What the Default Mode Network Does
The brain spends about a fifth of the body's energy whatever it is asked to do. The default mode network is the pattern found inside that expense, and the evidence supports it as an organised network, though not as a way of doing nothing.
Answer to the question
What does the brain do when you do nothing?
It stays busy. The brain uses about 20 per cent of the body's energy, and tasks add 5 per cent or less to local blood flow. A set of regions called the default mode network falls in activity when attention turns outward and is active during internally directed thought, including mind-wandering. Whether it is the mind-wandering network is disputed.
Key takeaways
- The brain is about 2 per cent of body weight and uses about 20 per cent of the body's oxygen, and local blood-flow changes during most cognitive tasks are often 5 per cent or less (Raichle and Gusnard, 2002).
- Raichle and colleagues (2001) defined a baseline by oxygen extraction and found consistent decreases in the same regions during goal-directed tasks, which they called an organised default mode.
- Greicius and colleagues (2003) showed the regions are coupled at rest, and Mason (2007) and Christoff (2009) tied the network to reported mind-wandering, with executive regions recruited alongside it.
- Morcom and Fletcher (2007) argued that activity at rest has no privileged status, and Smallwood and colleagues (2021) show the network rising in memory and abstract thought, so the word default records how it was found, not what it does.
- The Distracted Mind lists daydreaming's activation of the default mode network among its break advice (p. 229, note 24), an item the vault audit grades as mechanism and not as a test of whether breaks help.
The objection came six years after the network was named, and it was a fair one. If rest is only the absence of a task, Alexa Morcom and Paul Fletcher argued, then activity found at rest has "no privileged status as a fundamental metric of brain functioning". The default mode network, on that view, describes the scanner room and tells us little about the brain. The case for the network is stronger than that objection allows, and it begins with an energy bill that no task can explain.
A fifth of the body's energy, spent before tasks
The human brain is about two per cent of body weight and uses about a fifth of the body's oxygen and so of its calories, ten times what its weight alone would predict (Raichle and Gusnard). Whatever a task adds is small beside that. In the same commentary the authors point out that local changes in blood flow during most cognitive tasks, measured with PET, are often five per cent or less, and that attempts to detect a whole-brain change during intense mental effort have failed.
If effort is a few per cent on top of a large fixed cost, the fixed cost is the thing to explain. Calling it idling explains nothing, because an idling organ would not be paying it.
Calling it idling explains nothing, because an idling organ would not be paying it.
The default mode network was first noticed as a decrease. In early PET studies that set attention-demanding tasks with no bearing on the self against relaxed rest, a consistent set of regions in the parietal, medial prefrontal and temporal cortices fell in activity when the task started, as Raichle's later review describes it. The discovery came largely by accident.
The baseline got a physiological definition in 2001, built on the oxygen extraction fraction, the share of the oxygen delivered by blood that tissue takes up (Raichle and colleagues). In people awake with their eyes closed it was remarkably uniform across the brain. "Areas of activation were conspicuous by their absence." What did show up were consistent decreases during goal-directed tasks, which the authors read as an organised default mode, suspended when behaviour is aimed at a goal. A definition written in oxygen answers part of the objection before it is made: nobody had to be told to think of nothing.
Connections come first, then the thoughts attached to them
A map of decreases is not yet a network. The connections came next. Two regions that dropped during a working-memory task, the posterior cingulate cortex and the ventral anterior cingulate, were the starting point, and the question was what moved with them at rest (Greicius and colleagues). The posterior cingulate was strongly coupled with the other default regions.
A simple visual task left that pattern almost unchanged.
Three lateral prefrontal regions, busier during working memory, correlated negatively with the posterior cingulate at rest, which the authors read as a possible suppression of default activity during cognitive work. They presented it as the first resting-state connectivity analysis of the default mode, and as strong evidence for a cohesive default mode network.
A network defined by scans still needed a thought attached to it. The first test combined thought sampling with imaging. Stimulus-independent thought engaged the default network, and people who reported a greater tendency to let their minds wander showed more activity in it (Mason and colleagues, Science, 2007). How often minds wander, and what that costs in mood, is a separate question.
A study that sampled experience during fMRI looked at the moments before each probe (Christoff and colleagues). Default regions were active when people reported mind-wandering, and the medial prefrontal ones were also active when errors on the task rose. Executive regions were recruited as well.
Activity in both sets peaked when people did not realise their minds had wandered.
Two systems usually treated as rivals may, the authors proposed, work together.
The design matters here: participants were doing a task while being probed, so the wandering was caught alongside performance and not in a quiet room.
| Study | Design | What it added |
|---|---|---|
| Raichle and colleagues, 2001 | PET, oxygen extraction fraction | A physiological baseline, and consistent decreases in the same regions during goal-directed tasks |
| Greicius and colleagues, 2003 | Resting-state connectivity | Default regions coupled at rest, with lateral prefrontal regions correlating negatively |
| Mason and colleagues, 2007 | Thought sampling with imaging | Stimulus-independent thought engaged the network |
| Morcom and Fletcher, 2007 | Critical analysis | Activity at rest has no privileged status as a metric of brain function |
| Christoff and colleagues, 2009 | Experience sampling during fMRI | Default and executive regions both active during reported mind-wandering |
| Smallwood and colleagues, 2021 | Review | Activity drops in demanding attention tasks and rises in memory and abstract thought |
The Distracted Mind reaches the same ground from the practical side. Its list of evidence for taking breaks includes the statement that daydreaming activates the default mode network (p. 229, note 24). The page does not name the study, and the audit of the book grades the item as a mechanism, which is a different thing from a test of whether a break helps. How people interrupt themselves, the book's other contribution, and why ten quiet minutes feel so long are neighbouring questions treated elsewhere.
