In 1975, two researchers at Stanford, Richard Atkinson and Michael Raugh, gave the people in their experiment a list of Russian words and an odd instruction. For each word, find an English word that sounds like part of it. Then picture that English word doing something with the meaning of the Russian one. The people who used this trick got 72 percent right on the most important test. A control group, who could study as they liked, got 46 percent.
So yes, mnemonic devices work for vocabulary, and the effect can be large. But later studies found a catch: words learned through a mnemonic can also be forgotten faster, unless you review them. Used together with regular review, they are a useful tool for the words that refuse to stick.
How the keyword method works
Atkinson and Raugh described their method, published in the Journal of Experimental Psychology, as a chain with two links. The first link is sound: the foreign word reminds you of a word you already know, the keyword. The second link is an image: you picture the keyword together with the meaning, in one scene.
Here is how it might look for a learner of Spanish. Pato means duck. It sounds a little like the English word "pot". So you picture a duck sitting in a cooking pot, looking annoyed. Next time you hear pato, the sound leads to the pot, and the pot leads to the duck.
The method is simple. What makes it work is that it gives a meaningless sound something to hold on to. A new foreign word is, at first, just noise. The keyword connects it to something already in your memory, and the image connects that to the meaning. Germans have a good word for such a mental shortcut: an Eselsbrücke, a donkey bridge.
The catch: fast in, fast out
In the 1990s, the psychologist Alvin Wang and his colleagues looked more closely at what happens after the first test. Most earlier studies had tested the same people immediately and again later, which mixes up how much they learned with how much they forgot. Wang's team gave the immediate test and the delayed test to different groups.
The results were less flattering. In four experiments published in 1992, learners who used the keyword method forgot more over time than learners who simply repeated the words. In a 1995 follow-up, keyword learners did better on an immediate test. Two days later, the order reversed: people who had learned the words by thinking about their meaning in a context remembered more. This happened with rare English words and with second-language vocabulary.
The same study also found the fix. When learners got more chances to practise, the keyword group gained more from it than the comparison group did. The authors concluded that keyword memories are "especially fragile over time" and "will benefit from repeated testing and rehearsal". In plain words: the image gets the word in quickly, but it needs reviews to stay.
The memory palace
The memory palace, also called the method of loci, is the other famous mnemonic device. You imagine a route you know well, such as your flat, and place the things you want to remember at points along it. To recall them, you walk the route in your mind.
Memory athletes use techniques like this to memorise hundreds of words or digits within minutes. In a 2017 study in the journal Neuron, researchers in Munich and Nijmegen compared 23 of the world's top memory athletes with matched controls. Seventeen of the athletes also took a word test: twenty minutes after studying a list of 72 words, they recalled about 71 on average. Their matched controls recalled about 40.
Then came the more useful part. The team recruited 51 men with no experience in mnemonics and split them into three groups. One group practised the method of loci for 30 minutes a day, on 40 days. A second group trained their working memory with a demanding computer exercise. A third group did no training. Only the memory palace group improved significantly more than the others in recalling new word lists, both after 20 minutes and after 24 hours. The advantage was still there four months later.
Two limits matter for language learners. The study tested lists of words to recall in order, not foreign words with their meanings. And all the trainees were men, so we can't say how well the results apply to everyone. The memory palace is a strong tool for ordered lists; for linking a foreign word to its meaning, the keyword method is the closer fit.
What this means for you
- Save mnemonics for the stubborn words. Many words stick through normal review. When a word has failed three or four times, build a keyword image for it.
- Make the image do something. The method asks for the keyword and the meaning together in one scene, not standing side by side.
- Review within two days. That is where Wang's studies saw keyword memories fall behind. A quick test the next day protects the image while it is still fresh.
- Let the bridge go. Use the word in a sentence as soon as you can. Once you understand pato without thinking of a pot, the keyword has done its job.
- Use a memory palace for things in order. It suits lists you need in sequence, such as the seven Polish cases or the steps of a phone call to a doctor's office.
If you build a keyword image for a word, Qlango supplies the part that keeps it: review. Every finished lesson comes back after one, two, four and eight days and then once more, and examples you needed a hint for return later in harder questions. Because many exercises ask you to write the answer yourself, you find out quickly whether the image still leads to the right word. Our guide to spaced repetition explains why this timing works.
The kind of picture Atkinson and Raugh asked for still works fifty years later. It just was never meant to work alone. A mnemonic gets a word through the door; review is what keeps it in the house. For the other tools in the box, see our comparison of vocabulary learning methods.
Frequently asked questions
What are mnemonic devices for learning vocabulary?
Mnemonic devices are memory aids that link new information to something you already know. For vocabulary, the best-studied one is the keyword method: you connect a foreign word to a familiar word that sounds similar, then picture that word together with the meaning.
What is the keyword method?
It is a two-step technique tested by Atkinson and Raugh in 1975 with Russian vocabulary. Step one: find a known word that sounds like part of the foreign word. Step two: imagine a scene that joins that word with the meaning. In their study, it raised scores on the main test from 46 to 72 percent correct.
Does the memory palace work for learning a language?
It works very well for remembering lists in order: in a 2017 study, six weeks of training clearly improved word-list recall, with effects still visible after four months. It was not tested on linking foreign words to their meanings, so for single words the keyword method is usually the better choice.
Do mnemonics help you remember words long-term?
Only with review. Studies by Wang and colleagues found that words learned with keywords were remembered better at first but forgotten faster, and that extra practice helped keyword learners more than others. Combine a mnemonic with a review a day or two later, then at growing intervals.
