What Is a Mnemonic Device?
A mnemonic device is any technique that makes information easier to remember by tying it to something your memory already handles well: a word, a sentence, a tune, a place or a vivid image. The word, pronounced ni-MON-ik, comes from the Greek mnēmonikos, "of memory," a relative of Mnemosyne, the Greek goddess of memory.
You almost certainly use a few already. If you learned the colors of the rainbow as ROY G BIV, or the lines of the treble clef as "Every Good Boy Does Fine," you've used the most common kind: a first-letter mnemonic.
Mnemonics work in two broad ways:
- They give random material a shape. Nothing in the names of the planets tells you their order, and nothing in "Huron" tells you which four lakes go with it. A sentence, a rhyme or a route through your house gives an arbitrary list a structure your memory can hold.
- They give recall a place to start. Much of what feels forgotten is still stored but hard to find. In a classic 1966 experiment, Endel Tulving and Zena Pearlstone showed that people recalled more of a word list when they were given category cues. A first letter, a rhyme or an imagined location is a cue you build on purpose.
That second point explains why mnemonics are brilliant at some jobs and useless at others. A mnemonic helps you retrieve something; it doesn't help you understand it. "Please Excuse My Dear Aunt Sally" can remind you of the order of operations, but it can't tell you why the rule exists, and, as you'll see, it doesn't even state the rule correctly.
So here's the rule of thumb this guide builds on: use a mnemonic when the information is arbitrary, and build understanding when it isn't. If a fact follows from a principle, learn the principle. If it's a name, a number or a sequence that someone simply decided on, a mnemonic is often the fastest way in.
9 Types of Mnemonic Devices, With Examples
Most mnemonics fall into two families. Verbal mnemonics work with letters, words and sounds: acronyms, acrostics, rhymes and songs. Visual mnemonics work with mental images: the method of loci, the peg-word system, the keyword method and the story method. Chunking sits in between, since it reorganizes information rather than decorating it.
| Type | How it works | Classic example | Best for | Strength of evidence |
|---|---|---|---|---|
| Acronym | The first letters of the items spell a word | HOMES: Huron, Ontario, Michigan, Erie, Superior | Short lists you already know | Modest; mainly helps with order |
| Acrostic | The first letters start the words of a sentence | "Every Good Boy Does Fine" for E, G, B, D, F | Ordered lists | Modest; mainly helps with order |
| Rhyme | Facts set in rhyme and rhythm | "Thirty days hath September, April, June, and November" | Short rules and facts | Little direct testing |
| Song or jingle | Words set to a simple melody | The alphabet song | Exact wording | Helps only when the tune repeats |
| Chunking | Items grouped into fewer, meaningful units | A phone number split into 555-867-5309 | Numbers and long strings | Strong |
| Method of loci (memory palace) | Images placed along a route you know well | A shopping list stationed through your apartment | Long ordered lists, speeches | Strong |
| Peg-word system | Items attached to rhyming number pegs learned in advance | "One is a bun, two is a shoe, three is a tree" | Numbered lists | Strong in lab studies |
| Keyword method | A sound-alike word plus an image links a term to its meaning | Spanish pato (duck) sounds like "pot," so picture a duck wearing a pot | Vocabulary and terminology | Strong for recall, mixed after delays |
| Story (link) method | Items chained together in a vivid story | Each item on a list hands off to the next in one connected scene | Short lists | Strong in lab studies |
A few notes on each:
Acronyms and acrostics. An acronym uses the letters themselves (ROY G BIV for red, orange, yellow, green, blue, indigo, violet). An acrostic turns each letter into a word, which helps when the letters don't spell anything you can say: "King Philip Came Over For Good Soup" for kingdom, phylum, class, order, family, genus and species.
Rhymes and songs. "In fourteen hundred ninety-two, Columbus sailed the ocean blue." Rhyme, meter and melody limit which words can come next, so a wrong word sounds wrong. Some of these are very old: a British Library prayer book from the first half of the 1400s already has "Thirti dayes hath November, April, June, and September."
