How Stories Help You Remember People, Places, and Things

The roof came down while he was outside. Simonides, a poet, had been reciting at a banquet in Thessaly when a servant called him out to meet two visitors waiting at the door. There was nobody at the door. Behind him, the hall collapsed onto everyone still at the table.

The bodies were crushed past recognition. Families arrived and could not tell which of the dead belonged to them, and without identification there could be no burial rites. Simonides closed his eyes and walked around the table in his mind, seeing each guest in the seat they had taken, and named every one of them.

The story is almost certainly a legend. Cicero told it centuries later, and it has been repeated ever since because it captures something true: he had not memorized a list of names. He had memorized a room, with people in it, and a thing that happened.

What Simonides worked out

Human memory is bad at bare items and good at places, faces, movement, and consequence. A list of ten words is a set of unrelated things your brain has no reason to keep. Ten guests around a table, in their seats, in a room you were standing in an hour ago, is a scene, and scenes hold.

The technique that came out of that insight is called the method of loci, or the memory palace. You walk a route you know well, your childhood home, your street, and you leave what you need to remember at particular spots along it. Later you walk the route again and collect them. Competitive memory champions still use almost nothing else.

The experiment that put a number on it

In 1969, two psychologists at Stanford, Gordon Bower and Michal Clark, ran a study so clean it still gets taught.

They gave people twelve lists, ten nouns each. Half were told to study and rehearse in the ordinary way. The other half were told to invent a story that strung the ten words together, however silly. Story-makers got as long as they needed, usually a minute or two, and their partners in the control group were given exactly the same amount of time.

Straight after each list, both groups were tested. Both were essentially perfect. Ninety-nine percent either way. At that point there is no story advantage at all.

Then all twelve lists were done, and the real test came.

Six or seven times as much, from the same words in the same amount of time. The storytellers also made fewer mistakes, rarely dragging in words that had never been on any list, and they got quicker at building stories as they went.

The shape of that result is the useful part. Stories did not help people learn faster. Both groups learned it. Stories stopped them losing it.

The version that runs for thousands of years

Simonides used one room. Aboriginal Australian songlines use a continent.

A songline is a sung route across country, where each significant place along the way carries knowledge attached to it: which plants grow there, which animals, what is safe to eat, where water sits, the law, the genealogies, the stars overhead. Sing the route and you walk through the knowledge in order. The landscape holds the sequence, so the sequence cannot be lost.

Lynne Kelly, working closely with Aboriginal Elders and later co-authoring Songlines: The Power and Promise with Margo Neale of the National Museum of Australia, describes each location as functioning like a subheading. Every place is a chapter of a much larger body of knowledge, and because you can always add another layer at a location you already use, the system never runs out of room.

Kelly has argued further that ancient monuments like Stonehenge and the Nasca lines functioned as memory spaces of the same kind, which remains a hypothesis and is debated among archaeologists. The songlines themselves are not in doubt.

Someone tested it on medical students

David Reser, a neuroscientist at Monash University, read Kelly’s work and had an idea. Medical students spend years memorizing lists of names with no inherent meaning, the bones of the wrist, the ligaments of the knee. He tried teaching them the Aboriginal technique, using narrative and place, and compared it against the classical memory palace.

It worked. Students who learned the songline-style method performed well on recall, and reported understanding something about Aboriginal knowledge systems they had not encountered before. A technique developed over tens of thousands of years turned out to be applicable to anatomy exams.

The people who memorized entire libraries

The largest memory feats on record were not performed by champions on a stage. They were performed by priests and monks, and the reason is simple: for most of history, if a text was not inside somebody’s head, it did not exist.

In India, the Vedic hymns were transmitted orally for thousands of years before anyone wrote them down, and the reciters were not merely repeating them. They developed a set of recitation patterns called pāṭhas, which take the same words and run them through fixed permutations.

Recite a text forward, then backward, then forward again, and any word that has drifted out of place announces itself. The patterns are an error-correcting system built entirely out of the human voice, and they are the reason a body of text survived thousands of years of copying by mouth with its word order and even its pitch accents intact.

In Myanmar, the same instinct produced an examination. Since 1948 the government has held the Tipiṭakadhara examinations, testing monks on the entire Pāli Canon, and they are held over thirty-three days each December and January in the Mahāpāsāṇa Cave at the Kaba Aye Pagoda in Yangon.

