Experts forget what ignorance feels like
Explaining something you know well feels easy. You say what the thing is, mention the two details that matter, and the other person nods. Later it turns out they understood almost nothing, and the part that lost them was the step you did not consider worth mentioning. The name for this comes from economics rather than teaching. In 1989, Colin Camerer, George Loewenstein and Martin Weber ran experiments in which better-informed participants had to reproduce the judgments of people who knew less than they did. Standard economic models assume this is straightforward: if you know more, you can always imagine knowing less. The participants could not do it, and they lost money by failing. Market competition cut the effect roughly in half, but never removed it.
"Better-informed agents are unable to ignore private information even when it is in their interest to do so; more information is not always better." They called it the curse of knowledge, and the crucial word is unable. This is not arrogance or impatience. Knowledge is not something we can put down for a moment in order to see the problem as a beginner sees it. A doctoral study by Elizabeth Newton at Stanford in 1990 shows the mechanism almost too neatly. Participants tapped the rhythm of a well-known song on a table while someone listened and tried to name it. The tappers expected about half of the listeners to succeed. In reality, roughly one tune in forty was identified, a rate below even the most pessimistic prediction any tapper made. The reason is that the tapper does not hear tapping. They hear the whole song: the melody, the words, the arrangement. The listener hears knocking on wood. The tapper cannot subtract the music from their own head, so the listener's failure looks like inattention rather than a completely different experience. This explains a lot of ordinary frustration. The senior engineer whose documentation assumes the reader already knows why the service exists. The doctor who explains a diagnosis in language that has no meaning outside the profession. The manager who briefs a task by describing the last third of it. In each case the explanation genuinely felt complete to the person giving it, because it was complete against everything they already knew. The most treacherous part is that the feeling of clarity is produced by our own knowledge, so it cannot tell us whether the explanation works. Which means asking "is that clear?" is close to useless: it tests politeness, not understanding. What helps is external evidence. Give the instruction to someone unfamiliar and watch where they stop, instead of asking whether they followed it. Say the obvious step out loud, especially the one that feels insulting to mention. Keep notes of what confused you when you were learning, because that record is the only honest access you will ever have to your own earlier ignorance. Lead with a concrete example and let the general rule come second. Expertise does not automatically make you a better explainer. It quietly removes your ability to tell whether you are one.
Source: Camerer, Loewenstein and Weber, The Curse of Knowledge in Economic Settings, Journal of Political Economy (1989) Elizabeth Newton, The Rocky Road from Actions to Intentions (1990)
- C. Camerer, G. Loewenstein, M. Weber