Changes in how brain regions “talk” to each other from childhood to adulthood follow and support a fundamental organizational pattern of the brain
Audrey Luo was the lead author on this study. Audrey is a 7th year MD/PhD student fascinated by brain development, which happens to be the topic of her graduate work with Ted Satterthwaite. Having recently defended her PhD, Audrey is heading back to the wards to finish her medical degree. She plans to pursue psychiatry and hopes to combine her research with clinical practice to better understand the human brain and treat mental illness.
or technically,
Functional connectivity development along the sensorimotor-association axis enhances the cortical hierarchy
[See Original Abstract on Pubmed]
Authors of the study: Audrey Luo, Valerie Sydnor, Adam Pines, Bart Larsen, Aaron Alexander-Bloch, Matthew Cieslak, Sydney Covitz, Andrew Chen, Nathalia Bianchini Esper, Eric Feczko, Alexandre Franco, Raquel Gur, Ruben Gur, Audrey Houghton, Fengling Hu, Arielle Keller, Gregory Kiar, Kahini Mehta, Giovanni Salum, Tinashe Tapera, Ting Xu, Chenying Zhao, Taylor Salo, Damien Fair, Russel Shinohara, Michael Milham, Ted Satterthwaite
The human brain supports a wide range of functions. Different regions of the outermost layer of the brain, the cortical surface, support two categories of functions—sensorimotor regions are involved in sensation and movement and association regions are involved in functions like self-reflection and emotion.
However, these different cortical regions can be distinguished from each other further beyond these two broad categories. We can use a framework known as the sensorimotor-association (S-A) axis. Instead of just categorizing each region as either a sensorimotor or association region, the S-A axis ranks the different brain regions from a sensorimotor end to an association end. Think of this as the difference between simply categorizing people as either “Gen Z” or “Millenial” versus ranking them from youngest to oldest— the ranking method is much more useful for meaningfully distinguishing between people!
Previous studies have shown that the S-A axis summarizes how many characteristics, including the brain’s physical structure, blood flow, and cell type composition, differ across the cortical surface. Because of the prevalence of this cortical surface pattern across so many biological brain measures, neuroscientists believe that the S-A axis is essential to the human brain’s ability to carry out its wide variety of functions.
In this paper, NGG student Audrey Luo and her colleagues in the Satterthwaite lab were interested in how different parts of the brain communicate with each other during the transition from childhood to adulthood. Imagine each brain region has a nonstop video call link that any other region can join at any time. When two regions’ activities are synchronized, it’s as if they have connected on video, communicating while their “screens” are in sync—neuroscientists call that functional connectivity (FC). Audrey wanted to know how the change in functional connections between different cortical regions during development related to the S-A axis. To do this, Audrey used data from participants ages 5 to 23 scanned with functional magnetic resonance imaging (fMRI). fMRI is an imaging research tool that can tell us when different areas of the brain are getting oxygen-rich blood, which roughly reflects which parts of the brain are activated.
Because Audrey included 200 cortical regions in her analysis, there could be an astonishing 19,900 possible FC pairs, or 19,900 “video calls”. To summarize across all these connections, Audrey gave each region a single score called FC strength—the average of a region’s connectivity to all the rest of the cortical regions. A high score means the region’s activity tends to stay in sync with many other regions (e.g., long calls with many partners, shared screen, very social); a low score means its activity tends to differ from other regions (e.g., short calls with few partners, distinct screen, less social). With this one-number summary of FC strength, Audrey tracked how each region’s connectivity to the rest of the brain changed from childhood to adulthood to ask the question: how is FC strength development across the cortical surface related to the S-A axis?
What Audrey found was that as children grew older, sensorimotor regions showed increased connectivity and association regions showed decreased connectivity. More specifically, Audrey showed that the pattern of change in FC strength of each brain region was directly related to its S-A axis rank: the closer the brain region was to the “sensorimotor end” of the axis, the more likely it was to increase in connectivity (i.e., more social), and the closer the brain region was to the “association end” of the axis, the more likely it was to decrease in connectivity (i.e., less social). In other words, functional connections don’t develop uniformly across the cortical surface. Instead, they follow a specific pattern that matches the fundamental organizational framework of the S-A axis.
To further understand how functional connectivity during development related to the S-A axis, Audrey asked whether the change in FC between a pair of regions depends on their identities as either sensorimotor or association regions. Bringing back the video call analogy, we can think of sensorimotor and association regions as two friend groups. Essentially, Audrey is asking—how does video call frequency between and within the two friend groups change over development? What she found was that as children grew into adults, connectivity between sensorimotor regions increased, connectivity between association regions decreased, and connectivity between association and sensorimotor regions also decreased. This paints a more specific picture of how functional connectivity develops: sensorimotor regions become more social (but just with each other), while association areas become more independent and specialized.
Overall, this work suggests that the S-A axis is not only a framework of how the brain is organized in adulthood, but also how the brain develops from childhood onward. Audrey argues that this specific pattern of changes in how sensorimotor and association areas communicate ultimately helps establish the S-A axis as the dominant cortical framework of the brain, enabling the vast diversity of functions observed in adulthood.
Perhaps the most impressive aspect of the paper is it reports the same results from not just one dataset, but four large datasets that were collected from different sites and in different ways. Indeed, some of Audrey’s findings are replications of results found in a paper by a graduated NGG student, Adam Pines: https://www.upennglia.com/briefs/bib-personalized-brain-networks— an impressive study conducted in one large dataset. Now, with Audrey’s rigorous work in four large datasets, we can have even more confidence as a scientific community that these findings on the S-A axis are not just dependent on the specific circumstances of a particular sample, but instead, capture something real about human biology. This is a concept known as generalizability—the extent to which findings can be applied to the entire population beyond a specific sample. This study is an excellent example of generalizable science and reveals a rigorous framework of functional brain development. Hats off to Audrey and the rest of the team!
About the brief writer: Kevin Sun
Kevin is co-mentored by Drs. Aaron Alexander-Bloch and Ted Satterthwaite, working with functional neuroimaging data to ask questions about development, genetics, and transdiagnostic psychopathology risk. Outside of research, Kevin enjoys film, weightlifting, art museums, and speculative fiction.