Chunking. Ten loose digits are hard to hold; three groups of digits aren't. Our guide to the chunking technique covers it in depth.
The method of loci. Picture a route you know well and leave a vivid image of each item at a stop along the way; to recall, walk the route again. It's the technique behind most memory championships, and it has its own deep dive on the memory palace.
The peg-word system. Memorize a fixed set of pegs once ("one is a bun, two is a shoe, three is a tree") and hang each new item on its peg with an image. If item three is "hammer," picture a hammer stuck in a tree. Because each number is its own cue, you can jump straight to item seven.
The keyword method. It ties a new word to its meaning in two links: a sound link to a familiar word that sounds like it (pato and "pot"), and an image link from that word to the meaning (a duck wearing a pot). Richard Atkinson and colleagues at Stanford developed it in the 1970s for foreign-language vocabulary, and the same two-step trick is often used for technical terms and names.
The story method. Chain the items into one story, each doing something to the next. It sounds childish, but in Bower and Clark's 1969 study, students learned 12 lists of 10 nouns, and on a final test the median recall of those who wove each list into a story averaged 93% across the lists, against 13% for students who studied them normally.
Do Mnemonic Devices Work? What the Research Shows
Short answer: yes, for the right kind of material, and some kinds work far better than others.
The keyword method. In Atkinson and Raugh's 1975 experiment, 52 Stanford undergraduates learned 120 Russian nouns over three days. On a final test on day four, students given keywords recalled 72% of the meanings, against 46% for students who studied however they wanted. On a surprise test about six weeks later, the keyword group still led, 43% to 28%. Atkinson also reported a Spanish experiment against instructed rote rehearsal: 88% to 28%.
Memory palaces and peg-words. In Henry Roediger's 1980 comparison of four mnemonics, 150 undergraduates learned 20-word lists. Scored for words recalled in their exact positions right after study, a control group that repeated the words while thinking about their meanings averaged 5.8, the story (link) method 9.6, peg-words 12.5 and the method of loci 13.6. A day later, the story, peg-word and loci groups still recalled about four times as many words in position as the rehearsal group (4.9 to 5.8, against 1.3). Picturing each word on its own, without linking the images, did no better than rehearsal at keeping words in position.
Training effects last, too. Martin Dresler's 2017 study split 51 men with no memory training into three groups, and the 17 in the loci group practiced the method for 30 minutes a day, 40 sessions over six weeks. In a later analysis of the same data, their recall of a 72-word list rose from about 25 words to 62, far more than in the groups that trained working memory or did no training. At a follow-up four months later, after studying a word list again, the loci group recalled about 50 words, against about 30 and 27 for the other two groups.
Superior memory looks trainable rather than inborn. When Eleanor Maguire and colleagues studied ten people renowned for memory feats in competitions such as the World Memory Championships, nine of the ten used the method of loci. As a group they matched their controls on tests of general ability, and their brain scans showed no significant structural differences.
In real classrooms. When educational psychologist Joel Levin graded twenty years of classroom research in 1993, two areas earned an A: "mnemonic materials specially designed for factual content" and "mnemonic applications with educationally handicapped students." Research on whether students keep using the strategies and apply them on their own earned much weaker grades. The second A has held up: in Thomas Scruggs and Margo Mastropieri's 2000 research synthesis, mnemonic instruction for students with learning and behavior problems produced an average effect size of 1.62, enormous by education standards.
For students in general, the estimate is smaller but still solid. A 2021 meta-analysis by Gregory Donoghue and John Hattie found an average effect of d = 0.50 for mnemonics across 107 effect sizes, drawn from the studies behind a well-known 2013 review, and mostly measured on surface-level, factual learning.
The skeptical view. That 2013 review, by John Dunlosky and colleagues, rated the keyword mnemonic "low utility," along with using imagery to learn from text. Not because they fail, but because "their effects are limited to materials that are amenable to imagery" and "they may not work well for students of all ages." It also flagged the time they take to learn and open questions about how long the benefits last.
