The London Taxi Driver Study (Maguire 2000) Explained

    How Memory Works

    The London Taxi Driver Study (Maguire 2000) Explained

    London cabbies have a larger back part of the hippocampus and a smaller front. A later study scanned trainees before and after the Knowledge, and only those who qualified changed. What both studies found, what they did not, and what it means for your memory palace.

    By Published Updated

    Try a small exercise: getting from Holborn to Regent's Park using reference stories, small vivid pictures that each stand for one piece of information, instead of a map. A rat stands for right, leaves for left, and a headbutt for forward. Each instruction becomes something you can picture.

    London's taxi drivers solve the same problem on a vast scale. So what did the taxi driver hippocampus study find? In 2000 the neuroscientist Eleanor Maguire and colleagues scanned licensed London cabbies and found that the back part of their hippocampus, a brain structure central to memory and finding your way, was larger than in people who did not drive taxis, while the front part was smaller. The longer someone had driven a taxi, the larger the back part.

    Eleven years later, Woollett and Maguire tackled the obvious question: does the job change the brain, or do people with that kind of brain become cabbies? They scanned trainees before and after their years of learning the Knowledge, the training in which would-be drivers learn London's layout. Only the ones who qualified showed growth in the back of the hippocampus.

    I have written about the cabbies before, and I simplified it. I will put that right further down.

    Why it matters

    Two reasons, one for everybody and one for anyone who uses a memory palace, a set of places in your mind, walked in order, where you leave pictures of what you want to keep.

    For everybody: this is some of the clearest evidence that the adult brain still changes with sustained, demanding learning, and that the change can show up in the brain's structure, not just in how busy it looks on a scan.

    For palace users: the hippocampus supports spatial memory, the kind that lets you find your way. A memory palace hangs information on places. So the question of why places work so well as a filing system leads straight to the cabbies.

    The first study: cabbies and non-cabbies

    The research: Maguire et al, 2000

    • In short: Licensed London taxi drivers had a larger back part of the hippocampus than people who did not drive taxis, and a smaller front part. The back part was larger the longer they had been in the job.
    • What they did: Structural brain scans compared the size of the hippocampus in taxi drivers and in matched controls, and checked it against the number of years each driver had spent driving a cab.
    • Who was involved: Licensed London taxi drivers and matched controls. The exact numbers were not recorded when the paper was checked, so none is given here.
    • What they found: The drivers had larger posterior (back) hippocampi and smaller anterior (front) hippocampi than the controls. Posterior volume rose with time in the job.
    • What they concluded: The hippocampus, which supports spatial memory, may change with intensive navigation experience.

    What kind of evidence is this? A single study comparing two groups at one moment in time. It is famous and the finding has been replicated, which makes it moderate evidence. The limit is built into the design. A snapshot cannot rule out that people who already had a larger back hippocampus were the ones drawn to the job, or the ones who stuck at it. The link with years of driving is suggestive, but on its own it does not prove the job caused the difference. That needed a different kind of study.

    The second study: before and after the Knowledge

    The research: Woollett and Maguire, 2011

    • In short: Trainee London cabbies were scanned before and after learning the Knowledge. Only the ones who qualified showed growth in the back of the hippocampus.
    • What they did: The researchers scanned trainees at the start of their training and again afterwards, then compared those who qualified, those who did not and a control group, on brain structure and on memory tests.
    • Who was involved: Trainee London taxi drivers, both those who went on to qualify and those who did not, plus controls. As with the first study, the numbers were not recorded in our check.
    • What they found: Only the trainees who qualified showed an increase in grey matter in the posterior hippocampus. The qualifiers also did slightly worse on one visual memory test.
    • What they concluded: Intensive spatial learning can change the structure of the adult brain, and in this case the gain came with a small trade-off on another memory test.

    This is stronger evidence. Scanning the same people before and after is the design that answers the chicken-and-egg question, and the trainees who did not qualify make a useful comparison: they started the same training but did not complete it. Put together with the 2000 study, it makes a consistent case that the learning drove the change.

    What I take from it

    First, a correction to myself. In How to Remember Anything I wrote that tests show London cab drivers have an enlarged hippocampus. It is the version repeated in talks, articles and books, mine included, and it is a simplification. The accurate version is more interesting. The back part was larger and the front part was smaller. It was a shift within the hippocampus, not the whole thing swelling up like a balloon. If I am asking you to trust what I write, I need to get that right.

    Second, finishing matters. The growth turned up in the trainees who qualified, not in the ones who dropped out. That is the most useful line in either paper. A smaller version of it shows up in everyday learning: the people who change are the ones who keep going past the point where it gets hard. That comes from practice, and these studies do not measure it, but it fits.

    Third, the link to memory palaces. These papers did not test this. A memory palace borrows the brain's skill for finding its way around. You take the system that lets a cabbie get from one side of London to the other and ask it to carry something else: a speech, a list, a chapter. In that sense the Knowledge is one enormous memory palace, learned for its own sake. It is a helpful way to explain why places work so well. But nobody in these studies learned a memory palace, so treat it as an idea, not a finding.

    Cabbies were the benchmark long before I read the papers. On a television show I devised in the nineties, one of the experts our contestants had to face was a London cabbie who had done the Knowledge. Nobody needed a scanner to know that was a trained memory.

    What the studies do not show

    The taxi driver hippocampus study is one of the most quoted results in memory science, which means it is one of the most bent. Here is where it stops.

