This article was first published in the September 2015 issue of WIRED magazine. Be the first to read WIRED's articles in print before they're posted online, and get your hands on loads of additional content by subscribing online
In May 2015, during the final match of the group stages of the Extreme Memory Tournament, Johannes Mallow, an engineer from Germany, attempted to break the speed record for memorising 80 digits in the correct order. Mallow, who claims to frequently misplace his keys and forget his appointments, can, for instance, memorise 1,080 binary digits in five minutes, 380 cards in ten minutes and 492 abstract images and words in 15 minutes. He can also memorise 501 digits in five minutes -- a world record he established in 2012 when he became world champion in London. He has developed a system that allows him to visualise every number from zero to 999 as a mental image. For instance, 090 corresponds to a pizza, 122 to plums, 559 to a tractor. "It took me half a year to develop a system with meaningful images," Mallow explains. "I memorised it in two weeks, but it was only after around two years that I became really fast at using it."
Inside the large room where the tournament was taking place, there were four tables, each with two laptops facing each other and two large flat-screens on the sides, turned outwards, facing an audience of 20 people. The first day of the competition took the form of a round-robin tournament. Twenty-four mental athletes, divided into six groups, competed against each other in four-minute matches of memorising -- one minute to memorise and three to recall -- in the categories of images, playing cards, random words, names matched to faces, and numbers. As the athletes memorised, the spectators sat quietly, watching the screens, which displayed the competitor's moves with a ten-second delay.
Mallow, who had already qualified in Group D, now faced Norwegian Ola Kåre Risa, who is seventh in the world rankings. Mallow was wearing his customised sunglasses with handmade cardboard black lens inserts with a small pinhole to block any peripheral visual distractions. Risa, sitting on the other side of the table, had red earmuffs and a cap with side flaps attached to the visor. At that point, three other matches were taking place. Also for Group D, Marwin Wallonius, from Sweden, faced Enkhmunkh Erdenebatkhaan, a member of a five-strong Mongolian team, all wearing blue polo shirts with sponsor logos and the Mongolian flag. His compatriots, Tuuruul Myagmarsuren and Akjol Syeryekkhaan, competed on the other side of the room, as Filipino Mark Anthony Castaneda played one of the favourites, Boris Konrad, a German neuroscientist with a PhD in the science of superior memory and a penchant for outbursts.
This year's Extreme Memory Tournament (XMT) was hosted at the headquarters of sponsors Dart NeuroScience, a pharmaceutical company developing memory-related therapeutic drugs, on the outskirts of San Diego. Apart from a few isolated studies, psychologists and neuroscientists have taken little interest in studying memory athletes, despite the fact that their memory feats falsify much of our current understanding about the limitations of human memory. For instance, one of the common assumptions about our memory span -- our ability to recall a list of digits, letters or words in order, having heard it only once -- is that it's fundamentally limited to seven items. "Johannes Mallow once got 364 correct, presented to him at a rate of one per second," says Henry Roediger, a cognitive psychologist who was sitting in the XMT audience and has been working with Dart for the past four years. "A lot of people still seem to think that our memory span can't be trained. Well, these guys have done it."
In 2010, Roediger was approached by Dart with a curious proposition: to find and study people with extraordinary memories. The assumption is that a close scientific scrutiny of these individuals might unravel new neurological or genetic discoveries that can help Dart develop new drugs and therapies. According to the company's estimates, it would need to screen approximately five million people to find 100 with superior memorisation skills; so, as its first project, it ran an online survey with a memory test that was taken by 50,000 people. Roediger also suggested a short cut to find extraordinary memories. They could test mental athletes. "Dart wanted us to find the best memorisers in the world so that it can run genetic tests and potentially find genes behind superior memories," Roediger says. "There are already a few tournaments around, but we thought it would a good idea to bring the athletes here."
Mallow is one of the seven mental athletes who, for the past three years, has travelled to the Memory Lab at Washington University's Department of Psychology in St Louis to be tested by Roediger, as part of his collaboration with Dart. When Mallow wants to memorise a long sequence of objects, he might mentally walk through his flat in Madgeburg, northern Germany, placing the objects he wants to memorise in specific places.
