The University of Chicago is engaged in a project to produce an atlas of the human brain, aiming to shed light on mental illnesses, energy shortage and creating better machines
“For a long time, there was this belief our brains were made up of a little person sitting in our heads,” says Dr. Narayanan “Bobby” Kasthuri, Neuroscience researcher at Argonne National Laboratory and Adjunct Assistant Professor at the University of Chicago. “It’s called the ‘homunculus’. It has eyes, arms, and legs just like us and as it moves its arms and legs, our arms and legs move too. It is an instantiation of us. But over time, we gradually discovered cells are the only things in our brains. There is nothing magical.
“A human brain probably contains 100 billion brain cells or neurons,” Dr. Kasthuri continues. “These neurons are different from any other cell in our body as they make long connections. On average, each human neuron makes 10,000 connections. If you take the 100 billion neurons and multiply by 10,000 connections, you will arrive at a big number like a quadrillion. One way to think about it is there are more connections happening inside our brains than the stars in the Milky Way galaxy. My dream is to map every single one of them.”
Dr. Kasthuri kick-started the brain-mapping project in 2018, hoping to create an “atlas of the brain”, to figure out all the brain cells, their connections, and their meaning in the context of normal brain functions. The project leverages the most advanced photon source – a form of powerful x-ray offered by the Argonne National Laboratory – to search all the wirings of the brain and reconstruct brain circuits to generate new information.
“I think this could be one of the most difficult things that humans have ever attempted to do,” Dr. Kasthuri calmly states. “We are getting to a physical basis for our thoughts, actions, emotions and behavior. Mapping every neuron and how each of them is connected to other neurons not only speaks of all the information about you, your personality, your memories, and so on but also many of the things that we call illnesses. All the mis-routings and miswiring of that diagram.
“Although we think that many illnesses like autism or schizophrenia are missed connections, neurons that are connecting with inappropriate neurons, we aren’t sure where to start correcting these errors. These are puzzles that haven’t yet been solved by science. However, if we have a map of the brain and compare it to a map of a schizophrenic brain, we’d start to understand what the physical basis of these mental illnesses are and this, naturally, will allow us to treat them more rationally.”
On top of a better understanding of mental health and curing degenerative diseases, the project also aims to study the brain as an energy-efficient system. “In our lab, I learned that by 2030, the single largest growing fraction of our energy budget is not the oil that we put in our cars or our aeroplanes but uploading and downloading pictures from the internet or moving data to and from the Cloud,” Dr. Kasthuri adds. “So, manipulating information is a task that requires a lot of energy. We understand that a brain can move information and do calculations for 20 watts of energy. We think unlocking some of the secrets of the brain might lead to the development of more energy-efficient computers or significantly help us in our energy crisis.”
Ultimately, Dr. Kasthuri is looking forward to building a “less artificial but more intelligent” AI environment. “At a remarkably high level, if we can build and decipher the brain map, we will be able to have insights into a person’s thoughts, skills, fears, etc. We can also reverse engineer any of the human qualities, like humor, creativity, empathy, and effectively transfer them to computers and AI.”
Dr. Kasthuri also points out that, unlike machines, human brains continue to change throughout our lives, and we learn from past experiences. Scientists don’t quite understand how this works and how to make this work in AI.
“This is the fundamental nature of the human condition,” Dr. Kasthuri says. “We are born completely helpless, completely useless, and completely hapless. Every one of us grows into adults to learn a million things about the world around us, learn how the world works and how to manipulate it. And I realized if I could figure that out, I could decipher the fundamental thing it means to be intelligent and finally, what it means to be human. I think as we try to understand the brain, we are also unfolding certain societal implications. These include the future of computing, the future of energy, and the future of AI.”