It is 11.40pm, the lights are off, and your brain has just remembered the email you never answered. Not the forty you dealt with today. That one.
It feels like proof of something everybody seems to know: the mind clings to unfinished business. Psychology even has a name for it, the Zeigarnik effect, and it turns up in productivity blogs, screenwriting guides and marketing decks as settled science.
Here is the twist. When two researchers at the University of Bern pooled almost a century of experiments in 2025, the famous memory advantage for unfinished tasks more or less vanished. What survived was something different, and arguably more useful to understand: not a better memory for loose ends, but a stubborn urge to go back and tie them.
That distinction changes what actually helps at 11.40pm.
A Berlin café, a young researcher and a famous result
The origin story is usually told with a waiter. In Berlin in the 1920s, so the often-repeated tale goes, the psychologist Kurt Lewin and his students noticed that a waiter could recall unpaid orders in detail, then forget them once the bill was settled. It is a lovely anecdote and hard to pin to any first-hand source, so treat it as legend.
The experiments are real. Bluma Zeigarnik, one of Lewin’s students, published her study in 1927. She gave people a series of small tasks, such as puzzles, and interrupted some of them before they were finished. Asked afterwards what they had done, her participants recalled more of the interrupted tasks than the completed ones.
A year later, her colleague Maria Ovsiankina published a companion finding that gets far less attention. When people were interrupted and then given a free moment, they tended to drift back and finish what they had started, even when nobody asked them to.
Two effects, then. One about memory. One about behaviour. For most of the twentieth century, the first got the fame.
What 59 publications really show
In a meta-analysis published in July 2025 in Humanities and Social Sciences Communications, Romain Ghibellini and Beat Meier gathered the research on both effects. Their search turned up 38 publications on the Zeigarnik effect and 20 on the Ovsiankina effect, plus one that tested both.
For memory, the headline number was a ratio of 0.99. In plain terms, people recalled interrupted and completed tasks about equally well. Across the studies that reported it, roughly 49% of the tasks people remembered were interrupted ones, which is what you would expect if interruption made no difference at all.
Resumption told another story. Across the studies, people went back to interrupted tasks about 67% of the time when they had the chance, well above the 50% you would expect by chance. The authors concluded that the Ovsiankina effect is a general tendency, while the Zeigarnik effect “lacks universal validity.”
Two caveats keep this honest. The resumption figure is indicative rather than precise, because older studies measured “going back” in very different ways. And the authors suggest the memory effect may appear under particular conditions, such as how involved people feel in a task or how much authority the experimenter projects, which may explain why it showed up more in early research.
So the nag you feel at night is less like a sticky note glued to your memory and more like a hand on your sleeve.
Zeigarnik, Ovsiankina and the cliffhanger, side by side
The three ideas get blurred together constantly, so it helps to see them laid out.
| Idea | What it claims | How well it holds up |
|---|---|---|
| Zeigarnik effect (1927) | Interrupted tasks are remembered better than finished ones | Weak: a pooled recall ratio of 0.99 in the 2025 meta-analysis |
| Ovsiankina effect (1928) | People feel pulled to resume interrupted tasks | Robust: about 67% resumption across studies |
| Cliffhanger or “open loop” | An unresolved ending makes you want the next instalment | Supported in recent experiments, mostly via desire to continue |
Why a plan quiets the noise before you have done anything
If unfinished goals tug at us, the obvious fix is to finish them. The more interesting finding is that you often do not have to.
In 2011, psychologists E.J. Masicampo and Roy Baumeister ran a series of experiments for the Journal of Personality and Social Psychology. They had students bring an important unfinished goal to mind and then do something unrelated, such as reading a passage or solving anagrams.
Students left with a live, unfinished goal reported more intrusive thoughts while reading, and goal-related words were more mentally accessible for them. They also did worse on the anagrams. But when students made a specific plan for the goal, spelling out when, where and how they would deal with it, those effects disappeared. They had made no progress on the goal itself. They had only decided what to do about it.