Mallow suffers from facioscapulohumeral muscular dystrophy, a degenerative muscle disease that confined him to a wheelchair in 2011. In his memory journeys, however, he sees himself walking. He starts with his front door, then he goes left, where his wheelchair sits, then he proceeds to the living room where he finds a cupboard to his left, then to his kitchen, where he places something in his refrigerator, his sink and his window, and another cupboard. He segues to his sofa, then inserts a disc into his DVD player and moves to the balcony. He moves to his lamp, to his TV, and, to its right, a set of candles. Next he goes into his bedroom, where he places objects on a chair, in his bed, on his bedside table, on the heater. There's a window with a potted plant. On his desk, his laptop. His bookshelf. He then goes into another room, where he has another plant, a coffee machine, his bike. Finally, he moves into his bathroom, placing the final objects in the toilet, the shower, the basin and the washing machine.
During a competition, Mallow can only use this journey once, so he has around 15 other "journeys", including one around his mother's flat and one around the campus at the Otto-von-Guericke University Magdeburg, where he is finishing his PhD in MRI tomography. "If I use one journey twice, there's a chance of mixing different images from different memorisations," Mallow says. "They are very hard to erase from the memory. I usually have to wait three to five days to be able to use that location again."
Mallow's final group match was numbers: competitors are presented with a string of 80 digits which they have to memorise. "Just imagine you are at a bar and you meet seven attractive people and you have to memorise their phone numbers," said Nelson Dellis, XMT's creator and founder, and the four-times USA Memory Champion and US memory-record holder, as he announced the round. Mallow looked at the digits displayed on his computer screen. At his front door, he placed a pizza with plums on it. He turned his wheelchair into a tractor and used a compass to navigate it. He saw the Digedags, protagonists of a popular German comic, bowling inside the cupboard. Inside his refrigerator, an alligator burned. In the sink, a frog climbed on to an anchor. On his windowsill was an oven full of fish fingers. He then saw himself using a hand plane to dry his laundry on the sofa. Between his video games, there was a cup with an igloo inside. On his balcony, a carousel with a drag queen. Shiva drove a car on top of his lamp. His TV showed a whiteboard struck by a gigantic fork. A nappy burned atop his candles while a stranger threw bread. Finally, he saw a tongue licking the chair, then somehow going online to search for an LP. After 20.01 seconds, he stopped the timer and closed his eyes. Then, from memory, he typed an 80-digit sequence on his laptop: 090122559241307722473262984627476516388218601050743002074416854297860438768731359. Even before the results were confirmed, Mallow knew he had broken the record.
The first World Memory Championships, officially called the First Memoriad, took place in October 1991, at the Athenaeum Club, a private members' club in Pall Mall, London. The seven contestants, all British, arrived dressed in tuxedos. They had been invited by Raymond Keene, a chess grandmaster, and Tony Buzan, author of a series of brain-training books including Use Your Head. Keene and Buzan wanted to establish an event that would pit the world's greatest mental athletes against each other in competitions that involved memorising abstract images, binary numbers, random words, even poetry. To find potential competitors Keene and Buzan scoured newspapers for stories related to memory feats. This is how they found Dominic O'Brien.
As a child, O'Brien had been diagnosed with dyslexia, which he suspects is the result of an accident he had when he was eight months old, when his pushchair was dragged by a train at St Leonards Station in East Sussex. The accident left O'Brien with a bump on the left side of his head. His parents didn't tell him about it until he was 16, at which point he had already left school, frustrated with his inability to cope in class. His school reports read: "Tends to dream in the middle of a calculation, which leads him to lose track of the thought." "Terribly slow. Often cannot repeat the question."
O'Brien got his first job working for an oil company. Later, he became involved in silver recovery, setting up a factory that processed 20 tonnes of photographic film a week. In 1987, O'Brien, then 30, watched an episode of TV show Record Breakers, featuring mnemonist Creighton Carvello. On the show, Carvello memorised a deck of cards in 179 seconds, viewing each card only once. "I had always been interested in card tricks, and that was the ultimate card trick," O'Brien says. "I didn't know whether he was a genius or if it was a trick." O'Brien went to a library to try to find information on how Carvello had done it. He found nothing, so he decided to work it out for himself. "I first thought he may have been using his body," O'Brien says. "If the first card was the two of clubs, he'd put his left foot at two o'clock; if the second card was a four, he'd move it to four o'clock. But you'd have to be a contortionist to do that." Mathematical formulas also proved a dead end. O'Brien thought of turning the cards into mental pictures -- an object, an animal or a person -- and placing them along a pre-planned route. "Making everything a journey was the eureka moment," O'Brien says. "The jack of hearts reminded me of my uncle, the ten of diamonds was the door at Downing Street, and so on. Then all I had to do was to mentally walk through the journey and meet all these characters." Using his journey method, O'Brien started memorising not only cards but sequences of numbers, words, names and faces, telephone numbers, poetry. His first deck took him 25 minutes to memorise, but he soon broke Carvello's three-minute record. In 1990, he memorised 35 decks with only two errors, entering the Guinness Book of World Records for the first time.
