You Snapped at Someone for Not Understanding You. Your Brain Set You Up for It.
You're walking your dad through his new phone for the fourth time. You've done this exact sequence a hundred times on your own device. He taps the wrong icon again. Your voice gets short before you've even decided to let it. He goes quiet. You feel the guilt land half a second after the tone already left your mouth — and the worst part is you genuinely don't understand how he's not getting it. It's right there.
He's not slow. He hasn't built the map yet, the one you built so long ago you forgot you ever had to build it. And that forgetting isn't a character flaw. It's a documented feature of how your brain stores what it knows.
Economists found it first, in a place you wouldn't expect
In 1989, economists Colin Camerer, George Loewenstein, and Martin Weber published "The Curse of Knowledge in Economic Settings: An Experimental Analysis" in the Journal of Political Economy. They weren't studying phone tutorials or family arguments. They were running trading experiments, and they found something that shouldn't happen in a rational market: traders who had superior information couldn't fully strip that knowledge out of their predictions of what less-informed traders would do. Even when it was in their financial interest to model the other side accurately — even when they were being paid to get it right — knowing more made them worse at simulating what it felt like to know less.
That's the core of the bias, stated as plainly as it gets: once information is in your head, you lose reliable access to what your mind looked like before it arrived. You can't voluntarily un-know something to check your explanation against a beginner's version of yourself, because that version of you no longer exists to consult.
The experiment that made it visible: tapping a song nobody could hear
A year later, Stanford graduate student Elizabeth Newton ran the study that made this bias impossible to un-see. She had one group of participants tap out the rhythm of well-known songs — "Happy Birthday," things everyone would recognize — on a table, while a second person tried to guess the song from the tapping alone. Before the guessing started, Newton asked the tappers to predict how often the listener would get it right. Their average guess: about 50% of the time.
The actual result: listeners identified the song correctly in roughly 2.5% of trials. The tappers weren't being arrogant. Something stranger was happening — while they tapped, they could hear the song in their own head, the full melody riding along with every beat. It was so obviously "Happy Birthday" to them that a 50% guess rate felt conservative. The listener heard only knocking. There was no melody in the room. There was only ever a melody in the tapper's head, and the tapper had lost the ability to imagine a version of the moment without it.
That's the curse in its purest form. Knowledge doesn't just inform what you say — it silently overwrites your ability to model what the other person is working with, and it does this without asking permission or announcing itself.
So actually — you're not failing to explain. You're standing somewhere they haven't reached yet.
Here's the reframe that should have landed on you back in the phone-tutorial scene, and didn't, because the curse doesn't announce itself in the moment: when you explain something and watch confusion spread across the other person's face, your instinct reads it as their failure to keep up. It's closer to the opposite. You are skipping steps you don't know you're skipping, because those steps stopped being conscious the day you learned them. You think you're being clear. You're actually several steps ahead, describing a map from a location the other person hasn't arrived at, and you've forgotten the walk it took to get there.
This is where the bias turns from an academic curiosity into something that quietly damages the people closest to you. The short tone with your dad. The flash of contempt for a junior colleague who "should just get it." The reflexive "it's not that complicated" aimed at someone who is, in that exact moment, doing real cognitive work you did years ago and no longer remember doing. None of that is arrogance in the way it feels like arrogance from the outside. It's a brain that genuinely cannot locate its own earlier, less-informed state to use as a reference point — and mistaking that inability for the other person's deficiency is the actual cruelty here, dressed up as impatience.
The fix isn't more patience. It's a specific, answerable question.
Patience is the advice everyone gives and almost nobody can sustain, because patience asks you to override a bias you can't see happening in real time. The better move is mechanical, not emotional: the next time someone doesn't follow you, stop and ask yourself one specific question — what step did I just skip without saying it out loud? Not "why aren't they getting this." That question keeps you inside the curse, still assuming your map is the only map that exists. The better question forces you to locate the actual gap, the exact step that lives so deep in your automatic knowledge that you stopped narrating it years ago.
Struggling to understand something isn't stupidity. Struggling to explain something isn't arrogance. It's a brain that forgot the road it traveled to get here — and the person across from you is standing exactly where you used to stand, tapping out a song they can't hear either.
Next time you feel that flash of "how is this not obvious," stop. Ask what step you forgot to say out loud. Say that step next.
Follow The Shadow Self Media on TikTok · Instagram · Facebook