You think you understand how everyday things work. You don't. Until asked to explain step-by-step, we confuse familiarity with comprehension.
Pick an everyday object. Rate your understanding. Then try to explain how it actually works.
Describe the mechanism of in detail.
What happens physically when you use it? What are the components? How do they interact?
In the original Rozenblit & Keil (2002) study, participants rated their understanding at 4.1 on average before explaining, and 2.9 after—a drop of 1.2 points. The act of explaining forced confrontation with actual ignorance. We confuse exposure (seeing zippers daily) with understanding (knowing how they work).
We see objects every day and assume this exposure means we understand them. Recognition is not comprehension.
We can describe what something DOES easily. Describing HOW it works is another matter entirely.
We think in terms of inputs and outputs. The mechanism in between is a "black box" we never examine.
Real mechanisms involve multiple steps. We imagine we know them—until forced to list them.
Yale researchers asked participants to rate their understanding of everyday devices (toilets, zippers, locks, speedometers) on a 1-7 scale. Then they asked for detailed, step-by-step explanations. Then they re-rated. Every single category showed significant drops—averaging 1.2 points. The effect was specific to causal/mechanical knowledge; memory for facts and procedures didn't show the same illusion.
People with extreme political positions also showed IOED. Those asked to explain HOW policies would work (not just list reasons FOR them) became significantly more moderate. The mechanism: explaining exposes the gaps in understanding that confident opinions had papered over.
Strong opinions without understanding HOW policies work. Ask for mechanisms, not reasons.
Executives confident in strategies they can't explain step-by-step. The Feynman Technique works.
Students think they understand after reading. Writing explanations reveals gaps.
Users assume they know how products work—until error. Design must accommodate false confidence.
Public "understands" evolution, vaccines, climate—but can rarely explain mechanisms.
We think we know ourselves. Ask "how" not "why" your emotions work—humility follows.
Richard Feynman's learning method: explain a concept as if teaching a child. If you can't, you don't understand it. The IOED is the problem; forced explanation is the solution. Writing is thinking—the act of explaining exposes gaps that confident intuition hides.