Building a Synthetic Brain

[ubermenu config_id=”main” menu=”84″] NEWSROOM Building a Synthetic BrainFeb 3, 2011 For electrical engineer Alice Parker ’70, ’75 PHD, simulating the machine on your shoulders is more complicated than building the latest computer chip. Forget the Pent …

[ubermenu config_id="main" menu="84"]


Building a Synthetic Brain

Feb 3, 2011

For electrical engineer Alice Parker ’70, ’75 PHD, simulating the machine on your shoulders is more complicated than building the latest computer chip.

Forget the Pentium. As you read this, you’re using a machine that puts the latest processor to shame–your brain. Its 100 billion neurons each have 10,000 synapses exchanging messages. Alice Parker ’70, ’75 phd hopes to one day duplicate that power with a synthetic brain. It may take a decade or more, she says, but eventually a synthetic brain could be used in prosthetic devices that could wire around damaged parts of the brain, or in brainlike systems that could drive robotic vehicles.

Other researchers simulate the workings of a few neurons using computer software, but building a whole brain that way would require too many processors, says Parker, a professor of electrical engineering and former vice provost for research at the University of Southern California. "Computer chips like the Pentium are considered comparable to a brainlike structure, but they’re nowhere near the complexity of the brain," she says.

For example, a Swiss university has simulated a brain using software and an IBM multiprocessor that uses about 8,000 processors. But it can do the work for only 10,000 neurons. "To scale up to the size of a brain, you would need millions of these IBM multiprocessors," Parker Says.

Alice Parker, Professor at the University of Southern California (Photo Courtesy of: Mark Berndt/University of Southern California)

In her vision, transistors will stand in for the neurons and synapses, and the electrical properties of voltage and current will simulate brain chemicals (neurotransmitters). Consider that it would take hundreds of transistors to simulate just one synapse, and you’ll get some idea of the complexity of her task. "[The brain has] a quadrillion synapses, if I do the arithmetic right," Parker says. In other words, 10 to the 15th power, or a one with 15 zeros behind it. "It’s a lot of circuitry," she says.

Parker has tackled daunting tasks before. She studied engineering at NC State in the early 1970s when few women did. In her first electrical engineering class, there were 120 students but just one other woman. "It was a bit isolating," she says. "The men might have been a little afraid to befriend me."

She sometimes made friends with fellow students’ girlfriends, and she made other friends while singing in the chorus. She also got support from professors like Wayland P. Seagraves ’32, ’33 ms, who chose to take Parker to a regional conference sponsored by the Institute of Electrical and Electronics Engineers, making her the first woman in the region to attend. "He was always trying to make opportunities for us. I don’t know if I would have been as successful if it hadn’t been for him," says Parker, who has been interested in synthetic-brain work since she was a graduate student. She began serious work on it several years ago when she saw that nanotechnology and other fields were improving fast enough to make the task feasible.

Parker and her collaborators are starting small, trying to simulate one neuron using transistors made out of carbon nanotubes, which are molecules of carbon that can be as much as 100,000 times narrower than a human hair. Because nanotubes are so small, even if each transistor used several, the synthetic brain wouldn’t be too big to be practical.

"I’m trying to get my students launched so they can ultimately get there," Parker says. She keeps her eye on neuroscience developments that will influence her next move, such as an October 2009 study showing that a particular type of cell helps the retina adapt when light changes from bright to dim. "All of a sudden we know about these cells that nobody was factoring into these circuits before," she says. "It’s like chess, where you’re always thinking about the end game. We can’t build this whole complicated system now, but every little piece that we make today has to make sense later in the final structure."

Angela Spivey, NC State Alumni Magazine
This article originally appeared in the Winter 2009 issue of NC State. The alumni magazine is a benefit of membership in the NC State Alumni Association.
Join the Alumni Association

Filed Under

Share This