The researchers’ reading is that the mind is not demanding completion so much as reassurance that the task is handled. A concrete plan seems to count.
This is the mechanism behind if-then planning, and it explains why a scribbled note can settle a racing head.
Your brain is not asking you to finish everything tonight. It is asking whether anyone is in charge.
The open loops in your wardrobe and your craft basket
You can see the pull of unfinished business in everyday corners of life, not just in inboxes. Knitters joke about their piles of half-done projects for a reason: an abandoned jumper with one sleeve has a way of catching your eye every time you pass the basket. If that rings true, our piece on the revival of handmade and heritage craft is a reminder of why those projects are worth returning to.
Small daily decisions can work as open loops too. Deciding what to wear while half asleep is one more unresolved choice, which is one reason people who build a capsule wardrobe often say mornings feel lighter. That is a practical observation rather than a lab finding, but it fits the Masicampo and Baumeister logic: decide once, and stop re-deciding.
Cliffhangers: what streaming knows and what the science says
Television writers did not need a meta-analysis to learn that an unresolved ending brings viewers back. The question researchers have started asking is how cliffhangers work, and what they cost.
In a study published in Media Psychology in 2024, Kaitlin Schibler, Lindsay Hahn and Melanie Green at the University at Buffalo ran two experiments with 202 and 273 participants. People read a short story that ended in a cliffhanger, a win for the hero or a win for the villain.
The cliffhanger readers wanted a further instalment more than those who saw the hero win, in both studies. Surprisingly, the cliffhanger did not make the story feel more suspenseful or more enjoyable. It simply left people wanting more. That looks a lot like Ovsiankina’s resumption urge, wearing a costume.
A 2025 study in Acta Psychologica added a twist from the other direction. Researchers led by J. Baldwin found that series and books consumed in long consecutive stretches were more memorable, and more likely to be replayed in people’s imaginations afterwards. It is correlational, so it cannot say binge-watching causes this, but it suggests that the stories we keep thinking about are often the ones we immersed ourselves in, not only the ones left hanging.
What a cliffhanger does to your sleep
The bedtime episode is where this stops being academic. At the University of Fribourg in Switzerland, Sandrine Baselgia, Björn Rasch, Andreas Fahr and colleagues brought 50 healthy young sleepers into the lab for two nights, reported in Sleep Medicine in 2023.
On one night they binge-watched three to four episodes of a suspenseful series before bed. On the other they watched a calm documentary. Some saw the last episode end on its cliffhanger; for others it was edited to stop at a calm moment.
The overall effects on sleep were small, and people actually fell asleep faster after the suspenseful series, which the team had not expected. But only the cliffhanger endings raised heart rate and cortisol, and after them participants spent less time in deep sleep during the first part of the night.
The researchers’ advice was practical: stop before the cliffhanger, not after it. Watch the episode, then skip the final two minutes and catch them tomorrow.
If late-night viewing is a habit that keeps stretching, our piece on revenge bedtime procrastination looks at why that last episode is so hard to refuse.
The bedtime to-do list that worked in a sleep lab
Back to that 11.40pm email. In 2018, Michael Scullin and colleagues at Baylor University’s Sleep Neuroscience and Cognition Laboratory tested a five-minute fix, published in the Journal of Experimental Psychology: General.
They monitored 57 healthy adults aged 18 to 30 overnight with polysomnography, the standard way of measuring sleep through brain activity. Before lights out, half spent five minutes writing a to-do list for the coming days. The other half wrote about tasks they had already completed.
The to-do list writers fell asleep faster, about nine minutes faster on average, according to Baylor. And the more specific their lists, the quicker they dropped off. For the group writing about finished tasks, more detail went with slower sleep.
The limits matter. This was one night, one small sample of healthy students, and a lab with phones and homework banned. Scullin himself said the team did not know whether the results would apply to people with insomnia. Still, it lines up neatly with the plan-making research: offloading the tasks onto paper seems to reassure the part of the mind keeping watch.
