Science Advances
February 1, 2019
Athena Demertzi, Enzo Tagliazucchi, Stanislas Dehaene et al.
545 citations
Consciousness depends on the brain's ability to sustain rich, dynamic patterns of signal coordination. Using functional magnetic resonance imaging, a complex pattern of coordinated and anticoordinated signals characterized healthy individuals and minimally conscious patients. Unresponsive patients showed low interareal phase coherence mainly mediated by structural connectivity, with fewer transitions between patterns. This complex pattern was also seen in patients with covert cognition who could perform mental imagery tasks, validating its link to consciousness. Anesthesia increased the probability of the less complex pattern to levels seen in unresponsive patients, confirming its role in unconsciousness. These results establish generalizable fingerprints of conscious and unconscious states after brain damage.
Communications Biology
June 29, 2022
Anira Escrichs, Yonatan Sanz Perl, Carme Uribe et al.
67 citations
Different brain states—resting, meditating, deep sleep, and disorders of consciousness after coma—are underpinned by distinct spatiotemporal dynamics that can be characterized using turbulence theory. Non-conscious states tend to be more synchronous, while conscious states are more asynchronous, but the work goes beyond this simple dichotomy. A model-free analysis of human neuroimaging data applied Kuramoto's turbulence framework with coupled oscillators and measured information cascades across spatial scales. A complementary model-based approach used exhaustive computer simulations of whole-brain models fitted to those measures to study information encoding. The framework shows that turbulence theory provides excellent tools for describing and differentiating between brain states.
Network Neuroscience
December 23, 2022
Gustavo Deco, Yonatan Sanz Perl, Laura de la Fuente et al.
43 citations
The default mode network (DMN) may coordinate the recruitment and scheduling of brain networks for solving cognitive tasks, supported by evidence that DMN regions are physically and functionally distant from sensorimotor areas. Using a thermodynamics-inspired, deep learning-based Temporal Evolution NETwork (TENET) framework to measure the 'arrow of time'—a marker of nonreversibility and hierarchy in brain signals—analysis of Human Connectome Project data from nearly a thousand participants suggests the DMN orchestrates hierarchy levels that shift between rest and seven cognitive tasks. This hierarchy differs significantly in health versus neuropsychiatric disorders, offering insights into brain dynamics for cognition.
Interface Focus
April 14, 2023
Elvira G-Guzmán, Yonatan Sanz Perl, Jakub Vohryzek et al.
34 citations
Living systems must constantly work against equilibrium to survive, a property that can be measured through temporal asymmetry in brain signals. Using statistical physics, researchers analyzed reversibility in functional magnetic resonance imaging data from patients with disorders of consciousness. They found that decreased asymmetry and reduced non-stationarity in brain signals characterize impaired consciousness states, consistent with previous findings in sleep and anesthesia. The work aims to identify biomarkers for patient improvement and classification, and to deepen mechanistic understanding of consciousness disorders.
Communications Biology
September 30, 2024
Pablo Castro, Andrea I. Luppi, Enzo Tagliazucchi et al.
21 citations
Brain activity during unconsciousness, whether from general anaesthesia or slow wave sleep, is dominated by a recurrent functional connectivity pattern primarily mediated by structural connectivity and with a reduced capacity to transition to other patterns. Conscious awareness is characterized by richer brain dynamics measured by entropy and a greater repertoire of connectivity states. These findings suggest that the dynamic exploration of functional connectivity states provides robust and generalizable markers for the state of consciousness across different conditions.
bioRxiv (Cold Spring Harbor Laboratory)
July 4, 2021
Gustavo Deco, Yonatan Sanz Perl, Jacobo Sitt et al.
16 citations
preprint
The brain operates far from equilibrium due to forces from the body and environment. Using a deep learning framework called Temporal Evolution NETwork (TENET) applied to large-scale neuroimaging data from over a thousand participants, researchers show that the arrow of time—a thermodynamic measure of non-reversibility—varies with cognitive state. Non-equilibrium levels are higher during tasks than at rest and differ across seven distinct cognitive tasks. In a separate dataset of 265 participants, the framework distinguishes resting-state brain activity in healthy controls from that in schizophrenia, bipolar disorder, and ADHD, with higher non-equilibrium levels in health. This thermodynamics-based approach offers new insights into how brain dynamics orchestrate behavior-environment interactions.
