Your Notes Are a Transcript. Make Them a Test.

Most notes are a record of what was said. That is a reasonable thing to own, though it does less for you than a page you can be tested from. Converting one into the other takes about twenty minutes.

A record and a test are different objects

A transcript answers one question: what was covered. Open your page from Tuesday and you can confirm that the lecturer went through buffer systems, in that order, with that example. Useful for checking you did not miss a topic.

What the page cannot tell you is whether you can produce any of it with the page shut. That is the only question an exam asks. Reading a correct page and thinking "yes, that is right" is a recognition task. Sitting in a hall producing it from nothing is a retrieval task. They feel similar while you are studying and they come apart on the day.

So the test worth applying to a page of notes is whether someone could hand it to you and quiz you from it without giving the answers away. Accuracy and neatness matter less than that.

What retrieval buys that reading does not

Roediger and Karpicke ran the comparison directly. Students read a prose passage, then either studied it again or sat a free-recall test on it. On an immediate test, the group that had restudied came out ahead. After a delay of a week, the ordering flipped, and the group that had been tested remembered more.

The uncomfortable part of that result is the mismatch between how the two activities feel and what they do. Restudying is smooth. Recall is halting and exposes how little you have. Students consistently expect the smooth one to work better, and on a delayed test it does not.

The practical version: minutes spent with the answer in front of you are worth less than minutes spent trying to produce it. Reading gets the material in front of you, and retrieval is what makes it stay.

Why the transcript happens in the first place

Mueller and Oppenheimer compared notes taken by hand and on a laptop during lectures. The laptop notes contained more of the lecturer's exact wording. Handwriting is slower than speech, so the people writing had to choose what mattered, compress it and put it into their own words while the lecture continued. The people typing could keep up, and many of them therefore made no decisions at all.

The headline that came out of that work, that handwriting beats laptops, has not held up cleanly. Later attempts to reproduce the memory advantage have had mixed results, and the tool probably matters less than what you do with it. The mechanism is the part worth keeping. Verbatim capture is a way of postponing the thinking until later, and later is usually the night before. It produces a page that reads well and tests badly, whichever device you used.

Turning one page into questions

Take the page. Read a line. Ask what question that line is the answer to. Write the question somewhere separate from the answer, and mark the line so you can check yourself afterwards.

Take a line like this one, from second-year statistics. "A 95% confidence interval is a range constructed so that, over repeated samples, 95% of such intervals contain the true parameter."

Reading that a fourth time does nothing. Three questions you can be tested from:

• What does the 95% describe: this particular interval, or the procedure that produced it?
• A classmate says there is a 95% chance the true mean sits inside this interval. What is wrong with that, precisely?
• Two studies report the same estimate but very different interval widths. What differs about the studies?

The first checks the definition. The second targets the misreading nearly everyone makes at least once. The third asks you to use the idea rather than restate it. None of them can be answered by looking at the original line, which is the point.

Three kinds of line, three kinds of question

Most notes are made of three sorts of line, and each converts differently.

Definitions and terms. Make them two-way. Term to meaning, and meaning to term. If you can only go one direction, you have recognition. These are also the lines that survive well as Flashcards, because the answer is short and unambiguous.

Processes and sequences. Convert them to a blank sheet: write the stages in order and say what causes each move to the next. A sequence you can follow with your finger is not a sequence you can produce.

Distinctions and comparisons. These are where marks actually go missing, because two ideas that were clearly separate in the lecture blur together by week nine. Convert them to a discrimination question. Not "define active and passive transport", but "transport across the membrane stops when ATP production is blocked. What does that tell you, and what does it not tell you?"

If a line does not fit any of the three, it is probably context rather than content, and it does not need a question.

Keep the wrong answers

The first time you answer your own questions you will get some of them wrong. Write down what you said, next to what the notes say, and keep it. A wrong answer is more specific than a correct one: it names the exact place your understanding bends, and it tends to bend the same way again a fortnight later.

A list of your own errors is not pleasant reading, which is presumably why this step gets skipped. It is also short, and everything on it is something you have already demonstrated you need. In SymbioLearn this is what Mistakes holds, with the Source line attached to each one, so the page number, section or timestamp that settles the question is one step away rather than something you have to hunt for again.

What to do next

Pick one page from this week that you have already re-read twice, and spend twenty minutes turning it into questions rather than reading it a third time. Answer them tomorrow, not tonight, and write down what you got wrong. If you would rather have the questions drawn from your own Sources than write them yourself, a Quick Quiz in SymbioLearn does that, and every question keeps the page or timestamp it came from so you can check the answer against the original.