# Cognitive shuffling, explained

> Cognitive shuffling is a way of falling asleep by thinking deliberately disconnected thoughts: one small, neutral image after another — a ladle, a harbour, a bolt of felt — with nothing joining them. It was developed by the cognitive scientist Luc Beaudoin at Simon Fraser University, who calls the structured version *serial diverse imagining*, and the idea behind it is that the brain lets sleep in once your thinking stops looking like work.

Source: https://dreamsheeps.space/cognitive-shuffling
Published: 2026-09-28 · Updated: 2026-10-14

## What is cognitive shuffling?

Cognitive shuffling — also called the cognitive shuffle — is a pre-sleep mental exercise. Instead of trying to empty your mind, you fill it with a slow run of simple, unrelated images, each held for a few seconds and then let go. A pear. A lighthouse. A wool glove. A spoon. None of them leads anywhere, and that is the point.

Luc Beaudoin, a cognitive scientist at Simon Fraser University in Canada, set out the idea in a 2014 paper, and the same year built it into an app, mySleepButton, through his company CogSci Apps. He describes the kind of thinking the exercise produces as *super-somnolent*: thought with the properties that thinking tends to have just before sleep anyway.

- *Diverse* — successive images do not connect, so the mind cannot bind them into anything.
- *Concrete* — things you could see or touch, not ideas or arguments.
- *Emotionally neutral* — no personal significance, pleasant or otherwise.
- *Absorbing, not gripping* — enough to hold attention, never enough to make you wonder what comes next. Curiosity is a small form of alertness.
- *Goalless* — nothing to solve, finish, remember or get right.

The structured, research version is called serial diverse imagining, or SDI. Where the do-it-yourself method has you generate the words, SDI hands them to you: one relatively concrete word at a time, with about eight seconds between them, while you picture each one until the next arrives.

## How do you do cognitive shuffling?

There are two ways. The first needs nothing but a word; the second needs something — an app, a recording, a list — to supply the words for you.

### The seed-word method

1. Get into bed, ready to sleep, lights out.
2. Think of a random, emotionally neutral word of at least five letters, with few repeated letters. *Bedtime* works; *banana* has only three different letters.
3. Take its first letter. Think of a word that starts with it — *b*: bread — and picture the thing for a few seconds. Then another *b* word: a bicycle. A bee. A bucket.
4. Keep going with the same letter until you get bored of it or run out, then move to the next letter: *e* — an egg, an elm, an envelope.
5. If a word brings up something you would rather not think about, or you cannot easily picture it, drop it and pick another.
6. If you reach the end of the word and are still awake, choose a new one and start again.

Those steps follow the do-it-yourself instructions published by mySleepButton, Beaudoin’s app. There is nothing to score and no way to do it wrong: losing track of which letter you were on is not a failure of the exercise, it is the exercise working.

### The serial diverse imagining method

The structured form removes the effort of finding words. Something plays one concrete word, pauses for roughly eight seconds, and plays another word unrelated to the first; you picture each one until the next arrives. Eight seconds is about how long it takes to build an image unprompted and let it fade. It is the version the research tested, and it suits anyone who finds generating their own words turns into a task. There is more on the timing in [the eight-second gap](https://dreamsheeps.space/learn/eight-second-gap), and ready-made sets in [the word lists](https://dreamsheeps.space/cognitive-shuffling/words).

## Why would thinking nonsense help you sleep?

Two explanations are on offer, and they reach the same practical advice from different directions.

### The shape of the thought, not its content

Beaudoin’s account is that the brain runs something like a gate on the way into sleep, which he calls the sleep onset control system. The gate does not check whether your thoughts are pleasant. It checks their *shape*. Orderly thinking — one thing following from another, a problem being turned over, tomorrow being planned, a conversation being rehearsed — is evidence that something still needs handling, and the gate stays shut. Thinking that wanders from a ladle to a harbour to a bolt of felt with nothing joining them is evidence that nothing does.

> The gate is not reading your thoughts for how nice they are. It is reading them for whether anything still needs doing.

That is also what thought looks like on its own as people drift off. Reports collected at sleep onset are fleeting, fragmentary, mostly visual and without a storyline, and a 2016 linguistic analysis of them found that future-directed language falls away as sleep approaches (Speth, Schloerscheidt and Speth, 2016). Cognitive shuffling imitates the state you are trying to reach, which is a more reliable route into it than waiting for it to arrive. [Why thoughts break up as you fall asleep](https://dreamsheeps.space/learn/shorter-sentences) goes further into that research, and [hypnagogia](https://dreamsheeps.space/hypnagogia) into the state itself.

