bioRxiv
February 1, 2022
Hardik Rajpal, Pedro A. M. Mediano, Fernando E. Rosas et al.
5 citations
preprint
Schizophrenia and drug-induced states from LSD and ketamine both increase neural signal diversity, but they differ in how information flows in the brain. In schizophrenia, transfer entropy from the front to the back of the brain is increased, whereas under both drugs it is reduced overall. These differences can be modeled by altering Bayesian inference within a predictive processing framework: drug effects correspond to reduced precision of prior beliefs, while schizophrenia corresponds to increased precision of sensory information. The findings clarify similarities and differences between these altered states, with potential implications for understanding consciousness and developing mental health treatments.
bioRxiv Preprint Server
August 3, 2020
David J. Schwartzman, Aleš Oblak, Nicolas Rothen et al.
4 citations
preprint
Grapheme-colour synaesthesia (GCS) involves automatic, consistent colour experiences triggered by letters or numbers. Two recent studies showed that extensive associative training can produce behavioural, neurophysiological, and phenomenological markers of synaesthesia in non-synaesthetes, but they did not deeply compare the induced experiences to natural synaesthesia. This study analyzed interview transcripts from participants who underwent such training and from natural synaesthetes. Both groups shared several experiential categories, including stability, location, shape, relative strength, and automaticity of colour experience. However, automaticity differed significantly: natural synaesthetes mostly reported automatic experiences, while induced synaesthesia-like experiences were mostly described as wilful. Additional categories emerged only in natural synaesthetes, highlighting heterogeneity. The results indicate that intensive training can alter conscious perception, producing phenomenology substantially resembling natural synaesthesia.
Behavioral and Brain Sciences
January 1, 2016
Anil K. Seth
4 citations
Consciousness may be linked to voluntary action selection, but a narrow focus on skeletomotor control overlooks how interoception and autonomic regulation contribute to conscious selfhood and subjectivity. From the perspective of predictive processing, internal 'action'—the regulation of the body's internal state—plays a crucial role in shaping conscious experience. The argument suggests that understanding consciousness requires integrating both external action selection and internal bodily regulation.
Neurosci Conscious
April 7, 2026
Romy Beauté, David J. Schwartzman, Guillaume Dumas et al.
3 citations
No Summary
bioRxiv Preprint Server
July 16, 2025
Borjan Milinkovic, Anil K. Seth, Lionel Barnett et al.
2 citations
preprint
Consciousness depends on neural activity across many scales. A new measure, dynamical independence (DI), quantifies these multi-scale relationships. Applying DI to EEG data from people under three anaesthetics, the authors found that propofol and xenon—which abolish conscious report—produce more emergent but highly variable dynamic structure, indicating fragmented macroscopic organisation. Ketamine, which preserves dream-like states, shows reduced overall emergence but partial preservation of macroscopic structure similar to wakefulness. Regional brain contributions varied. The results reveal drug-specific reconfigurations of emergent dynamics, dissociate the amount of emergence from its organisation, and caution against equating emergence with consciousness level.
bioRxiv Preprint Server
November 3, 2017
Keisuke Suzuki, Warrick Roseboom, David J. Schwartzman et al.
2 citations
preprint
A tool called the Hallucination Machine simulates visual hallucinatory experiences using deep convolutional neural networks and panoramic virtual reality videos of natural scenes. It induces visual phenomenology qualitatively similar to classical psychedelics, but does not evoke the temporal distortion commonly associated with altered states. This technique allows researchers to study altered consciousness without the confounding physiological and cognitive effects of psychoactive substances or psychopathological conditions, offering a valuable method for consciousness science and psychiatry.
bioRxiv (Cold Spring Harbor Laboratory)
February 18, 2026
Ethan Grove, Trevor Hewitt, Anil K. Seth et al.
