The New York Times used to publish a column called “Think Like a Doctor” by Lisa Sanders, a doctor and author. Each posed a medical mystery—the print version of an episode of House M.D.

Here’s an example mentioned by Anders Ericcson in “Peak”:

  • A thirty-nine-year-old male police officer had an earache, and his right pupil was smaller than his left.
  • He took antibiotics to treat an ear infection.
  • When the earache reappeared two months later, the antibiotics didn’t help. The doctor thought it was probably just a sinus infection, but because of the pupil issue, referred the patient to an eye doctor.
  • The eye doctor referred the patient to a specialist.
  • The specialist immediately recognized the small pupil as a symptom of Horner’s syndrome but had no idea of the cause and the relation to the ear pain. So he asked some questions until the symptoms seemed to align.

Solving what caused the pupil to be smaller than the other was straightforward: it only required having learned about the syndrome at some point and recalling its symptoms. But it was piecing that information together with the ear pain that proved difficult.

Less experienced doctors tend to diagnose and jump quickly to conclusions and discard seemingly irrelevant information. They fail to generate multiple options. They don’t dive deeper when things don’t seem to add up.

Expert diagnosticians can consider many different facts as pieces of larger patterns, not isolated information.

Experienced diagnosticians come up with possible diagnoses and then analyze the various alternatives to select the most likely one.

The solution to the medical mystery described in the New York Times required that sort of approach:

  • Stroke and shingles were both possible of causes, but the patient displayed no symptoms of these illnesses.
  • A third possibility was a tear in the carotid artery wall, which runs right alongside the nerve affected in Horner’s and passes near the ear. A slight tear in the artery can cause this nerve to press into the face and, in rare cases, press on a nerve to the ear.
  • This possibility prompted the specialist to ask the patient questions about lifting weights. Weightlifting can sometimes tear the carotid artery, which is typically associated with headaches. When the patient answered yes to lifting weights and having a headache, a tear in the carotid artery was the most likely diagnosis. An MRI scan verified this, and the patient was put on blood thinners to prevent the formation of a blood clot while the vessel healed.

The key to the successful diagnosis was having the necessary knowledge organized and accessible to allow the doctor to come up with possible diagnoses and zero in on the most likely. The doctor pressed on until the clues added up.

A simple way to apply more complex and integrated knowledge structures to your thinking is by using “if…then” structures.

If these things are true, then it must be or must not be the case that…