The bicycle you can't draw
Here is a small, humbling experiment you can run in your own head. Picture an ordinary bicycle. You've seen thousands of them; you may have ridden one this week. Now imagine someone hands you a page with only two wheels drawn, and asks you to fill in the rest—the frame, the pedals, the chain—so that the thing could actually roll.
Most people are quite sure they can. Then they try. In a well-known demonstration, the psychologist Rebecca Lawson asked people to complete exactly this kind of sketch, and a striking number produced bicycles that could never work: frames connecting the wrong points, chains looped around both wheels, pedals floating off the frame. These weren't people who had never seen a bicycle. They were people who had confused seeing a bicycle with understanding one.
That gap has a name, and once you know it you start noticing it everywhere.
What the illusion of explanatory depth actually is
In 2002, the Yale researchers Leonid Rozenblit and Frank Keil ran a series of studies on how well people think they understand everyday things—zippers, flush toilets, door locks, how a helicopter flies. First they asked participants to rate their own understanding on a scale. Ratings were high; people felt they basically got it. Then came the catch: write out, in detail, exactly how this works, step by step.
Under that instruction, confidence collapsed. Having to produce an actual mechanistic explanation—this part turns, which pulls that, which releases the other—revealed that most people were carrying around a vague, cartoon-level sketch rather than a working model. When they rated themselves again afterward, the numbers dropped. They had discovered, in real time, that they knew less than they'd assumed.
Rozenblit and Keil called this the illusion of explanatory depth: the feeling of understanding how something works that far outruns the actual ability to explain it. Crucially, it's specific. We don't hold the same illusion about raw facts (the capital of a country) or procedures (the steps of a recipe). It clusters around causal, explanatory knowledge—the how and the why. That's the knowledge most likely to feel solid and most likely to be hollow.
Why your brain lets you get away with it
The illusion isn't stupidity. It's an efficient shortcut that usually serves us well.
Part of it is that a vivid, familiar image stands in for a working model. Because you can instantly picture a bicycle in crisp detail, your mind quietly accepts that vividness as evidence of understanding. The picture is sharp, so the knowledge feels complete. But a photograph of an engine is not the same as knowing what fires when.
The deeper reason, argued by the cognitive scientists Steven Sloman and Philip Fernbach in The Knowledge Illusion, is that we live inside a community of knowledge. You don't need to understand how your phone, your medication, or your city's water system works, because someone does, and you can borrow their competence. Understanding is distributed across people, books, and tools—and the mind is not very good at drawing the line between what it knows and what it merely has access to. Knowledge that lives in the room feels like knowledge that lives in your head.
Most of the time this is fine. It becomes a problem the moment you actually need the model yourself—to make a decision, explain your reasoning, teach someone, or notice that your confident opinion rests on almost nothing.
The fix is embarrassingly simple: try to explain it
Here is the useful part. The same act that exposes the illusion is also what dissolves it. If forcing yourself to explain how something works reveals the gaps, then explaining is exactly the tool that fills them.
Education researchers have studied this for decades under the name the self-explanation effect. When learners pause and explain a concept to themselves in their own words—narrating the causal steps rather than rereading—they understand more deeply and transfer it to new problems better than learners who simply review. The effort of generating an explanation forces you to connect the pieces you'd otherwise let sit as a blurry whole.
The effect reaches past textbooks. In a 2013 study, Fernbach and colleagues asked people to explain, mechanistically, how a contested policy would actually produce its effects. People who attempted the explanation—and ran into how little they could say—became more moderate and less certain of their positions. Simply listing reasons they already believed did nothing. It was the specific demand to explain the mechanism that punctured the false confidence. Explaining doesn't just teach; it calibrates. It tells you what you really know.
Writing is the version that works
Explaining out loud helps, but writing is where this gets its teeth. Speech is forgiving—you can trail off, gesture vaguely, and your listener fills in the rest. A blank page does not fill anything in. It sits there until you produce the next actual link in the chain, and the sentence you can't finish is precisely the gap in your understanding, made visible.
This is why "write it out" is such reliable advice for anything you think you grasp but have never tested. Trying to explain a decision you're about to make. A concept from something you read and nodded along to. How a part of your job actually works, end to end. The idea you'd defend confidently in an argument. Give yourself a page and the single instruction Rozenblit and Keil used: explain how this works, step by step, as if to someone who knows nothing.
One of two things happens, and both are wins. Either the explanation flows, and you've earned the confidence you already felt—now backed by a real model. Or you stall halfway through, discover the joint you can't articulate, and you've located, for free, the exact thing worth learning next. The illusion thrives on staying unspoken. Written down, it can't survive contact with its own gaps.
Give the gap somewhere to show up
The reason most of us never run this test isn't disagreement—it's friction. The illusion only breaks when you actually start writing, and if capturing a thought means hunting for the right app, waiting for it to load, and deciding where a half-formed explanation belongs, the moment passes and the comfortable feeling of understanding stays intact.
That's the quiet case for keeping a fast, open page within reach. Pagebox opens in under a second and holds a plain space to think in—a note to explain a concept to yourself, a journal entry to reason a decision all the way through, a running page for the things you assumed you understood until you tried to write them down. Local-first and instantly synced, it's less a filing cabinet than a place to catch the sentence you can't finish—the one that shows you where your real understanding ends. If you want somewhere to test what you actually know, you can start writing at https://pagebox.lumenlabs.works.