    • Not "an enlarged hippocampus". Only the back part was larger. The front part was smaller. If you hear the overall version, now you know better.
    • Not cause, from the 2000 study alone. A one-off comparison of two groups cannot tell you which came first. The before-and-after design of 2011 is what points to the training.
    • Not a free upgrade. The qualifiers did slightly worse on one visual memory test. A small cost, but a real one, and a reason to be wary of anyone who sells brain change as pure gain.
    • Not quick. The trainees learned over years. Nothing here suggests a weekend course will change your brain's structure.
    • Not about memory palaces. No one in either study used one. For what memory palace training does to the brain, see The Memory Athlete Brain Study (Dresler 2017) Explained, where the change was in how networks connect.
    • No street counts. You will often see a number attached to how many streets the Knowledge covers. None has been checked against a primary source here, so it is left out.

    How to use it this week

    You are not about to spend years learning every street in London. The principles still carry, and you can start small.

    1. Learn one real route with reference stories. Pick a walk or a drive you do not know well. Give every instruction a picture, a rat for right, leaves for left, a headbutt for forward, and fix each one where the turn happens, by the pub sign or the postbox. Run the whole route in your head twice before you set off, then follow it without your phone.
    2. Build a palace from a street you have never walked. You can do it by hand on Google Street View, the way I built my first virtual palace on George Street in Aberdeen years ago, or let LOCI build the palace from Street View for you. Stand in one spot, turn all the way round and pick ten locations in a circle, then move a few clicks up the road and do it again.
    3. Walk it away from the screen. Close the laptop, stand in the room and imagine yourself on that street. Name each location in order, out loud. If one is missing, go back and make it more distinctive.
    4. Put something you need on it. Ten points of a talk, the steps of a process, a list for an exam. LOCI adds a flashcard with the real information on the back of each image, and spaced retrieval practice, recalling it at widening gaps, is built in, if you want the reviews handled for you.
    5. Review on purpose. Walk the route and the palace again the next day and again a week later. A cabbie's routes get used every shift; yours need a reason to come back.
    6. Pick one thing to finish. The lesson from the trainees is that the change came to the ones who completed the learning. Choose one learning project, give it a finish date and see it through.

    For what memory palace training does to the brain, read The Memory Athlete Brain Study (Dresler 2017) Explained. And to put your own navigation skill to work, What is a Memory Palace? A Beginner's Guide to the World's Oldest Memory Technique and How to Build a Memory Palace: A 7-Step Beginner's Guide (With Worked Examples) show you how.

    To put a palace to work where it counts, How to Memorise a Speech with a Memory Palace (No Notes Needed) shows you how to place a whole talk along a route and give it without notes.

    Memory map

    1. Reference stories turn directions into pictures: a rat for right, leaves for left, a headbutt for forward. Image: a rat scurrying right past autumn leaves blowing left while a goat headbutts straight ahead.
    2. Licensed London cabbies had a larger back part of the hippocampus (Maguire, 2000). Image: a black cab with a huge, stretched boot and a tiny squashed bonnet.
    3. The front part was smaller, so it was a shift, not overall growth. Image: a seesaw tipped down at the back end.
    4. The longer they drove, the larger the back part. Image: a cabbie whose grey beard grows longer as a pile of worn road maps grows taller beside him.
    5. A one-off snapshot cannot show cause. Image: a single photograph of a chicken and an egg side by side on a kitchen table.
    6. Woollett and Maguire scanned trainees before and after the Knowledge (2011). Image: the same learner with a folded map standing at the start and at the end of a long winding road.
    7. Only those who qualified showed growth. Image: one runner breaking the finishing tape while others sit by the roadside.
    8. A small cost: slightly worse on one visual memory test. Image: a pair of scales with a city map on one side and a single missing jigsaw piece on the other.
    9. "Enlarged hippocampus" is a simplification, and I corrected my own book. Image: an author rubbing out a line in an open book with a big pink eraser.
    10. In my view, the Knowledge is one enormous memory palace. Image: a whole city seen from above with glowing objects placed on every corner.
    11. Build a palace from a real street with LOCI: ten locations in a circle, then a few clicks on. Image: a person standing in the middle of a street turning slowly, ten glowing objects in a ring around them.
    12. Finish what you start: give it a date. Image: a black cab pulling up at a chequered flag.

    Sources

    • Mark Channon, How to Remember Anything (Hodder Education): the Holborn to Regent's Park directions as reference stories; the taxi driver point in chapter 10, Lifestyle, pages 132 to 133.
    • Maguire, E. A., Gadian, D. G., Johnsrude, I. S., Good, C. D., Ashburner, J., Frackowiak, R. S. J., and Frith, C. D. (2000). Navigation-related structural change in the hippocampi of taxi drivers. PNAS, 97(8), 4398 to 4403. https://doi.org/10.1073/pnas.070039597
    • Woollett, K., and Maguire, E. A. (2011). Acquiring "the Knowledge" of London's layout drives structural brain changes. Current Biology, 21(24), 2109 to 2114. https://doi.org/10.1016/j.cub.2011.11.018
    • Dresler, M., Shirer, W. R., Konrad, B. N., Müller, N. C. J., Wagner, I. C., Fernández, G., Czisch, M., and Greicius, M. D. (2017). Mnemonic training reshapes brain networks to support superior memory. Neuron, 93(5), 1227 to 1235.e6. https://doi.org/10.1016/j.neuron.2017.02.003 (named here in passing; explained in full in The Memory Athlete Brain Study (Dresler 2017) Explained)

    Loci

    Build the memory palace in this article inside Loci — store your locations, link images, and rehearse with spaced repetition.

    Try it in Loci

    Memory School community

    Learn directly with Mark inside the Memory School community on Skool — monthly lessons, practice, and Q&A.

    Join Memory School

    About the author

    Grand Master of Memory (1995). Author of three books on memory published by Hodder and Nick Hern. Creator of BBC1's Monkhouse's Memory Masters. Memory coach on Channel 4's Can I Improve My Memory?

    Read full bio →