Default Mode Network: Brain Activity at Rest and During Thought
Author: Ukrainian Psychological Hub · Published: October 12, 2026 · Editorial Policy
For complementary background, see Neural Circuits: How Brain Networks Generate Behavior.
The default mode network (DMN) is a distributed system of interconnected brain regions whose activity and functional relationships become especially apparent during quiet wakefulness and many forms of internally directed thought. It is involved in remembering personal experiences, imagining possible futures, considering other people's perspectives, building meaning from knowledge, and the spontaneous movement of thought. The network also participates in some demanding, goal-directed tasks. Its name describes a historical pattern observed in neuroimaging; it does not designate a single brain center, a switch that turns thinking on, or a scientifically established measure of mental health. The modern synthesis by Menon (2023) places the DMN within a dynamic organization of cooperating and competing large-scale networks.
What Is the Default Mode Network?
The DMN, also called the default network, is a large-scale functional brain network. A functional network is identified partly through coordinated variation in activity across physically separated regions. In resting-state functional magnetic resonance imaging (rs-fMRI), the slow fluctuations of the blood-oxygen-level-dependent (BOLD) signal in many DMN regions are statistically correlated, even when a participant is lying still and has no explicit cognitive task to perform. In task-based imaging, many of these regions also show reduced BOLD signal during externally demanding tasks relative to a less constrained reference condition. These two observations helped establish the network's identity. See the foundational synthesis by Buckner, Andrews-Hanna, and Schacter (2008).
Rest in this context means that investigators have withheld a specific task or instructed a participant to remain awake while looking at a fixation point, keeping eyes open, or closing them. Rest does not mean that neurons stop firing, consciousness becomes empty, or mental experience disappears. People may remember, anticipate, worry, plan, notice bodily sensations, or drift among topics. The measurable network pattern is a feature of brain organization; the content of consciousness must be assessed separately.
In a concise scientific formulation, the DMN is an intrinsically organized network with prominent medial and lateral association-cortex components that contributes to multiple internally oriented and conceptually integrative cognitive processes. There is broad evidence for the network's existence and recurring anatomical organization. The proposal that it performs one single psychological operation remains debated. Andrews-Hanna, Smallwood, and Spreng (2014) describe why its cognitive contributions must be divided into interacting components.
How Scientists Discovered the Default Mode Network
During the 1990s, functional imaging experiments repeatedly showed that some regions were more active in passive baseline conditions than during demanding perceptual or attention tasks. This pattern challenged a tempting assumption: that the baseline period was a biologically uninteresting control against which the real action of the brain should be measured. Across tasks, a recurring set of decreases suggested organized activity that was present before participants received experimental instructions.
In 2001, Marcus Raichle and colleagues formalized the idea of a default mode of brain function, drawing on convergent patterns of neuroimaging and the brain's substantial ongoing metabolism. Their paper, A default mode of brain function, is foundational. Subsequent functional connectivity research demonstrated that many of the implicated areas exhibited correlated spontaneous signals. A landmark study of healthy individuals documented reproducible resting-state network patterns across participants, including the default network: Damoiseaux et al. (2006).
Research then shifted from asking why the network appears during rest to asking how it contributes to thought during both rest and tasks. Studies of autobiographical memory, mental simulation, social reasoning, semantic processing, and spontaneous thought established a more extensive research program. The 20-year review by Menon (2023) traces this development while arguing for a network-level synthesis rather than a one-region account.
Where Is the Default Mode Network Located?
The DMN is distributed across both cerebral hemispheres. Its frequently identified cortical hubs include the posterior cingulate cortex and adjacent precuneus on the medial posterior surface, regions of medial prefrontal cortex toward the front of the brain, and lateral parietal association regions around the angular gyri. Depending on the parcellation, experimental state, and analytic approach, descriptions also include medial temporal structures, lateral temporal cortex, parts of the inferior parietal lobule, and other association areas.
Posterior cingulate cortex and precuneus
Posterior midline regions are prominent in many resting-state DMN maps and contribute to communication among broad cortical systems. Research associates them with aspects of internally directed cognition, autobiographical context, memory-related processing, and shifts among functional configurations. This does not justify assigning the whole network's self-related functions to the posterior cingulate cortex alone. The apparent boundaries of this region and its subdivision vary across imaging atlases and individuals.