To pass, a candidate recites more than two point four million words aloud with correct pronunciation, and writes out over two hundred texts from memory, and answers on doctrine and comparative philosophy on top of that. Out of a monastic community of several hundred thousand, only fifteen monks had passed both the oral and written portions as of 2020.

The first monk to earn the title, the Venerable Mingun Sayadaw, left an account of his method, and it is startlingly ordinary. He took the canon passage by passage. He made sure he understood each passage in both Burmese and Pāli before trying to hold it. He broke it into chunks sized by how hard they were. He read each chunk aloud five times, closed the book, and said it back.

Understand it, break it up, say it out loud, test yourself with the book shut. Every serious memory tradition arrives at some version of that.

Put it to a tune

Ask an adult to recite the alphabet and watch what happens to their voice. Almost nobody says it. They sing it, to a tune they were given before they could read, and it has not faded in forty years.

Wallace tested this properly in 1994. People heard the same text either spoken or sung, then tried to recall it word for word. The sung version won, and the melody was doing more than keeping time. It chunked the words into phrases, marked how long each line should be, fixed the stress pattern, and told the listener where the emphasis fell. Some people who forgot the words still produced the right number of syllables, humming the shape of a line they could no longer fill.

Two follow-ups make it concrete. Karen Ludke and colleagues in Edinburgh taught sixty English speakers phrases in Hungarian, one of the harder languages to get a foothold in, under three conditions: speaking, rhythmic speaking, and singing. After fifteen minutes, the singers were ahead on every test. When Ecuadorian children learned an English passage as a song rather than as a poem, they beat the spoken group on recall, on pronunciation, and on translation, with the advantage still there six months later.

The modern memory industry

A whole commercial field now teaches memory and mental performance, and it is worth knowing what in it is old wisdom repackaged and what is a claim running ahead of its evidence.

Neuro-linguistic programming, developed by Richard Bandler and John Grinder in the early 1970s, popularized the idea of encoding material through several senses at once rather than one. That instinct is sound, and it is the same instinct behind a Vedic reciter using voice, rhythm and pattern together. The wider framework has not held up well. A review conducted for the United States Army by the National Research Council found little evidence for its assumptions or its effectiveness, and the visual-auditory-kinesthetic learning-styles idea that spread out of it has been tested repeatedly and has not been supported. A 2015 meta-analysis of NLP psychotherapy did find a modest effect, drawn mostly from observational studies rather than controlled trials.

The Silva Method, built by José Silva from the 1940s onward, trains a countdown into deep relaxation, then visualization from within that calmer state. Relaxation before study and vivid mental imagery are both genuinely useful, and anyone who has tried to memorize anything while agitated knows why. The claims layered on top, that this state raises IQ or opens psychic perception, do not have supporting evidence.

Joe Dispenza teaches mental rehearsal, guided meditation and repeated visualization. Mental rehearsal is real: imagining a sequence activates much of the same circuitry as performing it, which is why musicians and athletes practice in their heads. His broader claims about reversing disease sit well ahead of the published research. He has one peer-reviewed paper to his name, on brainwave changes in retreat participants, and the studies on his site are largely case reports. None of this should stand in for medical care.

Why a story sticks when a list does not

Four things happen inside a narrative that never happen inside a list.

• Causation. A list is ten items. A story is one thing making another thing happen, so recalling the first pulls the second along behind it.
• Pictures. Stories force images, and images engage far more of the brain than words alone.
• Feeling. Anything absurd, frightening, or funny is tagged as important. This is why the silliest chains work best.
• Order. A story cannot happen out of sequence, so the order comes free.

How this applies to you

All of this is usable tonight, and none of it needs equipment.

• Spelling and vocabulary lists. Have the child build one ridiculous story containing all the words in order. Absurd is better. This is precisely what Bower and Clark tested.
• Facts that will not stay. Attach them to places in your house. Walk the route to retrieve them.
• Names at a gathering. Do what Simonides did and fix people to where they were standing.
• Anything sequential. Kings, elements, the water cycle. Turn the order into a plot where each step causes the next.
• Read the same book again. A child asking for it a fortieth time is running the retrieval that makes it permanent.

Everything above was invented by people who had no choice. Before writing, a culture’s entire medical knowledge, law, navigation and food supply had to live inside human heads and survive the death of everyone who held it. What they built was not a filing system. They built stories, songs, and routes through country, and some of that knowledge has now traveled further in time than most written records ever will.