Little kids, big insights: What childhood can teach us about how the brain supports cognition

Ursula Tooley was the lead author on this study. Ursula is a postdoctoral research scholar at Washington University in St. Louis. Her research examines functional brain network development in neonates and toddlers, with a focus on the pace of brain maturation and how neuroplasticity changes across development. She received her Ph.D. in Neuroscience in 2022 from the University of Pennsylvania, under the direction of Dr. Allyson Mackey and Dr. Dani Bassett, where she studied functional brain network development in children and adolescents. She received her B.S. in Neuroscience from the University of Arizona, where she conducted research on sleep disruption in children with Down syndrome.

or technically,

The age of reason: Functional brain network development during childhood

[See Original Abstract on Pubmed]

Authors of the study: Ursula A. Tooley, Anne T. Park, Julia A. Leonard, Austin L. Boroshok, Cassidy L. McDermott, Matthew A. Tisdall, Dani S. Bassett, and Allyson P. Mackey.

Early and middle childhood (4-10 years old) are full of developmental milestones. How children speak, move, learn, and play is constantly evolving and improving. Kids build social networks, become better able to control their attention, and begin to develop cognitive skills, like reasoning. However, despite the rapid cognitive development happening during this early childhood period, neuroscientists have very little information about how brain function is changing. 

This is because getting a clear picture of brain activity, like getting a clear picture of anything, requires that the subject stays almost perfectly still. If you’ve ever watched a 4-year-old sit at the dinner table, it comes as no surprise that they don’t make the best neuroimaging subjects. Functional magnetic resonance imaging (fMRI) scanners, which take many consecutive snapshots of brain activity, are even more sensitive to motion than cameras. The tiniest movements, even just a few millimeters, can blur the images and make it impossible for neuroscientists to tell what brain activity belongs to which brain region. So, most neuroimaging work to date uses subjects over 7 years old, which means that while researchers work to understand how the brain develops to support cognition, they’re missing many of the first pieces of the puzzle. 

Here’s where recent Neuroscience Graduate Group alumn Ursula Tooley and collaborators from the Robust Methods for Magnetic Resonance group stepped in. The team engineered a way to monitor and correct for head motion inside the scanner, allowing them to collect high-quality neuroimaging data from wiggly subjects during this critical early childhood period. Specifically, this motion-tracking technology gave the researchers a way to record exactly how much and in which direction kids were moving at any given point during the scan. Ursula could then use this information to correct (think: realign) the images of brain activity or exclude the child from the study if they moved too much. The ability to precisely monitor head position in real time also created an opportunity for kids to practice the correct behavior. Before the scanning session, children came to the lab to watch a movie while laying in a mock scanner that made the same whirring noises and beeps as the real deal. Each time they moved their head more than 1 millimeter, the movie paused. Incorporating this period of exploring the scanner and the scanning expectations meant that most of the kids who enrolled in the study stayed still enough for usable images of brain activity to be collected. This is a huge feat for Ursula and the team as well as a huge win for neuroscience, making it possible to take an earlier look at the developing brain.

Over the course of the study, Ursula and her colleagues scanned a diverse group of 92 children ages 4-10 from the Philadelphia community. Each child completed an fMRI scan as well as a series of cognitive tests (which they did outside of the scanner) designed to measure the strength of their cognitive reasoning abilities. What is cognitive reasoning? Reasoning is an umbrella term describing the ability to process information, problem solve, and make predictions based on pattern recognition (Fig. 1). Successful cognitive reasoning involves much of the brain and improves dramatically during early and middle childhood. Research suggests that how kids perform on cognitive reasoning tasks is predictive of their academic achievement — even years down the road! By combining a child's cognitive reasoning ability with information about their brain activity, Ursula was able to ask whether and how changes in brain function might support this shift in cognitive performance.

Figure 1

An example question from the cognitive reasoning test, which was administered at different difficulty levels to children in the study depending on their age. Here, we see the red rectangle switches from the background (left) to the foreground (right). To answer the question correctly, the child has to understand this spatial relationship for the rectangles and extend it to the pentagons.