Medial prefrontal cortex
Medial prefrontal regions are often engaged when people evaluate information in relation to themselves, consider socially relevant meanings, or make judgments with personal significance. Their contribution depends on exact location, task demands, and interactions with other areas. In particular, ventral, anterior, and dorsal medial prefrontal territories are not interchangeable anatomical units. The relationship between medial prefrontal signaling and self-relevant thought was investigated in early experiments, including Gusnard et al. (2001).
Angular gyrus and lateral temporal regions
Regions in the inferior parietal lobule, frequently including the angular gyrus, participate in distributed networks important for integrating information across contexts and modalities. Lateral temporal regions are relevant to conceptual and semantic representations. Evidence therefore connects portions of the DMN to meaning-making and knowledge-based thought, alongside its better-known autobiographical functions. A BOLD correlation cannot by itself determine whether a region supplies a memory representation, integrates it, or responds to processing elsewhere.
Medial temporal lobe and hippocampal interactions
Medial temporal structures, including the hippocampal formation and parahippocampal regions, interact with DMN components during episodic remembering, scene construction, and future simulation. Anatomically defined hippocampal circuits and cortex-wide DMN maps are related but should not be treated as identical systems. Different mapping methods include different medial temporal components. Experimental work on the network's subdivisions highlights these distinctions: Andrews-Hanna et al. (2010).
One Network, Several Subsystems
The DMN has distinguishable subcomponents. A highly influential account identified a midline core involving anterior medial prefrontal and posterior cingulate regions, a medial temporal subsystem implicated in constructing scenes and memory-based simulations, and a dorsomedial prefrontal subsystem associated with aspects of social and conceptual cognition. These functional distinctions emerged from convergent connectivity and task experiments, rather than the claim that each subsystem implements a single isolated mental faculty. Andrews-Hanna et al. (2010) provide the experimental basis for this fractionation.
Later high-resolution mapping of individual brains complicated the picture further: network territories can form fine-grained, adjacent and interdigitated patterns that are blurred by conventional group averaging. Braga and Buckner (2017) demonstrated parallel distributed networks within individuals, emphasizing the precision needed when assigning a cortical parcel to a particular network. A network map derived by averaging many people is a useful scientific summary, yet it may not reproduce one person's exact boundaries.
These findings clarify why an atlas image with several colored blobs is only a visualization of a network model. Borders depend on spatial resolution, recording quality, the chosen number of networks, and preprocessing. The DMN has hubs and relatively stable large-scale motifs, but its detailed topology is not a fixed stencil stamped onto every human brain.
What Does the Default Mode Network Do?
Autobiographical memory and personal context
Remembering a lived event involves reconstructing sensory, spatial, emotional, temporal, and personal information. DMN regions regularly participate in autobiographical retrieval alongside hippocampal and other networks. The network can help integrate pieces of a remembered episode with enduring knowledge about one's life. This interpretation is supported by convergent imaging studies and reviews, including Buckner et al. (2008). A functional imaging association, however, neither identifies an exact stored memory nor demonstrates that a particular DMN hub contains the memory itself.
Imagining the future and constructing possible events
People often plan by simulating scenes that have never occurred. Mental simulation can recombine remembered details into possible future episodes, including where something might happen and how another person might respond. Such tasks recruit overlapping cortical and medial temporal structures associated with the default network. This overlap supports a relationship between remembering and constructing possibilities, while the networks involved in attention, decision-making, and action control also shape useful planning. Spreng et al. (2010) demonstrated how default-network involvement in autobiographical planning can coexist with goal-directed control.
Self-referential processing
The network is frequently recruited during judgments about the self, such as evaluating whether an adjective describes one's personality or reflecting on personally meaningful events. Self-reference is a heterogeneous class of tasks: an experimental judgment about a trait, recognition of one's body, narrative identity, and the subjective experience of being conscious are different questions. DMN involvement in some self-related tasks supports a neural association with these operations; it does not establish that the network alone produces the self or that its activity is an objective meter of self-awareness. See Menon (2023).
