How this works

You are not bad at this.
You were taught to study wrong.

Almost everything that feels like learning isn't. Rereading, highlighting, and watching someone else write code feel productive. None of them do much, and there's a hundred and forty years of evidence saying so.

This page breaks down the learning science this course is built on, and it's worth the ten minutes, because knowing why a method works is what keeps you using it on the day it feels unpleasant.

1. You forget most of it, fast

In the 1880s, Hermann Ebbinghaus set out to measure his own memory. He invented nonsense syllables so no prior knowledge could help him, memorised long lists of them, and later measured how much effort it took to relearn each list. The gap between the two is what he had lost.

The shape he found is steep, then flat. Most of the forgetting happens almost immediately, in the first hours and the first day. Whatever survives that first drop hangs around far longer.

Trust the shape, not the exact numbers. Both studies rest on a single subject, and the material was nonsense syllables, the least memorable thing imaginable. So the curve tells you how forgetting behaves, not exactly how much of Tuesday's lesson is gone by Wednesday.

The useful part isn't the numbers, it's the timing. Forgetting starts immediately, which means the moment you close the tab feeling like you've got it is the moment decay begins. That feeling is not evidence of anything.

2. Recall is not recognition

Your memory can do two very different things, and confusing them is the most common way people mislead themselves about what they know.

Recognition is looking at something and knowing it's right. You read a line of code, it makes sense, you move on. Easy, fast, and it feels like understanding.

Recall is producing the thing from nothing. Blank editor, nothing open, and you have to write it. Slower, uncomfortable, and much closer to what the job actually asks of you.

You can recognise a for loop for months without being able to write one. Reading the solution and thinking "yes, obviously" is recognition wearing the costume of understanding.

It's why the course is honest about its own exercises. Multiple choice checks whether you hold an idea, not whether you can produce code. Predicting output and writing a function are recall, and those are the ones that build the skill. They're also the ones that will feel worse.

3. Retrieval is the thing that works

If forgetting is the problem, going over the material again seems like the fix. It mostly isn't.

In 2006, Roediger and Karpicke gave students passages to learn. One group studied, then studied again. The other studied, then took a test with no feedback. Everyone was tested later.

After five minutes, the restudy group did better. After two days, and again after a week, the tested group did substantially better, and the gap grew with time.

Read that twice, because the whole design of the course sits in the gap between those two results. The method that felt better immediately performed worse later. Judge your studying by how it feels at the time and you'll reliably pick the weaker one.

Dragging something out of your own head isn't a test of learning. It's what causes it. That's why the course is almost entirely exercises, and why there are no videos.

4. Why rereading fools you

Read something a second time and it goes down easier. That smoothness is real, but it's a fact about the text becoming familiar, not about you being able to produce it later. Psychologists call it fluency, and it's a terrible signal.

This isn't a hunch about rereading. In 2013, Dunlosky and colleagues reviewed ten common study techniques against the whole body of evidence and rated each one. Rereading and highlighting, the two most popular things students do, both came out low utility. Practice testing came out high, and so did spacing the practice out.

Worth sitting with: the two techniques almost everyone reaches for first are the two that were rated worst. Not useless, just far weaker than the effort spent on them suggests. Nobody taught most of us the other ones.

Retrieval feels the opposite. It's effortful, you stall, you get things wrong. That difficulty isn't a sign it's going badly. Here, a wrong answer is the most useful thing that can happen, because it's the one moment where the specific thing you misunderstand becomes visible and can be answered directly.

So: if it feels easy, be suspicious. If it feels like work, you're probably in the right place.

5. A wrong answer is information, if something reads it

Everything above is about getting you to attempt things. This part is about what happens the moment an attempt fails, which is the moment most learning products stop talking to you.

Say you typed this, and the property is misspelled:

for (let i = 0; i < flips.legnth; i++) {

Both of those tell you the same fact. Only one of them tells you what you believed. The sentence on the right is the whole difference, and it is the reason this course exists.

The distinction has a name in the research. Being told an answer is incorrect is verification. Being told what went wrong and why is elaborated feedback. In 2008, Valerie Shute reviewed the formative feedback literature for the Review of Educational Research. The pattern across it: feedback that explains tends to beat feedback that only marks. Specific beats general, and describing the error beats scoring it.

Two things that review does not say

It does not say feedback always helps. Shute is direct that it can backfire, and the two ways it does are the two this course is built around. Feedback arriving before you have genuinely tried removes the retrieval that does the learning, which is why nothing here reveals an answer until you ask for it. And feedback that reads as a judgement of you rather than of your answer makes people avoid the work, which is why nothing here is scored or ranked.

It also does not say anything about this course. That research is about feedback in classrooms and on tests, not about auto-graded JavaScript exercises with hand-written explanations. The idea is well evidenced. This particular implementation of it is not, and will not be until enough people have used it to measure.

The explanations are written by hand, one per mistake worth naming. When the product recognises what you did, it says so in a sentence. When it does not, it says less rather than guessing, and the hints and the worked solution are always there. A confidently wrong explanation would be worse than none.

6. Come back after you've started to forget

Put the two ideas together. Forgetting starts fast, and retrieval is what makes things stick. So the best moment to retrieve something is after forgetting has begun but before it's gone.

Practise something ten times in one afternoon and by the fifth go it's mostly recognition. Practise it four times across two weeks and that's four real acts of recall. Same effort, very different result.

In 2006, Cepeda and colleagues pooled 254 experiments covering more than fourteen thousand people. Spreading practice out beat cramming the same amount of practice together, consistently. The gap was not small and it was not a quirk of one lab.

The practical part is that the total amount of work doesn't change. Same number of repetitions, same minutes. Only the calendar moves, which makes this the cheapest improvement available to you.

It's why revealing an answer here doesn't finish an exercise. It gets marked as revealed, not done, and comes back later so you get to do it yourself. Reading a solution is recognition. Writing it afterwards is the part that counts.

What to actually do

• Try the exercise before you reveal the answer. A wrong attempt teaches you more than a confident read of the solution.
• Sit with being stuck for one minute, not ten, then take a hint. They cost nothing, and nothing is scored. Being trapped is what makes people quit.
• Study in short bursts, often. Twenty minutes four times a week beats a four-hour Sunday, and it isn't close.

None of this is about talent. It's method, and method is learnable, which means you can learn it too.

Free to start. No card.

Sources

• Ebbinghaus, H. (1885). Über das Gedächtnis. The original forgetting curve, measured by the savings method on a single subject: himself.
• Murre, J. M. J., and Dros, J. (2015). Replication and analysis of Ebbinghaus' forgetting curve. PLOS ONE, 10(7).
• Roediger, H. L., and Karpicke, J. D. (2006). Test-enhanced learning: taking memory tests improves long-term retention. Psychological Science, 17(3).
• Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., and Rohrer, D. (2006). Distributed practice in verbal recall tasks: a review and quantitative synthesis. Psychological Bulletin, 132(3). 254 experiments, over 14,000 participants.
• Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., and Willingham, D. T. (2013). Improving students' learning with effective learning techniques. Psychological Science in the Public Interest, 14(1). Ten techniques rated. Rereading and highlighting came out low utility; practice testing and distributed practice came out high.
• Shute, V. J. (2008). Focus on formative feedback. Review of Educational Research, 78(1). A review of what makes feedback work, and of the conditions under which it does not.