Default Mode Network

The default mode network (DMN) is a large-scale brain system that increases its activity during passive states and decreases its activity during external goal-directed tasks. The network was identified by Marcus E. Raichle and colleagues at Washington University in St. Louis in a 2001 paper in Proceedings of the National Academy of Sciences that proposed an organized baseline state of brain function. The principal anatomical components are: medial prefrontal cortex (anterior midline), posterior cingulate cortex and precuneus (posterior midline), lateral parietal cortex (angular gyrus and adjacent inferior parietal lobule), and medial temporal lobe structures (including the hippocampal formation). Functional connectivity among these regions is robust at rest and replicates across populations, species, and imaging modalities.

The original framing positioned the DMN as a “task-negative” network — defined by what deactivates during external tasks rather than what it does. The contemporary picture is substantially more developed. The 2008 synthesis by Buckner, Andrews-Hanna, and Schacter in Annals of the New York Academy of Sciences consolidated thirty years of evidence showing that the DMN is preferentially active during specific internal cognitive operations: autobiographical memory retrieval, envisioning the future, theory of mind (inferring others' perspectives), and self-referential processing. The network is not merely deactivated by external tasks; it is recruited by tasks that require internally focused thought. This shift from “task-negative” to “internal mentation network” is one of the more substantial conceptual revisions in cognitive neuroscience of the past two decades.

The DMN's clinical relevance is well-established. Altered DMN connectivity and reactivity are observed in major depressive disorder (hyperconnectivity and failure to downregulate), Alzheimer's disease (decreased connectivity, particularly in the posterior components), schizophrenia, autism spectrum disorder, and ADHD. The network has become a major analytic target in resting-state functional MRI research and a substrate for understanding several psychiatric and neurodegenerative conditions.

The DMN matters at three substantive levels with substantial supporting evidence.

For understanding internal cognition. The DMN's functional profile maps onto the kinds of thought that constitute much of waking mental life when not focused on external tasks: remembering personally significant events, imagining future scenarios, considering what others are thinking, and reflecting on the self. These are not idle activities; they are central to identity formation, social functioning, autobiographical continuity, and prospective planning. The DMN gives this internal cognition a substrate in the brain that can be measured, perturbed, and modeled. Before the network's identification, the “wandering mind” was treated as either uninteresting (a baseline against which task-related activity was measured) or pathological (rumination, intrusive thinking). The contemporary view recognizes spontaneous internal cognition as a major neural function in its own right.

For understanding psychiatric and neurodegenerative conditions. The DMN is altered in a striking range of clinical conditions. Sheline and colleagues (2009) showed in PNAS that major depression involves both heightened DMN activity during negative stimulus processing and a failure to normally downregulate DMN activity, providing a network framework for the disordered self-referential thought characteristic of depression. In Alzheimer's disease, the posterior components of the DMN (particularly the posterior cingulate and precuneus) show some of the earliest pathological changes, with reduced connectivity preceding overt cognitive decline. Altered DMN connectivity is observed in schizophrenia, autism spectrum disorder, ADHD, and post-traumatic stress disorder. The network has become one of the most actively studied substrates in psychiatric neuroscience.

For the basic neuroscience of large-scale brain organization. The DMN was one of the first reliably identified large-scale resting-state networks and helped establish that the brain's functional organization can be characterized in terms of distributed networks rather than only as task-activated regions. This shift has been substantial: contemporary cognitive neuroscience routinely characterizes brain function in terms of multiple coexisting large-scale networks (the DMN, the salience network, the executive control network, sensorimotor networks, etc.) with specific patterns of cooperation and competition. The DMN is the most-cited and best-characterized of these networks and serves as a methodological exemplar for the broader research program.