Social cognition and thinking about other minds
Making sense of another person's beliefs, intentions, and personal circumstances often requires constructing an internal model of a perspective that is not immediately visible. Studies of mentalizing and social inference implicate portions of the default network, particularly some medial prefrontal, posterior midline, and temporal-parietal territories. These regions also participate in other cognitive functions. A shared neural system is compatible with overlapping representational demands and does not imply that empathy, social understanding, and autobiographical thought are identical processes. Menon (2023) reviews this broader social-cognitive evidence.
Mind-wandering and spontaneous thought
During mind-wandering, thought shifts away from an immediate external task or stimulus toward internally generated material. Some experiences arise spontaneously; others involve deliberate daydreaming or purposeful reflection. Neuroimaging studies using intermittent experience-sampling questions have related DMN activity to reports of task-unrelated thought. Yet thought also depends on executive, attentional, perceptual, and affective systems. Christoff et al. (2016) present mind-wandering as a dynamic phenomenon with variable constraints and multiple contributing networks.
Mind-wandering also has different psychological outcomes. It may support episodic reflection, prospective planning, and creative association, or accompany task distraction and repetitive negative thought. Content, context, and degree of control matter. A restless or distressed inner life cannot be diagnosed from DMN activation. Likewise, a person who daydreams often cannot be presumed to have an unusually strong or unhealthy default network.
Conceptual knowledge, semantic processing, and an internal narrative
Research increasingly connects the DMN with integrating conceptual knowledge, language-related meaning, and representations across time. In his synthesis, Menon (2023) proposes that the network helps coordinate an internal narrative connecting memory, meaning, and personal context. This is a theoretical synthesis supported by diverse results; it is not an established demonstration that one identifiable neural computation explains every DMN-linked task. Alternative frameworks emphasize event construction, situation models, semantic integration, or network-level coordination in different proportions.
What Happens in the Brain at Rest?
Even when the surroundings are quiet and no explicit task is being performed, the brain continues to consume substantial energy and coordinate neural activity. Resting-state networks are patterns of organized intrinsic activity seen against that background. DMN regions show correlated slow BOLD fluctuations; regions associated with other networks show different but partly overlapping patterns. Raichle et al. (2001) introduced the default-mode concept, and Damoiseaux et al. (2006) documented consistent resting-state functional networks.
The everyday question 'What is my brain doing when I do nothing?' therefore has several answers. It maintains bodily regulation and responsiveness, continuously processes information, retrieves and combines memories, and may produce spontaneous conscious experiences. The relative prevalence of any particular operation depends on vigilance, emotional state, sensory conditions, recent experience, and the instructions that define rest. Researchers generally cannot identify the precise content of an individual's thoughts solely from a resting-state DMN connectivity map.
Is the Default Mode Network Active Only When the Mind Is Idle?
No. Early resting-versus-task contrasts made the network famous, but many active tasks recruit DMN components when they demand internally constructed or personally meaningful representations. Autobiographical planning, social inference, narrative comprehension, and episodic retrieval are examples. The useful distinction is the task's cognitive demands and their changing interactions with other networks, rather than a permanent division between rest and work. Spreng et al. (2010) demonstrated network cooperation during goal-directed internal planning.
Task-related decreases in mean BOLD signal are commonly called deactivations. They indicate a difference relative to an experimental baseline, not that an entire network has shut down or that its neurons have become silent. The result may depend on baseline choice, task difficulty, vigilance, and analysis. Local increases and decreases can occur simultaneously in different DMN subdivisions; one average value may obscure functional diversity.
How the DMN Interacts With Attention and Control Networks
The DMN is part of a broader organizational system. Externally oriented attention often involves dorsal attention circuitry, while flexible task management recruits frontoparietal control regions. Salience-associated circuitry, including insular and cingulate territories, is implicated in detecting behaviorally important information and regulating transitions among brain states in some models. These systems do not operate as a universal three-position switch. Patterns of coupling vary with task and analytic design. A complete description requires interactions among networks rather than treating the DMN as the brain's exclusive center for internally generated cognition.