Cortex; a journal devoted to the study of the nervous system and behavior
September 1, 2021
Santiago Alcaide, Jacobo Sitt, Tomoyasu Horikawa et al.
13 citations
Waking up from early sleep involves a two-stage brain process. First, subcortical and sensorimotor structures activate before most cortical regions, followed by rapid whole-brain activation, with frontal regions engaging slightly later. A second, slower stage may then occur, where cortical regions activate before subcortical structures and the cerebellum. This pattern suggests subcortical structures play a key role in initiating and maintaining conscious states.
PNAS nexus
December 1, 2024
Emilia Fló, Laouen Belloli, Álvaro Cabana et al.
10 citations
Directing attention toward the body's internal signals (interoception) versus external sounds (exteroception) produces distinct brain activity patterns. Exteroceptive attention flattened overall brain wave power, while interoceptive attention reduced brain signal complexity, increased frontal connectivity and theta oscillations, and modulated the heartbeat-evoked potential (HEP). Classifiers using HEP features correctly identified the attentional state in 17 of 20 healthy participants; power spectral density features classified all 20. In five brain-injured patients, one with unresponsive wakefulness syndrome and one with locked-in syndrome showed willful modulation of the HEP, suggesting they could follow commands. These findings highlight how attention shapes sensory processing and may aid diagnosis in disorders of consciousness.
bioRxiv (Cold Spring Harbor Laboratory)
August 19, 2024
Naji Alnagger, Paolo Cardone, Charlotte Martial et al.
3 citations
preprint
Disorders of consciousness, such as unresponsive wakefulness syndrome (UWS) and minimally conscious state (MCS), have few treatments. Using whole-brain computational models built from individual patients' fMRI and diffusion-weighted imaging data, this virtual clinical trial simulated the effects of LSD and psilocybin. The psychedelics shifted the brains of patients with disorders of consciousness closer to a critical dynamical state, with a larger effect in MCS patients. In UWS patients, the treatment response depended on structural connectivity, whereas in MCS patients it aligned with baseline functional connectivity. These results provide a computational foundation for considering psychedelics in treating disorders of consciousness and highlight the role of computational modeling in drug discovery and personalized medicine.
Advanced Science
November 20, 2025
Paolo Cardone, Charlotte Martial, Yonatan Sanz Perl et al.
2 citations
Simulated administration of LSD and psilocybin in computational models of patients with disorders of consciousness (DoC), including unresponsive wakefulness syndrome (UWS) and minimally conscious state (MCS), shifted brain activity closer to criticality—the phase transition between order and chaos. The effect was greater in MCS patients. In UWS patients, the treatment response correlated with structural connectivity, while in MCS patients it aligned with baseline functional connectivity. These results provide a computational foundation for using psychedelics in DoC treatment and highlight the potential role of computational modeling in drug discovery and personalized medicine.
bioRxiv (Cold Spring Harbor Laboratory)
October 3, 2025
Marian Martínez-Marín, Jakub Vohryzek, Anira Escrichs et al.
1 citation
preprint
Consciousness levels after coma can be assessed by measuring how far the brain's dynamics are from equilibrium. Using fMRI data and individualized whole-brain models, researchers found that patients with disorders of consciousness—those in a minimally conscious state or unresponsive wakefulness syndrome—show brain activity closer to equilibrium than healthy controls, with the shift increasing as consciousness decreases. Disruptions in hierarchical drive were identified in default-mode network regions and subcortical hubs like the thalamus. Recovery of near-control hierarchy in the visual network distinguished minimally conscious from unresponsive patients, while limbic areas showed similar abnormalities in both groups. Deviation from the fluctuation-dissipation theorem offers a model-based biomarker for clinical stratification.
bioRxiv (Cold Spring Harbor Laboratory)
July 1, 2026
Fernando Lehue, Iván Mindlin, Carlos Coronel-Oliveros et al.
