If you want to remember the book months from now, spread it out. If you need it tomorrow, one sitting wins. The catch the usual advice skips: spreading a book across days only pays if every session starts by pulling back the last one, with the book shut.
If you are reading to keep what the book says:
- Read in chapter-sized sittings. One chapter, or one real section of a long one. Stop at a joint in the argument, not mid-paragraph.
- Before you open the book again, recall the last sitting. Two minutes, book closed: what was the claim, what was it arguing against, what came next. Write it or say it.
- Then check it against the page and read on. The gaps you find are the part you would have lost.
- Bring each chapter back once more after a delay, days later, the same way. How far apart those returns should be has its own page: a recall schedule for a book, without flashcards.
The rest of this page is what the evidence behind that looks like, and where one sitting is the right call.
Is it better to read a book in one sitting or spread it out?
Spread it out if you want it to last; read it in one go if you need it tomorrow. That is the pattern in the closest experiment anyone has run, and it used a television series, not a book.
In a study published in First Monday on 4 September 2017, Jared Horvath, Alex Horton, Jason Lodge and John Hattie at the University of Melbourne had 51 students watch all six hour-long episodes of a BBC drama none of them had seen (the paper). Seventeen watched one episode a week. Seventeen watched one a day. Seventeen watched all six in a single day, with three-minute breaks and a lunch break after episode three. Everyone then took 60-question memory quizzes one day, seven days and 140 days after their final episode.
The binge group remembered the most at one day and lost it fastest. In the authors’ words, the weekly viewers “demonstrated the weakest memory performance 24 hours after watching the program, but this performance showed the shallowest decline over the ensuing 140 days.” Daily viewers sat in between. The binge group also reported enjoying the show less.
That is the answer most readers are after, and it matches what a lot of people describe about a book finished in one weekend: vivid on Monday, gone by the next month.
Hold it loosely. Seventeen people per group. Television, not text. And every group took two quizzes before the 140-day one, so each had already had to pull the show back twice. That matters for the next section.
Why spreading it out is not the same as spacing
The spacing effect, the finding that the same material studied twice with a gap sticks better than studied twice in a row, is one of the best-replicated results in learning research. It is also a finding about repetition. Reading a book once, across ten evenings, repeats nothing. Chapter 7 is met once, whether chapter 6 was an hour ago or a week ago.
The Melbourne authors say this themselves. The spacing research “is almost exclusively conducted exploring practice (repeated exposure to and revision of the same set of stimuli)”, and whether there is a similar effect for new material that is never repeated “is simply a theory at this stage.”
The cleaner evidence is on rereading. In the Journal of Educational Psychology (February 2005, volume 97, pages 70–80), Katherine Rawson and Walter Kintsch reported two experiments with 423 college students who read a text once, twice in a row, or twice a week apart (the abstract). Tested immediately, reading it twice in a row beat reading it once, and the week-apart group did not. Tested two days later, the week-apart group beat reading once, and reading twice in a row “no longer differed significantly” from reading once. The second reading in a row had bought a better score right away and then stopped counting.
So the spacing benefit is real for material you meet more than once. A single spread-out pass only gets it if something in the schedule makes you meet the material again. A television series does that for you. It keeps reminding you what happened before. A book does not, and a reader who opens it on page 140 and carries on has not met page 120 again. That is why step 2 above exists: two minutes of recalling the last sitting before you read on is the second encounter. It is also retrieval, which does more for memory than a second look at the page does. That argument is made properly in why you forget what you read.
Why the spread-out schedule feels worse while it works
If you have tried it, you know the complaint. Coming back to a book after four days feels slow. You have lost the thread. It seems like you are doing it wrong.
That feeling is in the data. In a preregistered classroom experiment published in Frontiers in Psychology on 9 January 2019, Carla Greving and Tobias Richter had 191 seventh-graders reread a 977-word text either straight away or a week later (the paper). The week-later group “reread the text more slowly, predicted their learning success to be lower, and reported a lower on-task focus.” Minutes later, the straight-away group scored higher. A week later, the week-later group had shown “no forgetting,” and the straight-away group had lost its lead.
These were twelve-year-olds rereading one short text, not adults with a book. The shape is the useful part: the schedule that lasts felt like the worse one at the time. In the Melbourne study, viewing schedule made no difference to how well people thought they understood the show, or how well they thought they would do on the quiz. Your sense of how well a sitting went is not reporting what will still be there in a month.
When one sitting is the right call
The pull toward a single long session is not a character flaw, and sometimes it is right.
- A novel you are reading for the story. The momentum is the point. If you would not want to be quizzed on it at 140 days, the binge costs you nothing you care about. The Melbourne binge group did report less enjoyment, but the authors warn this may be about spending a whole afternoon in a lab booth, so do not let that result decide your Saturday.
- A short book you will discuss tomorrow. One sitting is what the next-day score rewards. Just know it will fade faster after that.
- A first skim before a slow read. Going through a dense book fast to see its shape, then slowly chapter by chapter, is a deliberate way of reading it twice. The second pass is what gets spaced.
What one sitting should not be is the only encounter with a book you want to keep. If you did binge it, the fix is the same as step 4: close it, and a few days from now write down what each part argued, from memory. Then check.
Where a course fits
The method above is free and needs nothing but a pen. The friction is that nobody reminds you to do step 2, and step 4 happens only if you remember to remember.
Study Junkie turns the book you upload (PDF, DOCX or a text file) into a course of chapters built from the full text, never a summary, so the chapter-sized sitting is already cut for you. Each chapter has assignments: you write an answer, and the feedback is on what you wrote. That is the “book closed, your words” move with something checking it. Once you switch on scheduled recall, marking a chapter as read starts its recall sessions, on days 1, 3 and 7 by default, and you can change the days. The reminder arrives with the questions already written from that chapter. It does not replace the book. It gives each chapter the second and third encounter a single pass does not.
How the chapters, assignments and recall fit together takes a minute to read. Verifying your email gives you 1,500 free Study Credits, with no card. After that it is one-time packs from $10 that do not expire.
If you would rather do it by hand, you already have everything you need. Read one chapter. Tomorrow, before you read the next, tell yourself what the last one said.
Frequently asked questions
How many chapters a day should I read to remember a book? No study gives a number, and the pages that offer one are guessing. The unit that matters is the sitting you can recall afterwards. For dense nonfiction that is usually one chapter, sometimes half of one. For a light book it can be three. If you cannot say what the last sitting argued without looking, the sitting was too long.
Does reading a book in one sitting hurt your memory of it? Not tomorrow. In the closest experiment, the one-day group remembered the most after 24 hours. It fades faster afterwards. A single pass, binged or spread out, is also only one encounter. What protects it is meeting the material again later, preferably by recalling it rather than rereading it.
Is rereading a chapter straight away worth it? It helps on a test that same day and not much after. In Rawson and Kintsch’s 2005 experiments, reading twice in a row beat reading once on an immediate test, and the difference had gone two days later. If you have time for a second look, spend it a few days later, and try to recall the chapter before you reread it.
Does any of this apply to fiction? The TV study was a drama, so the fade applies to stories too. Whether that matters depends on why you are reading. If you want to keep the plot, the characters’ choices or a book’s argument about people, the same recall-before-you-continue habit works. If you want the experience, read it however you like.