The DMN was identified by Marcus E. Raichle and colleagues at Washington University in St. Louis. The pivotal paper was Raichle, MacLeod, Snyder, Powers, Gusnard, and Shulman (2001) in Proceedings of the National Academy of Sciences, titled “A default mode of brain function.” The paper proposed that the brain has an organized baseline state — a default mode — that is suspended when goal-directed external tasks are performed and that this baseline state involves a specific set of regions with consistent activity patterns. The argument was anchored in measurements of brain oxygen extraction fraction (OEF) and consistent patterns of task-induced deactivation across many different cognitive tasks.

The intellectual prehistory matters. As Raichle and Snyder noted in their 2007 retrospective in NeuroImage, evidence for the network had been accumulating for decades before the 2001 synthesis. Early regional blood-flow studies by David Ingvar in the 1970s had identified a baseline pattern of frontal activity. Subsequent imaging research consistently identified a similar set of regions that decreased their activity during task performance compared with passive baseline conditions. The 2001 paper consolidated these observations into a coherent theoretical framework with a specific claim: the deactivation pattern reflects an organized baseline state, not noise or methodological artifact.

The next conceptual milestone was the recognition that the DMN is not merely defined by what suppresses it. Buckner, Andrews-Hanna, and Schacter (2008) in Annals of the New York Academy of Sciences synthesized thirty years of evidence to argue that the DMN is preferentially active during specific internal tasks: autobiographical memory retrieval, future thinking, theory of mind, and self-referential processing. This reframing from “task-negative” to “internal mentation” was a substantial conceptual development. The network was no longer just what turned off during work; it was what turned on during internally directed thought.

The next major development was functional fractionation. Andrews-Hanna, Reidler, Sepulcre, Poulin, and Buckner (2010) in Neuron used functional connectivity analysis to show that the DMN is not a unitary system but comprises three distinguishable subsystems with distinct functional profiles. This functional-anatomic fractionation has been refined in subsequent work but the three-subsystem framework remains the dominant contemporary model. Andrews-Hanna's 2012 review article in The Neuroscientist elaborated the framework into the form that is now standard.

Clinical applications followed rapidly. Greicius, Srivastava, Reiss, and Menon (2004) showed in PNAS that DMN activity distinguishes Alzheimer's disease from healthy aging, opening a line of research that has documented DMN alterations across multiple neurodegenerative and psychiatric conditions. Sheline and colleagues (2009) extended the framework to major depression, showing both heightened reactivity and failed downregulation. By the mid-2010s, DMN alterations were documented across a broad range of clinical conditions, and the network had become one of the most actively studied substrates in psychiatric and cognitive neuroscience.

The DMN spans midline and lateral cortical regions with consistent functional connectivity. The contemporary framework, drawing on Andrews-Hanna and colleagues (2010) and Andrews-Hanna (2012), identifies three functional subsystems within the broader network.

1. Midline core hubs. The anterior medial prefrontal cortex (aMPFC) and posterior cingulate cortex (PCC) constitute the network's core. These hubs show the strongest functional connectivity to all other DMN regions and are preferentially recruited during self-referential processing — tasks that ask the participant to consider personal traits, autobiographical information, or their own preferences. The PCC and precuneus together form the most consistently identified posterior component and show some of the earliest pathological changes in Alzheimer's disease. The aMPFC is preferentially active during evaluative self-judgment and emotional processing of personally relevant content.
2. The dorsal medial prefrontal cortex (dMPFC) subsystem. Includes the dMPFC, temporoparietal junction (TPJ), lateral temporal cortex, and temporal pole. This subsystem is preferentially recruited during social cognition tasks — theory of mind, perspective-taking, mentalizing, and considering other minds. Activity in this subsystem is increased during tasks that require inferring others' beliefs, intentions, or emotional states. The dMPFC subsystem's connectivity to the temporoparietal junction is the most consistently replicated finding linking the DMN to social-cognitive function.
3. The medial temporal lobe (MTL) subsystem. Includes the hippocampal formation, parahippocampal cortex, retrosplenial cortex, posterior inferior parietal lobule, and ventromedial prefrontal cortex. This subsystem is preferentially recruited during memory-based scene construction — recalling specific autobiographical events, imagining future scenarios, navigating mentally through spaces, and constructing detailed internal representations. The MTL subsystem's involvement explains the DMN's consistent activation during autobiographical memory retrieval and future thinking, which both require constructing detailed scene-based mental representations.

