When expert intuition locks on, ask: 'What if it is NOT X? What would I see if it were Y?'
When expert pattern recognition locks onto a familiar solution, force consideration of alternatives by asking 'I see X—but what if it is not X? What would I expect if it were Y?' before committing.
Why This Is a Rule
The Einstellung effect (Luchins, 1942): expertise in recognizing familiar patterns prevents experts from seeing better solutions when the familiar pattern isn't optimal. A chess master sees a known 5-move sequence and commits to it, missing a simpler 3-move checkmate that a fresh-eyed player would find. The expert's pattern recognition — normally an enormous advantage — becomes a trap when it locks onto a familiar but suboptimal solution.
This happens precisely because expert pattern recognition is fast and confident. The familiar solution arrives fully formed, feels obviously right, and triggers the commitment response before alternatives are considered. The Einstellung effect is invisible from inside — the expert believes they've found the best solution when they've found the first recognizable one.
The disruption question — "I see X, but what if it is NOT X? What would I expect if it were Y?" — breaks the lock by forcing consideration of an alternative frame before commitment. This doesn't slow expert performance significantly (the question takes 10 seconds) but catches the cases where familiar-but-suboptimal solutions would otherwise be committed to without examination.
When This Fires
- When your expert pattern recognition produces an immediate, confident answer
- When a familiar solution to a new problem presents itself within the first minute
- During debugging when "I know what this is" arrives before thorough investigation
- Any situation where expertise makes you feel certain before you've considered alternatives
Common Failure Mode
Asking the disruption question rhetorically and immediately dismissing the alternative: "What if it's not a database issue? Nah, it's definitely the database." The question only works if you genuinely consider the alternative — which means specifying what evidence you'd expect to see if the alternative were true, then actually checking for that evidence.
The Protocol
When expert pattern recognition locks on: (1) Notice the lock-on: "I'm confident this is [X]." (2) Ask: "But what if it's NOT X? What would I expect to see if it were [Y — a specific alternative]?" (3) Check: is the evidence for Y present? (4) If Y's evidence is absent → X is strengthened. Proceed. If Y's evidence is present → your initial recognition missed something. Investigate further before committing. The 10-second question occasionally prevents the 30-minute investigation of the wrong cause.