In Isaac Asimov's 1987 novel Fantastic Voyage II: Destination Brain, Russian researchers send miniaturized humans into the brain of Pyotor Leonovich Shapirov, who's in a coma. Inside his head is the science that would make miniaturization safer and more widely available. Their mission is to extract his memories. At the heart of story is a basic human desire to know what people are thinking and to understand the source of their creativity and intelligence. Three decades after Asimov published the story, scientists are a little closer to realizing that desire, though they've found a better technique than miniaturization.
In a new study in the journal Nature Neuroscience, researchers report they're able to ID people by looking at patterns of activity inside the living brain. One of the study authors, Yale University neuroscientist Emily Finn, said the study shows how "individuality is reflected in brain function."
It's a small, but important, step toward understanding how our brains make us unique. The study has implications not just for basic neuroscience, but also for personalized mental health care, cognitive enhancement and neuroethics: how society and legal systems might treat people according to their 'brain fingerprints,' though that is much further into the future.
To look inside active brains, the scientists used a technique that measures changes in blood flow, called functional magnetic resonance imaging, or fMRI for short. If you're using your auditory cortex to process sound, for instance, blood heads to that part of your brain, and it would "light up" in an fMRI image.
Because fMRI is expensive and time-consuming, most fMRI studies only test a handful of people—20 or so. But the researchers needed a lot more data than that so they borrowed some from the Human Connectome Project (HCP), a global research collaboration to map the networks of neurons inside the human brain, which has banked tons of brain scans.
The researchers used data from 126 healthy volunteers between the ages of 22 and 35 in the HCP database who had their brains scanned over two consecutive days while they were just lying still or completing tasks related to language, motor skills, emotion and working memory, which is transient or very short-term memory, inside the fMRI scanner. Finn and her team used these scans to see what patterns of activity emerged inside individuals' brains during each session and, more importantly, to see if these patterns were different enough from person to person to use them as a brain-based identity check.
So, let's say Sally was one of the participants. The researchers would take Sally's whole brain scans from day 1 and "ask" an algorithm trained to read them to find Sally's day 2 images. They did this with all the participants. The algorithm was correct more than 90% of the time when it compared the "lying still," or rest, sessions and 87% of the time when it compared scans from sessions where Sally had been asked to perform one of the four tasks.
"Brain connectivity patterns are pretty dynamic, but they don't change so much that they're obscuring the identity of the person," Finn said. In other words, "the same brain doing two different things looks more similar than two different brains doing the same thing."
The researchers also wanted to test whether certain connections in the brain were providing the algorithm with bigger clues as to a person's identity. They found that the medial frontal and frontoparietal network, which are involved in cognition, navigation and movement, were especially important. When they had the algorithm use data from just from these networks to make its prediction, it was 100% correct under some conditions.
Other recent studies have reported similar findings, but this one is potentially more far-reaching because "nobody's been able to do this on this scale," said Damien Fair, a neuroscientist at Oregon Health & Science University, who does similar research but wasn't involved in this study. That's what makes it so important, especially as scientists and clinicians aim to understand why some individuals respond to some treatments and others don't. The research is part of the precision—or personalized—medicine movement.
Finn and Fair envision the technology described in the paper evolving into a type of diagnostic brain test. The idea is that in the future clinicians would be able to take a new patient's brain data, assess their risk for schizophrenia (or another disorder) and give them the treatment they need to stave off the disease.
But for that to happen, we need a lot more data. This study only looked at brains over two days. While brain patterns were stable enough over that time to allow scientists to ID one person from another, they don't think it would be enough to ID someone's brain over a lifetime. "We would expect the connectivity [of the brain] to change over the course of someone's life significantly," Finn told me, in response to traumas or the development of neurological conditions like Parkinson's, Alzheimer's or schizophrenia.
Finn's next project is looking at how the brain's connectivity patterns change over time in people with a high risk for developing schizophrenia. They have data for several hundred people who were scanned during adolescence or early adulthood. Some of them went on to develop the disease years later and some didn't. Using that collection of scans, they want to see whether they can teach an algorithm to predict who will develop schizophrenia and how severely.
Likewise, if we could understand what connections in the brain give us our ability to excel at things like math, science or art, then maybe we could devise "brain training" programs to make us better at those things. But that's "all sci-fi" for now, Finn said.
The study didn't directly test what exactly makes us unique, just that our brains behave differently enough while we're using them. But it's possible that they're related to how our own individual experiences and environment shape our brain during our lives. The brain, after all, is a very plastic organ.
The scientists want their research to be used to help people lead healthier, more productive lives. But, they admit, it does raise some ethical questions. How will we treat people whose brains predispose them to disease? Will we need to implement legal protections, like we did for genetics, that prohibit insurance companies from using genetic information to deny people coverage? How much data will be collected about our brains and who will have access to it?
These are all good questions to keep in mind, but before you start freaking out, remember this: the technology necessary for passive collection of your brain activity doesn't exist. fMRI machines like the ones used for this study are massive scanners that require a pretty large, powerful magnet to work. Only a handful research institutions have access to these things. They're by no means portable.
There are some tools that allow scientists to track brain activity more cheaply, outside huge scanners, like portable electroencephalogram, which measure the brain's electrical activity. But these have much, much lower resolution, so it's not clear that they would ever provide enough information to be able to tell one person from the next. They're also not passive. They require you to wear a funny skull cap-looking contraption.
If that doesn't assuage your discomfort, maybe this will. The brain is a super complex organ, and because of that, neuroscientists still know very little about how patterns of activity manifest as behaviors or brain-related disorders.
Today, "we can see a specific gene and know with high certainty whether someone's going to have a disorder or not. We're nowhere close to that for the brain," said Fair, the neuroscientist from Oregon Health & Science University. "For some people [the study] would seem scary, but we're ways off from anything like fingerprinting for the brain."
Daniela Hernandez is a senior writer at Fusion. She likes science, robots, pugs, and coffee.