What is the brain doing when it is being asked to do nothing in particular?
During the past five years, Marcus Raichle and his team at the Washington University School of Medicine in St. Louis have looked at that question. While working on other neuroimaging studies, Raichle noticed an interesting trend in brain activation when experimental participants began a cognitive task.
While some areas of the brain would “light up” during the task, other areas would show marked deactivation. Even more surprising, the deactivation pattern seemed to be very consistent in the medial regions of the brain, including the posterior cingulate and precuneus.
“This was not random noise,” Raichle says. “And given that the majority of what the brain does is just flat-out function—just the brain running itself, really—we started thinking about what was really going on in there.”
Raichle suggests that this highly organized state of brain areas is a “default network,” responsible for the intrinsic functional activity of the brain, as opposed to thought specifically evoked by a stimulus or cognitive task.
“Your brain is a very expensive gadget to run,” he says. “If you aren’t engaged in a task, your brain doesn’t just turn off. It wouldn’t make any sense.”
The use of imaging techniques such as positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) have provided neuroscientists with the ability to map specific areas of the brain and correlate them with behavior. But what about when there is no specific behavior to map?
“We don’t know much of anything about thought that is undirected or how much thinking humans do at a moment’s notice,” says Malia Mason, a postdoctoral fellow at the Massachusetts General Hospital Martinos Center for Biomedical Imaging. “We assume that everything neuroimaging shows us is related to a particular task, but I would guess that a lot of it is extra.”
Mason and colleagues reported in the Jan. 19 issue of Science that areas of the default network were activated in an fMRI scan of study participants letting their minds wander when they were between cognitive tasks.
“Daydreaming is a loaded term,” Mason says. “But I’m operationally defining it as a thought that is not directly related to an action or task—
a stimulus-independent thought.”
A Link to Memory
The default network may have a more critical role than just daydreaming. Raichle believes that a lot can be learned about disease by examining resting states in the brain.
“The notion that a disorder like depression or schizophrenia may be related to a problem with the fundamental organization of the brain is a completely plausible idea,” he says.
Michael Greicius, a neurologist at the Stanford School of Medicine, uses intrinsic activity connections as a way of studying Alzheimer’s disease. Many of the brain regions in the default network show some of the earliest degeneration in patients with the disorder.
“The default mode is specifically targeted in Alzheimer’s,” Greicius says. “Recently our group and Dr. Raichle’s group have shown that the hippocampus is also a part of this network. So we can make an assumption that it has something important to do with memory.”
Raichle agrees. “The brain is in the prediction business,” he says. “It’s trying to figure out where we’re going and how to respond to what’s happening in the world. We have memory, not just to muse about the past, but to predict how to respond to the present and future.”
Greicius’s work has also identified six to ten other potential resting state networks, related to other key cognitive categories including visual, auditory, sensory-motor, and executive control. But the first identified resting state network, currently referred to as the default network, seems to be dominant.
“The conjunction of brain regions in the default network are well-suited to mediate stream of consciousness processing,” Greicius says.
Implications in Depression
Helen Mayberg, a leading depression researcher at the Emory School of Medicine, has found differences in the baseline metabolic activity of brain areas in the default network in people with depression. By looking at the resting state, she does not run the risk of potentially confounding the disease signal with activation from a task or difficulty equating depressed people and a control group. It also affords her the ability to compare the brain of a person with the disease both before and after treatment.
“Depression doesn’t require you to stress the system to see its impact,” Mayberg says. “Once you have the illness, it’s pervasive, always there. By definition, it is distorting your resting state.”
And that altered resting state has great consequences. “Your brain is a machine in idle at all times, ready to do any one of a gazillion subroutines,” says Mayberg. “You should be able to move in any direction with great agility when doing nothing, but the disease hijacks that ability.”
Researchers in other labs are now looking at the default network in disorders such as autism and schizophrenia. But Raichle says there is still plenty to learn about the normal brain as well. His lab recently found the default network in an anesthetized macaque monkey and will soon report the finding in Nature.
This finding raises philosophical and methodological questions for the neuroscience community: What other function might be obscured by the methods researchers use in neuroimaging studies? What are the adaptive purposes of the default network and the brain’s other less dominant resting states? How can we apply our understanding of this underlying brain organization to the clinical treatment of disease?
“The more we learn, the more possibilities, the more questions open up,” Raichle says. “There’s lots of fun stuff to do here.”