You finish a chapter. You understood every paragraph. You feel like you learned it.
Three weeks later, you remember the topic and almost none of the content.
Nothing went wrong with you. This is how memory normally works. The feeling of understanding while you read is a poor guide to what you will be able to recall later.
The good news is that the fixes are well known. The bad news is that they are not what most people do. And the methods that work best usually feel worse while you are doing them.
This page is short, and it will save you more time than any other page in the book.
Rereading and highlighting are the two most popular ways to study. They are also two of the least effective.
Here is why. When you reread a passage, it feels familiar. Familiar feels easy. Your brain takes that ease as proof that you know the material.
But familiarity is only recognition. It means "yes, I have seen this before." What you will need later is recall. That means producing the idea yourself, with no page in front of you. Recognition and recall are different skills, and rereading trains only the first one.
You can check this right now. Close this page and say the main points out loud. The gap between what you expected to say and what you actually said is the size of the illusion.
Highlighting has a similar problem. It feels like work, but it is passive. A highlighted page is just a decision about what would be worth learning, put off until later.
Trying to recall something strengthens the memory more than reading it again does. This is called the testing effect. It is one of the most solid findings in the science of learning.
It helps to understand how it works. Recall does not just measure a memory. It changes it. Each time you pull an idea out of your head with effort, the next pull gets easier and lasts longer. A test is not only a check. It is a learning event.
Effort is the key. If the answer comes instantly, you gain little. If you have to struggle for a few seconds, you gain a lot. So if it feels easy, it is probably doing less.
In practice:
Before moving on, close the page and summarise what you just read.
Answer the self-test questions at the end of a chapter without looking.
Try to recall something before you look it up. Even a wrong guess makes the right answer stick better. This is called the generation effect.
Reviewing something over several days beats the same total time in one sitting. This is spaced repetition, and it is the second solid finding.
The reason is simple. Forgetting a little, and then remembering successfully, is what makes a memory strong. If you review while the memory is still fresh, the recall is easy and does little. The best moment to review is just before you would have forgotten.
As the memory gets stronger, you can wait longer. A typical schedule is one day, then three days, then a week, then three weeks, then two months.
You do not have to track this by hand. A Leitner box is the paper version. Cards you recall correctly move to a box that is reviewed less often. Cards you forget go back to the first box. SM-2 and FSRS are software versions that do the scheduling for you. Any of them beats reviewing everything equally.
What about cramming? It really does work for tomorrow, and it fails by three weeks from now. Cram for an exam you will never need again. Do not cram anything you plan to use.
Studying A, then B, then C, then A again beats finishing all of A before starting B. This is called interleaving. It works even though blocked practice feels more productive.
Here is why. When every problem in a block uses binary search, you never practise the hardest step, which is noticing that binary search is the right tool. Mixed practice forces you to choose the approach each time. That choosing is the skill you will need on a real problem. The problem chapters in Part 4 are built on this idea.
Interleaving feels worse and works better. You will see that pattern again on this page.
Ask two questions as you read: "Why?" and "How does this connect to something I already know?"
Memory works by association. A fact with many connections has many ways to be found. An isolated fact has only one.
The strongest version of this is self-explanation. After a paragraph, explain it in your own words, out loud or on paper. The moment you cannot, you have found a gap. You could not see the gap while reading, because the text was supplying the words for you.
Concrete examples help too. A rule with no example is almost impossible to retrieve. The same rule attached to a real failure, such as a customer refunded twice or a database that slowed down at three million rows, has a hook to grab. This is why every chapter in this book opens with a real situation.
Pictures help as well. A diagram plus an explanation is remembered better than either alone, because the two are stored differently. Try drawing the diagram yourself, from memory, even badly. That combines two good methods at once, and it takes five minutes.
Here is the idea behind everything above. Conditions that make learning feel slower and harder often make it last longer. Researchers call these desirable difficulties.
Spacing, self-testing, mixing topics, and rebuilding an idea instead of rereading it all belong to this family. All of them make practice feel worse and make memory better.
