technique

The Link and Story Method

The first mnemonic technique to learn: turn a list into a chain of images, and remember twenty words in five minutes with no system to build.

hero illustration for The Link and Story Method

Here is the promise, and it is a testable one: in the next five minutes you will memorise twenty unrelated words, in order, and you will still have most of them tomorrow. No system to build first, no palace, no peg list. You need one skill, and you already have it, because you use it every time you remember a story someone told you at dinner.

The technique is called the link method (or, in its more narrative form, the story method). It is the oldest trick in the book, the one nearly every memory champion learned first, and the one that most beginner guides skip in their hurry to talk about palaces. That is a mistake. The link method is where the actual skill of mnemonics lives: turning a word into a picture, and joining two pictures so that one pulls the next one out.

What you are actually doing

Your memory for lists is poor because a list has no structure. Twenty words on a sheet are twenty separate facts with nothing holding them together, so each one has to be retrieved on its own and most of them are not. Your memory for events is excellent, because an event has structure: one thing causes the next, and once you have the beginning you can replay the rest.

The link method converts a list into an event. You take the first word and the second word, turn each into a concrete image, and then make the two images interact in some way that is impossible to picture without noticing. Then you take the second word and the third, and do it again. You never hold the whole list in your head. You only ever hold one join at a time. Recall is a chain: the first image reminds you of the second, the second of the third, and so on to the end.

Two things make a join stick. The first is interaction: the two objects must do something to each other, not merely sit side by side. A dog next to a lamp is forgettable; a dog chewing through the cord of a lamp, which sparks, is not. The second is specificity: not "a dog" but a wet, muddy dog with one ear inside-out. The extra detail is not decoration. It is what gives your recall something to grab.

The five-minute exercise

Do this now, before reading on, because reading about the link method does almost nothing and doing it once does almost everything. Here are twenty words. Cover the list after you have built the chain.

  1. apple
  2. bicycle
  3. candle
  4. dolphin
  5. envelope
  6. fountain
  7. guitar
  8. helmet
  9. iceberg
  10. jellyfish
  11. kettle
  12. ladder
  13. mirror
  14. necklace
  15. octopus
  16. pillow
  17. quilt
  18. rocket
  19. saxophone
  20. telescope

Now build it, one join at a time. Here is a chain you can borrow, but a chain you make yourself will hold better than one you read, so use this only as a demonstration of the kind of join that works.

An apple is jammed into the front spokes of a bicycle, and the wheel grinds it into pulp as it turns. The bicycle's headlamp is not a lamp but a lit candle, guttering in the wind. The candle drips hot wax onto the back of a dolphin that is swimming alongside, and the dolphin squeals. In its mouth the dolphin carries a soggy envelope. You open the envelope and a fountain gushes out of it, soaking you. Floating on the fountain is a guitar, strings up, and someone is playing it with a helmet instead of a pick, scraping the strings. The helmet is made of ice; it is a tiny iceberg, and it is melting on your head. Clinging to the iceberg is a jellyfish, tentacles trailing, and the jellyfish is whistling like a kettle coming to the boil. The kettle's spout is so long it is a ladder; you climb it. At the top is a mirror, and in the mirror you are wearing a necklace made of tiny octopus arms that are still moving. The octopus is sleeping on a pillow, and the pillow is stitched into a patchwork quilt. The quilt is tied to the nose of a rocket and flaps as the rocket launches. The rocket's exhaust makes a noise like a saxophone, and the saxophone's bell is pointed at the sky like a telescope.

Cover the list. Start at the apple. What was it jammed into? Follow the chain. Most people who actually build the images (rather than skim the paragraph) get eighteen to twenty on the first attempt and are surprised. Then do the thing that matters: come back tomorrow, before you re-read anything, and run the chain again.

If you would rather do this with a list you might actually use, the ten-grocery challenge walks you through exactly this technique with a shopping list, scores your recall before and after, and saves your attempt so you can check yourself again later.

