A New Age of Exploration
“Reverse engineering the brain is not a defined task with a fixed endpoint,” said Gerald Loeb, USC Viterbi professor of biomedical engineering and neurology.
I had been assigned to write a story, a feature story no less, on reverse engineering the brain. (Spoiler alert: You’re reading it now.) You can imagine my frustration when one of the first responses I received from one of the leading brain researchers at USC was basically the academic version of “We don’t know where we are or where we’ll end up.” How was I supposed to write 2,000 words on a topic the National Academy of Engineering has deemed one of its Grand Challenges when the subject remains undefined and lacking direction?
After speaking with several people at USC Viterbi, I realized this was not a story of discovery, but of exploration. Consider this lesson from history: When European explorers first arrived in the Americas, they didn’t know what they had bumped up against. It took years for them to realize they had discovered two entire continents, and centuries before they mapped them out. Five hundred years later, we are still struggling to truly understand the world they found and the profound implications of their actions.
If we consider reverse engineering the brain more a voyage of exploration than a specific scientific mission, then let’s meet some of our explorers.
Charting a Course
To better understand a new frontier, it’s good to have a map. That’s where the research of people like Richard Leahy and Manuel Monge come in. Leahy, Dean’s Professor of Electrical Engineering in the Ming Hsieh Department of Electrical and Computer Engineering, uses his expertise in signal and image processing to build better images of the brain’s cerebral cortex. His team uses magnetic resonance imaging to create some of the most detailed images of the brain we have. “By breaking down the brain in such detail, we can get a better understanding of the overall architecture to a degree we never have before,” said Andrew Li, one of Leahy’s Ph.D. students.
“Leahy and his team have created three-dimensional imag.es of the brain that can even be observed by researchers in real time with augmented-reality glasses. “It is our hope that with our brain mapping technology, other researchers will have the tools necessary to make decisions and build devic.es to solve a host of questions about the brain,” Leahy said.
While Leahy improves our ability to map the brain from the outside, Monge, assistant professor of electrical and computer engineering, is doing the same from the inside. He develops novel integrated circuit techniques for neural interfaces, which are electrodes and integrated circuits implanted in the brain.
He hopes to make the circuits small and low-power enough that we can learn more about the brain in action. The result could be a better sense of how the brain processes information.
“In order to reverse engineer the brain, we need to be better at observing the brain up close in real time, without damaging it or altering its function. These devices are just one step among many in revealing more of the brain map to us,” Monge said.
Orienting Ourselves
If Leahy’s and Monge’s work represents one of our maps, then Francisco Valero-Cuevas, professor of biomedical engineering and of biokinesiology and physical therapy, provides a compass. As a mechanical engineer by training who works on “neuromechanics,” he wants people to think differently about our relationship with our brain.
“It’s important to remember that our nervous system originally evolved to deal with the physics of interacting with the world, not to think,” Valero-Cuevas explained. But in evolution, things get repurposed all the time. There’s even a catchy term for it: “evolutionary repurposing.” Feathers originally evolved to regulate heat. Reptilian jawbones were repurposed as mammalian inner ear bones.
The human brain may be the ultimate example of evolutionary re-purposing run amok. It’s humbling to realize that human consciousness might be nothing more than the accidental byproduct of a system built to move simple organisms around in some primordial soup — but it’s an important perspective to have when trying to better understand the brain.
Stepping into the Unknown
With a map and a sense of direction, we can start on the journey. This is where the work of researchers like Alice Parker, Han Wang, Gerald Loeb and Maryam Shanechi, among many others at USC, comes in. Much of their work involves designing and building biomimetic devices — inventions built to mimic the way our brain does things — and brain-machine interface technologies that aim to decode or regulate abnormal brain activity to treat some of its most debilitating disorders.
Parker, Dean’s Professor of Electrical and Computer Engineering, is a proponent of positive reinforcement. She builds electronic circuits that mimic neurons (as well as astrocytes, brain cells that contribute to computation). These circuits, called neuromorphic circuits, learn and adapt just like biological neurons do, by responding to positive stimulation when they do something correctly. The technology represents an important step in addressing one of the biggest technology challenges out there: getting artificial intelligence to think and “learn” without forgetting.
When humans are exposed to something new and useful, neurons get a spike of dopamine and the connections surrounding those neurons strengthen. “Think of an infant sitting in a high chair,” said Parker. “She might be waving her arms around wildly because her undeveloped neurons are just randomly firing.” Eventually one of those wild movements leads to a positive result — say, knocking over a cup and making a mess. All of a sudden, the neurons that made that motion get a response and strengthen. Done regularly enough, the baby’s brain begins to associate that spike with something worth internalizing. And just like that, she has learned that an arm motion causes an entertaining result and that learning persists over time. This is exactly what neuromorphic computing is trying to do: teach AI to learn from real-world experiences.