The three subsystems are functionally connected and frequently co-active, but they show distinguishable activity patterns that vary with task demands. Social-cognitive tasks preferentially recruit the dMPFC subsystem; autobiographical memory and future thinking preferentially recruit the MTL subsystem; self-referential evaluation recruits the midline core. This functional differentiation is important because it explains why the DMN can be involved in such a diverse range of cognitive operations: the network is not a single processor but a coordinated system of subsystems, each with distinct functional specializations.

The DMN typically shows anticorrelated activity with the dorsal attention network and other task-positive systems. During external goal-directed tasks, the task-positive networks increase activity while the DMN decreases activity; during passive rest or internally directed cognition, the pattern reverses. This anticorrelation pattern was one of the original observations that established the DMN's existence and remains one of its most reliable signatures. However, the anticorrelation is not absolute: the DMN can be co-active with task-positive networks during tasks that require integration of internal and external information, such as evaluating an external stimulus in terms of personal relevance.

Cross-species evidence supports the network's evolutionary depth. DMN-like networks have been identified in non-human primates and rodents using similar resting-state methods, suggesting the network is an evolutionarily conserved feature of mammalian brain organization rather than a uniquely human feature. The functional implications of this conservation are still being worked out, but the anatomical homology is well-established.

The DMN is measured through several complementary methods, with the choice depending on whether the research question concerns the network's spatial organization, its functional connectivity, or its task-related activity patterns.

Resting-state functional MRI (rs-fMRI). The dominant contemporary method. Participants lie in the scanner without performing any task, typically with eyes open and fixated on a crosshair, while BOLD signal is recorded for 5-15 minutes. Functional connectivity analysis identifies regions whose activity fluctuates together over time; the DMN is one of the most consistently identified networks across populations and analytic approaches. Standard methods include seed-based correlation analysis (using one DMN region as a seed and identifying correlated regions) and independent component analysis (ICA, which decomposes the data into independent spatial-temporal components). The DMN is reliably identified across these methods, supporting its status as a real network rather than a methodological artifact.

Task-induced deactivation paradigms. The original method, still useful for confirming DMN identification and for studying state-dependent modulation. Participants perform external goal-directed tasks (working memory, attention, sensory processing) and the regions showing consistent deactivation across tasks are identified as the DMN. This method was central to the original 2001 paper and remains useful for studying how task demands modulate DMN activity.

Task-induced activation paradigms. Used to study what the DMN actively does rather than what suppresses it. Participants perform internally directed tasks (autobiographical memory recall, future thinking, theory of mind, self-referential judgment) and DMN regions showing increased activity are identified. These paradigms were central to the conceptual shift from “task-negative” to “internal mentation network.”

Diffusion-weighted imaging and structural connectivity. Used to characterize the anatomical white-matter connections underlying DMN functional connectivity. Diffusion tensor imaging and diffusion spectrum imaging show that the major DMN regions are connected by identifiable white-matter tracts, providing anatomical evidence for the network's organizational coherence. Structural connectivity analyses have become increasingly important in linking DMN findings to disease processes, particularly in conditions involving white-matter pathology.

Electrophysiological methods. EEG and MEG can identify DMN-like activity patterns through analysis of low-frequency power fluctuations and connectivity in specific frequency bands. These methods complement fMRI by providing better temporal resolution, though their spatial resolution is lower. Convergence between hemodynamic and electrophysiological identifications of the DMN supports the network's status as a real neural phenomenon rather than a methodological artifact of any particular imaging modality.

