Does the memory palace technique actually work?

Yes, for ordered material, and this is unusually well tested for a study technique. In a 2017 experiment reported in Neuron, people with no memory training who practised the method for six weeks improved, and the change in their brain's functional connectivity predicted how much better they still were up to four months later. The effect is on recall of the material you encode, not on general mental ability.

How long does it take to learn a memory palace?

The idea takes minutes; useful fluency takes weeks. The 2017 Neuron training study ran for six weeks at about half an hour a day and produced gains that were still measurable four months on. A 2012 experiment found that people could build a workable palace out of a virtual environment they had studied only briefly, so the starting cost is lower than most guides suggest.

Do people with high IQ have better memory?

Not in the way the question assumes, and the best evidence on this comes from studying the people who are best at it. When Eleanor Maguire and colleagues examined competitors renowned for memory feats in 2003, they reported that superior memory was not driven by exceptional intellectual ability or by structural differences in the brain. What set the group apart was strategy: they were using a spatial learning method.

Does the place have to be somewhere I know well?

Apparently not. Nearly every guide insists on a deeply familiar building, but a 2012 study in Acta Psychologica tested that directly: participants who used a virtual environment they had studied briefly did as well as those using highly familiar real places. The authors concluded that for people new to the technique, a highly detailed environment does not support substantially better memory for word lists.

Can a memory palace make you smarter?

It makes you better at remembering what you put into it, which is not the same thing. The classic demonstration of how far trained memory can go is a 1980 Science report in which one participant raised his digit span from 7 to 79 digits over more than 230 hours of practice — an extraordinary gain, achieved with digits. Treat any claim that a memorisation technique raises general intelligence as a claim about transfer, and transfer is the part that usually fails.

Memory Palace (Method of Loci): Does It Work?

Q: What is a memory palace?

A memory palace — the older name is the method of loci — is a way of remembering a list by placing each item at a specific spot along a route through a place you can picture. To recall the list you walk the route in your mind and collect the items in order. The technique is ancient, and it is still the method competitive memorisers use.

A memory palace is a way of remembering things in order by placing each one at a particular spot along a route through a place you can picture. To recall the list, you walk the route in your mind and pick the items up as you pass them. Its older name is the method of loci, and unlike most study techniques, it has been tested rather than merely recommended.

How the method of loci works

The mechanics are simple enough to learn in one sitting. What makes the technique work is that it converts a kind of material people are bad at holding — an arbitrary ordered list — into a kind they are good at: a walk through a place.

Choose a route, not a room. A sequence of distinct stopping points, in a fixed order — front gate, hallway, first stair, landing. Order is what the technique buys you, so the route has to have one.
Turn each item into an image. Something you could see and, ideally, something odd. An abstract fact remembered as an abstract fact has not been encoded into anything.
Put one image at each stop and interact with it — have it block the door, spill down the stairs.
Recall by walking. Re-imagine the route from the start. The order of the places gives you the order of the items, which is why serial recall is the thing this method is best at.

The technique is old. It was already a standard part of classical rhetorical training, taught as a way for a speaker to hold the sequence of a long argument without notes, and it has been rediscovered and renamed repeatedly since — journey method, mind palace, memory journey.

Does a memory palace actually work?

This is the first thing Google's own readers ask about it, and it is the question the popular guides skip. The strongest evidence is a 2017 study in Neuron by Martin Dresler and colleagues, which did something most memory research does not: it studied both the experts and what happens when beginners are taught the same method.

The team scanned 23 of the world's most successful memory athletes and matched controls, at rest and while encoding.
Separately, untrained controls did six weeks of mnemonic training. The changes in functional connectivity that training produced were correlated with the network organisation that distinguishes athletes from everyone else — the beginners' brains moved in the direction of the experts'.
How closely a trainee's connectivity came to resemble the athletes' predicted their memory improvement up to four months after the training ended. The gain was not a trick of the testing session.

The authors' conclusion is worth stating precisely, because it is narrower and more interesting than “it rewires your brain”: mnemonic training drove distributed rather than regional changes — reorganisation across visual, medial temporal lobe and default mode networks rather than growth in one memory centre. How the brain changes with practice in general is covered in neuroplasticity.

Are memory champions smarter — or just using a better method?

This is the question an intelligence site should be able to answer, and the evidence is unusually direct, because someone went and tested the champions.

In 2003, Eleanor Maguire, Elizabeth Valentine, John Wilding and Narinder Kapur studied people renowned for outstanding memory feats in forums such as the World Memory Championships, using neuropsychological tests alongside structural and functional brain imaging. Their finding, in their own words, is that superior memory was not driven by exceptional intellectual ability or structural brain differences. What distinguished the group was that they used a spatial learning strategy, engaging the hippocampus and other regions critical for spatial memory.

That is close to the most encouraging result in this whole field. The people who memorise decks of cards for sport are not a different kind of human with a different kind of brain; they are ordinary people running a method, and the method is the one described above. It also means the reverse inference fails: someone with a startling memory for one kind of material has demonstrated a trained skill, not a high score. What an IQ test does measure is set out in how IQ tests work.

What a memory palace will not do

The honest limit of this technique is the same limit that defeats most brain training: the gains attach to the material you trained on.

The best-known demonstration of the ceiling is a 1980 report in Science by Anders Ericsson, William Chase and Steve Faloon. One participant, practising in the laboratory for more than 230 hours, raised his memory span from 7 to 79 digits — and the authors noted that his performance on other memory tests with digits matched that of memory experts with lifelong training. An enormous, genuine, hard-won gain, built inside one kind of material. Whether that sort of practice moves a general score at all is the transfer question, and it is dealt with directly in how to increase your IQ.

The related idea that some people simply store pictures of pages is a separate claim, and a weaker one: photographic memory examines what the evidence for it actually shows. The difference matters here. A memory palace is a procedure anyone can run and that visibly takes effort; a photographic memory is a supposed capacity you either have or do not.

The advice that turns out to be wrong

Almost every guide insists the place must be one you know intimately — your childhood home, your daily walk. That instruction has been tested, and it did not hold up.

Eric Legge, Christopher Madan, Enoch Ng and Jeremy Caplan had participants build palaces from a virtual environment they had studied only briefly, then apply the method to ten lists of eleven unrelated words. Measured against an uninstructed control group, the briefly studied virtual environments worked as well as the traditional highly familiar ones. Their conclusion: at least for people new to the technique, a highly detailed environment does not support substantially better memory for verbal lists.

The practical consequence is that the usual first obstacle is imaginary. You do not need to wait until you have an intimately known building with exactly twenty distinct stations. Something briefly studied is enough to start, which is also why running out of palaces is a smaller problem than it is usually made to sound.

Trust and scope notes

This page is educational. It summarises published research on memory and mnemonic technique and is not medical or psychological advice. Memory problems that are new, worsening or interfering with daily life are a matter for a qualified clinician, not a study technique.

Our IQ test measures reasoning performance. It does not measure trained memory skill, and practising a mnemonic technique is not preparation for it — the two are different targets, which is the point of the Maguire finding above.

IQ Revealed is not affiliated with, endorsed by or connected to the World Memory Championships or its organisers, University College London, the Max Planck Institute of Psychiatry, Radboud University, Stanford University, the University of Alberta, Springer Nature, Elsevier, the American Association for the Advancement of Science, or any of the universities, journals or researchers named on this page. They are named so that each claim can be traced to the work that made it. Quoted phrases are taken from the published titles and abstracts of the papers linked in the sources.

Memory Palace: How the Method of Loci Works, and Whether It Works