1 citation
Visual hallucinations (VHs) occur in psychedelic states and various psychiatric and neurological conditions, but their phenomenology is hard to characterize due to a lack of large-scale datasets. Stroboscopic light stimulation (SLS) with closed eyes reliably induces VHs in healthy people, producing vivid colors and dynamic geometric patterns similar to simple VHs in other contexts. Researchers developed an unsupervised computer-vision pipeline to analyze 10,598 drawings made after hallucination-inducing SLS at a public installation. Most drawings contained geometric forms, consistent with prior observations, but novel patterns like concentric squares, crosses, and hyperbolic shapes also appeared. The pipeline organized the drawings into interpretable classes, mapping the diversity of simple geometric VHs and placing new constraints on theoretical accounts.
PLoS One
December 4, 2025
Jonathan Robinson, Andrew W. Corcoran, Christopher J. Whyte et al.
1 citation
Active inference, a framework for modeling how sentient agents behave, is being tested as necessary for changes in conscious content. In an adversarial collaboration, active inference will be contrasted with two other theories that do not require it for consciousness. This study protocol describes an adaptation of the motion-induced blindness paradigm: an active condition where participants direct their gaze toward a target after it disappears from consciousness and report its reappearance, versus a passive condition where participants fixate centrally while the stimulus array moves in a replay of active eye-tracking data. Two experiments will compare target reappearance across conditions to evaluate active inference's contribution to conscious awareness.
April 22, 2025
Anil K. Seth
1 citation
preprint
Consciousness in artificial intelligence is unlikely on current technological trajectories because computation alone is insufficient to generate it. Instead, consciousness depends on our nature as living organisms—a view called biological naturalism. People may mistakenly think AI could become conscious due to cognitive biases. Conscious AI becomes more plausible only as systems become more brain-like or life-like. Ethical considerations arise from AI that either is, or convincingly appears to be, conscious. Overestimating machine consciousness risks underestimating our own selves.
December 13, 2024
Trevor Hewitt, Ioanna Amaya, Romy Beauté et al.
preprint
Exposure to rapid and bright stroboscopic light can induce vivid visual hallucinations of color and geometric forms, a phenomenon first documented by Purkinje over 200 years ago. Despite centuries of scientific, therapeutic, and cultural interest, fundamental questions remain about its phenomenology, physiological origins, and potential clinical applications. This narrative review summarizes the historical research on stroboscopic light stimulation, its use in recreational and lay-therapeutic settings, and discusses the phenomenology of these experiences. It also examines current perspectives on the neural mechanisms that may underlie stroboscopically induced experiences and outlines directions for future research.
medRxiv Preprint Server
June 17, 2026
Danny Nacker, Luise Kalus, Anil K. Seth et al.
preprint
Supervised stroboscopic light stimulation (SLS) was safe, tolerable, and feasible in adults with depressive symptoms, but efficacy was not established. In a staged program, 31 participants tested 11 SLS parameter sets; no severe adverse reactions occurred, and mean discomfort was low (0.49 out of 10). A subsequent randomized trial assigned 84 participants to four weekly 31-minute sessions of SLS or a low-phenomenology control. Retention was 83.3% (70 of 84 participants), with higher retention in the intervention arm (39 of 42) than the control arm (31 of 42). Exploratory depressive-symptom changes suggested a possible signal on the BDI-II but do not confirm efficacy. The next step is a Phase 2a feasibility trial with a locked protocol.
arXiv Preprint Archive
April 13, 2026
Adam B. Barrett, Borjan Milinkovic, Pedro A. M. Mediano et al.
The integrated information theory of consciousness (IIT) proposes a mathematical formula, derived from fundamental properties of conscious experience, to describe the quantity and quality of consciousness for any physical system. This article clarifies common misunderstandings. A high value of the measure Φ does not mean 'more consciousness'; Φ might be replaced with a suite of quantities for a multidimensional characterization. IIT implies a distinct panpsychism where space and time are tiled with substrates of proto-consciousness, which the authors find unproblematic. Φ is not well-defined for real physical systems and has never been computed on one; only proxies have been computed, not approximations. For IIT to align with fundamental physics, a reformulation in continuous fields would be needed.
Neuroscience of Consciousness
January 1, 2026
Lancelot da Costa, Anil K. Seth, Karl Friston et al.