Put together, mnemonics produce big gains in recalling arbitrary information, especially for learners who struggle, but their reach is narrow.
Acronyms, Acrostics, Rhymes and Songs: What Word Mnemonics Can Do
First-letter mnemonics are the ones everyone knows, and the evidence for them is surprisingly thin. In a survey of 481 psychology students at 64 institutions, Jennifer McCabe and colleagues found acronyms and acrostics were the most familiar mnemonics (4.43 on a 5-point scale) and the most used (3.53), far ahead of the memory palace (2.91 and 2.19).
Here's what the experiments show:
- For unrelated words, they barely help. In Boltwood and Blick's 1970 study, students who made first-letter mnemonics for 19 unrelated nouns did no significantly better than students who just repeated them, either right away (16.26 against 15.73) or eight weeks later (8.26 against 6.77). A story condition in the same study did beat repetition at the delayed tests.
- They protect order. Nelson and Archer (1972) found that a supplied first-letter code word improved serial recall, and tied the gain to keeping words in their right positions rather than recalling more of them.
- They help most when you already know the items. Morris and Cook (1978) found no benefit for unrelated words, but a clear one for remembering a random order of the days of the week. Their conclusion: first-letter mnemonics "aid recall when the items are known but their ordering presents problems."
Stalder (2005) found that psychology students who used acronyms scored higher on the related exam questions, though the study tracked students' own use rather than assigning it. Treat an acronym as a retrieval checklist: great for naming all five Great Lakes or the steps of a procedure in order, useless for learning what the items are.
Songs come with a catch. Wallace (1994) found that singing a text helped people recall it only when its verses repeated one simple melody, so the tune itself was easy to learn. With a single verse, or a new melody for each verse, plain speech worked better. Calvert and Tart (1993) found a similar pattern with the Preamble to the US Constitution: after eight exposures over four weeks, college students who learned it as a song recalled more of it word for word than students given a spoken version (40.3 words against 25.4 five weeks later), but after a single exposure the two came out about even.
When Mnemonics Backfire
A mnemonic is a shortcut, and shortcuts can take you somewhere wrong. Here are six ways it happens.
1. The memory can fade fast. In four experiments by Wang, Thomas and Ouellette (1992), keyword learners forgot more over a week than rote learners did, and in two of them the keyword group's head start was gone by the delayed test. In a 1995 follow-up, Wang and Thomas saw the keyword advantage reverse after two days against learning words from sentence context. The fix is review, not abandonment: in that same paper, extra study opportunities helped keyword learners' long-term retention more than the context group's.
2. The mnemonic states the wrong rule. PEMDAS, or "Please Excuse My Dear Aunt Sally" (parentheses, exponents, multiplication, division, addition, subtraction), implies that multiplication always comes before division and addition before subtraction. In fact each pair shares a rank and runs left to right, so 7 - 3 + 11 is 15. When Glidden tested 381 future elementary school teachers in 2008, error rates on order-of-operations problems ran from 21.7% to 78.5%. Among those who correctly multiplied before adding, 30.9% added before subtracting and 38.0% multiplied before dividing instead of working left to right: exactly the mistakes the mnemonic invites.
3. The rule has too many exceptions. "I before E except after C" falls apart at weird, their and science. England's 2009 spelling guidance for primary schools, Support for Spelling, was blunt: "The i before e except after c rule is not worth teaching." It only holds when the letters make a clear "ee" sound, and the few words that fit are easier to learn directly.
4. The facts change. "My Very Educated Mother Just Served Us Nine Pizzas" lost its pizzas when the International Astronomical Union reclassified Pluto as a dwarf planet in August 2006. National Geographic's contest for a replacement was won in 2008 by 10-year-old Maryn Smith, whose sentence kept Pluto and added the dwarf planets Ceres and Eris: "My Very Exciting Magic Carpet Just Sailed Under Nine Palace Elephants."
