Why Every Topicly Discovery Ends With Three Quick Questions

Every discovery in Topicly ends with "Three quick questions" because recalling something is a stronger way to remember it than reading it twice. The questions take under a minute, they come only from the article in front of you, and they turn a pleasant read into something you are more likely to still know next week. If you would rather not, you can switch them off.

That is the short version. The rest of this post explains what the quiz actually does, why we built it that way, and where it falls short.

#What happens when you finish a discovery?

A discovery starts as an "In brief" summary, somewhere between 90 and 150 words. Below it sits the quiz. When you start it, the app opens the full story first, so the questions are never about text you have not had the chance to see.

Then you get three multiple-choice questions. Each has between three and five options. After you pick one, the app tells you "That is right" or "Not quite", and in both cases it adds a short explanation. The explanation matters more than the verdict. A wrong answer followed by one clear sentence about why is often the moment the idea actually lands.

At the end you see "You answered 2 of 3 correctly", or whatever your number is. There is no leaderboard to climb.

The questions are drawn only from that article and written in its language. If you read a discovery titled "The Moon is slowly leaving us." in German, the questions are in German and they ask about what that article said, not about lunar science in general. We made that rule on purpose. A quiz that tests things the article never mentioned feels like a trick, and it teaches you to distrust the questions instead of engaging with them.

#Why test someone who just read the thing?

It sounds redundant. You read it thirty seconds ago, of course you know it.

Except you often don't, or not in a way that lasts. Reading produces a feeling of familiarity, and familiarity is easy to confuse with knowing. The text was clear, every sentence made sense, so the brain files it as understood. Ask a specific question and the gap shows up fast.

The research behind this is old and fairly solid. Hermann Ebbinghaus showed in the 1880s that memory for new material drops steeply in the first day or so and then levels off. Much later, Henry Roediger and Jeffrey Karpicke (2006) had students either reread a passage or take a recall test on it. On a test given right away, the rereaders did about as well or slightly better. On a test days later, the students who had practised recall remembered noticeably more. This is usually called the testing effect, or retrieval practice, and it has been replicated many times across different kinds of material.

We are not claiming three questions will make you an expert. The studies involve more practice than a three-question quiz, and the size of the effect varies. But the direction is consistent, and the cost to you is maybe forty seconds. That trade seemed worth making the default.

If you want the broader picture of how to use this outside the app, our post on active recall goes into techniques you can use with any book or course.

#What is "Today's questions"?

The three questions after an article test what you just read. "Today's questions" tests what you read over the past few days.

It is a per-topic quiz of 10 questions, drawn from your recent discoveries in that topic, with a new set each day. So if you follow space and read about stardust on Monday, the Moon's drift on Tuesday, and something about Jupiter's moons on Wednesday, Thursday's set might mix questions from all three.

This is where most of the memory benefit probably lives. The first quiz catches misunderstandings while the article is fresh. The daily one asks you to pull something back after a delay, which is harder and, according to the spacing research, more useful. Our post on spaced repetition covers why a gap between reading and recalling helps.

Here is how the two compare:

Neither quiz costs anything against your plan's daily allowance. That was deliberate. If quizzes ate into the number of discoveries you could read, people would skip them to save reads, and the one habit most likely to help memory would be the first thing to go.

#What does the quiz not do?

A few honest limits.

It checks recall, not truth. Questions come from the article, and the article was written by AI. Every discovery carries the line "This discovery was generated by AI and may contain errors. Check original sources for important decisions." If the article got something wrong, a question can test you on that wrong thing. On higher plans, independent AI reviewers look over a draft before it is stored, which catches some problems. It is still AI review, not fact-checking against sources. For anything that matters, look it up.

It is multiple choice. Recognising the right answer among four options is easier than producing it from nothing. Free recall is stronger practice. If you want more out of a discovery, try this before tapping any option: cover the choices with your thumb and answer in your head first. Then check.

Three questions cannot cover a 1,000-word article. If your reading time is set to 5 minutes, the full story runs to about 1,000 words, and three questions will touch only a few of its ideas. That is fine. The goal is to make you retrieve something, not to audit everything.

It is not a course. Topicly is for steady, broad curiosity. If you need to pass an exam in Java, you need exercises and a real study plan. The quizzes help you keep what you read. They do not replace practice.

#When should you turn it off?

The quiz is on by default. You can switch it off in Settings, and some people should.

Maybe you read your discovery on the train and just want the story. Maybe you follow a topic like cinema purely for enjoyment and have no interest in remembering which year a film was released. Being quizzed on something you read for fun can make it feel like homework, and a habit that feels like homework tends not to survive.

A reasonable middle path: leave it on for the topics you actually want to learn, and simply don't tap the quiz on the ones you read for pleasure. Nothing breaks if you skip it. Your streak counts opening a discovery, not finishing a quiz.

#A small way to get more from it

Try this for one week. Pick the one topic you care most about learning. After each discovery, take the three questions. When you get one wrong, read the explanation out loud, or say it back in your own words. Then open "Today's questions" for that topic the next morning, before you read anything new.

At the end of the week, see whether you can explain two or three ideas from those discoveries to someone else without looking. That is the real test, and it is a better measure than any score.

If you are deciding how long each discovery should be, our guide to Light, Balanced or Deep explains how depth and reading time change what you get. And for how the article behind the questions is produced, see how Topicly writes a discovery.

Topicly is coming soon to iPhone and Android. You can see the download page for when it lands.