You’ve probably had the experience of missing a typo even after carefully re-reading something you’ve written or barely spotting a pedestrian who steps unexpectedly onto the road in front of you.
These kinds of errors are a result of the way our brains try to simplify the vast amounts of visual information that our eyes capture, researchers reported in a new paper published in Trends in Cognitive Sciences.
“We’ve got very substantial limits on what we can attend to at any given moment, and there’s way too much stuff coming in for our brains to attend to everything,” the paper’s lead author, Dr. Jeremy Wolfe, a professor of ophthalmology and radiology at the Harvard Medical School and the head of the Visual Attention Lab at Brigham and Women’s Hospital, told TODAY.
The brain has various mechanisms to pick and choose what might be worth paying attention to, Wolfe said. One of those mechanisms is to bypass what is rare, he added.
Take proofreading, for example: We might read past a typo because it’s rare and also because, in the interest of efficiency, our brains are always anticipating what comes next.
“So as we’re desperately reading along hunting for typos and missing words, our brains are filling in the blank part,” Wolfe explained. “Our brains are saying, ‘I know what the next word is,’ and that’s very hard to overrule.”
That’s why a person unfamiliar with the text is going to be better at spotting errors, Wolfe said. Also, it’s why it might be easier to find typos and misspelled words in a young child's compositions because they won't be so rare.
There are famous experiments showing how easy it is for people to miss the unusual — even when it’s right in front of them. Wolfe described one in which participants were asked to watch two teams of actors passing a ball around. The study participants were asked to count the number times the white-shirted team passed the ball. In the middle of the experiment, another actor, in a gorilla suit, walked into the room, pounded on her chest and then left. Later, when the participants were asked questions about what went on, only 50% of them said they saw the gorilla.
This problem is of particular interest to Wolfe because he’s a radiologist, and the same brain processes sometimes lead to mistakes in reading scans and X-rays. If a radiologist is looking specifically for a lung tumor, for example, pneumonia might be missed.
Wolfe and colleagues ran a similar experiment years ago, embedding an image of a gorilla in a lung CT scan. The vast majority of the radiologists, more than 80%, missed the gorilla as they focused on the small white nodules they had been told to identify.
For some radiologists, a solution that seems to work is to look at the scans before checking what the patient’s doctor thinks might be wrong. Another strategy could be having the scans read by two experts, Wolfe said.
For people proofreading their own writing, reading aloud slows you down enough that you can’t just jump over errors, Wolfe said.
When the brain skips over something it's not actively looking for, it's doing exactly what it’s designed to do, so Wolfe and his co-authors have decided to call this feature “normal blindness.”
“The brain is limited to attending to about 20 to 30 things per second,” he said. “If you are paying attention to every leaf on a tree and every pebble on the ground, you’re going to be stuck in place. These mechanisms help us get through the world, and inevitably we miss stuff.”
Most of the stuff we’re missing isn’t important to us, like what your significant other wore to work today. “But if you’re doing things like radiology, you want to cut down on errors,” Wolfe said.
Like all evolutionary systems, the way the brain “sees” the world has drawbacks, David Calkins, Ph.D., professor of ophthalmology and visual sciences and director of Vanderbilt Vision Research Center, told TODAY. He likened it to looking but failing to see.
“This article is important because it uses common mistakes, like a proofreader missing a typo, to make a far broader point,” Calkins said. “There are errors that are a lot more nefarious than typos.”
“You don’t necessarily see what you’re not looking for, and that has implications for navigating the world, for example, ducking at the right moment when you don’t see something headed your way,” Calkins added. “It also has implications for how we use scans, which are made at increasingly high resolution, to make diagnoses. It highlights the need for multiple pairs of eyes.”