by Kevin Ashton
Within a year, Warren and Marshall had one hundred clean samples. They found that 90 percent of patients who had the bacteria had ulcers. Every patient with a duodenal ulcer—erosion in the lining of the acidic passage at the start of the intestine, immediately after the stomach—had the bacteria.
The two men tried and failed to grow the bacteria in the hospital’s lab. For six months they started with live samples and ended with nothing.
Then, during Easter 1982, a drug-resistant superbug contaminated the hospital and overwhelmed its lab. Warren and Marshall’s samples were forgotten for five days. The lab staff had been discarding the samples after two. The bacteria grew in the extra three. All they had needed was more time.
The bacterium was new. It was eventually given the name Helicobacter pylori, or H. pylori for short. Warren and Marshall wrote about their discovery in a 1984 letter to the Lancet, one of the world’s highest-impact medical journals. They concluded that the bacteria, which “appeared to be a new species, were present in almost all patients with active chronic gastritis, duodenal ulcer, or gastric ulcer and thus may be an important factor in these diseases.”
The Lancet’s editor, Ian Munro, could not find any reviewers who agreed. Everybody knew bacteria could not grow in the stomach. The results had to be wrong. Fortunately for Warren and Marshall—and all of us—Ian Munro was no ordinary journal editor; he was a radical thinker who, among other things, campaigned for human rights, nuclear disarmament, and medicine for the poor. In an unusual and impactful moment of science as it should be, Munro published the letter over the objections of his reviewers, even adding a note saying, “If the authors’ hypothesis should prove valid this work is very important indeed.”
Warren and Marshall went on to show that H. pylori causes ulcers. Others, building on their work, learned how to cure ulcers by killing H. pylori with antibiotics. In 2004, Warren and Marshall won the Nobel Prize “for their discovery of the bacterium Helicobacter pylori and its role in gastritis and peptic ulcer disease.” We now know that there are hundreds of species of bacteria in the stomach and that, among other things, they play an essential role in keeping the digestive system stable.
There is something strange about this story.
What Robin Warren saw on that cold, wet Monday was not something no one had seen. It was something everyone had seen. The only thing he did that no one else had done was believe it. By 1979, Warren had spent seventeen years mastering the complex science of pathology—the careful preservation and examination of human tissue—especially analyzing stomach biopsies. These became common in the 1970s, after the invention of the flexible endoscope—a tube with a light, camera, and cutting tool that doctors could feed down the throat of a patient and use to extract tissue. Before this, most samples were either from whole stomachs that had been removed or from cadavers. These were difficult to process. Information was lost while the samples were made ready for analysis. It was these bad samples that had led to every doctor and scientist being taught that bacteria do not live in the stomach. Warren said, “This was taken as so obvious as to barely rate a mention.” His biopsies told a different story.
“As my knowledge of medicine and then pathology increased, I found that there are often exceptions to ‘known facts,’ ” he said.
Also, “I preferred to believe my eyes, not the medical textbooks or the medical fraternity.”
He makes it sound simple. Yet flexible endoscopes were being used all over the world. Thousands of pathologists were looking at stomach biopsies. H. pylori was staring them all in the face. But they saw dogma, not bacteria.
In June 1979, the month Warren first noticed H. pylori, a group of American scientists published a paper about an epidemic of stomach disease among participants in a research study. The volunteers were healthy at the start of the project; then half of them became ill with stomach pain, followed by a loss of stomach acidity. The illness was almost certainly infectious. The scientists tested the patients’ blood and stomach fluid. They looked for a virus—because they knew bacteria could not grow in the stomach—and they did not find one. Their conclusion was: “We have been unable to isolate an infectious agent, nor have we been able to establish a viral or bacterial cause.” These were not beginners; they were led by a decorated professor of medicine who was also editor in chief of the journal Gastroenterology. After Warren and Marshall’s work was published, these scientists revisited their biopsies. H. pylori was clearly visible. They had seen it and not seen it. Their patients had been suffering from an acute infection of the bacteria. One of the scientists said, “Failing to discover H. pylori was my biggest mistake.”