An objection to treating rest as the baseline
Nothing in the objection denied the default mode network.
Morcom and Fletcher questioned whether studies of the resting state are valuable and interpretable. Their target was the step from cost to baseline: that resting activity is expensive does not, they argued, make rest the absolute reference point against which every other state should be measured.
The stakes are practical. Resting-state studies now play a major role in research on the healthy and the diseased human brain, and the default mode network is central to that work (Raichle, 2015). If rest is a poor reference condition, a good deal of that work stands on soft ground.
The case has two replies. Within the oxygen data, the decreases were consistent across goal-directed tasks, so the claim concerns organisation and survives without rest being special. And in the connectivity work the same regions stayed coupled during a visual task, so the default mode network does not need a rest condition to exist.
What the word default leaves out of the story
A 2021 review by Smallwood and colleagues puts the pattern more broadly than the first PET work did. Activity in the default mode network tends to drop during demanding attention tasks and to rise during many complex forms of thought, especially those involving memory or abstract reasoning, in cortex lying farthest from the sensory and motor systems. The word default records how the network was found, by comparison with a rest condition. It does not describe what the network does.
The authors argue that mapping where the network sits in the cortex can help explain how it supports cognition and behaviour.
The word default records how the network was found, by comparison with a rest condition. It does not describe what the network does.
The evidence supports an organised, energetically costly network that recedes when attention is aimed outward and is active during internally directed thought, and it does not support "doing nothing", nor yet the claim that the default mode network is the mind-wandering network.
Which reading holds decides what a pause in the day is for. If those regions are where memory and abstract thought get worked on, a quiet quarter hour has a function. If its activity is only what remains when attention stops, it has none. Advice built on either reading runs ahead of the evidence.
Common questions
What does the brain do when you do nothing?
It stays busy. The brain uses about 20 per cent of the body's energy, and task-related changes in blood flow are often 5 per cent or less, so most of the cost is already being paid before any task begins. A set of regions called the default mode network falls in activity when attention turns outward and is active during internally directed thought.
What is the default mode network?
A set of linked regions in the parietal, medial prefrontal and temporal cortices whose activity tends to fall during demanding, outward-directed tasks. Marcus Raichle and colleagues named it in a 2001 paper in PNAS.
Is the default mode network the same as mind-wandering?
No. Imaging studies find the network active when people report that their minds have wandered, but it is also active during other forms of internally directed thought, especially those involving memory or abstract reasoning. The two are associated and not identical, and whether the network is the mind-wandering network is still disputed.
Who discovered the default mode network?
The regions were first noticed as decreases in early PET studies, largely by accident (Raichle, 2015). Raichle and colleagues called the pattern a default mode of brain function in a 2001 paper in PNAS, and Greicius and colleagues showed in 2003 that its regions are coupled at rest.
Sources
Raichle, M. E., MacLeod, A. M., Snyder, A. Z., Powers, W. J., Gusnard, D. A., & Shulman, G. L. (2001). 'A default mode of brain function.' Proceedings of the National Academy of Sciences, 98(2), 676-682. doi:10.1073/pnas.98.2.676
Raichle, M. E., & Gusnard, D. A. (2002). "Appraising the brain's energy budget." Proceedings of the National Academy of Sciences, 99(16), 10237-10239. doi:10.1073/pnas.172399499
Raichle, M. E. (2015). "The brain's default mode network." Annual Review of Neuroscience, 38, 433-447. doi:10.1146/annurev-neuro-071013-014030
Greicius, M. D., Krasnow, B., Reiss, A. L., & Menon, V. (2003). 'Functional connectivity in the resting brain: a network analysis of the default mode hypothesis.' Proceedings of the National Academy of Sciences, 100(1), 253-258. doi:10.1073/pnas.0135058100
Mason, M. F., Norton, M. I., Van Horn, J. D., Wegner, D. M., Grafton, S. T., & Macrae, C. N. (2007). 'Wandering minds: the default network and stimulus-independent thought.' Science, 315(5810), 393-395. doi:10.1126/science.1131295
Christoff, K., Gordon, A. M., Smallwood, J., Smith, R., & Schooler, J. W. (2009). 'Experience sampling during fMRI reveals default network and executive system contributions to mind wandering.' Proceedings of the National Academy of Sciences, 106(21), 8719-8724. doi:10.1073/pnas.0900234106
Morcom, A. M., & Fletcher, P. C. (2007). 'Does the brain have a baseline? Why we should be resisting a rest.' NeuroImage, 37(4), 1073-1082. doi:10.1016/j.neuroimage.2006.09.013
Smallwood, J., Bernhardt, B. C., Leech, R., Bzdok, D., Jefferies, E., & Margulies, D. S. (2021). 'The default mode network in cognition: a topographical perspective.' Nature Reviews Neuroscience, 22, 503-513. doi:10.1038/s41583-021-00474-4
Gazzaley, A., & Rosen, L. D. (2016). The Distracted Mind: Ancient Brains in a High-Tech World. Cambridge, MA: MIT Press