Wang, the Robert G. and Mary G. Lane Endowed Early Career Chair and associate professor of electrical and computer engineering, takes a different approach. From his perspective as an engineer with a background in nanoelectronics, the first step in building an electronic version of the brain is to develop new electronic components to work more like our own biological systems.
“Our brain has numerous different sections, all responsible for different tasks,” Wang said. “In each of these sections, the neurons are organized in their own unique way. In fact, the neurons in different sections take different forms.” Each section of the brain has its own architecture and neurons. That diversity is what allows it to focus on different tasks and accomplish them with only tiny amount of energy. Wang believes that to emulate the brain, we will need just as many unique manmade neural network architectures as the brain has biological ones.
One of the many materials Wang works with is called oxidized hexagonal boron nitride. That tongue twister of a material is used to produce nanoscale electronic devices. Because the geometry is so different from traditional 3-D silicon, it can be used to produce extremely thin memory devices that operate at a power level much closer to the brain’s own memory system.
While Parker and Wang build devices that mimic the brain’s ability to think, Loeb is focusing on re-creating its ability to interact with the physical world. Much of his work focuses on a part of the neurological system sometimes left out of the conversation on reverse engineering the brain: the spinal cord.
The spinal cord plays a huge role in motor and sensory function. Among many other inventions, Loeb has built neural prosthetics that help deaf patients hear and devices that mimic human touch and motor functions. “What we’re trying to do now is to understand how the developing nervous system self-organizes to recognize and control our musculoskeletal system and the parts of the external world with which it interacts,” Loeb explained. By better understanding where and how the brain and spinal cord compute motor functions, even more effective and targeted prosthetics can be built in the future.
Of course, any discussion about exploring the unknown wouldn’t be complete without touching on emotions. Our complex emotions may be our most uniquely human quality. Unfortunately, we often struggle to control or understand our own emotions, let alone of the people around us.
Shanechi, the Andrew and Erna Viterbi Early Career Chair and assistant professor of electrical and computer engineering, designs advanced signaling and control technology to do just that. By developing new neurotechnologies and brain-ma.chine interfaces, she has been able to read patients moods in real time by interpreting electric signals given off by the brain.
“Our brain’s functions and dysfunctions are actually encoded in the electric signals it generates,” Shanechi said. “The better we understand how emotions or other brain functions influence these signals and how we can regulate abnormal signals, the better we can diagnose and treat a wide range of debilitating neurological and neuropsychiatric disorders such as depression.”
From implants that bring sight to the blind to microsensors that improve brain mapping; from algorithms that help robots learn to move on their own to technology that can interpret human emotion from reading brain signals, the list of brain research undertaken at USC Viterbi is a long one. Our understanding of the brain has grown exponentially but, as Loeb cautions, “Exponential growth in science and exponential growth in technology are not the same thing.”
This is the ultimate challenge: We learn something new and create technology to take advantage of that knowledge. We then discover that in the time the technology was conceived, designed, patented and produced, our understanding of the very thing it was meant to solve has changed drastically.
“The brain is so complex that we may never have a complete understanding of how it does everything it does … but that’s OK so long as we can use the knowledge we glean at every step,” Valero-Cuevas said.
The Blind Men and the Elephant
“The challenge of the brain reminds me of the poem ‘The Blind Men and the Elephant,’” Parker said. In the poem, several blind men are situated around a mysterious animal called an elephant. They each come up against different parts of the animal and begin to feel around. The man near the trunk is positive the strange beast resembles some kind of snake. The man near the elephant’s flank insists that the elephant most resembles a wall. Another man, feeling the tusks, scoffs at the others’ interpretation and is convinced an elephant is most like a spear. All of the blind men have some understanding of what is before them, but none of them get the whole picture.
The story’s first lesson is clear: Our inability to see the bigger picture can lead us to false assumptions. But the second is even more important. If those blind men had talked to each other and worked together, they could have discovered what an elephant actually was.
The brain is not an elephant. Solving its mysteries will re.quire more collaboration, more communication and more creativity than anything else we’ve ever accomplished. It will require countless engineers building new devices, each of which reveals just a sliver of a map we are still drawing.
As big as the challenge is, Valero-Cuevas remains optimistic. “When engineers were given chemistry, we created the industrial revolution. When given physics and electricity, we brought about the information age,” he said. “Now, we have been given biology, and I have no doubt that engineers will bring about a new technological age in medicine and neuroscience.”
As USC Viterbi Dean Yannis C. Yortsos often says, “Engineering is the ability to use phenomena for useful purposes.” Today, we are being presented with new, fascinating and at times confounding biological phenomena. The world that engineers reveal and build with those phenomena may not yet be fully realized, but one thing is certain: It will be built.