Explain the mechanism, not the function

When you think you understand something, try to explain how it works step-by-step — not just what it does.

Why it works

The IOED arises because narrative knowledge ("a refrigerator keeps food cold") is fluent, accessible, and feels like understanding. Mechanistic knowledge ("the compressor circulates refrigerant which absorbs heat at the evaporator") requires explicit representation of causal steps. Most people have the narrative without the mechanism, but cannot detect the gap until explanation is demanded. Requiring a mechanistic explanation surfaces the gap immediately and precisely.

How to do it

1. Pick a concept you believe you understand well.
2. Set a timer for two minutes and explain out loud or in writing: "Step 1 is X because… Step 2 is Y because…"
3. Stop explaining at the first step you cannot fill in with a causal reason.
4. That stopping point is your actual level of understanding — not your initial rating.

Evidence

Rozenblit and Keil (2002) showed across multiple studies that self-rated understanding dropped significantly after subjects attempted to produce mechanistic explanations, demonstrating that subjective understanding vastly overestimates actual depth. Fernbach et al. (2013) extended this to political attitudes, finding that asking people to explain policy mechanisms — rather than justify their positions — moderated extremism and revealed the same overconfidence gap. (observational)

The IOED was established for mechanical devices and social policies; its magnitude may vary by domain and individual expertise. Experts in a domain show smaller IOED effects for their own field.

Sources

• Rozenblit & Keil (2002), "The misunderstood limits of folk science," Cognitive Science
• Rozenblit, L., & Keil, F. (2002). The misunderstood limits of folk science: An illusion of explanatory depth. Cognitive Science, 26(5), 521-562.
• Fernbach, P. M., Rogers, T., Fox, C. R., & Sloman, S. A. (2013). Political extremism is supported by an illusion of understanding. Psychological Science, 24(6), 939-946.
• Chi, M. T. H., de Leeuw, N., Chiu, M.-H., & LaVancher, C. (1994). Eliciting self-explanations improves understanding. Cognitive Science, 18(3), 439-477.

Common mistake

Explaining what something does or what it is for rather than how it produces its effect — which satisfies the feeling of explanation without exposing the mechanistic gap.