The Spontaneous Brain
When people sit peacefully at rest, doing and thinking nothing in particular, their brains still buzz merrily along. In scans called functional MRIs, they light up in characteristic patterns. No one knows the purpose of this spontaneous chatter, but it accounts for up to 98 percent of the brain’s activity and burns about three-quarters of the brain’s energy. To help unravel its origins and significance, researchers at Indiana University built a new computational model of a macaque monkey brain, which they describe in the June 12 issue of the Proceedings of the National Academy of Sciences.
“With this work, we can shed some light on what is actually driving the pattern of activation and deactivation that is seen in the resting brain,” says Olaf Sporns, PhD, associate professor of psychology, who worked on this project with his graduate student, Christopher J. Honey. The brain’s activity at rest may ultimately influence how individuals think and behave and how the brain responds to injury and disease.
Sporns and Honey chose the macaque brain because its wiring diagram is well understood. Researchers have done hundreds of tracer experiments—where they inject dye into one area of the brain and trace its spread to other areas—to establish the connectivity patterns of the macaque brain. From these data, Sporns and Honey built a “connection matrix” that specifies which of 47 brain areas are connected and which are not. On top of this roadmap, they superimposed differential equations that describe the electrical activity of each brain area. Then they ran a simulation to see how their virtual brain lights up when it is just talking to itself, with no external inputs.
The resulting patterns of brain activity closely resembled those seen in imaging studies of the human brain at rest. Interestingly, though the model operates at a very fast time scale (sub-millisecond resolution) it generates the slower fluctuations seen on fMRI (seconds to tens-of-seconds resolution). “Despite the fact that we have fast dynamics, we get these very slow processes to unfold,” Sporns says.
When they randomly scrambled the connection matrix in their model, they no longer saw the characteristic activity patterns of the resting brain. “So we have a good argument that what we see is actually because of the specific pattern of the connectivity,” he says.
“Their work makes this very important step of linking the anatomy—the connections between the brain areas—to the patterns of spontaneous activity. I think this is really the first study that makes this link explicitly,” comments Giulio Tononi, MD, PhD, a professor of psychiatry at the University of Wisconsin. “They are able to explain many of the features that are observed in studies using fMRI.”
The next step is to apply this modeling approach to the human brain, Sporns says. Though people cannot undergo invasive tracer studies, a new non-invasive technique—diffusion tensor imaging—is providing the connectivity data for human brains.
Using human models, Sporns plans to study how brain lesions interrupt the brain’s network—its connectivity, spontaneous activity, and ultimately performance. “There is great potential here for understanding brain injury and recovery processes,” he says.
He also plans to study how the resting brain’s activity influences people’s thoughts and behaviors. Every person has a unique pattern of spontaneous activity. “The open question is whether this spontaneous activity actually colors or somehow interacts with our ability to do a task,” Sporns says. “If that were the case, that would be really interesting.”