Interactive lessons

Three short lessons where you make a prediction before you see the answer. Each has one control that changes the underlying cause, and you watch the answer change before anything explains why.

Made
–
Built with
HTML, CSS and a little JavaScript
Status
Live, three lessons

The question

Can five minutes on a web page make you understand something you'd otherwise only read about? The test I care about is the moment of surprise: you commit to an answer, the evidence disagrees, and you see exactly why.

How it works

Every lesson opens with a question, before any explanation, and your first answer is final. The buttons lock once you choose, so you can't click around until the right one lights up, and the end of each lesson lists what you predicted and whether you were right first time. You only find out after the simulation, the split table or the proof, never the moment you answer.

Each lesson has one control that changes the cause. You can make the Monty Hall host guess instead of know, move a hospital's patients between groups, or shrink the squares. The formula comes last, as shorthand for something you've already watched happen, and the last question is a different problem with the same structure: the three prisoners, university admissions, √3.

Each page reads as a complete explanation without JavaScript; the script only adds the interaction. The lessons work with a keyboard, announce results to screen readers, and hold still if you've asked your system for reduced motion.

The first version

An earlier version had eleven hand-built lessons and a generator that wrote new ones. An adversarial review of it was blunt. Most lessons were screens to click through, with quiz checkpoints on the way, and retrying until a gate opened counted as passing. Generated lessons were more expository than the hand-built ones: they told you the mechanism before letting you discover it. And nothing showed that anyone learned anything.

These three are rebuilt from scratch to answer the first three points. Answers lock, the verdict comes after the evidence, and each lesson is designed around the cause you can change, where the old ones were a sequence of screens. The fourth point still stands (see below).

Evidence

There are three lessons, each about five minutes, with four to six steps. Each loads under 7 KB of gzipped JavaScript: a shared 2.7 KB runtime plus 2 to 4 KB for the lesson itself. Only these pages run scripts; the rest of the site has none.

I walked each lesson through end to end in headless Chrome, every step and widget, and the page with JavaScript switched off. The controls are native buttons, radio buttons and a slider, plus a 100-door grid you can move around with the arrow keys. The simulations behave as the maths says: over 1,000 Monty Hall games, switching wins about 67% with a host who knows and about 50% with one who guesses.

What didn't work

There's no learner evidence yet. I tested that the lessons work, but I haven't tested whether they teach. A real test needs people who haven't seen the answer, a new problem afterwards, and a check a week later.

The first build gave the answer away. It said "Not quite" the moment you answered, before you'd played a single game. Holding the verdict back was the fix, and it changed how every lesson is structured.

I designed these by hand. The first version's finding holds: generating lessons that let you discover things, instead of telling you, is the hard part, and it isn't solved here.

They only suit problems with a mechanism. Probability, statistics and proof have a fact to be wrong about. History and other argued subjects would need a different shape, built around evidence and counterfactuals, and none of these try that.