What the LBL Brain Age Index captures. The Brain Age Index estimates the modifiable portion of the brain-age gap from 25 self-reported questions, 23 of them scored in six domains of potentially modifiable risk factors (Cardiometabolic, Lifestyle & Movement, Sleep & Recovery, Cognitive Reserve, Mind & Mood, and Sensory & Protective), drawing on the 2024 Lancet Commission and the broader cognitive aging and brain-age-gap neuroimaging literatures. It includes no cognitive tasks and does not measure DMN connectivity, which would require neuroimaging; for users interested in direct DMN measurement, neuroimaging methods are necessary.

The DMN sits in a busy conceptual neighborhood and is commonly confused with related but distinct ideas.

• vs. resting-state networks generally. The DMN is one of several reliably identified resting-state networks, alongside the salience network (anchored in anterior insula and dorsal anterior cingulate), the executive control or frontoparietal network, the dorsal attention network, sensorimotor networks, and visual networks. These networks are partially independent and partially interactive. “Default mode network” refers specifically to the network involving medial prefrontal cortex, posterior cingulate, lateral parietal cortex, and medial temporal lobe; it should not be used as a generic term for resting-state activity.
• vs. mind-wandering. Mind-wandering is the phenomenological experience of spontaneous, internally directed thought that is not currently linked to external task demands. The DMN is the proposed neural substrate of mind-wandering, but the two are not identical. Mind-wandering can occur with varying degrees of DMN activity, and DMN activity can occur during deliberately directed internal cognition (such as instructed autobiographical recall) that would not typically be called mind-wandering. The empirical relationship is robust but not deterministic.
• vs. consciousness. The DMN is not the “seat of consciousness” despite occasional popular claims. Consciousness involves multiple coordinated brain systems; the DMN appears important for specific aspects of self-referential and autobiographical experience but is not the whole picture. The DMN is reduced in many unconscious states (deep sleep, anesthesia) but is also reduced during many engaged conscious activities (external attention, focused task performance). The relationship between the DMN and consciousness is real but more complicated than the popular framing suggests.
• vs. self. The DMN is consistently active during self-referential processing, and some researchers have proposed it as a neural substrate of self-related cognition. This is a reasonable hypothesis with empirical support, but the equation of DMN with self is too simple. Self-referential processing is one of several functions the DMN supports; the network is also active during memory retrieval, future thinking, and theory of mind, which are not straightforwardly self-related. The DMN-self relationship is one piece of a more complex picture.
• vs. neuroplasticity. Neuroplasticity refers to the brain's capacity for structural and functional change in response to experience, injury, or training. DMN connectivity can change with training (meditation, cognitive interventions) and with disease (Alzheimer's, depression), making the DMN a substrate on which neuroplasticity operates. The two concepts are at different levels: neuroplasticity is a general capacity; the DMN is a specific network that exhibits that capacity.
• vs. cognitive reserve. Cognitive reserve is the capacity to maintain cognitive function despite age-related or pathological brain changes. DMN integrity is one factor that contributes to cognitive reserve, particularly in aging populations. The Alzheimer's literature shows that DMN alterations precede overt cognitive decline, and DMN integrity is one predictor of trajectories. The constructs are conceptually distinct but empirically related.
• vs. rumination. Rumination is the repetitive, often unwanted internal focus on negative content that is characteristic of depression and several anxiety conditions. The DMN is implicated in rumination through its general role in internally directed thought, and DMN hyperconnectivity is observed in depression. But rumination is not synonymous with DMN activity; it is a specific maladaptive form of internal cognition that involves the DMN among other systems. The reduction of rumination is a treatment target in CBT and mindfulness-based therapies; the DMN is a relevant substrate but not the whole story.

Example 1 — The commute

A 41-year-old accountant drives the same 35-minute route to work each morning. The drive is familiar and largely automatic. By the time she arrives at the office, she has often had no conscious awareness of the last few miles — her attention was elsewhere. Where was it? Most commonly: thinking about a conversation with her sister yesterday that she has been turning over, considering whether to apply for a different role at her firm, imagining what her daughter's presentation tomorrow will be like, replaying a moment from last weekend that she found awkward.