Ursula used resting-state fMRI data (data collected while the kids laid “at rest” in the scanner) to explore the brain’s functional organization. In other words, she inferred how much different brain regions talk to each other based on how their activity fluctuates together over time. As such, regions with activity that rises and falls together are likely functionally connected. These groups of connected brain regions are called “systems.” The brain has many of these functionally-connected systems, and neuroscience research shows that they can rewire and reconfigure themselves. For example, another neuroimaging study of older kids and young adults (ages 8-22) from Philadelphia showed that the organization of these brain systems changes with age [1]. Specifically, our brain systems become more segregated and more modular as we move towards adulthood, with weaker connections between systems and stronger connections within systems (Fig. 2). Ursula found the same trends in her data with older kids tending to have more segregated brain systems than younger kids, suggesting that our brain’s functional architecture is flexible and continues to refine as we age.

Figure 2

As we age and develop, our brain systems (red, green, and blue ovals) reorganize, moving from more integrated (e.g., many connections between systems) to more modular (e.g., more connections within systems and fewer connections between systems). Ursula’s work shows that this brain system separation supports the development of cognitive skills, like reasoning.

Do some systems remodel more than others? Ursula found that changes in connectivity were largest in brain systems involved in abstract cognition, visual processing, and attention. As it turns out, these are the same systems involved in cognitive reasoning. For instance, reasoning is supported by the brain’s visual areas taking in information from the world while attention systems focus the brain’s resources on what’s important to the task at hand while ignoring distractors. Given what we know about the blossoming of cognitive reasoning during childhood, Ursula wondered if there could be a relationship between these changes in brain connectivity and cognitive ability. To test this, Ursula compared the patterns of brain system connectivity for each child with their scores on the cognitive reasoning test (Fig. 1). She found that the remodeling of cognition, visual processing, and attention systems was associated with increased cognitive ability! In other words, kids who had more mature patterns of brain system connectivity were better equipped to reason about the world and their place in it.

Taken together, Ursula’s work suggests that the massive restructuring of brain systems as kids age might be happening to support the rapid development of cognitive abilities emerging during these early and middle childhood years. Beyond offering a new perspective on healthy brain development, this relationship between brain organization and brain function offers new ways to think about -- and potentially treat -- various neurodevelopmental or neurological disorders.

About the brief writer: Kara McGaughey

Kara is a PhD candidate in Josh Gold’s lab studying how we make decisions in the face of uncertainty and instability. Combining electrophysiology and computational modeling, she’s investigating the neural mechanisms that may underlie this adaptive behavior.

Citations:

1. Baum, G.L., Ciric R., Roalf, D.R., Betzel, R.F., Moore, T.M., Shinohara, R.T., … & Satterthwaite, T.D. (2017). Modular segregation of structural brain networks supports the development of executive function in youth. Current Biology, 27(11). doi: 10.1016/j.cub.2017.04.051.

Want to learn more about how brain function supports the development of cognitive reasoning during childhood? You can find Ursula’s full paper here!

How different levels of brain development help adolescent cognition - or don’t

Adam Pines was the lead author on this study. Adam is a postdoctoral fellow in the Stanford PanLab for Precision Psychiatry and Translational Neuroscience. He completed his Ph.D. in Neuroscience at UPenn in 2022. His other research interests include developmental neuroscience, brain-environment interactions, and adaptive plasticity in the brain.

or technically,

Dissociable multi-scale patterns of development in personalized brain networks

[See Original Abstract on Pubmed]

Authors of the study: Adam R. Pines, Bart Larsen, Zaixu Cui, Valerie J. Sydnor, Maxwell A. Bertolero, Azeez Adebimpe, Aaron F. Alexander-Bloch, Christos Davatzikos, Damien A. Fair, Ruben C. Gur, Raquel E. Gur, Hongming Li, Michael P. Milham, Tyler M. Moore, Kristin Murtha, Linden Parkes, Sharon L. Thompson-Schill, Sheila Shanmugan, Russell T. Shinohara, Sarah M. Weinstein, Danielle S. Bassett, Yong Fan & Theodore D. Satterthwaite

You don’t need to be a scientist to know that kids get smarter as they grow up - they get better at things like problem-solving, thinking flexibly, and remembering information. But what exactly is changing in the brain to make these cognitive skills, which researchers call “executive function” easier?

