William Osler kept a copy of Thomas Browne's *Religio Medici* on his nightstand for thirty years — not for its theology, but for its 1643 argument that the physician's chief instrument is memory: the practiced habit of holding prior patients fully in mind while meeting the one in front of you. Osler called this 'the educated sense of the past,' and what he meant was something precise. Memory in medicine isn't archive retrieval — it's the way accumulated encounters actively shape perception, the way a clinician who has seen a hundred presentations of a disease sees the hundred-and-first differently than a clinician who has seen ten. The philosopher Mary Warnock, building on Henri Bergson, made the same point from a different direction: memory is not storage but structured anticipation, a set of primed templates that determine what counts as salient. The clinical danger this points to is one with no common name — not bias exactly, not overconfidence, but something closer to perceptual narrowing: the more fluent your memory, the more it selects for familiar signal and filters out the feature that doesn't fit the template. The discipline, then, is not to remember less, but to treat each additional case as evidence that could revise the template itself, not merely confirm it.

What is the most recent case where the template you brought to the room actively prevented you from seeing something that was there?

Drawing from British empiricist philosophy of memory (Bergson, Warnock) synthesized with Oslerian clinical pedagogy — Mary Warnock (Imagination, 1976; Memory, 1987) synthesized with William Osler (Aequanimitas, 1904) and Henri Bergson (Matter and Memory, 1896)