The Midnight Scan Club sheds new light on the human brain

Science Friday
Midnight Scan Club

Back in 2013, Dr. Nico Dosenbach had a problem. He wanted to study the brain activity of individual people, but as a junior-level researcher, he couldn’t afford the daytime hourly rate to use the scanning machine he needed. That’s when the Midnight Scan Club was born — a group of curious researchers who were willing to scan their own brains late at night, starting with Dr. Dosenbach himself.

Dosenbach, an assistant professor of neurology at Washington University School of Medicine in St. Louis, says he has been obsessed for quite a long time with studying individual brain activity.

“Most functional MRI research has been done on group average data,” he explains. “Usually, you would get 20, 50, 100 or 1,000 people. You take those brain scans and average all the images together. Then you can say something with clarity about a group of people — typical people or people with a certain psychiatric disease.”

“We thought that wasn’t an image of real brains as they are in the world,” he continues. “There’s no such thing as the ‘average brain.’ There’s no average of your brain and my brain. There’s my brain, and there’s your brain. We wanted to study the real thing because we thought it would look quite different, and our study definitely shows that it does. An individual map looks nothing like a group map.”

But with only about $10,000 to work with, there wasn’t much chance of getting time on the MRI machines during the day to collect enough data or pay test subjects. Dosenbach’s eureka moment came when his colleague Steve Nelson told him that, after midnight, scan charges were 90 percent off. “I slapped him on his back really hard, and I said, ‘That’s it. That’s how we’re going to do it,’” Dosenbach relates. “He probably thought I was just kidding at the time, but that’s how we all got started.”

At first, Dosenbach had trouble finding recruits to help out with the experiment. Two people are needed for each scan: a participant and someone to work the scanner. What’s more, the amount of scanning that he and his colleagues wanted to do “sounded pretty gruesome,” Dosenbach says.

“The two main problems are people moving too much — and I’m talking about if you move more than 0.2 millimeters it can hurt the study,” Dosenbach says. “So, no moving at all. And falling asleep. That’s the most difficult one, because one of the scans we did was what’s called the resting state scan. You can think of it as meditating with your eyes open. You lie in complete darkness. You’ve got a little white dot on the screen that you are allowed to look at, and you’re not allowed to move and you’re not allowed to fall asleep. We started each scan off with 30 minutes of that.”

Dosenbach decided the best way to find out if this was even possible was to try it himself. At first, he was apprehensive about it, he says, but once he tried it, he found it wasn’t that bad. He then recruited a couple pairs of graduate students and junior faculty who were willing to volunteer their time in the middle of the night.

“By the time we’d done two data sets — I was the first one, Steve Nelson was the second one — a lot of curious grad students at Wash U had heard about it, and we essentially started getting requests — like, ‘Hey, can I be in the study? I want to know what my brain looks like.’ So, to get to 10 people willing to do this wasn’t all that hard,” Dosenbach says.

The group’s first data set consisted of 12 two-hour scans, or about 24 hours per person, enough data to reach a couple of interesting conclusions about brain activity in individuals, Dosenbach says.

“The thing we suspected, and also found, was that [every] person’s brain was different, even though we’re all night-owl neuroscientists in this cohort, who like to stay up late and volunteered for this,” he explains. “[This] makes sense, because even identical twins have different personalities. There isn’t another Nico Dosenbach in this world. But it was really fascinating to actually see that.”

He and Steve Nelson, for example, are the same age, have a similar job and a similar interest in neuroscience, and their brains look nothing alike, Dosenbach points out. But the biggest surprise was the possible discovery of what Dosenbach calls “common variants in global brain organization.”

“For the basic backbone of the brain, eight of the people looked kind of like the group data and similar to each other, [in terms of] things like attention, executive control, vision, movement,” he explains. “They were all arranged in sort of a big circle, so that information could flow quickly [through] the giant network — the brain is essentially a giant network.”

In two people, however, this ring was broken. Not only could they see it, they could even prove it mathematically, Dosenbach says. “There’s a mathematical property called global efficiency, and for those two people it was lower,” he explains. “I’m one of those two people. Obviously, I don’t think it means that my brain is slow, but it’s fascinating to see that there’s that much variation across people who are otherwise pretty homogeneous. And the only way to see it is with high-density MRI scans.”

Considering that there are hundreds of MRI machines sitting idle after midnight at research institutes across the country, Dosenbach wonders about the possibility of crowdsourcing the study of individual brain scans to increase the amount of available data. So, if you are a medical researcher interested in joining the Midnight Scan Club via your university’s scanner, Dosenbach can be contacted by email at or on Twitter @club_scan.

This article is based on an interview that aired on PRI’s Science Friday with Ira Flatow.

Sign up for our daily newsletter

Sign up for The Top of the World, delivered to your inbox every weekday morning.