A Rosetta Stone hypothesis from predictive processing proposes that beliefs serve as a central hub linking phenomenology, behavior, and neural dynamics. Under the assumption that phenomenology is a function of beliefs, the authors derive mathematical predictions for subjective similarity judgments, cognitive metabolic cost, subjective cognitive effort, and time perception. They review the connection between beliefs and neural dynamics to complete a generative passage for neurophenomenology, omitting the belief–behavior link as already documented. Testing these predictions will inform the validity of the central assumption and advance the neurophenomenology research program.
The Behavioral and brain sciences
April 21, 2025
Anil K. Seth
The author argues that consciousness is unlikely to emerge from current artificial intelligence systems because it depends on biological life, not computation alone. People may mistakenly believe AI could become conscious due to biases, but true consciousness requires the organic nature of living organisms—a position called biological naturalism. The author considers scenarios where artificial consciousness might become more plausible if AI becomes more brain-like or life-like, but concludes this is improbable along current trajectories. Ethical issues arise from AI that either is or convincingly appears conscious. Overestimating machines risks underestimating human consciousness.
arXiv Preprint Archive
February 25, 2025
Romy Beauté, David J. Schwartzman, Guillaume Dumas et al.
Stroboscopic light stimulation on closed eyes typically induces simple visual hallucinations—vivid, geometric, and colorful patterns. An analysis of 862 open-ended reports from the Dreamachine immersive experience, using large language models and topic modeling, confirmed these simple hallucinations and also revealed altered states of consciousness and complex hallucinations. This computational approach enables systematic study of subjective experiences beyond standard questionnaires, capturing subtle patterns not readily identified through closed-form questions. The findings broaden understanding of stroboscopically induced phenomena and demonstrate the potential of natural language processing in computational neurophenomenology.
Neuroscience of Consciousness
January 1, 2025
Trevor Hewitt, Ioanna Amaya, Romy Beauté et al.
Exposure to rapid, bright stroboscopic light has been known for over 200 years to induce vivid visual hallucinations of color and geometric shapes. This narrative review summarizes the history of research on stroboscopic light stimulation (SLS), its recreational and lay-therapeutic uses, the phenomenology of the experiences it produces, and current understanding of potential neural mechanisms. Despite progress, fundamental questions remain about the full characterization of its phenomenology, its precise physiological origins, the conditions under which it leads to altered states of consciousness, and possible clinical applications.
arXiv Preprint Archive
October 9, 2024
Christopher J. Whyte, Andrew W. Corcoran, Jonathan Robinson et al.
Subjective experience is multifaceted, making consciousness hard to study because traditional theories often focus on isolated aspects like perception or wakefulness and are difficult to compare. This work starts from active inference—a first-principles framework that models behavior as approximate Bayesian inference—and builds toward a minimal theory of consciousness derived from shared features of computational models under active inference. Reviewing models applied to consciousness, the authors argue that these models imply a small set of theoretical commitments pointing to a minimal, testable theory of consciousness.
arXiv Preprint Archive
September 30, 2024
Lancelot da Costa, Anil K. Seth, Karl Friston et al.
A Rosetta Stone hypothesis from predictive processing proposes that beliefs serve as a central hub linking phenomenology, behavior, and neural dynamics. If phenomenology is a function of beliefs, then specific predictions follow for subjective similarity judgments, cognitive metabolic cost, subjective cognitive effort, and time perception. The connection between beliefs and neural dynamics completes the generative passage for neurophenomenology, while the belief-behavior link is already well-documented. Testing these predictions will inform the validity of the central assumption and advance the neurophenomenology research program.
Trends in Cognitive Sciences
March 1, 2024
Tim Bayne, Anil K. Seth, Marcello Massimini et al.
New tests for consciousness (C-tests) are urgently needed to resolve uncertainty about when consciousness arises in human development, when it is lost due to neurological disorders and brain injury, and how it is distributed in nonhuman species. This need is amplified by recent developments in AI, neural organoids, and xenobot technology. Existing C-tests are of limited use, and no tests exist for many critical populations. The paper identifies challenges facing any attempt to develop C-tests, proposes a multidimensional classification of such tests, and identifies strategies for validating them.
