Computing methods used to study 'servers' in human brain

For years, engineers have been able to chart traffic flow in computer networks like the Internet. Now, a team of Toronto researchers has adapted those computational techniques to study "servers" in the human brain.

They suspect that parts of the brain act as hubs, collecting and distributing information and controlling how distinct parts of the brain work together to see and hear, reason and remember. They hope to discover how these "servers" vary from person to person and how aging affects them. Their goal is to shed light on diseases that damage the brain, including Alzheimer's and stroke.

"These are the way stations, the bridges between different communities in the brain," said Bratislav Misic, lead author of a research paper published on Thursday in the Public Library of Science journal PLoS Computational Biology. He is part of a team at Baycrest's Rotman Research Institute.

We often compare our brains to computers and vice-versa. In their paper, the researchers describe how tools from telecommunications research can be adapted to study server configuration and traffic flow in the brain.

"The cool thing is this technology was developed to look at information processing and computer networks," said neuroscientist Randy McIntosh, who is also director of the Rotman Research Institute.

There is growing evidence the brain has "small world" properties; areas that are physically close have more connections with each other than they do with parts that are farther away. But there are also signs that a number of regions collect information and route it to more distant destinations.

Dr. McIntosh and his colleagues want to learn more about these suspected "servers," including an area called the precuneus, towards the back and middle of the brain. It is particularly well-connected to distant regions, Dr. McIntosh said, although it is not yet clear exactly what it does. There is evidence, he said, that it is one of the first areas hit in Alzheimer's disease.

In the case of stroke, it may be that damage to a hub is far more devastating than to another area. Understanding the flow of traffic in the brain might help researchers understand why some people make a full recovery and others don't.

Neuroscientists are already using different brain imaging techniques to understand the connections between regions of the brain. This new approach allows them to study suspected hubs in action, to see, for example, how information flowing from a "server" changes when someone is performing a specific mental function, like listening to music.

It is based on assessing the output of a region, the electric or magnetic signals it sends out.

It is the timing of the signals that matters. If they are slow but steady, then not much is going on. But if the signals come all at once, then slow down, then speed up again, it is a sign of a lot of activity. Variable output indicates that the processing in these brain regions is changing to meet the demands of incoming traffic, said Mr. Bratislav, who is a doctoral student at the University of Toronto.

This is similar to what happens in computer networks, he said.

"We think we can identify the flow of neural traffic using this method."

In the paper, the researchers describe a simple experiment in which they asked volunteers to rest with their eyes shut, and then open them. Electrodes on the scalps of each volunteer allowed the researchers to measure the electrical activity in the parts of the brain that process visual information.

They were able to measure how the "traffic" through these regions increased once the volunteers opened their eyes.