In 1967, Susumo Ito, a professor at Harvard Medical School, had biopsied his own stomach and used an electron microscope to take a perfect photograph of H. pylori. It appeared labeled as a “spirillum,” but without further comment or attempt at identification, in that year’s American Physiological Society Handbook of Physiology. Tens of thousands of scientists saw the picture. None of them saw H. pylori.
In 1940, Harvard researcher Stone Freedberg had found H. pylori in more than a third of ulcer patients. His supervisor told him he was wrong and made him stop his research.
Only Robin Warren believed and would not be dissuaded. He maintained a lonely vigil over H. pylori for two years, until Marshall arrived.
H. pylori has now been found in medical literature dating back to 1875. When Robin Warren discovered it, it had been seen and not believed for 104 years.
2 | WHAT YOU SEE IS NOT WHAT YOU GET
H. pylori’s tiny boomerangs hid in plain sight for more than a century because of a problem called “inattentional blindness.” The name comes from perception psychologists Arien Mack and Irvin Rock, but the best definition comes from novelist Douglas Adams:
Something that we can’t see, or don’t see, or our brain doesn’t let us see, because we think that it’s somebody else’s problem. The brain just edits it out; it’s like a blind spot. If you look at it directly you won’t see it unless you know precisely what it is. It relies on people’s natural predisposition not to see anything they don’t want to, weren’t expecting or can’t explain.
Adams uses this definition in his book Life, the Universe and Everything in a scene where nobody notices that an alien spacecraft has landed in the middle of a cricket match. The story is comic, but the concept is real: the brain is the secret censor of the senses. It takes steps between sensing and thinking that we do not notice.
The path from eye to mind is long. Each eye has two optic nerves, one for the right half of the brain and one for the left. They travel as far back as they can possibly go—to an outside layer at the back of the brain called the visual cortex. Touch the back of your head, and your hand is next to the part of your brain that connects to your eyes. The visual cortex compresses what your eyes see by a factor of ten, then passes the information to the center of the brain, the striatum. The information is compressed again, this time by a factor of three hundred, as it travels to its next stop, the basal nuclei at the striatum’s core. This is where we discover what the eyes have seen and decide what to do about it. Only one three-thousandth of what is rendered on the retina gets this far. The brain selects what gets through by adding prior knowledge and making assumptions about how things behave. It subtracts what does not matter and what has not changed. It determines what we will and will not know. This preprocessing is powerful. What the brain adds seems real. What it subtracts may as well not exist.
This is why it is a bad idea to have a phone conversation while driving. Using a phone halves the amount of sensory information that enters our mind. Our eyes stare at the same things for the same length of time, but our brain edits out most of the information as unimportant. The information may be important for driving, but our brain preprocesses it based on what is important for our phone call. This does not happen when we are listening to the radio, because the radio does not expect us to talk back. It does not happen when we are talking to
a passenger, because the passenger is in the same space we are. But studies show that when we talk on the phone, we get inattentional blindness. The conversation is our problem. That child unexpectedly crossing the street in front of us is somebody else’s problem. Our brain does not let us see her. As Douglas Adams described it, our brain blinds our mind to the unusual.
This is also true when we are walking. In one study, researchers put a clown on a unicycle in the path of pedestrians. The researchers asked people who walked past the clown if they had noticed anything unusual. Everybody saw him unless they had been on their cell phone. Three out of every four people who had been using their phone did not see the clown. They looked back in astonishment, unable to believe they had missed him. They had looked straight at him but had not registered his presence. The unicycling clown crossed their paths but not their minds.