### Occupying the space a worry would use

The second explanation is simpler. A mind can only hold so much at once, and a task that fills it leaves no room to go back to the worries. In a 2002 study of people with insomnia, Harvey and Payne attributed the benefit of an engaging imagery task to exactly that: it occupied enough “cognitive space” to keep people from re-engaging with their thoughts. A later treatment, cognitive refocusing, was built to change the style and content of pre-sleep thoughts in the same way, and in a small 2013 trial with university students it reduced insomnia severity more than sleep-hygiene advice alone (Gellis, Arigo and Elliott, 2013).

The two accounts agree more than they first seem to. Beaudoin’s task is meant to engage: the 2016 poster of its trial calls it a more active, engaging technique than the monotonous imagery training that came before it, and the occupancy account wants the images absorbing enough to hold you. Where they meet is the practical advice — give the mind a steady run of concrete images, and withhold anything that makes it want to know what comes next.

### Why it is not counting sheep

Counting is monotonous, which looks helpful, but it is also orderly, sequential and too easy: every number has a next one, and it takes so little attention that the worry fits in around it. Cognitive shuffling is the opposite on each count. [Does counting sheep work?](https://dreamsheeps.space/counting-sheep) covers the difference, and the popular claim about sheep that the research does not support.

### Why trying harder backfires

The clinical account of how poor sleep entrenches itself — attention, intention, effort — says that sleep is inhibited by deliberately trying to produce it (Espie and colleagues, 2006). Part of what shuffling does is hand the mind a small, harmless job, which displaces the job of trying to sleep. That is also why it should never become a performance: see [why trying to sleep keeps you awake](https://dreamsheeps.space/learn/trying-to-sleep).

## Does cognitive shuffling work? What the research shows

The honest answer is that it is promising, low-risk and less proven than most articles about it suggest. Two studies are usually cited, and they show different things.

- *The serial diverse imagining trial*, presented as a poster at SLEEP 2016 by Beaudoin and colleagues at MacEwan University: 154 university students who complained of a racing mind at bedtime were randomly assigned to serial diverse imagining, structured problem-solving (a standard technique, done earlier in the evening), or both. Pre-sleep arousal, sleep effort and sleep quality all improved compared with where they started, with large effects, and serial diverse imagining did as well as problem-solving — in a trial its poster says was too small to tell the two apart — while the students who tried both rated it the more helpful and the easier to use. There was no untreated comparison group, and neither the abstract nor the poster reports a figure for how long people took to fall asleep.
- *Harvey and Payne (2002)*: 41 people with insomnia spent one night following one of three sets of instructions — distract yourself with imagery, distract yourself in general, or no instructions. The imagery group fell asleep sooner and had fewer, less distressing pre-sleep thoughts than the group given no instructions. This is the study behind the “fall asleep about 20 minutes sooner” figure — and it tested imagery distraction, not cognitive shuffling.

So the popular claim that cognitive shuffling makes you fall asleep a set number of minutes faster joins two different studies into one. What the evidence does support is narrower and still useful: occupying a busy mind with simple imagery at bedtime seems to reduce the arousal and effort that keep people awake. [What cognitive shuffling research actually shows](https://dreamsheeps.space/cognitive-shuffling/evidence) goes through both studies line by line.

## Who doesn’t cognitive shuffling suit?

### People who can’t picture things

Somewhere around four people in a hundred have aphantasia — little or no voluntary visual imagery (Dance, Ipser and Simner, 2022). Told to picture a pear, they get nothing, and reasonably conclude the technique is not for them. It may still be: the occupancy account does not need pictures, only something to fill attention. [Cognitive shuffling with aphantasia](https://dreamsheeps.space/cognitive-shuffling/aphantasia) covers what to try instead.

### People who find it effortful

For some people, generating word after word turns into exactly the kind of task the technique exists to remove. If you notice yourself checking whether you are doing it right, the structured version — or something that supplies the words — is the better fit.

### Waking at three in the morning

Cognitive shuffling is a technique for falling asleep. Waking in the night is a different problem with a different cause, and the first thing worth knowing about it is that brief waking between sleep cycles is normal: see [waking at 3 a.m. is normal](https://dreamsheeps.space/learn/waking-at-three).

### Insomnia that has lasted for months

Cognitive shuffling is a wellness technique, not a treatment. For chronic insomnia, the American College of Physicians recommends cognitive behavioural therapy for insomnia (CBT-I) as the first treatment to try (Qaseem and colleagues, 2016), and a doctor is the right place to start.