All of these activities engage the DMN. The autobiographical memory of yesterday's conversation activates the MTL subsystem and core hubs. The future-oriented consideration of the role application engages the MTL subsystem (constructing the future scenario) and the dMPFC subsystem (considering how others would respond). The mental simulation of her daughter's presentation engages the dMPFC subsystem (theory of mind: what is her daughter feeling and thinking) and the MTL subsystem (constructing the imagined scene). The replaying of the awkward moment involves self-referential processing in the core hubs. The DMN is doing genuine cognitive work during the “automatic” commute; the work is just internally rather than externally directed. This is much of what waking mental life looks like when external task demands are minimal.

Example 2 — The Alzheimer's trajectory

A 72-year-old retired engineer has been showing subtle changes over the past two years. He has trouble recalling specific recent events (what he had for breakfast yesterday, what was discussed at last week's family gathering). He gets confused about the temporal order of recent experiences. Family members notice he has become quieter at social gatherings and seems to track the back-and-forth of group conversation less well than before. His neurologist orders a comprehensive evaluation including structural and functional brain imaging.

The imaging shows what the Alzheimer's literature would predict. Resting-state fMRI reveals reduced functional connectivity within the DMN, particularly involving the posterior cingulate cortex and the medial temporal lobe components. The connectivity changes are bilateral and consistent with the pattern documented across many Alzheimer's neuroimaging studies. The cognitive profile maps onto the affected DMN subsystems: the MTL subsystem's role in episodic and autobiographical memory aligns with his difficulty recalling specific recent events; the dMPFC subsystem's role in social cognition aligns with his reduced engagement in group conversation. The DMN alterations precede the more dramatic clinical changes that will likely follow and are part of the rationale for early biomarker-based approaches to Alzheimer's research and intervention.

The DMN is one of the most reliably identified and well-characterized large-scale brain networks, but several real qualifications are worth naming.

• The original “task-negative” framing was incomplete. The contemporary literature has moved substantially beyond the original 2001 framing of the DMN as defined by deactivation. The 2008 Buckner, Andrews-Hanna, and Schacter synthesis explicitly argued that the network is preferentially active during internal cognitive tasks rather than just deactivated by external ones. Popular sources still occasionally describe the DMN as “the network that turns on when you're doing nothing,” which is misleading. The DMN is doing something specific during internal cognition; it is not idle.
• Functional interpretations are still being refined. The three-subsystem framework (midline core for self-referential processing, dMPFC subsystem for social cognition, MTL subsystem for memory-based scene construction) is the dominant contemporary view, but it is not the final word. Subsequent fMRI studies have proposed alternative functional fractionations, and the precise mapping from anatomy to function is still being refined. Researchers using the three-subsystem framework should treat it as the current best model rather than as established fact.
• Clinical implications are robust at the group level but variable at the individual level. Group-level DMN alterations in conditions like depression, Alzheimer's disease, and schizophrenia are well-replicated. But individual-level DMN measurements are noisier and less directly diagnostic. A single resting-state scan from one person cannot reliably diagnose depression or predict Alzheimer's onset, even though group differences are robust. Individual-level applications require careful validation and typically depend on multivariate prediction methods rather than single-network metrics.
• Causal interpretation is limited. DMN alterations correlate with many clinical conditions, but causal direction is often unclear. Reduced DMN connectivity in Alzheimer's could be a consequence of underlying pathology, an early contributor to cognitive decline, or a parallel process. Hyperconnectivity of the DMN in depression could be cause, consequence, or correlate of depressive cognition. Establishing causality requires longitudinal designs, intervention studies, or experimental manipulation (e.g., transcranial magnetic stimulation to specific DMN nodes), and this work is ongoing rather than settled.
• Cross-cultural and developmental variability is incompletely characterized. Most DMN research has been conducted in WEIRD samples (Western, educated, industrialized, rich, democratic). Cross-cultural studies suggest the network exists across populations, but the functional implications of cultural differences in self-construal (independent vs. interdependent), cognitive style, and content of internal cognition are not fully worked out. Similarly, the developmental trajectory of the DMN from childhood through adolescence to adulthood is an active research area; the network does not appear in mature adult form until late adolescence or early adulthood.
• The relationship to consciousness is more complicated than popular accounts suggest. The DMN has sometimes been described as the “neural correlate of self” or even “the seat of consciousness.” These framings overstate what the evidence supports. The DMN is important for specific aspects of self-referential and autobiographical experience but is not the whole of consciousness, and reduced DMN activity during focused external attention does not mean reduced consciousness. The popular conflation of DMN with self or consciousness obscures the more interesting actual picture.
• Meditation and intervention effects are real but modest. The popular literature on meditation has emphasized DMN downregulation as a putative mechanism for the benefits of meditation practice. Some empirical work supports this (e.g., reduced DMN activity during certain meditation states, altered functional connectivity in experienced meditators), but effect sizes are typically modest and the relationship between DMN changes and the subjective benefits of meditation is not fully established. The popular framing often overstates the strength and specificity of the evidence.