Writing down what is done can feel virtuous. Writing down what is next seems to be what lets you rest.
A ten-minute shutdown ritual to close the day
Put the research together and a simple end-of-day routine falls out of it. None of these steps has been tested as a package, so think of it as an evidence-informed habit rather than a proven treatment.
Try this tonight
- Empty your head onto paper. List every open task that is nagging you, big or small. Five minutes is enough.
- Give the top three a when and where. “Reply to Sam, 9.15am, before opening Slack.” Specific plans are what quieted intrusive thoughts in the lab.
- Park the rest. Anything without a slot goes on a “later” list you trust yourself to check.
- Leave a deliberate loose end. Stop a piece of work mid-sentence or mid-step so the pull to resume works for you tomorrow.
- End the screen at a calm moment. Pause a series before the cliffhanger, not after it.
That fourth step uses the Ovsiankina effect on purpose. If you struggle to start things, beginning a task for five minutes and walking away can be enough to create a pull back to it, which is a gentler cousin of the advice in our guide to why we procrastinate.
Loose ends are not a flaw
It is tempting to treat the nagging as a malfunction, a mind that will not switch off. The research suggests something kinder. The tug of an unfinished task is a reminder system, crude but well-meant, and it settles down when it hears a credible answer.
You do not have to finish everything to be allowed to rest. You only have to decide where each loose end lives.
Then put the pen down, leave the last two minutes of the episode for tomorrow, and let the list keep watch.
Frequently asked questions
The memory version is shaky. A 2025 meta-analysis found people remembered interrupted and completed tasks about equally well, though the effect may appear in some specific situations.
The Zeigarnik effect is about remembering unfinished tasks better. The Ovsiankina effect is about the urge to go back and finish them, and it is the better supported of the two.
An open goal tends to keep tugging at attention until it feels handled. Making a specific plan, or writing a short to-do list before bed, has reduced that pull in experiments.
Their overall effect looks small, but a 2023 Swiss lab study found cliffhanger endings raised physiological arousal and reduced deep sleep early in the night. Stopping before the final scene is a simple workaround.
Sources
- Ghibellini, R., & Meier, B. (2025). Interruption, recall and resumption: A meta-analysis of the Zeigarnik and Ovsiankina effects. Humanities and Social Sciences Communications, 12, 962. DOI
- Masicampo, E. J., & Baumeister, R. F. (2011). Consider it done! Plan making can eliminate the cognitive effects of unfulfilled goals. Journal of Personality and Social Psychology, 101(4), 667–683. DOI
- Scullin, M. K., Krueger, M. L., Ballard, H. K., Pruett, N., & Bliwise, D. L. (2018). The effects of bedtime writing on difficulty falling asleep: A polysomnographic study comparing to-do lists and completed activity lists. Journal of Experimental Psychology: General, 147(1), 139–146. Summary via Baylor University. Link
- Baselgia, S., Combertaldi, S. L., Fahr, A., Wirz, D. S., Ort, A., & Rasch, B. (2023). Pre-sleep arousal induced by suspenseful series and cliffhangers have only minor effects on sleep: A sleep laboratory study. Sleep Medicine, 102. DOI
- Schibler, K., Hahn, L., & Green, M. C. (2024). Investigating responses to narrative cliffhangers using affective disposition theory. Media Psychology, 27(1), 1–25. DOI
- Baldwin, J., Ulusoy, E., Durfee, M., Busselle, R., & Ewoldsen, D. R. (2025). Watching one more episode and reading one more chapter: What entertainment contexts lead to retrospective imaginative involvement? Acta Psychologica, 257, 105101. DOI
Updated 18 October 2026: we expanded this article with newer research, fresh figures and new sections. Cover photo: Bennett Frazier on Unsplash. Charts and graphics: Something Navy.
How we reported this. We went back to the original research for this story, 6 sources in all, listed above, and checked every figure against them. If you think we have got something wrong, tell us and we will look at it. Read more about how we work.