Why it works, briefly

There is a classic experiment that anyone teaching this should be able to cite. In 1969, Gordon Bower and Michal Clark had students learn twelve lists of ten unrelated nouns. One group was told to make up a story that linked the words of each list, in order. The other group simply studied. When tested on all 120 words afterward, the story group recalled most of them; the control group recalled only a small fraction. The difference was not marginal. It was one of the largest effects in the memory literature, and it came from an instruction that took a minute to give.

The explanation most researchers accept is that a story forces elaborative encoding: every word gets connected to something, which gives retrieval a path to follow. Allan Paivio's dual-coding theory adds that concrete, picturable words are stored both verbally and as images, and that two codes are more retrievable than one. This is also why the technique works far better for "dolphin" than for "justice": you cannot picture an abstraction, so you have to swap it for something you can (a judge's gavel, a pair of scales).

One popular belief is worth correcting. The advice to make images "bizarre" is everywhere, and the evidence for bizarreness on its own is mixed. What reliably helps is interaction and distinctiveness. A strange image that does not connect the two items is worse than a plain one that does. Aim for the join, not the weirdness.

Turning words into pictures

The bottleneck for beginners is never the chain. It is converting each word into an image fast enough that the chain does not stall. Three rules cover most cases.

Concrete nouns: picture the thing. Apple, bicycle, dolphin. Make it specific and give it a texture, a colour, a sound.

Abstract words: substitute. For "freedom" picture a bird leaving an open cage; for "economy" picture a coin, or a shopping trolley. Pick the first concrete thing the word suggests to you and commit. You are not looking for the perfect symbol, you are looking for any picture that will bring the word back.

Names and unfamiliar words: use sound. "Kowalski" becomes a cow on skis; "endoplasmic" becomes an end of a plastic bag. This is the same move the Major System makes for numbers, and the same move behind the keyword method for foreign vocabulary. Sound-alikes feel silly for about a week and then become automatic.

Speed comes from practice, and the practice is trivial: take any list (a menu, a table of contents, the credits of a film) and link it. Five lists and you will be building joins as fast as you can read.

The story variant

The pure link method joins each item only to the next. The story method does the same thing but lets the joins accumulate into a plot, so that item seven is not only linked to item six but also happening in the scene the earlier items established. Many people find stories easier to build and easier to replay, especially for lists under thirty items. The cost is that stories drift: you can end up remembering the plot and losing which word each scene stood for. If that happens to you, tighten the joins and drop the plot.

There is no correct answer here. Try both on the same twenty words and use whichever one your recall prefers. The grocery challenge uses the story form, because groceries have a natural setting (a kitchen, a shop) and the setting does some of the work for you.

What goes wrong

A broken link. You reach the ladder and cannot remember what was at the top. This is the classic failure of a chain: one lost join and everything after it is gone. The fix is not to try harder in the moment. It is to notice, when you built the join, that it was weak (no interaction, or a vague image), and to rebuild it. With practice you develop a feel for a join that will not hold, the way you can tell a knot is loose before you pull on it.

Two items merge. The kettle and the ladder blur into one kettle-ladder and you skip one. This happens when a join is a substitution ("the spout is a ladder") rather than an interaction. Substitutions are quick to build but easy to collapse. Prefer one object acting on another: the kettle boils over onto the ladder and the rungs steam.

You cannot start. You remember the whole chain but not the first item, because nothing points at it. Link the first item to something that will be present when you need the list: the door of the shop, the cover of the exam paper, the face of the person you will be telling. This is the seed of the memory-palace idea, and it is the usual reason people graduate from links to loci.

You need item fourteen, and only fourteen. The link method gives you sequence, not random access. If someone asks "what was number fourteen?" you have to walk the chain from the start. For lists where you need to jump straight to a position, use a numbered peg system, which is what the Major System and its simpler cousins are for.

Vague images. "A guitar" with no detail will hold for ten minutes and be gone by evening. If your next-day recall is much worse than your immediate recall, this is nearly always why. Make the images more particular, not more numerous.

Making the images faster

After a few lists the limiting factor is no longer whether a join holds but how quickly you can build one, and there are three habits that speed it up.

