Proceedings of the National Academy of Sciences of the United States of America
May 26, 2009
Kalina Christoff, Alan M Gordon, Jonathan Smallwood et al.
1,832 citations
Mind wandering, which occupies a large portion of waking life, involves parallel recruitment of both default and executive brain networks, two systems previously thought to work in opposition. Using fMRI with experience sampling during a task, activation in default network regions, particularly medial prefrontal cortex, was linked to subjective reports of mind wandering and to performance errors. Executive network recruitment also occurred, especially when participants lacked meta-awareness of their mind wandering. The findings suggest that mind wandering represents a unique mental state where these networks cooperate rather than oppose each other.
Neuroimage
February 25, 2015
Kieran C. R. Fox, R. Nathan Spreng, Melissa Ellamil et al.
717 citations
Spontaneous thought processes such as mind-wandering recruit both default mode network (DMN) regions and many non-DMN areas. A meta-analysis of 24 functional neuroimaging studies found consistent activation in DMN regions including medial prefrontal cortex, posterior cingulate cortex, medial temporal lobe, and bilateral inferior parietal lobule, but also in non-DMN regions such as rostrolateral prefrontal cortex, dorsal anterior cingulate cortex, insula, temporopolar cortex, secondary somatosensory cortex, and lingual gyrus. These findings indicate that DMN activation alone cannot fully explain the neural basis of spontaneous thought; frontoparietal control network and other areas are equally central. Progress in cognitive and clinical neuroscience of spontaneous thought requires a broader view beyond the DMN.
Frontiers in Human Neuroscience
January 1, 2013
Kieran C. R. Fox, Savannah Nijeboer, Elizaveta Solomonova et al.
297 citations
Mind wandering during wakefulness and dreaming during sleep share many features: both involve audiovisual, emotional, fantasy-tinged narratives tied to personal concerns, draw on long-term memory, simulate social interactions, and lack meta-awareness. Comparing neuroimaging data shows that both states activate default mode network regions such as medial prefrontal cortex, medial temporal lobe, and posterior cingulate, which support self-referential thought and memory. However, dreaming appears as an intensified version of mind wandering, with longer, more immersive, and more visual content, along with even deeper deactivation of prefrontal executive regions responsible for cognitive control and metacognition. This suggests dreaming amplifies the same features that distinguish mind wandering from goal-directed thought.
PLoS One
September 25, 2012
Kieran C. R. Fox, Pierre Zakarauskas, Matthew L. Dixon et al.
194 citations
People who meditate for many years are better at accurately reporting their own internal bodily sensations than beginners. The study compared subjective reports of touch sensitivity in 20 body regions during a body-scanning meditation with objective measures of tactile sensitivity from prior research. Expert meditators (1–15,000 hours of experience) showed significantly greater introspective accuracy than novices, and overall meditation experience predicted individual accuracy. The findings suggest that long-term meditation practice may enhance the ability to accurately introspect.
Neuroimage
April 30, 2016
Melissa Ellamil, Kieran C. R. Fox, Matthew L. Dixon et al.
150 citations
By studying experienced mindfulness practitioners who can closely observe when a thought first arises, researchers identified distinct brain regions that activate before versus after a thought emerges. Activity in the medial temporal lobe and inferior parietal lobule peaked before the thought began, suggesting these areas help generate spontaneous thoughts. In contrast, regions including the medial prefrontal, temporopolar, mid-insular, lateral prefrontal, and dorsal anterior cingulate cortices activated at or just after the thought's onset, indicating a role in elaborating or reflecting on the thought. These timing differences may explain why spontaneous thoughts are so varied and how they connect to memory and attention.
Neuroimage
March 8, 2020
Manesh Girn, Caitlin Mills, Leor Roseman et al.
139 citations
Creativity involves alternating between generating ideas and evaluating them. A refined Dynamic Framework of Thought describes creative generation not as one mental state but as a product of many varying states. The psychedelic state is proposed as a mental state with high potential for facilitating creative generation, making psychedelic substances useful experimental tools for studying the neuroscience of creativity.
Neuropsychopharmacology
September 20, 2021
Andre Zamani, Kalina Christoff, Robin Carhart-Harris
50 citations
The prefrontal cortex contains multiple subregions linked to different large-scale brain networks, supporting a wide range of mental phenomena from goal-directed thought and executive functions to mind-wandering and psychedelic experiences. A key dimension distinguishing conscious experiences is the stability or variability of mental states over time, which is central to the dynamic framework of thought (DFT) and the relaxed beliefs under psychedelics (REBUS) model. This review synthesizes these frameworks to explain how prefrontal subregions may differentially contribute to the stability and variability of thought and conscious experience, and suggests future research directions.
Philosophical Transactions of the Royal Society B Biological Sciences
December 13, 2020
Caitlin Mills, Andre Zamani, Rebecca White et al.