Like instruments in a band, different areas of the human brain have different roles and will perform together in different combinations to everything from processing what your eyes see, to controlling your muscles, to solving a crossword, to feeling emotions. A group of brain regions that work together is called a functional brain network. Some functional brain networks perform easier, or “lower-order”  tasks, like sensing pain when you get a cut. Others perform harder, more complex tasks, like solving physics equations or learning a language, which are considered “higher-order”. 

Dr. Adam Pines, who recently graduated from the Neuroscience Graduate Group, wanted to know how all these functional networks mature as kids age and how this pattern of development relates to kids’ improving executive function. To study this, Adam had two challenges. First, we don’t know how many functional networks there “really” are in the brain; you can divide the brain up into different numbers of chunks and still do a good job of grouping regions that activate together and separating those that don’t (Figure 1, Columns). Second, the layout of everyone’s functional networks is a tiny bit different: one network may take up a little more space in one person, for instance, or the parts of the brain that do a certain task on one person may be just a little bit more to the left on another (Figure 1, Rows). Therefore, Adam made personalized functional networks (PFNs), which are maps of a person’s unique functional network layout, for every subject in the study. He also tried grouping the brain into different numbers of networks to see whether this would change his results.

Figure 1: Illustration of personalized functional networks mapped for varying numbers of networks.

Adam mapped the unique functional networks of each person in the study (PFNs), as shown in the rows. He also divided the brain’s activity into different numbers of networks, with maps of 4, 7, and 13 networks pictured. Different colors show that the brain regions are part of different functional networks.

To make personalized functional networks (PFNs) for each subject (Figure 1, Rows), Adam and his colleagues mapped the layout of every functional network in the average person and mathematically tweaked the layout to fit each participant’s unique pattern of brain activation. Then, they repeated this step using different numbers of networks in their baseline map (Figure 1, Columns) and labeled whether each network did lower- or higher-order functions. In the end, they had 29 brain maps for each person (each dividing brain activity into 2 to 30 functional networks), that they could compare to each participant’s age and score on a test of executive function.

First, Adam compared PFNs across participants ages 8 through 23 and found that lower- and higher-order networks tended to develop differently. Lower-order networks (each of which does an easier task) became more interconnected over the course of adolescence, while higher-order networks (each of which does a harder task) became less interconnected. Next, he tested how these PFN patterns were related to kids’ executive function. Interestingly, he found executive function tends to be better when very low-order and very high-order networks are distinct, but networks that fall in the middle (ones that do medium-complexity tasks) are more interconnected. Dividing the brain into a greater number of PFNs, Adam saw this effect grow stronger, especially in lower-order networks.

Taken together, Adam’s results are surprising because, while aging makes higher-order networks more distinct (which is better for executive function), lower-order networks actually become more interconnected (which is worse for executive function)! This may mean that while increasingly distinct higher-order networks allow kids’ executive function to improve as they grow up, their brains’ lower-order networks are already starting to decline. These findings will be important for future scientists studying how kids’ executive function develops and may help uncover why some kids struggle with cognitive development.

About the brief writer: Margaret Gardner

Margaret is a PhD student in the Brain-Gene-Development Lab working with Dr. Aaron Alexander-Bloch. She is interested in studying how different biological and demographic factors influence people’s brain development and their risk for mental illnesses.

Want to read Adam’s work for yourself? You can find the full article (complete with equations and pretty brain pictures) here!