5. You remember the cue but not what it stands for. If you learn "King Philip Came Over For Good Soup" without learning the ranks, the C could stand for anything. Adam Putnam's 2015 review of mnemonics in education is skeptical of first-letter mnemonics despite their popularity, and recommends "treating mnemonics as a retrieval aid rather than a core learning strategy."
6. Rushed images don't form. In a 1968 peg-word study, Bugelski, Kidd and Segmen found the pegs helped when people had 4 or 8 seconds per item, but not when they had 2. If you're racing through a list, the images may not have time to form, and the mnemonic may not help at all.
How to Make Your Own Mnemonic in 5 Steps
Step 1: Check that the material is arbitrary. Ask whether you could work the answer out from something you understand. If you can, learn that instead. Mnemonics earn their keep on names, lists, sequences, vocabulary, symbols and dates.
Step 2: Pick the type that fits the job. Match the mnemonic to the job:
| What you need to remember | Best mnemonic | Why it fits |
|---|---|---|
| The order of a list you mostly know | Acronym or acrostic | The first letters cue each item and lock the sequence |
| A long list in order, like a speech or a procedure | Method of loci | A familiar route supplies a fixed series of cues |
| A numbered list you need to jump into | Peg-word system | Each number is its own cue, so you can start anywhere |
| Foreign words or technical terms | Keyword method | A sound link and an image tie the word to its meaning |
| Exact wording, like a poem or a definition | Song or rhyme | Meter and melody restrict what can come next |
| Long numbers | Chunking | Fewer, meaningful groups fit in memory |
Step 3: Make your images interact. A duck next to a pot is weak; a duck wearing the pot as a helmet is strong. Wollen, Weber and Lowry (1972) showed people pairs of objects drawn bizarre or ordinary, and interacting or side by side. Bizarreness made no difference. Interaction did: interacting pictures beat a words-only control, and side-by-side pictures did worse than no pictures at all. Strangeness helps mainly when it stands out: McDaniel and Einstein (1986) found bizarre images were remembered better when they were mixed in with ordinary ones.
Step 4: Borrow a good one, or build your own. It's often said that a mnemonic only works if you invent it yourself. The evidence is mixed. In an experiment with 363 high school students learning Latin words, Campos and colleagues (2004) found keywords beat rote learning only for vivid, easy-to-picture words, and in a second experiment, for the vivid words, keywords made up by other students beat the ones students made themselves. Putnam's review splits the difference: ready-made mnemonics help young, untrained or time-pressed learners, and making your own can pay off once you've practiced and have time.
Step 5: Test it, then space it. Cover the answer, look at the cue, and recall. Check yourself, then do it again tomorrow, in a few days and next week. The mnemonic got the information in; this is what keeps it.
Mnemonics, Active Recall and Spaced Repetition: Use All Three
The three techniques do different jobs, and they work best together:
- A mnemonic turns something meaningless into something you can picture or say, and gives you a cue to start from.
- Active recall makes you pull the answer from memory instead of rereading it, which strengthens the path from cue to answer.
- Spaced repetition brings each item back just as you're about to forget it, which keeps it for months rather than days.
Dunlosky's 2013 review rated just two of its ten techniques "high utility": practice testing and distributed practice. Mnemonics don't replace them. They give retrieval practice something easier to grab.
Miyatsu and McDaniel tested the combination in 2019 with 40 Lithuanian words. In one of their experiments, students who combined keywords with retrieval practice remembered more a week later than those who relied on keywords alone.
A simple way to run all three on flashcards: put the term on the front and the meaning on the back, with your mnemonic underneath. When you review, try to recall the answer without the mnemonic first, and use it only as a hint when you're stuck.
How to Turn What You Read Into Mnemonics
Most of what you'll need a mnemonic for turns up in something you're reading or watching: a list of risk factors in a textbook, a framework in a business book, the stages of a process in a lecture. Catch them as they pass instead of hunting for them the night before an exam.