The LBL Brain Age Index estimates the modifiable portion of the brain-age gap from 25 self-reported questions about potentially modifiable risk factors, scored in six domains, drawing on the 2024 Lancet Commission and the brain-age-gap neuroimaging literature. It includes no cognitive tasks and does not measure DMN connectivity, which would require neuroimaging. For users interested in direct DMN measurement, structural and functional neuroimaging methods are necessary — the BAI is a risk-factor estimate, not a measure of brain networks.

The default mode network (DMN) is a large-scale brain system that increases its activity during passive states and during internally directed cognitive tasks. The network was identified by Raichle and colleagues (2001) in PNAS, with subsequent synthesis by Buckner, Andrews-Hanna, and Schacter (2008) establishing that the network is preferentially recruited during specific internal cognitive operations: autobiographical memory retrieval, envisioning the future, theory of mind, and self-referential processing. The principal anatomical components are medial prefrontal cortex, posterior cingulate cortex and precuneus, lateral parietal cortex, and medial temporal lobe structures including the hippocampal formation. Andrews-Hanna and colleagues (2010) identified three functional subsystems within the broader network: midline core hubs (self-referential processing), the dMPFC subsystem (social cognition), and the MTL subsystem (memory-based scene construction). Clinical relevance is well-established: Sheline and colleagues (2009) showed both heightened reactivity and failed downregulation of the DMN in major depression, while Alzheimer's disease shows characteristic reduced posterior DMN connectivity that precedes overt cognitive decline. Altered DMN connectivity is also documented in schizophrenia, autism spectrum disorder, ADHD, and post-traumatic stress disorder. The principal contemporary qualifications are that the original “task-negative” framing has been substantially revised toward an “internal mentation network” framing, that functional interpretations of the subsystems are still being refined, and that individual-level diagnostic applications require careful validation despite robust group-level findings.

This entry is intended as a citable scholarly reference. Choose the format that matches your context. The retrieval date should reflect when you accessed the page, which may differ from the entry's last-reviewed date shown above.

LifeByLogic. (2026). Default Mode Network: Raichle, Subsystems & Disease. https://lifebylogic.com/glossary/default-mode-network/

LifeByLogic. "Default Mode Network: Raichle, Subsystems & Disease." LifeByLogic, 25 Oct. 2026, https://lifebylogic.com/glossary/default-mode-network/.

LifeByLogic. 2026. "Default Mode Network: Raichle, Subsystems & Disease." October 25. https://lifebylogic.com/glossary/default-mode-network/.

@misc{lbldefaultmodenetwork2026,
  author = {{LifeByLogic}},
  title = {Default Mode Network: Raichle, Subsystems & Disease},
  year = {2026},
  month = {oct},
  publisher = {LifeByLogic},
  url = {https://lifebylogic.com/glossary/default-mode-network/},
  note = {Accessed: 2026-10-25}
}