Do People Do the Same Jigsaw Puzzle Twice?

In October 2026 I rebuilt a puzzle I had last done six years earlier. It was amazing how much of the detail came back to me. Not only the picture: the joys of the first build, and the parts that were not much fun, almost replayed themselves. So I went looking for what the research says about doing a thing again.

Who does a puzzle twice

Puzzlers fall into two camps on this, and each tends to be slightly baffled by the other.

The first camp builds a puzzle once and moves it on. Stacey, who writes the blog My Jigsaw Journal, says that "almost every cardboard puzzle that I've assembled has been packed up and donated". Several of her readers do the same, and one buys second hand, does each puzzle once and gives it away again.

The second camp keeps them. In the same comment thread one reader keeps almost every puzzle they have assembled, and another wrote: "I don't really have a lot so I redo my puzzles often." A third sits in the middle and names the thing that usually settles it, which is storage: "I do enjoy redoing puzzles, but I keep fewer and fewer."

Then there are the speed puzzlers, who repeat a puzzle deliberately as training. One competitive puzzler, writing on the Puzzle Talk newsletter in 2025, admitted to starting out as a sceptic: "at first I didn't see a point in doing the same puzzle multiple times. Now I've jumped ship."

Why a second go is better than you expect

The strongest research on this is not about puzzles. It is about repeat experiences in general, and it points one way.

In 2019 Ed O'Brien, a behavioural scientist at the University of Chicago, published seven studies involving more than 3,300 people. Participants did something once, a museum exhibit, a film or a video game, and then predicted how much they would enjoy doing exactly the same thing again. Others actually did it again. The predictions were consistently too gloomy. People enjoyed the repeat more than anyone forecast.

The reason is the useful part. O'Brien found the gap was driven by novelty people did not expect to find. A first pass leaves you feeling you have seen all there is, when in fact you took in only some of it. The effect depended on complexity: the more there was in the thing, the more there was left to notice. And in later studies people paid real costs to avoid a repeat that they would have enjoyed as much as the new option.

It is worth being exact about what this does and does not say. It does not say a repeat matches the first time. It says we underrate it, and so we skip repeats we would have liked.

Two consumer researchers, Cristel Russell and Sidney Levy, came at the same question by interviewing people who chose to reread books, rewatch films and return to places. They described two things people get from going back. One they called emotional efficiency: you know what the experience will deliver, so you get the feeling you came for reliably. The other is a kind of stocktake. The book has not changed and you have, so the repeat shows you the difference. The literary scholar Patricia Meyer Spacks spent a year rereading novels from across her life for the same reason, and wrote a book about what she found.

Why so much of it comes back

This is the part that surprised me most at the table, and it turns out to be some of the most solid ground in memory research.

Memory for pictures is enormous. In a 2008 study at MIT, people looked at 2,500 pictures of objects over five and a half hours. Afterwards they were shown pairs and asked which one they had seen. They were right 92% of the time when the other picture was a different kind of object, and still right 87% of the time when the other picture was the same object in a slightly different state or pose. They had kept the detail, not just the gist.

It lasts far longer than you would think, and it does not need you to remember. In 2006 the psychologist David Mitchell tracked down people who had looked at simple pictures in a laboratory for one to three seconds each, 17 years earlier. He posted them fragments of pictures to identify. They identified fragments of the pictures they had seen more often than fragments of new ones, and that held even for people who had no memory of ever taking part. A fragment of a picture you once looked at is, more or less, a puzzle piece.

The key is the difference between recall and recognition. Before I opened the box I could not have described that picture in any detail. That is recall, and after six years it had mostly gone. But a piece in the hand is a cue, and recognition is a different and much more durable thing. This is why the feeling is so strange. You are not remembering the puzzle and then building it. Each piece is handing your own memory back to you as you go.

Skill at the task is stored differently again. In 1976 the psychologists D. N. Brooks and Alan Baddeley tested patients with amnesia on a range of tasks. On word-based memory tests they were severely impaired. On hands-on learning, which included assembling a jigsaw puzzle, they were far closer to normal. The knowing-how of a puzzle sits somewhere that the knowing-that does not reach.