Harvard researchers Trafton Drew and Jeremy Wolfe did a similar experiment with radiologists by adding a picture of a gorilla to X-rays of lungs. An X-ray section of a lung looks like a black-and-white picture of a bowl of miso soup. As radiologists flick through images, they see progressive slices of the lungs, as if they are looking deeper into the soup. In Drew and Wolfe’s images, a crudely cut out black-and-white picture of a man in a gorilla suit had been added to the top right corner of some of the layers, as if he were floating on his back there. The radiologists saw the tiny nodules that indicated whether each lung was cancerous, but almost all of them missed the gorilla, even though it was shaking its fist at them and would have occupied as much space as a matchbox if it had actually been present in the lungs. Each radiologist who did not see the gorilla looked at it for about half a second.
Inattentional blindness is not an experimental effect. In 2004, a forty-three-year-old woman went to the emergency room suffering from fainting and other symptoms. The doctors suspected heart and lung problems, so they put a catheter into her body using a guidewire that went from her thigh to her chest. The doctors forgot to remove the guidewire. Five days passed before anybody found it. The woman spent all that time in intensive care, where she had three X-rays and a CT scan as attempts were made to stabilize her. A dozen doctors looked at the images. The guidewire in her chest—which, fortunately, did not contribute to her condition—was obvious on all of them, but nobody noticed it.
3 | OBVIOUS FACTS
When Robin Warren accepted his Nobel Prize, he quoted Sherlock Holmes: “There is nothing more deceptive than an obvious fact.”
It was an “obvious fact” that bacteria do not live in the stomach, just as it was an “obvious fact” that emergency room doctors remember to remove guidewires and an “obvious fact” that there are no gorillas in pictures of lungs.
Radiologists are experts in seeing. Years of training and practice make what is invisible to us obvious to them. They can diagnose a disease after looking at a chest X-ray for a fifth of a second, the time it takes to make a single voluntary eye movement. If you or I were to look at an X-ray of a lung, we would scan the whole thing, searching for irregularities. This is also what novice radiologists do. But as they become more trained, they move their eyes less, until all they have to do is glance at a few locations for a few moments to find the information they need.
This is called “selective attention.” It is a hallmark of expertise. “Expert” has the same Latin root as “experience.” Aldous Huxley, writing in his 1932 book Texts and Pretexts, says: “Experience is a matter of sensibility and intuition, of seeing and hearing the significant things, of paying attention at the right moments, of understanding and coordinating.”
Adriaan de Groot, a chess master and psychologist, studied expertise by showing a chess position to players of different ranks, including grandmasters and world champions, and asking them to think aloud as they considered their move. De Groot had two expectations. First, that better players would make better moves. Second, that better players would make more calculations. What he found surprised him.
The first thing he noticed was the same problem-solution loop undergraduates used to solve the candle problem, Apple used to design the iPhone, and the Wright brothers used to invent the airplane.
A chess expert’s first step is to evaluate the problem. One master started like this:
“Difficult: this is my first impression. The second is that by actual numbers I should be badly off, but it is a pleasant position.”
The second step is to think of a move:
“I can do a whole lot of things. Get my Rook into it, at the Pawns.”
Each move is evaluated after it is generated:
“No, a touch of fantasy. Not worth much. No good. Maybe not so crazy.”
De Groot discovered several things. First, unfamiliar problems are solved with slow loops that are easily verbalized. Second, everybody revisits and reevaluates some solutions. This is not indecision: each evaluation goes deeper.
What surprised De Groot was how the problem-solution loop differed between players of different ranks. He’d expected grandmasters to make the best moves, and they did. But he had thought that this would be due to more analysis. What he found was the opposite. Grandmasters evaluated fewer moves and reevaluated them less often than other players did. One grandmaster evaluated one move twice, then evaluated another and played it. It was the best possible move. This was generally true: despite evaluating fewer moves fewer times, four of the five grandmasters in the study made the best possible move. The other grandmaster made the second best possible move. Grandmasters did not consider any moves that were not in the top five best moves. Lower-ranked players considered moves as poor as the twenty-second best. The less expert the player, the more options they considered, the more evaluations they made, and the worse their eventual move was.