27 citations
Abrupt transitions in thought content occur at surprising times but lead to unsurprising content, while wayward transitions occur at surprising times and also lead to surprising content. The authors distinguish these two types using probabilistic and predictive processing concepts, employing transition and occurrence probability metrics. They discuss how these transitions may contribute to mental function and their possible neural implementation.
bioRxiv (Cold Spring Harbor Laboratory)
April 1, 2020
Matthew L. Dixon, Manesh Girn, Kalina Christoff
14 citations
preprint
When people face anxiety, they can either mindfully accept their feelings or elaborate on them through narrative. Mindful acceptance, which broadens attention to present-moment sensations without judgment, was linked to a more globally interconnected brain network, greater integration of interoceptive insula regions, and a shift in striatal connectivity from the default mode to the salience network. It also shifted central hubs from the default network to frontoparietal control network regions, which support metacognitive awareness and cognitive control. These findings suggest that the neural landscape underlying mindful acceptance is more integrated and flexible, aiding emotion regulation.
arXiv (Cornell University)
April 20, 2017
Kieran C. R. Fox, Kalina Christoff
7 citations
Spontaneous thought—including mind-wandering, daydreaming, and creative insight—remains poorly understood despite being a universal mental experience. This volume assembles perspectives from neuroscience, psychology, philosophy, phenomenology, history, education, contemplative traditions, and clinical practice to address fundamental questions: what spontaneous thought is, why the brain engages in it, when it occurs, and how the brain generates, elaborates, and evaluates its own spontaneous creations. It represents the first effort to integrate such diverse viewpoints to begin answering these questions scientifically.
March 14, 2023
Remington Mallett, Yasmeen Nahas, Kalina Christoff et al.
2 citations
preprint
The flow of thought is persistent and sometimes unpredictable, and understanding what influences how thoughts shift from moment to moment has implications for disorders like schizophrenia and recurrent nightmares. This study examined whether cognitive control limits moment-to-moment content shifts across sleep and wakefulness. Thought variability was measured as semantic incoherence between sequential thought phrases in dreaming and waking reports. During wakefulness, on-task conditions showed reduced thought variability compared to off-task conditions, and thought variability was greater when thoughts wandered freely. During sleep, lucid dreams, which involve higher cognitive control, showed reduced thought variability compared to non-lucid dreams. These results suggest that cognitive control may limit thought variability across the 24-hour cycle of thought generation, consistent with the Dynamic Framework of Thought.
November 13, 2025
S. Sheth, Matthew L. Dixon, Lawrence M Ward et al.
1 citation
preprint
Human consciousness typically produces an ongoing, seemingly continuous stream of thought. Prior work has examined the neural correlates of thought content, but less is known about thought dynamics—how thoughts change over time. The Dynamic Framework of Thought proposes a neural model of how large-scale brain networks contribute to freely moving (spontaneous) versus deliberately constrained thought. In a neurophenomenological study, participants reported their thought stream in real time during fMRI scanning. Freely-moving thought was associated with greater activation of the default network (especially the medial-temporal subcomponent) than the frontoparietal control network, while deliberately constrained thought showed the opposite pattern. Distinct thought dynamics were also reflected in changes in functional coupling within and between these networks.
arXiv (Cornell University)
March 21, 2016
Kieran C. R. Fox, Matthew L. Dixon, Savannah Nijeboer et al.
Different styles of meditation produce distinct patterns of brain activity. A meta-analysis of 78 functional neuroimaging studies (31 experiments, 527 participants) found that four common meditation styles—focused attention, mantra recitation, open monitoring, and compassion/loving-kindness—reliably activate and deactivate different brain regions, with suggestive differences for three others (visualization, sense-withdrawal, and non-dual awareness). Some areas like the insula and anterior cingulate cortex are recruited across multiple techniques, but convergence is the exception. Medium effect sizes for activations (d = .59) and deactivations (d = -.74) suggest practical significance. The findings support the neurophysiological dissociability of meditation practices while highlighting methodological concerns.
arXiv Preprint Archive
May 23, 2016
Kieran C. R. Fox, Manesh Girn, Cameron C. Parro et al.
Psychedelic substances produce radical psychological effects and have been surrounded by political and legal controversy since their widespread adoption about 50 years ago. This review examines functional neuroimaging studies that investigate the neural correlates of the psychedelic experience, highlighting connections with the psychological and neural bases of creativity, daydreaming, and dreaming. The authors synthesize findings from these imaging investigations to show how brain activity during psychedelic states overlaps with patterns seen during imaginative and creative thought processes.
arXiv Preprint Archive
May 11, 2016
Kieran C. R. Fox, Yoona Kang, Michael Lifshitz et al.
Meditation and hypnosis can change how thoughts flow automatically, potentially making thinking more flexible and less rigid. Three mechanisms are proposed: reducing the tendency for thoughts to chain together automatically; making thought chains more varied and less habitual; and creating new, intentionally chosen mental habits. Evidence from behavioral and cognitive neuroscience shows these practices influence internal cognition, with possible benefits for mental adaptability.
arXiv Preprint Archive
March 21, 2016
Kieran C. R. Fox, Matthew L. Dixon, Savannah Nijeboer et al.
Meditation comprises diverse mental practices with distinct strategies. A meta-analysis of 78 neuroimaging studies (527 participants) found reliably different brain activation patterns for four common meditation styles—focused attention, mantra recitation, open monitoring, and compassion/loving-kindness—and suggestive differences for three others. Some brain regions (insula, pre/supplementary motor cortices, dorsal anterior cingulate cortex, frontopolar cortex) were recruited across multiple techniques, but convergence was the exception. Effect sizes were medium for both activations (d = .59) and deactivations (d = -.74), indicating potential practical significance. The findings support the neurophysiological dissociability of meditation practices while highlighting methodological concerns and future research directions.