MIT creates analogue 'brain' chip

Researchers have designed a new computer chip that mimics synaptic plasticity--how a brain's neurons adapt in response to new information.

Credit: MIT

The most common approach to modeling a human brain in silicon involves dumping as much processing power as possible to the problem until you reach brain-like capabilities. IBM's Watson is a good example of this.

Choosing a different route, a team of MIT researchers have created a silicon chip that can simulate the activity of a single brain synapse, mimicking the chemical, ion-based communication channels that flow between synapses.

To accomplish this, Chi-Sang Poon, a principal research scientist in the Harvard-MIT Division of Health Sciences and Technology, and his colleagues used 400 transistors and standard CMOS manufacturing techniques to build a chip that allows current to flow through not in a digital, on/off fashion, but a finely tuned analog one.

“If you really want to mimic brain function realistically, you have to do more than just spiking. You have to capture the intracellular processes that are ion channel-based,” Poon says. Spiking refers to how, in previous attempts, researchers had built circuits that could simulate the firing of an action potential, but not all of the circumstances that produce the potentials.

“We can tweak the parameters of the circuit to match specific ion channels,” Poon says. “We now have a way to capture each and every ionic process that’s going on in a neuron.”

The researchers anticipate the chip will help neuroscientists learn much more about how the brain works, and say that it could also be used in neural prosthetic devices such as artificial retinas.

The new chip represents a “significant advance in the efforts to incorporate what we know about the biology of neurons and synaptic plasticity onto CMOS chips,” says Dean Buonomano, a professor of neurobiology at the University of California at Los Angeles, adding that “the level of biological realism is impressive.

MIT's analogue chip could be used to simulate a simple brain circuit magnitudes faster than high-capacity digital computer, and even faster than the biological system itself, note the researchers.

While the chip represents a leap forward in the field of computational neuroscience, it's worth noting that the human brain is massively complex with around 100 billion neurons, and each neuron can be connected to thousands of others with synapses.

To actually simulate a human brain requires more than processing power and neural mimicry, but also a detailed circuit diagram of the brain (connectome) to use as raw data to build a simulation. Considering scientists are just now simulating the 302-neuron brain of a flatworm, for which a wiring diagram exists, you can appreciate the enormous technical difficulty of the undertaking, so don't count on it to happen any time soon.

Poon and his research team describe the chip this week in the journal Proceedings of the National Academy of Sciences.