July 22, 2020
Counting Brain Cells
In 2003, Suzanna Herculano Houzel (neuroanatomist from the Comparative Neuroanatomy Lab at Vanderbilt University) became curious about where the number came from and began to question senior neuroscientists. No one was able to point her to the original source. After an extensive search through the scientific literature, she wasn’t able to find a single source supporting the claim (Herculano-Houzel 2016). At that time, she was working at the Federal University of Rio de Janeiro and didn’t consider herself a neuroanatomist, didn’t have a lab, and didn’t have research funding. But her skepticism was strong, and she was interested in counting brain cells: Had anyone counted them in a reliable way?
When she asked Nobel laureate Eric Kandel, coauthor of Principles of Neural Science, for an original source for the claim, he couldn’t provide one even though the claim was made in Principles of Neural Science. Kandel said he wasn’t responsible for the chapter containing the 100 billion neuron claim; one of his coauthors wrote that chapter. “It was 2004, and no one really knew how many neurons could be found on average in the human brain.” When Roberto Lent, author of One Hundred Billion Neurons, was asked where the number came from, he couldn’t provide an answer either according to Herculano-Houzel. The title of the book was later changed to One Hundred Billion Neurons? with a question mark added to the title.
A key problem when trying to count neurons in the brain is the variability of their distribution throughout the brain. The density of neurons may vary by factors of up to 1,000 across structures. Even within a single structure, different layers can consist of different numbers of neurons.
After numerous attempts at developing a reliable, valid method for counting brain cells (which included counting neurons, non-neuronal-glia, and endothelia brain cells) the “isotropic fractionator” was complete. The method involves dissolving cell membranes while preserving nuclear membranes (each neuron consists of one nuclear membrane), producing what Herculano-Houzel refers to as brain soup. In the brain soup, there are free-floating nuclei that are relatively easy to count by sampling tiny amounts of the soup. All the nuclei from the cells are stained blue, collected, and counted. Counting the nuclei is simple and requires no special training. In the 2016 book The Human Advantage, Herculano-Houzel provides a description of what went into developing the technique, including detail on the early failed attempts to create brain soup. Other researchers, including Christopher von Bartheld from the University of Reno and Jon Kaas from Vanderbilt University, have shown this method to be fast, reliable, and relatively easy to apply (Bahney and von Bartheld 2014).
Results, from the isotropic fractionator indicate the human brain has an average of 86 billion neurons and 85 billion non-neuronal cells (glial cells and endothelial cells). For people who like to point out that “eighty-six is close to 100” and who claim the 100 billion is reasonable as an order-of-magnitude estimate, Herculano-Houzel counters by saying an entire baboon brain contains eleven billion neurons. Fourteen billion is a significant number of neurons. It becomes even more significant when considering each neuron may connect to thousands of other neurons.
Herculano-Houzel’s original research and the research influenced by her findings has led to changes in the understanding of the brain. She has counted brains cells in a range of different species, so her work is also important regarding nonhuman animals. Textbook writers and top brain scientists are acknowledging her groundbreaking work. In a recent conversation with Bryan Kolb, coauthor of Fundamentals of Human Neuropsychology (2009), he told me there would be no referrals to the 100 billion neuron myth in future editions of the text. I contacted Kolb after reading that the human brain contained about 100 billion neurons in a textbook he coauthored. It was a text, Fundamentals of Human Neuropsychology, I used in graduate school. I was happy when I read in Eric Kandel’s book The Disordered Mind: What Unusual Brains Tell Us about Ourselves (2018) that the brain “can perform its remarkably swift and accurate computational feats because its 86 billion nerve cells—its neurons—communicate with one another through very precise connections.” Recall from earlier in this article, Kandel coauthored a book that stated the human brain consisted of 100 billion neurons. The book he coauthored, Principles of Neural Science, is often considered a seminal work in the field of neuroscience.
The journey Herculano-Houzel made to scientific discovery is impressive; it demonstrates a consistent skepticism that is essential for the development of scientific knowledge. In addition to refuting the 100 billion neuron statistic, her research has refuted other long-held neuroscience dogmas. As an example, until recently textbooks promoted the claim that glial cells (a type of brain cell involved in numerous processes, including supporting nerve cells) outnumber nerve cells ten to one in the human brain. A search for an original source to support the claim failed. Counting brain cells allowed Herculano-Houzel to dispel that myth as well. A paper published by Herculano-Houzel and colleagues titled “Equal Numbers of Neuronal and Non-Neuronal Cells Make the Human Brain an Isometrically Scaled-Up Primate Brain,” which is now a heavily cited paper, was rejected by prominent journals, including Nature, Proceedings of the National Academy of Sciences of the U.S.A., Neuron, and the Journal of Neuroscience. The paper was eventually published in the Journal of Comparative Neurology (Azevedo et al. 2009). Even high-ranking science journals are sometimes resistant to change and do not necessarily practice the skepticism that is essential to scientific thinking.
Azevedo, F.A.C., et al. 2009. Equal numbers of neuronal and non-neuronal cells make the human brain and isometrically scaled-up primate brain. Journal of Comparative Neurology 513: 532–541.
Bahney, J., and C.S. Bartheld. 2014. Validation of the isotropic fractionator: Comparison with unbiased stereology and DNA extraction for quantification of glial cells. Journal of Neuroscience Methods 222: 165–174.
Herculano-Houzel, S. 2016. The Human Advantage: A New Understanding of How Our Brain Became Remarkable. Cambridge, MA: The MIT Press.
Kandel, E.R. 2018. The Disordered Mind: What Unusual Brains Tell Us about Ourselves. New York, NY: FSG.
Kandel, E.R. et al. 2000. Principles of Neural Science, 4th edition. New York, NY: McGraw-Hill.
Kolb, B. and I. Whishaw. 2009. Fundamentals of Human Neuropsychology, 6th edition. New York, NY: Worth Publishers.