Why You Don't Want Help With the Edge Pieces (But Love It for the Last Fifty)

I do not want anyone near a puzzle when it starts. The pile is a mess, I have a way of attacking it, and a second pair of hands feels like noise. But when the picture is nearly there and someone in the family wanders over to put in a few pieces, I love it. I wondered whether that was competitiveness, or a quieter fear of looking slow, or simply introversion turning up at the puzzle table. The research points somewhere more interesting.

The confession

For me a jigsaw is meditation. It is the one part of the day that runs entirely my way: my sort, my order, my pace, nobody asking me anything. I would not have described it as mindfulness until I noticed how much I guard it.

And I do guard it. I do not like doing a puzzle with someone else from the beginning, while the pieces are still a heap on the table and everything is chaos. A second person at that stage means a second plan running over the top of mine. And when they start finding pieces while I am still struggling to place one, I feel slow, and if I am honest, a little insecure.

Then, near the end, the feeling reverses. When my immediate family come over and help with a few pieces, I love it. It is the same table, the same puzzle and the same people. The only thing that has changed is where we are in the build.

I had two theories about myself, and I did not much like either of them. Either I am more competitive than I want to admit, or I do not enjoy feeling slow in front of people. A third, more flattering one followed: maybe I am just an introvert and this is what it looks like on a puzzle board.

I spent a long career studying how people make decisions, and the first lesson of that work is that the explanation you reach for first is usually the one that flatters you. So I went looking.

Every puzzle is two tasks

In 1965 the psychologist Robert Zajonc published a short paper in Science that tidied up decades of contradictory results. Some studies found people performed better when others were around. Others found they performed worse. Zajonc's resolution was that the presence of other people raises arousal, and arousal pushes you towards whatever response comes most easily to you. On a simple or well-practised task, the easy response is usually right, so company helps. On a complex or unfamiliar task, the easy response is usually wrong, so company hurts.

That idea has been tested a great many times since. In 1983 Charles Bond and Linda Titus pooled 241 studies in Psychological Bulletin. The effects are real but small: the presence of others tended to speed up simple work a little, and to make complex work both slower and less accurate.

Now look at a jigsaw with that in mind. It is not one task. It is two.

The opening is the textbook complex task. You are deciding how to sort, which features to chase first, what to hold back. Every one of those choices is half-formed and depends on the others. Someone else at the table is not just two more hands. They are a second plan running over the top of yours.

The endgame is the textbook simple one. There are twenty gaps and twenty pieces. Anyone can walk up and contribute without disrupting a thing. The puzzle has changed character under you, and your reaction to company changes with it.

Feeling watched, feeling slow, and why family is different

That explains the irritation. It does not quite explain the other feeling, the faint embarrassment of watching someone else find pieces while I struggle.

A few years after Zajonc, Nickolas Cottrell and colleagues ran an experiment that pulled his idea apart. They had students do a word task alone, in front of two attentive spectators, or next to two people who sat just as close but wore blindfolds, posing as volunteers waiting for a colour-perception experiment. The watching audience changed how people performed. The blindfolded people did not. Cottrell went on to argue that what matters is not the bare presence of others but the sense of being evaluated by them. That account has been debated ever since and is not settled, but it rings true at a puzzle table.

The second piece of the explanation is about who is doing the outpacing, and here my own experience does not fit the textbook neatly.

Abraham Tesser's self-evaluation maintenance model, set out in 1988, starts from a simple premise: we want to feel good about ourselves, and the people around us shape that. When someone does well, one of two things happens.

• If the area does not matter much to how you see yourself, you bask in it. Their success reflects well on you.
• If it is central to how you see yourself, you compare. Their success becomes a measure of your shortfall.

Tesser expected both effects to grow stronger the closer the person is to you. That is where I part company with him. When my immediate family succeed, in any shape or form, I am only too happy. I never feel judged by them, and if they are cleverer than me, good. The sting comes from the wider circle: young cousins, nieces and nephews, friends outside my immediate circle.

A second line of research helps explain why. In 1991 Arthur Aron and colleagues argued that in our closest relationships the other person becomes, in a real sense, part of the self. Their successes register partly as our own. If that is right, there is a point of closeness past which comparison stops working, because there is no longer a separate "them" to be measured against. My immediate family are past that point. A nephew or a friend is close enough that I care what they think, but not so close that their win feels like mine.

I run a puzzle shop, so puzzling is part of how I see myself. Being outpaced at it, in front of people whose opinion I notice, is about as relevant a comparison as it gets. That is the "feeling dumb" part, and it is not a character flaw. It is a comparison made in front of an audience, in the one area I care about most.

The endgame is different on both counts. There is no performance left to compare: the hard thinking is done and the last pieces are close to self-placing. And nobody walking past to drop in a piece is assessing anyone. The evaluation has drained out of the situation, and what is left is just company.

What I think is really going on: control

Social facilitation and self-evaluation explain a lot, but I do not think they are the deepest layer. That belongs to self-determination theory.

Richard Ryan and Edward Deci's framework, summarised in American Psychologist in 2000, holds that people are most motivated and most satisfied when three needs are met: autonomy (doing it your own way), competence (feeling capable) and relatedness (feeling connected to others).

Map that onto the puzzle.

• Puzzling alone at the start gives me full autonomy and an uninterrupted sense of competence. It gives no relatedness, but I am not looking for any at that stage.
• Someone joining at the start costs autonomy, because it is no longer my method, and risks competence, because now there is a comparison.
• Family joining at the end adds relatedness and costs nothing. My method has already won. The competence is banked. The company is pure gain.

