“A multi-modal parcellation of human cerebral cortex” (Nature 536 [7615]: 171-+ AUG 11 2016), was recently named a Fast-Breaking Paper for Neuroscience & Behavior in Essential Science Indicators. At present this paper has both Hot and Highly Cited Paper status, and shows 46 citations in the Web of Science.
In this interview, two of the paper’s authors, Matthew Glasser and David Van Essen, talk about their paper and its implications for the field of Neuroscience & Behavior.
Why do you think your paper is highly cited?
The paper provides what we consider to be the best answer yet to a century-old question: What is the map of cortical areas in the human brain? We used hundreds of precisely aligned normal healthy young adult brains from the Human Connectome Project to create the map. Multiple MRI modalities, including those that measure cortical architecture, function, connectivity, and ordered maps of visual space, helped to define the boundaries of areas. We focused on locations where more than one modality provided evidence for a boundary between areas. We painstakingly related these boundaries to what was already known about cortical areas in the literature and described each area in detail. We also trained a machine-learning algorithm to find each cortical area automatically in individual subjects based on its unique fingerprint of architecture, function, connectivity, and topographic maps.
Does it describe a new discovery, methodology, or synthesis of knowledge?
It describes all three. A total of 97 new areas were discovered, and 83 previously reported areas were confirmed. All 180 areas were identified in both the left and right hemispheres, and they were mapped much more precisely than had previously been possible. Many new methods were developed to make this possible. Then the areas needed to be related to the prior literature. As a result, the full paper—including supplemental material—contains 181 pages, including extensive sections on novel methods and synthesis in relation to the neuroanatomical literature.
Would you summarize the significance of your paper in layman’s terms?
To understand how the brain’s central cognitive structure, the cerebral cortex, works in health and disease, it is vital to have an accurate map of its major functional components—just as an accurate map of national boundaries and other political subdivisions is key to understanding the world around us. This paper provides an accurate map of human cerebral cortex, together with a set of methodological improvements that should accelerate progress in understanding how the brain works at a macroscopic scale.
How did you become involved in this research, and how would you describe the particular challenges, setbacks, and successes that you’ve encountered along the way?
David has been involved in mapping primate cortex for four decades, and improving our maps of human cerebral cortex has been a longstanding goal. Matt has been interested in this problem for over a decade and has devoted the better part of his research career to this question too.
Our common interests also stemmed from a general dissatisfaction with how the vast majority of brain imaging studies were relating their results to neuroanatomy. Common practice has been to map new results to the classical anatomical map made in 1909 by Korbinian Brodmann. Often, brain areas that have complex patch-like shapes on the sheet-like cerebral cortex are represented by single points in 3D space—a gross oversimplification. Additionally, brain-imaging studies typically don’t align brain areas accurately across subjects, causing blurring in average results of groups of subjects; many investigators intentionally blur their data even more. The result has been an over-reliance on statistical measures of significance (which recent papers have shown are sometimes fallible) and an under-reliance on basic neuroanatomy—the brain areas that we have mapped—for assessing whether brain-imaging results make sense.
The improved methods used in this paper ensure that brain areas are aligned across subjects and avoid deliberate blurring of the data. The result is much like the gains one gets from a space telescope above the earth’s blurring atmosphere relative to ground-based telescopes: a much clearer picture of brain organization and a much more precise map of the cerebral cortex.
Another major success involves data sharing. The Human Connectome Project was predicated on open sharing of data, tools, and resources, and it has been highly successful in this regard. Thousands of investigators have already tapped into the freely available data from this project, including the extensively analyzed data associated with our cortical parcellation paper.
Where do you see your research leading in the future?
The Human Connectome Project, which David led, has recently been extended to projects that span the human lifespan, including children and older adults, plus more than a dozen projects to study brain connectivity in various brain disorders. We’ll both continue to be involved in these projects for the foreseeable future. Matt is also interested in how these methods can be applied to clinical problems, such as mapping brain areas prior to neurosurgery to help minimize deficits after surgery.
Do you foresee any social or political implications for your research?
Not any specific ones, but we hope that political folks will reinvigorate their decades-old bipartisan support for scientific research that benefits humankind. We also hope that the public will continue to be interested in scientific advancement in general and for basic research that deepens our understanding of the brain.
Matthew F. Glasser, M.D., Ph.D.
Resident Physician
Internal Medicine St. Luke’s Hospital
Radiology BJC Hospital
Department of Neuroscience
Washington University Medical School
St Louis, Missouri, USA
David C. Van Essen, Ph.D.
Alumni Endowed Professor
Department of Neuroscience
Washington University Medical School
St. Louis, Missouri, USA