Why the feelings come back too

It was not only the image. The mood of the first build came back, in the same places.

There are two reasons for that, and the first is plain. The hard parts are hard because of the picture, and the picture has not changed. A stretch of sky or foliage that wore me down six years ago is still several hundred near-identical pieces. It was always going to be a slog the second time. The frustration did not come back as a memory so much as get made again, from the same ingredients. The same is true in reverse: the section that fell together fast the first time is built to fall together fast.

The second reason is how memory is retrieved. Endel Tulving and Donald Thomson showed in 1973 that a memory comes back most readily when the situation at recall matches the situation in which it was laid down. A rebuild is about as close a match as ordinary life offers. Same pieces, same picture, same posture over the table. So what returns is not just the image but the episode around it.

A book, a film, or something else?

What follows is our view, built on the evidence above, not a research finding.

Of the two, a film is the closer comparison. A puzzle is visual, it is taken in over hours and not weeks, and people return to both for the same thing Russell and Levy described: a known feeling, reliably delivered. But a jigsaw differs from both a book and a film in three ways, and all three favour doing it again.

On the first point there is even some evidence from fiction. In 2011 Jonathan Leavitt and Nicholas Christenfeld gave several hundred students short stories, some with the ending given away in advance. Readers generally rated the spoiled versions as more enjoyable, including the mysteries. Knowing where a thing ends does not ruin the getting there. A jigsaw has been built on that principle all along.

Is it easier the second time?

A little, and probably less than you expect after a long gap. We should be straight that nobody appears to have timed ordinary puzzlers rebuilding the same jigsaw years apart, so this section is reasoning from related evidence.

The related evidence is old and sturdy. In 1885 Hermann Ebbinghaus showed that material he could no longer recall at all was still quicker to learn the second time. He called the difference savings. A 2015 replication of his work found the same pattern. Something of a first attempt survives below the level of what you can consciously bring to mind.

Speed puzzlers see the practical version of this. The competitor quoted earlier found that repeating a puzzle taught them the maker's piece shapes, and that they "learned to identify the location of quite a few pieces", mostly near the frame and in strongly coloured sections. The middle stayed hard. Their targets tell the story: under 40 minutes for a first attempt at a puzzle, under 30 for a repeat. That is a real gain, and it comes from repeats done close together.

After six years it is a different matter. You get the recognition described above, and that helps you place a distinctive piece. It does not sort a thousand pieces for you, and it does nothing for the sky.

As for whether a rebuild still does your brain any good, nobody has tested a repeat against a new puzzle. The main study of puzzling and cognition, led by Patrick Fissler in 2018, found that jigsaw puzzling draws on several cognitive abilities at once and linked long-term puzzling experience to visuospatial skill. A rebuild still means sorting, rotating and fitting every piece. We see no reason it would stop counting, and we cannot prove that it does.

Which puzzles are worth a second build

This is our advice, drawn from the research above and from the table.

1. Keep the detailed ones. O'Brien's finding was that the bonus of a repeat depends on complexity. Busy illustrated scenes, collages, fine art and large piece counts hold more than one build can use up. A simple image with big areas of flat colour has less left to give.
2. Keep the ones you enjoyed, not the ones you survived. The hard parts will be hard again for the same reasons. If a puzzle was a slog the first time, a rebuild is the same slog with better lighting.
3. Leave it long enough to forget. There is no researched number. Our rule of thumb is to wait until you cannot picture the image in any detail, which for most people is a year or more. Research on tiring of things shows that enjoyment recovers with time, and faster when you call to mind the variety you have had in between. Other puzzles are what make an old one fresh.
4. Write the date inside the lid. And how long it took, and who you built it with. It costs ten seconds and turns a rebuild into a comparison with your earlier self.
5. Decide before you glue. Glue is the one choice that cannot be undone. If there is any chance you want to build it again, photograph it and box it instead.

We would like to measure this properly. Since nobody has counted, we would like to. If you redo puzzles, email info@premiumpuzzles.com.au and tell us which ones, how long you leave them, and whether the second build was what you expected. We will update this page when there are enough answers to be worth printing.

Common questions