Less thinking led to better solutions. More thinking led to worse solutions. Was this evidence of genius and epiphany? Were grandmasters making their moves by inspiration?
No. De Groot noticed something odd as he listened to grandmasters thinking aloud. Here is a typical comment from a grandmaster:
“First impression: an isolated Pawn; White has more freedom of movement.”
Compare this to a master, a skilled player just one rank down, talking about the same position:
“The first thing that strikes me is the weakness of the Black King’s wing, particularly the weakness at KB6. Only after that a general picture of the position. Finally, the complications in the center are rather striking: possibilities for exchange in connection with the loose Bishop on K7. Still later: my Pawn on QN2 is en prise.”
En prise means the piece is vulnerable to being taken—this is the “isolated Pawn” the grandmaster mentioned first. We do not need to understand chess to see that the grandmaster came to an instant conclusion where the master took more time. De Groot hypothesized that the grandmasters’ “remarks represent but a fraction of what has, in reality, been perceived. By far the largest part of what the subject ‘sees’ remains unsaid.”
Experts do not think less. They think more efficiently. The practiced brain eliminates poor solutions so quickly that they barely reach the attention of the conscious mind.
De Groot showed this with another experiment. Grandmaster Max Euwe (a world champion), a master (De Groot himself, with his wife acting as experimenter), an expert-level player, and a class-level player were shown a position for five seconds, then asked to reconstruct it and think about a move. For Euwe, the grandmaster, this was trivial—he reconstructed the board easily. De Groot, the master, put nearly all the pieces in the right place but argued with his wife because he thought she had made a mistake setting up the board: “Is there really a Black Knight on KB2? That would be rather curious!” The expert-level player remembered three-quarters of the board; the class-level player, less than half.
Was grandmaster Euwe a genius? Did he have a photographic memory? No. As De Groot suspected, forcing Euwe to reconstruct the position showed he was thinking in fast loops:
“First impressio
n: awfully rotten position, strong compressed attack by White. The order in which I saw the pieces was about King on K1, Knight on Q2, White Queen on QB3, Queen on K2, Pawns on K3 and his on K4, White Rook on Q8, White Knight on QN4, Rook on QN5—that funny Rook that doesn’t do anything—Knight on KB2, Bishop on KB1, Rook on KR1, Pawns on KR2 and KN3. I didn’t look at the other side very much, but I presume there is another Pawn on QR2. The rest for White: King on KN8, Rook on KB8, Pawns on KB7, KN7, KR7, and QR7, QN7.”
In the five seconds he was given to look at the board, Euwe had seen the pieces in priority order, understood the logic of the position, and started reasoning about his move. He was doing ordinary thinking extraordinarily fast. His speed came from experience. It enabled him to see similarities to other games and connections between pieces. He did not remember the positions of the pieces—he inferred them. For example, he reconstructed the position that De Groot had assumed was a mistake by inference and without doubt: “Another piece is on KB2—the King was completely closed in—that must be a Knight then.”
The position reminded him of another game—“there’s a vague recollection of a Fine-Flohr game in the back of my mind”—and all the similarities he saw gave him “a certain feeling of being familiar with this sort of situation.” Experience enabled him to find a solution almost instantly.
Grandmasters have not been grandmasters forever. When they were masters, they played like masters, evaluating more moves more times. When they were expert-level players, they played like expert-level players—evaluating even more moves even more times. Because they have evaluated so many moves and accumulated so much experience, grandmasters can pay very selective attention to a game. The expert’s first impression is not a first impression at all. It is the latest in a series of millions.
Creating is thinking. Attention is what we think about. The more we experience, the less we think—whether in chess, radiography, painting, science, or anything else. Expertise is efficiency: experts use fewer problem-solution loops because experts do not consider unlikely solutions.