Studies Skewed By Focus On Well-Off, Educated Brains
Brain imaging studies have a diversity problem.
That's what researchers concluded after they re-analyzed data from a large study that used MRI to measure brain development in children from 3 to 18.
Like most brain imaging studies of children, this one included a disproportionate number of kids who have highly educated parents with relatively high household incomes, the team reported Thursday in the journal Nature Communications.
For example, parents of study participants were three times more likely than typical U.S. parents to hold an advanced degree. And participants' family incomes were much more likely to exceed $100,000 a year.
So the researchers decided to see whether the results would be different if the sample represented the U.S. population, says Kaja LeWinn, an assistant professor at the University of California, San Francisco School of Medicine. "We were able to weight that data so it looked more like the U.S." in terms of race, income, education and other variables, she says.
And when the researchers did that, the picture of "normal" brain development changed dramatically.
For instance, when the sample reflected the U.S. population, children's brains reached several development milestones much earlier.
One of these milestones involved the total surface of the brain's cortex, which plays a key role in in memory and thought. The unweighted data showed that this surface area continued to increase until after a child's 12th birthday. The weighted data showed a much earlier peak — before age 10.
Unweighted data also showed several areas in the front and back of the brain developing at the same time. But the weighted data showed a different pattern. Areas toward the back of the brain, which do things like process visual information, developed first. Meanwhile areas toward the front of the brain, which are involved in thinking and judgment, developed later.
This study doesn't look at what those differences might mean for children's emotional and intellectual development. The key point is rather that researchers should make sure that they're looking at a representative sample when they're defining "normal."
The idea that the brain tends to develop earlier toward the back and later toward the front is "more consistent with our broader understanding of brain development," LeWinn says. And it is one reason many brain scientists argue that judgment and impulse control are not fully developed until people reach their 20s.
The study is a reminder that the brains of children from different backgrounds can develop differently, LeWinn says. "The brain is really responsive to experience," she says. "That's something we need to pay attention to."
The results also offer a reminder that brain imaging studies tend to attract an atypical group of people. Participants are likely to live near a major university, where the studies are usually conducted, LeWinn says. They are also "more likely to be white, more likely to be high income, more likely to have more education, and they may have different social networks," she says.
One reason for these differences is that university-based studies tend to attract people who have ties to the school. "Many of the studies we use to understand how the brain works included mostly college students," she says.
It's unrealistic to expect that every brain imaging study sample represents the full range of U.S. residents, Le Winn says. But even small studies should do a better job disclosing the characteristics of people being studied. And larger studies should consider weighting the results to more accurately represent the nation's population, she says.
At least one big study is already trying to address the diversity issue.
The federally funded Adolescent Brain Cognitive Development (ABCD) study is in the process of enrolling 10,000 children ages 9 and 10. Researchers plan to track participants until they become adults using brain imaging as well as information about sleep, attention, substance use, physical activity and sports injuries.
And to make sure the participants reflect the U.S. population, researchers are taking steps to enroll children of diverse races and ethnicities, education and income levels, and living environments.
The team is doing that by recruiting students from carefully selected schools from 21 sites across the country. "Our ultimate goal is to recruit a sample that matches the U.S. population," says Emily Giron, the ABCD project's communications manager.