- Highlight the arbitrary bits as you read. With Glasp's web highlighter, mark lists, sequences, names and terms on any web page or PDF. Give them their own highlight color so "memorize this" never gets mixed up with "this is interesting."
- Write the mnemonic on the highlight. Add it as a note on the highlighted passage, so the cue and its source stay together and you can check your acrostic against the original list.
- Pull in your books and lectures. Import your Kindle highlights to collect the lists from what you've read, and use YouTube Summary to get a lecture's transcript with timestamps, so you can highlight the moment the professor runs through the seven steps.
- Brainstorm candidates, then check them. Ask Glasp's AI chat for three possible keywords or acrostic sentences for a list you've highlighted. Pick one, swap in images you can picture easily, and check that every item is there in the right order, because a generated acrostic can drop or reorder one.
- Quiz yourself from the cue. Turn each highlighted list into a question, recall it, using the mnemonic only when you're stuck, and space the reviews out. If a mnemonic keeps failing, rebuild it.
Frequently Asked Questions
What are the main types of mnemonic devices?
The most common are acronyms, acrostics, rhymes, songs, chunking, the method of loci (memory palace), the peg-word system, the keyword method and the story method. Verbal mnemonics use letters and sounds; visual ones use mental images, and in lab studies the visual ones usually produce bigger gains.
What is an example of a mnemonic device?
ROY G BIV, for the colors of the rainbow, is one of the best known. Others include HOMES for the Great Lakes and "Thirty days hath September" for the lengths of the months. A less famous but better-tested example: picturing a duck wearing a pot to remember that the Spanish pato means duck.
Do mnemonic devices actually work?
Yes, for information that has no meaning of its own. In Atkinson and Raugh's 1975 experiment, keyword learners recalled 72% of 120 Russian words against 46% for a control group, and a 2021 meta-analysis put the average effect of mnemonics at d = 0.50. The catch: they improve recall, not understanding, and some of the benefit can fade within days unless you review.
What is the most effective mnemonic device?
In head-to-head lab tests, imagery methods win. In Roediger's 1980 comparison, students recalled an average of 13.6 words of a 20-word list in the right positions with the method of loci and 12.5 with peg-words, against 5.8 for a rehearsal control group. For foreign vocabulary, the keyword method has the strongest record. First-letter mnemonics, the most popular kind, mostly help with order.
Are mnemonics good for studying for exams?
They're good for exams that test recall: terms, lists, steps, symbols, names and dates. They do little for questions that ask you to explain or apply ideas, so use them for the arbitrary pieces and spend the rest of your time on practice questions, starting days before the exam.
How do you make a mnemonic?
Check that the material is arbitrary, then choose the type that fits: an acrostic for order, a memory palace for long sequences, the keyword method for vocabulary, chunking for numbers. Make the images interact, check the mnemonic against the original, and practice recalling from the cue over several days.
Can mnemonics help you learn a language?
Yes. The keyword method was designed for second-language vocabulary, and in Atkinson's Spanish experiment keyword learners recalled 88% of the words against 28% for rote rehearsal. Expect some of that gain to fade without review, and pair the method with retrieval practice: in a 2019 experiment, keywords plus self-testing beat keywords alone a week later.
What is the difference between a mnemonic and a memory palace?
A memory palace is one kind of mnemonic. "Mnemonic" covers every technique that ties information to an easier cue; a memory palace, or the method of loci, is the specific technique of placing images along a familiar route. It's among the most powerful mnemonics, and the most effort to learn.
Start With One List
Mnemonics aren't a study system. They're a precision tool for one stubborn kind of information: the arbitrary list, the foreign word, the sequence somebody else decided on. Used there, they have strong evidence behind them, especially the picture-based methods. Used everywhere, they waste time and can plant the wrong rule.
So don't try to mnemonic your way through a whole course. Pick one list you keep forgetting this week, choose the type that fits it, build one vivid image or sentence, and test yourself on it tomorrow and again next week.
Start with whatever you're reading now. Highlight the list with Glasp, write the mnemonic on the highlight, and let retrieval practice do the rest.