## Can something do cognitive shuffling for you?

Yes, in two forms. The first is the research form: audio that plays single words with long pauses between them, which is what serial diverse imagining is. Several apps do this, mySleepButton — Beaudoin’s own — among them; they are [compared side by side here](https://dreamsheeps.space/cognitive-shuffling/apps), and [cognitive shuffling, read to you](https://dreamsheeps.space/cognitive-shuffling/audio) covers when the guided form suits better than doing it yourself.

The second is narrated: a story built to the same rules, so that it holds your attention without giving you anything to follow. That is what Sheeps does, and it is ours, so weigh this paragraph accordingly. Each story is a run of harmless, concrete, unconnected scenes. The sentences start at fourteen to eighteen words and shrink to fragments of four to eight; the silence between scenes grows from about two seconds to about seven, converging on the research’s eight-second rhythm; and it ends in silence. Its Radio mode is the plain word-by-word form. [Plotless sleep stories](https://dreamsheeps.space/plotless-sleep-stories) explains the reasoning.

## Questions people ask

### Is cognitive shuffling the same as counting sheep?

No. Counting is orderly and needs almost no attention; shuffling is disorderly and needs just enough. The difference is the whole mechanism.

### How long does it take to work?

Nobody has a reliable number. The trial of the technique reported how wound up people felt and how hard they were trying, not minutes, and any figure you read in minutes comes from a different study.

### What words should I use?

Concrete, neutral, easy to picture: food, trees, tools, weather, kitchen things. Avoid anything with a personal story attached, anything unsettling, and abstract words you cannot see. There are [ready-made lists](https://dreamsheeps.space/cognitive-shuffling/words), grouped so that no two neighbours belong together.

### Is cognitive shuffling safe?

It is a way of thinking, so there is little to go wrong beyond it not helping. If a word brings up something distressing, drop it and move on — and if sleep has been a problem for months, see a doctor rather than relying on any technique.

### Does it work for children?

The research so far is in adults and university students. Nothing about the method is specific to adults, but there is no evidence either way for children.

## Sources

1. [Beaudoin (2014) — A design-based approach to sleep-onset and insomnia: super-somnolent mentation, the cognitive shuffle and serial diverse imagining](https://summit.sfu.ca/item/17237)
2. [Beaudoin, Digdon, O’Neill & Rachor (2016) — Serial diverse imagining task: a new remedy for bedtime complaints of worrying and other sleep-disruptive mental activity. Poster abstract, SLEEP 2016](https://summit.sfu.ca/item/16196)
3. [Simon Fraser University (2025) — Reprogramming Rest: Dr. Luc P. Beaudoin’s Science of Falling Asleep](https://www.sfu.ca/education/news-events/2025/april2025/reprogramming-rest-luc-beaudoin-cognitive-shuffling.html)
4. [mySleepButton — The cognitive shuffle: do-it-yourself instructions](https://mysleepbutton.com/en/support/do-it-yourself-cognitive-shuffle-sdi/)
5. [Harvey & Payne (2002) — The management of unwanted pre-sleep thoughts in insomnia: distraction with imagery versus general distraction. Behaviour Research and Therapy 40(3), 267–277](https://doi.org/10.1016/s0005-7967(01)00012-2)
6. [Gellis, Arigo & Elliott (2013) — Cognitive refocusing treatment for insomnia: a randomized controlled trial in university students. Behavior Therapy](https://doi.org/10.1016/j.beth.2012.07.004)
7. [Espie et al. (2006) — The attention–intention–effort pathway in the development of psychophysiologic insomnia: a theoretical review. Sleep Medicine Reviews](https://doi.org/10.1016/j.smrv.2006.03.002)
8. [Speth, Schloerscheidt & Speth (2016) — As we fall asleep we forget about the future: a quantitative linguistic analysis of mentation reports from hypnagogia. Consciousness and Cognition](https://doi.org/10.1016/j.concog.2016.08.014)
9. [Dance, Ipser & Simner (2022) — The prevalence of aphantasia (imagery weakness) in the general population. Consciousness and Cognition](https://doi.org/10.1016/j.concog.2021.103243)
10. [Qaseem et al. (2016) — Management of chronic insomnia disorder in adults: a clinical practice guideline from the American College of Physicians. Annals of Internal Medicine](https://doi.org/10.7326/M15-2175)

---

Sheeps does the shuffling for you, as a story: a new narrated sleep story, written for you, made of harmless, unconnected scenes, with sentences that shrink and pauses that lengthen until it ends in silence. It has not itself been clinically tested. Sheeps on the App Store: https://apps.apple.com/app/id6749510901