports throughput and prevents decision paralysis, while confidence may develop before the surgeon can explain why a familiar pathway fits the patient.
A surgeon may know a classification and standard treatment while lacking enough mechanistic understanding to recognize when a similar case requires a different pathway. Internal fluency can masquerade as transferable knowledge. The surgeon may apply incomplete principles broadly without recognizing the gap.
Before treating an infrequent or borderline case as routine, we as surgeons should state the causal chain, indications, reasoning, and operative steps in language. What structure failed? Which forces produced the deformity? Why should the reconstruction resist those forces? Which patient variables threaten the result? How will early failure be detected? Explaining the case to a patient, parent, resident, partner, or supervisor audits understanding. Questions from others expose hidden assumptions.
Vague phrases deserve examination. "Acceptable alignment" requires an account of acceptable for whom, under which growth conditions, within what tolerance, and with what consequence if the estimate is wrong. When an explanation becomes vague, the surgeon has located a weak link.
Surgical intelligence is often credited to speed, memory, and technical skill. Those traits alone do not explain why some surgeons detect trouble earlier or improve faster. High-performing trainees choose effortful thought. They recheck imaging, test whether a cast, brace, implant, or instruction will remain effective after swelling, fatigue, or imperfect adherence, and plan follow-up outside idealized assumptions. Like a trained runner who selects running repeatedly, we become surgeons more capable of difficult reasoning through use.
An attending cannot analyze every step at maximum depth. Constant deliberation would consume attention and slow care. Expert automaticity develops from years of earlier analysis. The expert detects the tripwire: a subtle fluoroscopic step-off, a history detail that conflicts with the examination, or a reduction that appears satisfactory within a poor splint. Metacognitive calibration permits a shift from unreliable automatic processing to deliberate analysis.
Residents and fellows should explain the path from evidence to action and the purpose of each action without hiding uncertainty behind jargon. Useful questions include: Which finding would change treatment? What supports the indication to operate? Which step is most likely to determine whether the case proceeds smoothly? These questions build reasoning structures that can later be recalled quickly.
Explanatory-depth testing produces direct feedback. When an account becomes terminology-heavy and mechanism-light, the deficiency becomes visible. The answer may remain correct, although confidence in reasoning and execution should decrease. Our training often rewards rapid answers because speed resembles command. Judgment deserves greater weight. Under time pressure, trainees can present the diagnosis, supporting and conflicting evidence, treatment mechanics, failure mode, follow-up interval, and finding that would change management. Faculty feedback should target omitted links.
The same method applies to morbidity and mortality review. Complications should be examined for points where the team believed it understood more than it did. Missed compartment syndrome, fixation loss, delayed infection recognition, wrong implant length, bracing failure, and revision construct problems seldom arise from ignorance or laziness alone. They often reflect an untested assumption.
Uncertainty has structure. It may arise from random variation, measurement error, bias, model error, incomplete information, or precise language that conceals disagreement. Cobb angle depends partly on end-vertebra selection, and pain interpretation may be affected by family context. Population data describe possible outcomes, while each patient experiences one. Probability estimates do not remove the need to prepare for an adverse result. Expected-value reasoning becomes hazardous when the downside exceeds the capacity to absorb it.
Strong command of fundamentals supports extrapolation into unfamiliar cases. Understanding buttress plating in the proximal tibia may guide application of the same principle to a similar femoral fracture. Bayesian reasoning provides a parallel discipline: prior beliefs should change when new evidence arrives, in proportion to its reliability and relevance. Orthopaedic decisions often follow this pattern implicitly; the reasoning should be explicit enough to inspect.
A practical sequence follows. First, recognize borderline or discordant cases and require a stepwise explanation. Second, identify dominant unknowns and obtain information when it can change the decision. Third, define failure modes and a planned response. Fourth, explain the mechanism, operative steps, and reasoning to yourself and the team. Fifth, use disagreement to compare causal chains, available information, tolerated downside, reversibility, lived experience, and thresholds for changing course. Sixth, translate tolerances, model outputs, and published averages to the patient's constraints.
A familiar case can feel understood after repeated exposure while exact mechanics remain concealed within intuition. Repeated, effortful thought exposes weak links before the patient bears the cost. Some analyses will consume several minutes without changing management. A decision that survives causal explanation, explicit uncertainty, patient-level translation, and a defined failure response has undergone stronger examination than recognition alone.
What is the last familiar case that taught you something? Share in the comments.