When he finally met Carvello at the Memoriad, the first thing O'Brien noticed were his shoes. "There were shiny to military precision," O'Brien says. "I thought, if this guy pays that sort of attention to his shoes, he's going to devastate me with his memory. But he was an oddity. He told me, 'Oh this is a bit of fun, isn't it?' And started giggling. That broke the ice." O'Brien beat Carvello in the final with the very skill that had first inspired him, by memorising a deck of cards in 149 seconds. O'Brien became the first World Memory Champion.
Dominic O'Brien retired from competition in 2002. By then he had won eight world championships and written eight books on the art of memory improvement. "Up until I wrote my first book, I had kept the journey method very much to myself," O'Brien says. "But then my co-author, who was doing some research, showed me these passages from Cicero and Parmenides that described how orators used cities and villages to plan out their speeches. They would mentally walk through these journeys and talk for hours from memory. It was virtually the same as my method. It was really fascinating to see I wasn't inventing anything new. These people had worked it out for themselves centuries before I did."
That same year, O'Brien, Carvello and six other mnemonists took part in a study at the UCL Institute of Neurology in London, conducted by cognitive neuroscientist Eleanor Maguire. She put her subjects inside an fMRI machine to look at their brain structure and examine what happens during memorisation. She found they predominantly used the hippocampus, a region in the brain where spatial memory is processed -- in essence, they were using a learning strategy based on how humans perceive space. The study also concluded that there was nothing special about their brains. The brains of mnemonists were structurally indistinguishable from the brains of the control group.
One of the first scientists to dissect the details of memorisation was a biologist called Tim Tully. In the 90s, while working at the Cold Spring Harbor Laboratory in New York, Tully found that memorisation was triggered by the formation of a protein called cyclic AMP response element-binding protein (CREB). CREB would bind to the DNA of brain cells, activating a set of genes needed to make new proteins that etch permanent connections between the cells that consolidate the memory. The more CREB available to the neuron, the faster the memorisation.
In 1994, Tully created a transgenic fruit fly with the CREB protein gene switched on. Tully ran experiments that conditioned flies to avoid a scented chamber where they would receive an electric shock. Normal flies would learn their lesson -- in other words, form a long-term memory that associated the odour to the shock -- after doing ten sessions with a 15-minute interval in between. The transgenic fruit fly would only require one session. "It was the first fruit fly with photographic memory," Tully says. "In fact, it was the first genetic manipulation in any animal, in history, that created a supra-normal memory. We proved that memory is a biological process that can be improved."
After that breakthrough, Tully decided to search for molecules that would have the same effect on the human brain, without direct genetic manipulation. "The mechanics of memory formation in humans aren't very different from those of fruit flies," Tully says. "Repetition is required. A normal person can memorise a list of 60 names and faces with sufficient repetition. When you up-regulate the CREB pathway, you reduce the amount of repetition required. We're trying to make small organic molecules that will act on the brain. People can take these orally and they will require less repetition to memorise information."
Finding people with superior memory was crucial to confirm the fruit-fly results, so in 2010 Tully approached Henry Roediger, one of the world's foremost experts in learning and memory. Roediger had spent decades dissecting the intricacies of average human memories. His research extends from the study of hypermnesia to researching the differences between cramming and spaced memorisation. Roediger was one of the originators of the Deese-McDermott-Roediger (DRM) paradigm -- the discovery that people can form false memories after an event. He is also one of the discoverers of the "testing effect", having showed that trying to recall information from memory re-enforces that same memory more effectively than re-reading that information. "The key to human memory is recoding: we take information in one form and we recode it in a different form," Roediger says. "Everybody does it, but memory athletes have perfected it. They build these journeys and they take in information and put it around it. Is CREB expressed in memory athletes? Sure. But we don't know if that's the main mechanism by which they memorise. The question remains: what are the neurological mechanisms that enable the memory athletes to do what they do?"