Some studies report anticorrelation between default-network and attention-network BOLD signals. Anticorrelation means that, under a particular preprocessing pipeline and observation window, the signals vary in opposite directions. It does not prove direct neural inhibition between every pair of areas or a psychological war between imagination and concentration. The estimated strength and even appearance of some anticorrelations depend on preprocessing choices, particularly global signal regression; see Murphy and Fox (2017).
In real planning, the DMN may cooperate with control systems. An imagined conversation, for example, can require memory-based simulation, deliberate evaluation of alternatives, monitoring of goals, and later outward action. Task studies such as Spreng et al. (2010) show that frontoparietal-control connectivity can flexibly accompany both internally focused and externally focused planning. Brain organization is dynamic, and functional networks describe recurring coordination patterns rather than rigid boxes of mental abilities.
How Scientists Measure the Default Mode Network
Resting-state functional MRI
Resting-state fMRI is one of the most common methods for mapping the DMN. Its BOLD contrast reflects changes in blood oxygenation influenced by local vascular and metabolic responses associated with neural activity. Investigators sample a time series from each brain location, correct for artifacts, and estimate statistical similarity across regions. Seed-based analyses select a starting region and map areas with correlated signals. Independent component analysis and network-parcellation approaches instead identify distributed patterns according to different mathematical assumptions. Each approach has strengths and can produce somewhat different boundaries.
A correlation between BOLD time series is called functional connectivity. It is not a direct recording of neuronal communication, an anatomical fiber tract, or proof that one region causes another to activate. Functional connectivity may arise from direct and indirect neural coupling, common inputs, changes in arousal, physiology, and measurement artifacts. Researchers use complementary experiments to refine causal hypotheses.
Task-based fMRI
Task fMRI compares BOLD responses under experimentally defined conditions. A scientist might contrast judgments about one's own traits with judgments about another person, or compare autobiographical retrieval with an alternative memory task. This can identify regions associated with carefully defined task contrasts. However, self-related evaluation and autobiographical retrieval also demand language, attention, memory, and decision-making. The statistical contrast isolates an operational difference between conditions, which may not correspond to one indivisible psychological function.
EEG, MEG, intracranial recording, and PET
Electroencephalography (EEG) measures differences in electrical potential at the scalp generated largely by coordinated postsynaptic currents; magnetoencephalography (MEG) measures extremely weak magnetic fields produced by neuronal currents. These methods offer substantially finer temporal resolution than hemodynamic fMRI but infer their neural sources indirectly and with anatomical uncertainty. Intracranial recordings can provide more localized timing and frequency information in selected clinical or research populations, with limited spatial sampling. Positron emission tomography (PET), depending on its tracer, can assess aspects of metabolism, perfusion, or receptor-related processes. These methods offer complementary measurements, and no single method delivers a complete map of the network's cognitive functions.
Experience sampling, behavioral experiments, and perturbation
A brain scan does not report what someone is thinking. Experience sampling can ask participants at unpredictable moments whether they were remembering, planning, focusing, or mind-wandering; the responses are then related to preceding measurements. Behavioral tasks help distinguish cognitive functions, while lesion evidence and carefully designed stimulation or perturbation studies can strengthen causal inference about particular regions or pathways. Each technique has limitations: retrospective reports can be inaccurate, lesions often affect several pathways, and stimulation influences networks beyond the nominal target.
What Resting-State Connectivity Does and Does Not Prove
At the group level, reproducible DMN structure is an established result. At the individual level, the exact network borders, average correlations, and changes over time are sensitive to data quality, preprocessing, head motion, breathing, cardiac signals, vigilance, and scan duration. Power et al. (2012) showed that even small head movements can produce systematic spurious functional-connectivity correlations. Motion correction, physiological modeling, quality control, and replication are essential.
Global signal regression is a particularly consequential processing choice. It removes a mean signal estimated across much of the brain, potentially reducing some widespread artifacts but also changing network relationships. The methodological discussion by Murphy and Fox (2017) explains why preprocessing choices can offer different insights and should be declared transparently. An anticorrelation that appears after one pipeline cannot simply be converted into a claim that two cognitive processes are mutually exclusive.
