For each surprise, write: 'What did I apparently believe that turned out wrong?'
For each captured surprise, write one sentence answering 'What did I apparently believe that turned out to be wrong?' to convert observations into explicit model gaps.
Why This Is a Rule
Surprise is the feeling of a prediction error — your mental model expected one thing and reality delivered another. The surprise itself is raw signal. But unless you extract the implicit belief that the surprise violated, the signal degrades into a vague memory of "that was unexpected" without producing any model update.
The diagnostic question "What did I apparently believe that turned out to be wrong?" performs the extraction. It converts a feeling (surprise) into a specific, falsified belief — which is the most valuable unit of learning. "I apparently believed that the database could handle this query volume without indexing" is a concrete model gap you can fix. "The database was surprisingly slow" is an observation that doesn't identify what needs updating in your mental model.
One sentence is the constraint that makes this sustainable. You don't need a full analysis — just the falsified belief. The sentence format ("I apparently believed X, but Y") forces precision without demanding time.
When This Fires
- After any event that surprised you — positive or negative
- During journaling or daily review when processing the day's events
- When something "shouldn't have happened" according to your understanding
- Any time reality diverged from your expectation, no matter how small
Common Failure Mode
Recording the surprise without extracting the implicit belief: "I was surprised the client said no." This captures the event but not the model gap. The extracted version: "I apparently believed the client's objection was about price, when it was actually about implementation timeline." Now you know which belief to update, which changes how you prepare for the next client conversation.
The Protocol
For each captured surprise: (1) Write: "I was surprised that [what happened]." (2) Ask: "What did I apparently believe that turned out to be wrong?" (3) Write one sentence: "I apparently believed [falsified belief]." (4) Optionally: "The more accurate model is [updated belief]." Steps 2-3 take 30 seconds and convert an ephemeral feeling into a permanent model update.