Use the first image, not the best one. Beginners stall looking for the perfect picture for "envelope". Take the first one that arrives, even if it is dull, and make it interact with its neighbour. Interaction rescues a dull image; searching does not.

Involve your own body. Joins in which you are the one doing the thing (you climb the ladder, you open the envelope, the wax drips on your hand) build faster and hold longer than joins you watch from outside, because they come with a sense of touch and effort attached.

Say the join under your breath. "Apple in the spokes. Candle for a headlamp." Naming the interaction in four or five words forces you to have actually decided what it is, which is the step people skip when they merely picture two things vaguely near each other. It also gives you a second route back to the image at recall time.

None of this is a substitute for the chain itself. It is what turns a technique that works into one that is fast enough to use in the moment, at a counter, with someone waiting.

Where the link method fits

Think of the three foundational techniques as answering three different questions.

You need to rememberUseWhy
A short list, in order, soonLink or story methodNo setup; the list is the structure
A list you will reuse, by positionA peg list (rhymes, shapes, or the Major System)Each position has a permanent hook
A long list, or many lists, for a long timeA memory palacePlaces are ordered, plentiful, and already memorised

The link method has no setup cost, which is why it is the right place to start and the wrong place to stop. Once you are comfortable making joins, everything else in this dictionary is just a more organised place to put them. Pegs give the joins fixed addresses. Palaces give them a route. But the join itself, the moment where one image grabs the next, is the same skill in every technique, and it is the skill you just practised.

A training plan for the first week

Day one. The twenty words above. Build the chain, test immediately, and note your score. Then do the grocery challenge with its starter list.

Day two. Before looking at anything, recall both lists. Whatever you dropped, look at why the join failed and rebuild it. Then link a new list of your own: today's to-do list, or the ingredients of a recipe.

Days three to five. One new list a day, twenty to twenty-five items. Try the story form on one of them and the pure link form on another. Time yourself building; you should be well under five minutes by day five.

Day six. Recall every list you have built this week, in order, without looking. This is the test that tells you whether your images are specific enough. Expect the early ones to be weaker; that is information, not failure.

Day seven. Read about the Method of Loci and build your first ten-location route. You will find that placing an image in a location is a join like any other, and that you already know how to do it.

If you want a drill that generates fresh word lists and scores your recall, the words practice does that, and your misses feed back into your personal dictionary so the words you drop most often come round again.

Lists that are not lists of words

Most of what you need to remember does not arrive as a tidy column of nouns, but nearly all of it can be made into one, and the link method takes it from there.

A speech or presentation. Reduce each point to one concrete image: the budget slide becomes a pair of scissors, the hiring plan becomes a queue of people at a door. Chain the images. You will speak from the chain rather than from cards, and you will be free to look at the room. Politicians and barristers have done this since antiquity for the good reason that it works.

Directions. Left at the church, second right after the petrol station, past the school. Each landmark is already an image; join them in order and add the turn as an action (the church swings left, the petrol pump points right).

Facts for an exam. The causes of a war, the stages of a process, the bones of the wrist. Turn each into a substitute image and chain them. For material you will need for months rather than days, build the chain and then move it into a palace, where it can be revisited on a schedule.

A to-do list. The everyday case, and the one that makes the technique a habit. Chain the day's errands over breakfast, then throw the paper away. Being able to do this on demand, with no pen, is a small daily proof that the method is real.

The honest limits

The link method does not make you remember things you did not encode. It does nothing for material you read passively; you have to stop and build the join, and in the first week that feels slow. It also does not, on its own, give you long-term retention. A chain you built once will fade in days unless you walk it again. The people who keep lists for months are not building better images than you; they are revisiting the chain the next day, then three days later, then a week later. That schedule is the second half of the skill, and there is a practice page for it.

But for the specific, common problem the technique is meant to solve, a list you need in order, soon, with no preparation, it is as close to a free lunch as memory offers. Twenty words in five minutes was the promise. Cover the list and check.