Reverse inference is another frequent error. The observation that the DMN tends to be engaged during autobiographical thought does not make DMN activation proof that a particular participant is recalling personal events. DMN territories participate in several processes, and group-level averages conceal variation. Similarly, network 'connectivity' does not establish the direction or strength of information transfer, and a change after training does not automatically establish that training caused a specific psychological benefit.
Is the DMN Related to Learning and Memory Consolidation?
The evidence linking internally oriented activity, hippocampal interactions, and memory processing is substantial, but precise consolidation mechanisms remain an active research area. Theoretical accounts propose that spontaneous neural reactivation during rest helps stabilize or integrate recently acquired information. A recent synthesis by Joshi, Tompary, and Kucyi (2026) examines how DMN-supported spontaneous thought could create conditions for reactivation and consolidation, including interactions with the hippocampus.
This synthesis is a developing mechanistic explanation, not a clinical rule that deliberately maximizing DMN activity improves memory or that every episode of daydreaming is consolidating information. Memory outcomes depend on encoding conditions, sleep and wakefulness, retrieval, interference, emotional relevance, and several interacting brain systems. Research examining spontaneous thought and neural reactivation requires careful separation of parallel associations from demonstrated causal pathways.
Default Mode Network, Mental Health, and Brain Disorders
Researchers have reported group-level differences in DMN connectivity or activity across numerous clinical populations, including studies of depression, anxiety-related conditions, neurodevelopmental differences, psychosis, and neurodegenerative disease. The patterns are heterogeneous; they vary by symptom dimensions, sample characteristics, medication status, arousal, comorbidity, methodology, and analytic decisions. Overlap among conditions and among healthy individuals limits the specificity of such measures. A 2023 network review surveys many of these clinical research questions.
An altered group-level resting-state correlation is not a diagnostic criterion, a validated personal brain test, a treatment target established for every condition, or evidence that an individual has an 'overactive DMN.' Clinical diagnosis rests on validated assessment procedures and disorder-specific criteria. Network measures remain research tools unless a particular test has demonstrated suitable individual-level validity, reliability, clinical usefulness, and regulatory standing for its specific intended use.
Rumination is sometimes discussed alongside default-network activity because it involves repetitive internally directed thinking. However, rumination is defined by the content, persistence, and functional consequences of cognition; the DMN is a measured brain-network construct. Neither term can substitute for the other. Mind-wandering, constructive self-reflection, worry, and clinical rumination must be distinguished through behavioral and clinical evidence. Overgeneralizing the literature can encourage self-diagnosis from a popular neuroscience diagram.
Meditation and mindfulness studies also report changes in DMN-associated signals in particular practices and populations. Experimental designs differ in instruction, expertise, attentional targets, and comparison conditions, and imaging findings cannot by themselves demonstrate a specific therapeutic outcome. Meditation is not equivalent to 'turning off' the DMN, and successful treatment for a mental health condition is not established by a single resting-state scan. Clinical implications require adequately controlled intervention studies with meaningful patient outcomes.
Does the Default Mode Network Change With Age, Sleep, and Experience?
Brain-network organization develops over childhood and adolescence, changes across adulthood, and is influenced by neurological and physiological states. Sleep and drowsiness alter patterns of functional coordination, while eye closure, fatigue, and waking vigilance influence resting-state recordings. Consequently, developmental and aging comparisons require controls for motion, vascular physiology, sleepiness, and other confounds. The presence of age-associated differences does not justify treating a single connectivity value as a biological age clock.
Training, repeated experiences, and learning may also be accompanied by changes in resting-state network measurements. Such findings show sensitivity to context and experience but should be interpreted with attention to test-retest reliability and potential confounds. A stronger causal conclusion usually requires randomized interventions, appropriate controls, replication, behavioral outcomes, and independent evidence regarding the biological process that changed.
Common Myths About the Default Mode Network
Myth: The DMN is the part of the brain that thinks
Thought depends on interacting neuronal systems. The DMN contributes to several forms of conceptual, episodic, and internally directed cognition, while sensory, attention, memory, control, language, and motor systems are also indispensable to many thoughts. A brain network should not be renamed a universal thinking center.
