Do memory palaces and mnemonics really work?
An old trick for remembering lists, tested in a real lab: after weeks of practice, ordinary volunteers nearly quadrupled how many words they could recall.
Yes, and the evidence for it is stronger than ’ancient trick that sounds nice’. The method of loci, better known as building a memory palace, has been used for over two thousand years by orators memorizing long speeches without notes. It survived this long not out of nostalgia. Modern labs have since put it to the test, and it keeps winning.
What the palace can actually do
In 2017, neuroscientist Martin Dresler and colleagues ran a controlled trial published in the journal Neuron. Ordinary volunteers who had never used the technique trained in the method of loci for 30 minutes a day over 40 days. Before training, they could recall around 26 to 30 words from a list of 72. After training, the average climbed to 62 words, roughly double. A comparison group doing generic short-term memory drills improved by only 11 words, and a no-training control group improved by 7, the kind of gain you would expect just from taking the same test twice. The trained group’s gains were still there four months later, with no further practice.
That is a genuinely large effect, and it deserves an honest boundary right next to it: the task was recalling a list of unrelated words in order. The technique is built for exactly that kind of demand: things that need to be held in a sequence, with nothing natural to connect them. It is not a general intelligence booster, and the study did not claim it made anyone better at reasoning, comprehension, or anything beyond structured recall.
What seems to happen, according to the researchers, is that the training reshapes how different brain networks talk to each other during encoding, borrowing patterns already seen in elite memory athletes rather than simply working harder at the same old strategy. That is a useful thing to know for anyone tempted to dismiss the palace as a party trick: the change it produces is measurable, and it does not evaporate the moment practice stops.
Acronyms: excellent at order, quiet on understanding
Acronyms and rhymes work on a similar principle with a narrower job. A 2019 study by Tijana Radović and Dietrich Manzey, published in Frontiers in Psychology, taught participants an eight-step procedural sequence, some with a mnemonic acronym to hang the steps on, some without. The acronym group learned the sequence about four minutes faster on average, and when the researchers interrupted people mid-task, those using the acronym recovered their place with fewer errors, especially when the interruption landed right at a natural break point the acronym had marked.
But once a sequence was fully learned, the acronym group performed the task no faster and no more accurately than anyone else. The device helps you get the order into your head and helps you find your place again if you lose it. It does not make you better at the task itself. Reviews of this literature note the same pattern again and again: mnemonic devices reliably help when the order of items has to be learned and retrieved, and reliably do less when the challenge is understanding what each item actually means.
A memory palace stores what you already understand. It does not do the understanding for you.
Building a small one with a child
None of this requires marble halls. A memory palace can be the walk from a child’s bedroom to the kitchen, five or six spots they already know without thinking.
- Pick a route the child can walk with their eyes closed: bedroom door, hallway, stairs, kitchen table, fridge
- Turn each item to remember into one vivid, even ridiculous, image placed at one stop along the route
- Walk the route out loud together once, then have the child walk it alone from memory
- Practice retrieval, not just placement: ask for the list a day later, not only right after building it
- Save the technique for things that are genuinely just a list: capitals, planets, a sequence of historical dates, not for a concept that needs to be understood
That last point matters more than it sounds. A child who has built a beautiful memory palace for the order of the water cycle can still recite it perfectly without grasping why evaporation leads to condensation. The palace guarantees the words come back in order. It says nothing about whether the ideas behind them have landed. Pair it with an ordinary question afterward, ask them to explain one step in their own words, and the technique keeps its rightful place: a genuinely useful tool for holding onto order, sitting right alongside the harder, slower work of actually understanding what the order means.