This leads to a warning. Your own sense of which method is working is unreliable, because you judge by how it feels in the moment. Trust the evidence over the feeling.
There is one honest limit. A difficulty is only useful if you can still succeed. Material far beyond what you know is not a helpful challenge. It is just a wall. That is why this book is ordered the way it is, with each idea built before the one that needs it.
Read one chapter in one sitting. They are sized for that.
At each section break, stop and recall the last section in one or two sentences. It takes ten seconds.
Answer the self-test questions without looking. Write the answers down. Writing forces you to be complete in a way that thinking "yes, I know that" does not.
Come back after a gap. The next day, read only the recall box and try to rebuild the chapter from it. A week later, try to rebuild the recall box from the title alone.
Turn what you got wrong into cards. Only what you got wrong. A deck of things you already know wastes your review time. It is also the most common reason people quit spaced repetition: the deck grows faster than it pays off.
Build something with it. Using an idea for a real purpose is the strongest memory aid there is. A small project, a debugging session or an explanation to a colleague all count.
Teach it. Explaining a topic to someone else is the hardest test of understanding. The point where your explanation gets vague is exactly where your understanding is thin.
Good flashcards follow four rules:
One fact per card.
Phrase it as a question with a specific answer.
Use your own words and include a concrete example.
Delete any card that is not earning its place.
Twenty good cards reviewed for a year beat four hundred abandoned in a month.
Everything so far is about knowledge. Skills such as writing code, debugging and designing systems need a different kind of practice, called deliberate practice. It has four parts:
Work at the edge of your ability, not comfortably inside it.
Get feedback fast, and make it specific.
Repeat with corrections, instead of redoing the whole task.
Pick one weakness and target it.
You have probably heard "ten years of experience." Often that is one year of experience repeated ten times. Doing a job is not the same as getting better at it. The difference is whether you are working on something you cannot yet do.
For design skills, the order that works is this. First read a complete worked design. Then do one where the structure is given and you make the decisions. Then do one from scratch. Jumping straight to "design a system" mostly produces frustration. That is why Parts 9 and 11 teach the method before they set the problems.
Memory is consolidated during sleep. This is not a wellness tip. It is how the mechanism works. A late night of study followed by four hours of sleep often leaves you with less than stopping early and sleeping.
Exercise also improves memory formation, and the effect is not small.
And attention is the input to all of it. Studying with constant interruptions works at a fraction of the normal rate. This is not a lecture about focus. A memory needs your attention to form, and there is no way around that.
Use three tests. Each one is harder than the last.
Can you recall it without the page? This is the minimum.
Can you explain it to someone who does not know it, without jargon, and answer their follow-up question? This is where most apparent understanding falls apart.
Can you use it on a problem you have not seen before? This is the real test, and it is the one that an interview and a real job both use.
Spotting the pattern in an unfamiliar problem is what "understanding" actually means. That is why the chapters in this book end by pointing to the general idea, not with a summary.
Recall
Rereading and highlighting train recognition, but you need recall. Close the page and try to produce the idea. The gap shows you what you really know.
The testing effect: pulling an idea out of memory with effort strengthens it. Try to recall before you look it up (the generation effect).
Spaced repetition works because forgetting a little and then remembering is what strengthens a memory. Cramming works for tomorrow and fails by three weeks.
Mixing topics (interleaving) trains you to pick the right approach, which is the part blocked practice never trains.
Explain it in your own words, attach a concrete example, and draw it from memory. The place where you get stuck is the gap.
Harder-feeling methods are often better, so your feeling about what works is unreliable. The one limit: you must still be able to succeed.
Using this book: recall at each section break, write your self-test answers, return after a gap, and make cards only from what you got wrong.
Skills need deliberate practice. Work at the edge of your ability, get fast feedback, and target one weakness. Sleep is how memory is stored, so protect it.
Self-test: Why does rereading feel effective and fail? · What does recall do to a memory besides check it? · When is the best moment to review something? · Why does mixing topics beat blocks of one topic? · What is the honest limit on "harder is better"? · Which of the three tests of understanding is strongest, and why?