Myth: DMN activity means someone is lazy or not concentrating
Resting-state activity is an organized feature of healthy brain physiology, and the network can also participate in difficult intentional mental work. Attention failures may sometimes accompany patterns involving DMN regions, but one correlation cannot evaluate effort, motivation, competence, or a person's current thoughts.
Myth: A powerful DMN makes a person creative, anxious, or mentally ill
These psychological outcomes involve differing processes and contexts. There is no accepted everyday 'DMN strength' score capable of measuring creativity, identifying anxiety, or diagnosing a mental disorder. Correlations reported between connectivity and behavior are usually statistical group findings subject to measurement and replication constraints.
Myth: The DMN and attention networks can never operate together
Internally oriented and externally oriented tasks often produce different group-level patterns, and some estimated signals are anticorrelated. Cooperation also occurs. Spreng et al. (2010) documented coordination with frontoparietal control regions during planned thought. Experimental state and processing methods determine what the observed network relation means.
Myth: A resting-state scan can reveal someone's private thoughts
Resting-state connectivity describes relationships among measured signals. It does not directly transcribe subjective content. Neural decoding research uses specialized models, training data, defined tasks, and statistical inference, with substantial constraints on generalization. A standard DMN connectivity map is not a tool for reading an individual's unreported mental narrative.
Why the Default Mode Network Matters
The DMN provides a research framework for understanding how the brain organizes activity when thought is not driven primarily by immediate sensory demands. It helps connect memory, semantic knowledge, self-relevant judgments, social simulation, and spontaneous cognition within a distributed system that interacts with the rest of the brain. The most useful insight is that large-scale coordination has recognizable structure across rest and active thought. At the same time, the meaning of any signal must be established through experiments that specify tasks, behavior, physiology, and analytic assumptions.
The next generation of work combines more precise mapping of individuals, simultaneous or complementary recording methods, better measurement of ongoing experience, and perturbations capable of evaluating causal hypotheses. In 2026, important questions include the relationship between DMN activity and memory reactivation during rest, how specific subsystems exchange information, and how cognitive states emerge from coordination across networks. Joshi et al. (2026) exemplify this developing research agenda.
Frequently Asked Questions
What does DMN stand for in neuroscience?
DMN stands for default mode network. It refers to a distributed set of brain regions identified through recurring task-related patterns and functional connectivity, especially in studies of quiet waking rest.
Where is the default mode network in the brain?
Commonly identified hubs include medial prefrontal cortex, posterior cingulate cortex and adjacent precuneus, and lateral parietal regions such as the angular gyrus. Other medial temporal and lateral temporal regions participate in various DMN subdivisions. The exact map varies with participant and method.
Is the DMN on when you daydream?
DMN regions are often involved in daydreaming and mind-wandering, but these experiences recruit multiple interacting networks, and DMN activation does not uniquely identify a person's thought content.
Does the default mode network turn off during focused work?
Many external attention tasks are associated with decreases in average DMN BOLD signal relative to rest. Those changes do not mean a complete neural shutdown, and some goal-directed tasks involving internal simulation recruit DMN components.
What is the difference between DMN activation and DMN connectivity?
Activation is usually a relative change in a region's signal across experimental conditions. Functional connectivity is a statistical relationship between signals recorded in two or more regions over time. They measure different properties, and neither alone establishes a specific mental experience or direct causal communication.
Can an overactive default mode network cause anxiety or depression?
Research has identified associations between DMN-related measurements and some clinical symptom patterns, but 'overactive DMN' lacks the precision required for individual diagnosis. Mental health conditions arise within complex biopsychosocial contexts and require condition-specific clinical assessment. Imaging correlations alone do not prove causation.
Can meditation switch off the default mode network?
Meditation studies have reported differing activity and connectivity patterns depending on practice and study design. The evidence does not support a universal claim that meditation switches the DMN off. Subjective benefits and clinical outcomes must be measured independently.
Is the default mode network the seat of consciousness?