Disorders of consciousness are linked to large-scale changes in brain dynamics, but the structural factors behind these changes are unclear. Using a whole-brain computational model constrained by diffusion MRI-derived connectivity, the authors show that a node's integration within the structural connectome, measured by a spectral integration metric, strongly predicts its impact on global brain dynamics. Lesions to highly integrative hubs, especially in posterior medial regions like the precuneus and posterior cingulate cortex, drive the system toward low-complexity dynamical regimes resembling disorders of consciousness. Increasing excitability in these regions restores healthy-like dynamics in silico. Perturbations to weakly integrated regions have limited global effects, explaining why damage to specific hubs disproportionately disrupts conscious brain activity.
bioRxiv Preprint Server
April 22, 2026
Nicolás Bruno, Federico Cavanna, Federico Zamberlán et al.
preprint
Spontaneous thoughts make up most of everyday inner experience, but studying them is difficult because traditional methods disrupt the natural flow of thinking or introduce motor artifacts. An alternative approach combined delayed verbal retrospective free reports with automated ratings from large language models. Twenty-two participants performed an eyes-closed free-thinking task, and their reports were evaluated on ten dimensions by four LLMs and human raters. Machine-learning models trained on EEG features achieved above-chance accuracy for predicting emotional valence. LLMs showed higher inter-rater agreement than humans, supporting their use for scalable annotation and suggesting that affective dimensions of spontaneous thoughts can be decoded from brain activity.
bioRxiv (Cold Spring Harbor Laboratory)
January 13, 2026
Iván Mindlin, Carlos Coronel-Oliveros, Jacobo Sitt et al.
A biologically grounded inhibitory homeostatic plasticity rule embedded into the Dynamic Mean Field (DMF) model creates a Homeostatic Dynamic Mean Field (HDMF) model that dynamically tunes local excitation-inhibition balance. The HDMF reproduces statistical observables of brain activity as well as the original DMF, can sustain neuromodulatory perturbations without overhead computations, and generates unprecedented sleep-like slow-wave activity that can coexist with wake-like asynchronous dynamics, permitting modeling of dissociated states of consciousness such as parasomnias. A single homeostatic rule broadens the stability and expressiveness of the DMF, providing a unified platform for studying how local adaptive processes shape the diverse global dynamics of the human brain.
Neuroscience of Consciousness
December 27, 2025
Yayla A Ilksoy, Alethia de la Fuente, Jacobo Sitt et al.
In a backward masking task, participants received transcranial alternating current stimulation (tACS) at 20 Hz or 40 Hz to test whether beta- or gamma-band oscillations causally influence conscious and unconscious visual perception. Contrary to expectations, 20 Hz-tACS selectively impaired objective visibility (correct categorization of masked targets) but not subjective visibility (self-reported conscious perception). 40 Hz-tACS showed no effect. Local beta power increased after 20 Hz-tACS, but inter-areal beta synchrony may have been disrupted. The findings suggest a potential causal role for beta-band activity in visual perception, particularly for objective discrimination, and point to future studies using other stimulation methods or model organisms.
bioRxiv Preprint Server
September 28, 2022
Yonatan Sanz Perl, Carla Pallavicini, Juan Piccinini et al.
preprint
Brain states are often described on a single scale from full consciousness to unconsciousness, but this ignores the complex, high-dimensional nature of brain activity. By combining whole-brain modeling, data augmentation, and deep learning, researchers mapped states of consciousness into a low-dimensional space where distances reflect similarities between states. They found an orderly trajectory from wakefulness to brain-injured patients, with coordinates related to functional modularity and structure-function coupling, both increasing as consciousness is lost. Model perturbations provided a geometric interpretation of state stability and reversibility. The work suggests conscious awareness depends on functional patterns encoded as a low-dimensional trajectory within the vast space of brain configurations.
bioRxiv Preprint Server
July 2, 2020
Yonatan Sanz Perl, Carla Pallavicini, Ignacio Pérez Ipiña et al.
preprint
The level of consciousness—how conscious someone is—is often measured by how similar their brain activity is to normal wakefulness. However, this approach misses important information about how stable that state is. Using computer models of the whole brain, the authors show that the stability of a conscious state—how easily it can be disrupted—provides additional, complementary information. They propose a new framework that sorts brain states by both their similarity to wakefulness and their stability, which helps distinguish between different types of unconsciousness: natural sleep, anesthesia, and brain injury. This framework offers a more complete way to characterize and differentiate states of consciousness.