Current Biology
August 1, 2023
J. LeDoux, Jonathan Birch, Kristin Andrews et al.
Leading experts in consciousness research address big questions about the relationship between AI and consciousness. The piece presents a multi-perspective discussion, with contributors including philosophers, neuroscientists, and cognitive scientists responding to challenges posed by Joseph LeDoux and Jonathan Birch. The experts explore whether AI systems could ever be conscious, what criteria might be used to assess machine consciousness, and how insights from animal and human consciousness research might inform these debates. No single conclusion is reached; instead the work maps the current landscape of scientific and philosophical positions on AI consciousness.
Cognitive Neuroscience
January 1, 2021
Anil K. Seth, Jakob Hohwy
Theories of consciousness often treat it as a single phenomenon to be explained, but this approach struggles to capture the diverse properties of conscious experience. Progress in consciousness science requires systematic mappings between physical or biological mechanisms and the functional and phenomenological features of consciousness. The authors argue for developing theories for consciousness science rather than a single theory of consciousness, and they highlight predictive processing as a highly promising framework for this purpose.
Entropy (Basel, Switzerland)
December 25, 2018
Pedro A. M. Mediano, Anil K. Seth, Adam B. Barrett
Integrated Information Theory (IIT) is a prominent theory of consciousness that centers on measures quantifying how much a system generates more information than the sum of its parts. This article provides clear descriptions of six distinct candidate measures of integrated information and explores their properties through simulations on networks of eight interacting nodes with Gaussian linear autoregressive dynamics. The results reveal striking diversity in the measures' behavior—no two measures show consistent agreement across all analyses. A subset of the measures appears to reflect some form of dynamical complexity, meaning simultaneous segregation and integration between system components. These findings help guide the operationalization of IIT and advance development of measures with more general applicability.
arXiv Preprint Archive
June 25, 2018
Pedro A. M. Mediano, Anil K. Seth, Adam B. Barrett
Integrated Information Theory (IIT) proposes that consciousness arises from a system generating more information than the sum of its parts, but multiple measures of this integrated information (Φ) exist with little comparative testing. This article describes six candidate measures and simulates them on eight-node networks with Gaussian linear autoregressive dynamics. No two measures agreed consistently across all analyses, and only a subset genuinely reflected dynamical complexity—simultaneous segregation and integration between components. The findings guide operationalization of IIT and development of more generally applicable measures of integrated information.
Neuroimage
February 15, 2018
Michał Bola, Adam B. Barrett, Andrea Pigorini et al.
Loss of consciousness during both natural sleep and anesthesia is linked to excessively correlated gamma-band activity across brain regions. Analyzing ECoG from 4 macaque monkeys under ketamine, medetomidine, or propofol and sEEG from 10 epilepsy patients during wakefulness, NREM, and REM sleep, resting wakefulness showed intermediate gamma coupling. NREM sleep and propofol anesthesia produced a robust increase in positive correlations among brain areas, while REM sleep and ketamine anesthesia—states often involving dream-like experiences—did not. Hyper-correlated gamma activity may be a general signature of unconsciousness independent of behavioral responsiveness.
Brain and neuroscience advances
January 1, 2018
Anil K. Seth
The mind and brain sciences began with consciousness as a central concern, but for much of the 20th century it was marginalized. Since the 1990s, consciousness science has regained legitimacy, now encompassing philosophical, theoretical, computational, experimental, and clinical perspectives, with neuroscience as its central discipline. Researchers have learned much about neural mechanisms underlying global states of consciousness, conscious versus unconscious perception, and self-consciousness. Further progress depends on specifying closer explanatory mappings between first-person phenomenological descriptions and third-person descriptions of embodied neuronal mechanisms. This will help reframe understanding of humanity's place in nature and accelerate clinical approaches to psychiatric and neurological disorders.