DMN research contributes to the neuroscience of internally directed thought and aspects of self-related cognition, but the network's involvement does not settle theoretical questions about consciousness. Consciousness research examines multiple interacting neural systems, levels and contents of experience, and competing explanatory models.
Does everyone have a default mode network?
Distributed patterns corresponding to the default network are repeatedly identified across studies of healthy human brains. There is meaningful variation in their precise anatomy, developmental course, state dependence, and measurement. A single individual's map cannot be interpreted as a personality profile.
Related Articles
For condition-specific discussions of what imaging research can and cannot show, see Depression and the Brain: What Neuroscience Actually Shows and Anxiety and the Brain: Threat Processing, Learning, and What Neuroimaging Can Show. These articles address their own clinical and psychological search intents. The foundational neuroscience articles on neural circuits, salience network, and frontoparietal control network will be connected when their English canonical pages are live.
References
Andrews-Hanna, J. R., Reidler, J. S., Sepulcre, J., Poulin, R., & Buckner, R. L. (2010). Functional-anatomic fractionation of the brain's default network. Neuron, 65(4), 550–562. https://doi.org/10.1016/j.neuron.2010.02.005
Andrews-Hanna, J. R., Smallwood, J., & Spreng, R. N. (2014). The default network and self-generated thought: Component processes, dynamic control, and clinical relevance. Annals of the New York Academy of Sciences, 1316, 29–52. https://doi.org/10.1111/nyas.12360
Braga, R. M., & Buckner, R. L. (2017). Parallel interdigitated distributed networks within the individual estimated by intrinsic functional connectivity. Neuron, 95(2), 457–471.e5. https://doi.org/10.1016/j.neuron.2017.06.038
Buckner, R. L., Andrews-Hanna, J. R., & Schacter, D. L. (2008). The brain's default network: Anatomy, function, and relevance to disease. Annals of the New York Academy of Sciences, 1124, 1–38. https://doi.org/10.1196/annals.1440.011
Christoff, K., Irving, Z. C., Fox, K. C. R., Spreng, R. N., & Andrews-Hanna, J. R. (2016). Mind-wandering as spontaneous thought: A dynamic framework. Nature Reviews Neuroscience, 17, 718–731. https://doi.org/10.1038/nrn.2016.113
Damoiseaux, J. S., et al. (2006). Consistent resting-state networks across healthy subjects. Proceedings of the National Academy of Sciences, 103(37), 13848–13853. https://doi.org/10.1073/pnas.0601417103
Gusnard, D. A., Akbudak, E., Shulman, G. L., & Raichle, M. E. (2001). Medial prefrontal cortex and self-referential mental activity: Relation to a default mode of brain function. Proceedings of the National Academy of Sciences, 98(7), 4259–4264. https://doi.org/10.1073/pnas.071043098
Joshi, D., Tompary, A., & Kucyi, A. (2026). The default mode network: Where spontaneous thought meets memory consolidation. Current Opinion in Behavioral Sciences, 67, 101622. https://doi.org/10.1016/j.cobeha.2025.101622
Menon, V. (2023). 20 years of the default mode network: A review and synthesis. Neuron, 111(16), 2469–2487. https://doi.org/10.1016/j.neuron.2023.04.023
Murphy, K., & Fox, M. D. (2017). Towards a consensus regarding global signal regression for resting state functional connectivity MRI. NeuroImage, 154, 169–173. https://doi.org/10.1016/j.neuroimage.2016.11.052
Power, J. D., Barnes, K. A., Snyder, A. Z., Schlaggar, B. L., & Petersen, S. E. (2012). Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. NeuroImage, 59(3), 2142–2154. https://doi.org/10.1016/j.neuroimage.2011.10.018
Raichle, M. E., et al. (2001). A default mode of brain function. Proceedings of the National Academy of Sciences, 98(2), 676–682. https://doi.org/10.1073/pnas.98.2.676
Spreng, R. N., Stevens, W. D., Chamberlain, J. P., Gilmore, A. W., & Schacter, D. L. (2010). Default network activity, coupled with the frontoparietal control network, supports goal-directed cognition. NeuroImage, 53(1), 303–317. https://doi.org/10.1016/j.neuroimage.2010.06.016