Neuroimaging and your child’s brain: A guide for the science-minded


© 2010 Gwen Dewar, Ph.D., all rights reserved

They’re the technologies that give us “snapshots” of the brain at work:

Functional neuroimaging shows us what parts of the brain “light up,” or become active, when we are thinking, learning, or feeling. And this permits researchers to map the brain—pinpointing where specific kinds of information get processed.

Reading music? A tiny spot at the back of your head–the right superior parietal cortex–becomes active (Schön et al 2002).

Feeling guilty? Several brain regions light up, including the medial prefrontal cortex (MPFC), an area associated with social emotions and the ability to consider another person’s perspective (Burnett at al 2009).

This sort of information can lead us to a new understanding of many things, including these:

Empathy in children


“I feel your pain” might be more than a figure of speech.

Neuroimaging studies suggest that the regions of the brain involved in processing first-hand experiences of pain are also activated when we observe pain in others. And empathy isn’t just an adult phenomenon. Neuroimaging studies demonstrate that children’s brains work the same way.

What’s going on in the minds of babies?

Psychologists have invented many ingenious ways to detect what pre-verbal infants are thinking. Neuroscience offers yet another approach: Watch brain activity to see if babies can distinguish between different kinds of stimuli.

Using neuroimaging, researchers found that 3-month old baby brains could tell the difference between 4 objects and 8 objects! Were the babies consciously thinking about quantity? We can’t be sure. But the results suggest that–at some level–baby brains possess “number sense.”

What’s different about the brains of kids with impairments and behavioral problems?


As neuroscientists build up a picture of normal brain function, we get a better idea of how the brains of kids with various problems (like dyslexia, ADHD, or autism) may differ from the brains of normally-developing kids.

And understanding these differences may suggest what sorts of remedial instruction would most benefit kids.

For example, some kids suffer from dyscalculia–a learning disability in mathematics. What’s going wrong in the brain? You might assume there is something wrong with the part of the brain that counts and performs precise calculations.

But brain imaging studies suggest something else.

When kids with dyscalculia were presented with a series of items to count, their brain activity was indistinguishable from that of normal kids.

The real brain difference showed up when kids had to use their “number sense,” an intuitive feeling for the relative magnitude of numbers.

When kids with dyscalculia had to make less precise, more qualitative judgments–estimating quantity without actually counting–their brains responded differently (Kucian et al 2006).

Perhaps, then, kids with dyscalculia would benefit from exercises that help them develop a better number sense. Future behavioral studies can test this hypothesis.

But there is a dark side…


People are terribly impressed by brain research, so much so that we sometimes suspend our sense of skepticism when we hear about new studies.

In some popular accounts, brain research is equated with a sort of mind-reading.

Neuroimaging is regarded as a direct window into the mind—revealing if an individual is smart or dull, happy or depressed, compassionate or self-absorbed.

So it seems to follow that brain imaging studies can replace other approaches to understanding the mind. Why talk to your subjects, if you can read their minds? Why give people tasks to solve, if brain imaging can tells us how clever they are?

It may also seem that brain studies are especially “true” or authoritative. If the mind is the product of the brain, then isn’t neurological research more important and relevant than psychological research? Isn’t neurology more scientific?

In fact, all of these notions are wrong.

Neuroimaging isn’t mind-reading

Neuroimaging studies tell us about the average brain. They may tell us that a particular pattern of activity is characteristic of most people suffering from depression. Or that another pattern is found in most kids with dyslexia.

But the patterns don’t sort perfectly. If you tried to predict what real people were like on the basis of brain activity alone, you’d make a lot of mistakes.

For example, when researchers tried using neuroimaging to detect deception, 33% of people telling the truth were misidentified as liars (Kozel et al 2010).

Measuring brain activity is no more “scientific” than measuring behavior


Neuroimaging is an exciting tool for understanding brain mechanisms. It enriches our understanding of the way brains process information.

But it’s wrong to think that brain research is going to trump, or replace, more traditional approaches to understanding behavior.

And people are mistaken if they think that brain research is somehow more rigorous, scientific, or likely to yield up “the truth.”

The reason? Knowing how the brain looks doesn’t tell us how people think or behave. Not by itself.

To interpret brain images of brain activity, we need to know what people are doing while the brain activity occurs. We need to measure their actual performance in the real world.

And neuroimaging can’t tell us which interventions make kids think, feel, or behave better.

Do naps help kids better memorize information? You could try comparing brain activity in well-rested and sleep-deprived kids.

But knowing patterns of brain activity won’t tell you how much a child remembers. To find that out, you have to measure behavior. Ask kids questions. Give them tasks to perform.

So brain research doesn’t render other fields–like cognitive psychology–obsolete. On the contrary. We need behavioral data to make sense of the brain data.

And science? What makes a study rigorous and authoritative—a reliable source of information—isn’t the tools you use. What makes a study “good science” is the way it’s designed and executed.

The right way to study the mind

Neuroimaging doesn’t make a study scientific. It’s merely a tool—like measuring blood pressure—that scientists can use.

And when it comes to answering practical questions—like identifying the best educational programs or parenting tactics—neuroimaging isn’t especially helpful. At least not now.

To see what I mean, let’s do a thought experiment.

Suppose you wanted to know if mild thirst–low levels of dehydration–makes kids perform worse on academic tasks.


