Think of the last time you explained something you know well, like how to get to your house, how your job works, or why a certain look from your mother means trouble, and watched the other person's face go blank. You probably felt a small flash of impatience, maybe even a quiet judgment: how are they not getting this? The uncomfortable truth is that the problem probably wasn't them. It was what you know. Once you understand something, your mind loses access to what it felt like not to understand it. You can't get that back by trying harder. You have to go looking for it.
Psychologists call this the curse of knowledge. It's one of the most common and least noticed reasons people feel misunderstood, by colleagues, by children, by the people they love.
A song nobody could hear
The clearest demonstration comes from a 1990 Stanford dissertation by Elizabeth Newton. She split people into two roles. "Tappers" picked a well-known song, like "Happy Birthday," and tapped its rhythm on a table. "Listeners" tried to name the song.
Before the listeners guessed, the tappers predicted how often they'd succeed. They expected about half the songs to be recognized. Out of 120 songs tapped, listeners identified three.
The gap between fifty percent and two and a half percent is the curse of knowledge. As the tappers tapped, they heard the melody in their heads, with the lyrics and the swing of the tune, and couldn't stop hearing it. The listeners heard knocking on wood. The tappers weren't careless or arrogant. They just couldn't hear their own tapping the way the listeners did, because the song was always playing for them.
Chip and Dan Heath later made this experiment famous in Made to Stick, and it has stuck because almost everyone recognizes themselves in it.
Where the idea comes from
The phrase itself comes from economics. In 1989, Colin Camerer, George Loewenstein, and Martin Weber published a paper showing that better-informed people had trouble predicting how less-informed people would judge something, even when getting the prediction right would have earned them money. Knowing more didn't help them model a less-informed mind. It got in the way.
This connects to an older finding. In the 1970s Baruch Fischhoff documented hindsight bias: once people learn how something turned out, they overestimate how predictable it was. He called it "creeping determinism." Knowledge seeps backward and rewrites your sense of what was knowable before.
Later work showed the effect in places you wouldn't expect. Developmental psychologists Susan Birch and Paul Bloom adapted a classic false-belief task, the kind used to test whether young children understand that other people can believe things that aren't true, and gave it to adults. When adults knew where an object had really been moved, they were more likely to judge that someone who hadn't seen the move would look there too, at least when that spot seemed plausible. Grown adults who pass these tests easily still let what they know leak into what they assume others know.
The mechanism is simple. Understanding isn't stored next to a memory of not understanding. It replaces it. When you learn what a word means, you can't go back to hearing it as noise.
Why this hurts more than it should
The curse of knowledge is usually discussed as a writing problem or a teaching problem. It's both. But it's also a relationship problem, and that's where it hurts most.
Think about the partner who says "you know what I mean" and gets hurt when you don't. Or the parent who has tied shoelaces ten thousand times and can't understand why a five-year-old keeps getting it wrong. Or the senior colleague whose instructions skip three steps because those steps stopped feeling like steps years ago.
In each case, the person who knows more experiences the gap as the other person's failure. They're not listening. They're not trying. They're slow. The person who knows less often feels it as their own failure. I should get this by now. Something is wrong with me. Both are wrong in the same direction. Neither sees that one of them is hearing a song and the other is hearing taps.
That's the quiet cost. The curse doesn't only make explanations worse. It pushes us toward contempt in one direction and shame in the other.
The fix is curiosity, not effort
It's tempting to think the answer is to try harder to imagine the other person's view. Research on perspective-taking suggests that's surprisingly unreliable. Simply imagining another person's mind often doesn't make our guesses about them more accurate, because we fill the gaps with our own assumptions. You can't imagine your way back into ignorance, because your imagination is made of what you know.
What works better is to get real information from the other mind: ask, test, watch. This is where curiosity changes everything. Curiosity is the urge to close an information gap, and the curse of knowledge hides a gap from you, the gap between what you know and what someone else knows. Once you treat that gap as something worth finding out about, instead of something you already understand, the conversation changes.
The question shifts from "Why don't they get it?" to "What are they hearing right now?" The first question is a complaint. The second is something you can actually go and find out.
Writing teachers have long noticed this. Steven Pinker, in The Sense of Style, calls the curse of knowledge the single best explanation he knows for why good people write bad prose. His advice isn't to imagine the reader harder. It's to show drafts to real readers and watch where they get confused. In other words, stop guessing and ask.
What a curious explainer actually does
People who explain well aren't usually more articulate than anyone else. They're more interested in the listener. They tend to:
- Start with a question, not an explanation. "What do you already know about this?" tells you where the other person's map ends.
- Treat confusion as data. A blank look isn't an insult or a failure. It shows exactly where your shorthand ran out.
- Use concrete examples. Abstraction is the native language of expertise. Examples are the native language of newcomers.
- Name the steps that no longer feel like steps. The moves that have become automatic for you are the ones most likely to go missing from your explanation.
None of this needs talent. It needs you to stay curious about something you'd rather assume: what it's like inside someone else's head.
Your next moves
- Run the tapper test on yourself today. Before explaining anything this week, whether a work process, a recipe, or directions, write down how clear you expect it to be on a scale of 1 to 10. Afterward, ask the listener to rate it. Notice the gap.
- Ask one "starting point" question before your next explanation. Try: "What do you already know about this?" or "Where does it start to get fuzzy for you?" Let their answer, not your assumption, decide where you begin.
- List the invisible steps. Pick one task you're expert at and write out every step as if for someone who has never done it. Circle the steps you would normally skip. Those are your blind spots.
- Swap one abstraction for an example. In your next message, email, or talk, find the most abstract sentence and add "for example..." right after it.
- Rewrite one "you know what I mean." The next time you feel that phrase coming, with a partner, friend, or coworker, stop and ask instead: "What did that sound like to you?"
The gift of remembering you didn't know
The deepest form of the curse of knowledge is forgetting that you were once a beginner, that the thing which now feels obvious once felt like taps on a table. The people who stay good at explaining, and stay gentle with others, tend to keep some connection to that earlier self. They stay curious about not-knowing, their own and everyone else's.
That's part of why we built CurioKit. It's a small place to hold onto your questions as they come up, before the answers smooth over what it felt like to ask them, and to keep practicing the kind of curiosity that reaches past your own head into other people's. If you want a daily habit that keeps your beginner's mind within reach, you can explore it at curiokit.lumenlabs.works.