Roediger put each of the mental athletes through three days of tests. He conducted not only memory tests, but also tests measuring other aspects of cognition, such as attention. In one exam, Roediger tested the DRM paradigm. He presented a list of 112 words at a rate of two per second, and tested them shortly after. "If you present a list with related words like bed, rest, awake, tired and dream, the probability of falsely recalling a word that's implied but not present in the list, like sleep, is about the same as the probability of recalling words that actually were on the list," Roediger says. "For the memory athletes, the probability of recalling words was seven times higher than the probability of recalling words that were merely implied. One of the athletes recalled all of the words without any mistakes." Roediger also tested their ability to recall words and non-words -- words which are pronounceable but meaningless, like flirp, mashua and lencils. The memory athletes recalled 80 per cent of the words in order, and about 20 per cent of the non-words. The control group, composed of Washington University psychology undergraduates, was worse at memorising words than the mental athletes were at memorising non-words. "The next day, I gave them a surprise test on those same words," Roediger says. "The memory athletes had told us that they could do this immediately but there was no way they would remember them a day later, but their memories were almost intact."
Another test was the Stroop task, a standard test for attentional control which discriminates how well a subject can switch back and forth between tasks while keeping the focus. The Stroop task requires subjects to look at words used to describe colours written in different colours. Answering is nearly instantaneous when the colour and the word match -- "red" displayed in red -- but slower when there's a mismatch. It's this so-called incongruent trial that most effectively measures focus. Memory athletes were faster at reacting, taking an average of 600 milliseconds, than the control subjects. "The difference between congruent and incongruent trials was halved," says Roediger. "Not only do they have great memories, they have developed a laser-like focus."
The second day of XMT started with the quarter-finals. Mallow faced 18-year-old Johannes Zhou from Germany. Nelson Dellis, the event's organiser, asked the athletes to remove their ear defenders, glasses, visors and anything they were wearing to prevent distraction. The event, Dellis announced, would consist of memorising a deck of cards while listening to an audio montage. Dellis then gave the competitors two minutes to come up with a strategy. "I do CrossFit, so I like to train for the unknown, under stress and exhaustion," Dellis tells WIRED. "I wanted to do a memory tournament similar to CrossFit. If I could, I would dump people into ice baths as they try to memorise something, but I don't want to piss them off."
When he wants to focus, Mallow shuts his eyes and imagines himself back in his flat. "When I find myself in a situation where I could get anxious, I just imagine sitting at my desk at home," he says. "There's nobody there and it's silent. My heart starts beating slower. Then I can focus." During memorisation, Dellis played an audio mix, including jumbled-up clips of Katy Perry, Stephen Hawking, Kanye West and Ricky Martin.
Despite the distraction, Mallow perfectly memorised the deck of cards in 37.58 seconds, beating Zhou's 50.87 seconds. Mallow then lost the next two matches, but won match four and five. He defeated Zhou in Match Six, in the category of names and faces. "Names and faces is definitely my weakest," Mallow says. "That's why competitors always choose to play that against me." Mallow memorised 17 names and faces in 59.98 seconds. Zhou got the same number of names right, but took 0.02 seconds longer. "The knockout stages were very competitive," Mallows says. "I kept losing the first match and had to come from behind in every round. I kept calm but by the end of the day I was knackered."
Last year, in the first Extreme Memory Tournament, Mallow made it to the final, where he was defeated 5-1 by his friend Simon Reinhard. This year, Mallow reached the final again, to face Boris Konrad. Unlike the earlier rounds, the final takes place on the stage of the main lecture room at Dart Neuroscience. The final was played at the best of nine matches. After match seven of the final, Mallow was ahead by 4-3. Match eight was names and faces, one of Konrad's strengths and Mallow's weakness.
Mental athletes don't use the journey method for names and faces. Mallow might try to find a facial characteristic, like a long nose or a birthmark, which he then connects to the name. For instance, in the final, one of the photos showed a man called Damian with a fuzzy beard, which Mallow imagined had somehow been given to him by a demon. Dennis was playing tennis; a woman called Connie wore a cone on her head. "I have fun and put emotions into it," Mallow says. "I saw Walter drawing Walt Disney comics, or Alana as the sister of Alan Harper from Two and a Half Men. It makes me smile when I imagine those two punching each other." Mallow got eight names and faces wrong, three fewer than Konrad. When Mallow saw the result, he yelled and punched the air. "It was my best ever score in names," Mallow says. "Four years ago, I was depressed. My disease was not only affecting my body but my mind. Memory tournaments made me feel confident again. I can't walk, I can't stand, but it doesn't matter."
This article was originally published by WIRED UK