That, to me, is the most convincing single explanation of the pattern, and it is what I mean when I say puzzling is mine and under my control.

It also fits two other findings. Mihaly Csikszentmihalyi's work on flow, the absorbed state where time disappears, lists a sense of control over what you are doing among its core features, and flow is easily broken by interruption. And Michael Norton, Daniel Mochon and Dan Ariely's IKEA effect paper (2012) found people value things they have built themselves more highly, and that the effect disappeared when people did not finish what they started. A puzzle you have mostly built yourself is worth more to you. A few family pieces at the end do not dilute that. Half the building done by someone else might.

So is it introversion?

Partly, at most. Here is why I do not think it is the main story.

A personality trait does not switch off halfway through a puzzle. If my preference were simply introversion, I would want to be left alone at the end as much as at the start. I do not. The preference changes with the stage of the build, which points to something about the task, not something fixed about me.

My sport tells the same story. I am very much a team player and I have played competitively in a lot of teams, but if I am being honest, I prefer singles tennis to doubles. I like solo performance. That is not a dislike of people, or I would never have stayed in all those teams. It is the same preference I bring to the puzzle table: when the performance is hard and it matters to me, I want it to be mine.

There is also a recent experiment that tested the introvert intuition almost directly. Dwight Tse, Ayodele Joseph and Kate Sweeny had people and their romantic partners play Perfection, the game where you race to fit shaped pieces into matching holes, under three conditions: alone in separate rooms, side by side without interacting, and together. They expected extraverts to struggle to reach flow on their own. On average, they did not. Extraversion helped people reach flow in the shared conditions, but it did not make solitary flow harder. What did matter was a separate trait, sometimes called an autotelic personality, the tendency to enjoy an activity for its own sake.

So the board does reveal something about you. I just do not think it is where you sit on the introvert to extravert scale. It shows how much control you want over the things you care about, and how much you mind being measured on them.

The people who love it from piece one

None of this means puzzling together is a lesser way to do it. For a lot of people it is the better way.

In 2010 Charles Walker published a set of studies in The Journal of Positive Psychology asking whether flow is better alone or together. People described and rated their own experiences of each, and shared flow, such as singing in a choir, came out as more joyful than solitary flow. In a follow-up experiment, students played a bat-and-ball game either with a partner or alone against a wall, adjusted to be equally hard. The paired version was rated more enjoyable. And games that needed more cooperation produced more joy still.

The puzzle world has its own evidence of this. The World Jigsaw Puzzle Championship, held since 2019, does not only crown individuals. It runs a pairs event, where two people complete a single puzzle against the clock, and a team event for groups of four. At the top level, puzzling together is a sport.

I have watched more than a few social media reels of puzzle teams racing each other, and they settled the question for me. It is very clear that the way I feel is not how everyone feels. Those teams are not putting up with each other, they are working as one, and there is clearly a place for teamwork in puzzles.

So if you are the person who wants everyone at the table from the moment the box is opened, nothing here says you are doing it wrong. You may simply get more from the relatedness than you lose in autonomy.

What kind of puzzler are you?

Three types, and I suspect most people will recognise themselves in one. You may also be a different type with different company, which is what the evaluation research would predict: the more someone's opinion of your puzzling matters to you, the less you may want them at the table early.

How to share a puzzle without the friction

If your household mixes types, and most do, the aim is not to convert anyone. It is to give each person the thing they need at the stage they need it.

1. Split by territory, not by turns. Divide the picture by colour zone or by section: one person takes the buildings, the other the water. Each of you keeps full control of your own patch, so nobody is running a second plan over your first.
2. Do the sort and the edge solo, then open the table. The Welcoming Finisher's rule. The complex half belongs to one person, the simple half to everyone. Say out loud when the table is open, so nobody has to guess.
3. Run two puzzles side by side. Two people, two puzzles, one table. You keep the company and the conversation and lose the competition for pieces entirely. It is also, roughly, the side-by-side condition in the Tse study.
4. Make the last stretch a ritual. The final fifty pieces are the easiest and the most satisfying, which makes them the natural moment to bring the household in. Leave them for an evening when everyone is home.
5. Let the finisher be agreed in advance. If the last piece matters to someone, settle it before the end rather than at it.
6. Give the Solo Puzzler a lid. A board or a case means a puzzle in progress can be moved somewhere private and closed between sessions, rather than sitting in the middle of the dining table inviting help.

This sits alongside something I wrote last month, in The Week I Could Not Stop Puzzling, where I suggested puzzling with someone as a natural brake on an absorbing habit. Both are true. Company at the end of a build is a pleasure. Company at the start is the thing I protect against.

What the research does not say

I want to be plain about this. No study I could find has looked at people doing jigsaw puzzles together. Everything above is general research applied to the puzzle table: word tasks, bat-and-ball games, a shape-fitting board game, and broad theories of motivation. The fit is good, and I think the argument holds, but it is an argument and not a finding.

What we do know about jigsaws specifically is that they work the brain hard. A 2018 study led by Patrick Fissler in Frontiers in Aging Neuroscience found jigsaw puzzling draws on several distinct cognitive abilities at once, including visual perception, mental rotation and working memory. That is consistent with the opening of a puzzle being the demanding, complex task described here. It does not tell us anything about company.

So if your experience at the table is different from mine, that is data, and I would like to hear it.

Common questions