Do thirsty kids have more trouble solving puzzles?

The most straightforward approach is to test the cognitive performance of mildly dehydrated kids.

You might begin by getting yourself a group of kids and making them sweat in the sauna. Then you’d randomly assign each child to one of two treatments.

• In the WATER-SUFFICIENT condition, kids drink enough water to replace what they’ve lost through perspiration.

• In the DEHYDRATED condition kids are given a little water, but not enough to replace what they’ve lost.

Next, you’d ask the kids to work on some puzzles. You’d make sure that the kids didn’t know what your hypothesis was. You’d also make sure that the people administering the tests didn’t know which treatment each kid received.

When you were done, you would compare the two groups. If the dehydrated kids performed more poorly on the puzzles, you’d do a statistical analysis. What is the chance that the experiment would turn out this way due to pure chance?

If the statistics suggest that the probability of a fluke result is small, you have a rather persuasive study.

This approach–a randomized, controlled experiment–is the gold standard of scientific investigation. Did we use neuroimaging? No. But we could.

Experiment replicated…with brains


Let’s suppose you perform the experiment again, this time taking the additional step of wiring kids up to brain-imaging equipment.

What parts of the brain are activated when puzzle-solvers are thirsty? Does brain imaging show any differences between thirsty and non-thirsty kids?

Assume that it does: Brain imaging reveals differences in brain activity between thirsty and non-thirsty problem solvers.

What have we learned from this?

Neuroimaging didn’t tell us that thirsty kids are worse at solving puzzles. It was the behavioral part of the experiment—giving kids real puzzles to solve—that showed that.

The brain imaging showed us something else: What brains look like when they are trying to solve puzzles under conditions of thirst.

That’s interesting, but it isn’t necessarily of practical importance. So if you want to know how to help kids learn–or improve other aspects of their lives–you probably don’t need neuroimaging.

And that’s why most useful information available today about how kids learn–including information marketed as “brain-based learning”– comes from studies of cognition and behavior, not brains.

So why bother with neuroimaging in studies of this kind?

Added value: What does neuroimaging tell us about behavior that other tools cannot?

My example makes it sound like brain imaging is totally superfluous if you want to predict how people actually behave. And in many cases, that’s true.

But as I noted at the beginning of this article, neuroimaging may indeed contribute insights of practical importance. Understanding brain activity may lead to new, testable hypotheses about learning.

Neuroimaging suggests that kids with dyscalculia might be especially benefited by training that helps them develop an intuitive sense of number. Investigators can now test this idea by conducting rigorous, controlled, behavioral studies.

And neuroimaging does something else, too. It may give us hints about mental processes that aren’t easy to detect through behavioral measures alone.


The baby study is one example of that. It’s hard to establish what very young babies are thinking. Patterns of brain activity can’t prove what’s in their minds, but it’s a clue.

Brain imaging can also suggest how hard our brains are working. Remember our thirsty kid experiment?

It turns out that several behavioral studies have been conducted along those lines. Most, but not all, confirmed that low levels of dehydration impair cognitive performance.

So Matthew Kempton and his colleagues decided to run another “thirsty puzzle solver” experiment that measured brain activity, too.

The results? Kempton’s team found no difference in puzzle-solving performance between conditions. Kids were just as good at solving problems whether they were thirsty or not.

But the team did find an interesting difference in their brains. When the kids were thirsty, they showed a pattern of brain activity associated with higher levels of oxygen consumption. Their brains seemed to be using more energy to accomplish the same task (Kempton et al 2010).

In other words, the thirsty brains had to work harder to achieve the same levels of performance.

And that’s a useful finding. It tells us that people may pay a cost even if it isn’t evident in their behavior. The kids solved the puzzles, but were less efficient doing so.

Would they have performed as well if they had additional problems to cope with–like the distractions we find in the average classroom? Perhaps not.

So neuroimaging can offer us important clues about mental processes. We just need to be careful about how we interpret these clues.

More information about the brain

Want to read more about recent discoveries in cognitive neuroscience? Brain research pops up in many Parenting Science articles. Here are a few you might want to check out:


Burnett S, Bird G, Moll J, Frith C, and Blakemore SJ. 2009.Development during adolescence of the neural processing of social emotion. J Cogn Neurosci. 21(9):1736-50.

Kempton MJ, Ettinger U, Foster R, Williams SC, Calvert GA, Hampshire A, Zelaya FO, O’Gorman RL, McMorris T, Owen AM, Smith MS. 2010. Dehydration affects brain structure and function in healthy adolescents. Hum Brain Mapp. 2010 Mar 24. [Epub ahead of print]

Kozel FA, Johnson KA, Grenesko EL, Laken SJ, Kose S, Lu X, Pollina D, Ryan A, George MS. 2009. Functional MRI Detection of Deception After Committing a Mock Sabotage Crime. J Forensic Sci. 54(1):220-31.

Kucian K, Loenneker T, Dietrich T, Dosch M, Martin E, von Aster M. 2006. Impaired neural networks for approximate calculation in dyscalculic children: a functional MRI study.Behav Brain Funct. 2:31.

Schön D, Anton JL, Roth M, Besson M. 2002. An fMRI study of music sight-reading. Neuroreport. 3;13(17):2285-9.

Content last modified 7/10