For the first time ever, neuroscientists have completed a comprehensive roadmap of the top-trafficked communication highways in the human brain.
This white-matter map not only charts the geography of these neural highways – it also plots out which of them interact with the most other paths, which are most crucial for supporting key brain functions, and which ones leave the whole brain most vulnerable to long-term damage if they’re disrupted.
“This map will give us more accurate insight into how severe the effects of a particular brain injury are likely to be,” says University of Southern California neuroscientist John Darrell Van Horn, who headed the project. “It’ll also give us more precise tools for predicting a patient’s ability to recover from a brain injury.”
Wiring in our brains falls into two main categories: Gray matter, which does most of the heavy computing; and white matter, the “highways” that enable gray-matter areas to communicate with each other. And just as a single under-construction highway onramp can gridlock traffic throughout a whole section of town, a single damaged white-matter highway can shut down cognitive and emotional processes in many gray-matter hubs.
In fact, whereas a patient who’s suffered a large amount of gray-matter damage may still be able to recover and live a normal life, a patient with even a small white-matter injury may end up with profound and lasting brain deficiencies. “This contrast raises some obvious questions,” Van Horn says. “For example, ‘Which connections in the brain are most sensitive to this kind of damage?’”
To find some answers to these puzzles, Van Horn and his team gathered data from 110 volunteers using a technique known as diffusion tensor imaging (DTI). Although previous studies have used DTI to map the brain’s white-matter pathways – and produce some beautiful rainbow-colored connectomic art in the process – Van Horn and his team had a very specific set of goals in mind this time.
“We wanted to do more than make pretty pictures of tractography,” he explains. “We wanted to use graph-theoretical approaches to describe patterns of connectivity in a large sample of subjects, and try to understand which of those connections are the main ‘freeways’ on which information travels in the brain.”
Graph theory is a mathematical field with a stunningly wide variety of applications. For example, it’s the type of math Facebook uses to suggest people you might know based on the people already in your friends list – and it’s also the kind of math your car’s GPS uses to calculate the shortest route from one place to another.
For this project, though, Van Horn and his team used graph theory to build predictive models of white-matter connectivity, based on the brain scans they’d gathered from their volunteers – then test out those computerized network models in a simulated environment. “We built models of all our subjects’ large-scale brain networks,” Van Horn explains, “and we investigated what happens – throughout an entire brain-wide network – when you strategically remove certain white-matter connections.”
Although these computerized models weren’t anywhere near the level of resolution necessary to simulate the activity of individual neurons, they were plenty high-res to pick out large-scale patterns – just as your GPS software can accurately render gridlocked freeway areas without needing to know the location of every car.
Watching for symptoms
As Van Horn and his team removed white-matter connections from their simulated brain networks, they focused specifically on simulating the kinds of brain damage associated with Alzheimer’s, multiple sclerosis, and other disorders that attack the brain’s major white-matter highways. “We know a lot about the effects of these diseases,” Van Horn says, “but we still don’t have a very detailed understanding of how damage to white matter affects the overall behavior of brain networks.”
What they found, though, made a lot of sense – and it’s only the beginning of this project’s long-term implications. “We noticed that white-matter pathways that link up a lot of brain areas also tend to link up with each other,” Van Horn explains. What this means, he says, is that even a microscopic injury to one of these busy neural highways may lead to much more overarching problems than a larger lesion in an area that’s not as well-connected. By way of example, a three-car pileup on a surface street may delay traffic through one part of town – but a crash on an interstate freeway can disrupt traffic between entire regions of the country.
But this discovery is one just one example of the insights this roadmap appears poised to provide in the near future. Armed with comprehensive brain-wide traffic maps like this one, doctors may soon be able to diagnose brain disorders like Alzheimer’s earlier and more accurately – and make better-informed treatment decisions.
Not to mention, of course, the obvious implications for longer-term brain-mapping efforts like the Human Connectome Project (HCP). “As neuroimaging data from the HCP, President Obama’s BRAIN Initiative, and similar efforts in Europe starts to become more widely available,” Van Horn says, “we’ll be able to use that data to build new and more detailed models of brain function.” And it’s likely that a lot of those studies are going to use the same type of approach that Van Horn’s team did here: Crunch the numbers, build a model, then break it systematically and watch what happens.