Neural basis of consciousness
How does the brain give rise to conscious experience?
Summary
Why it is open
For decades the field has searched for the neural correlates of consciousness (NCC), formally "the minimal set of neural events jointly sufficient for" a conscious state [9]. Researchers study both global states, such as wakefulness, dreaming and sedation, and local states, the contents of experience [9]. Anil Seth and Tim Bayne note that true correlates are hard to tell apart from the neural prerequisites and consequences of consciousness, and that theories are proliferating instead of being ruled out as data accumulate [9].
Measurement is part of the difficulty. Experiments usually learn what a person experienced by asking, so activity that serves the report can be mistaken for activity that serves the experience; "no-report" paradigms try to separate the two [4]. Method also shapes conclusions: across 412 experiments interpreted in the light of four leading theories, Itay Yaron and colleagues found that support for a given theory could be predicted "solely from methodological choices, irrespective of findings" [8].
No body sets a bar for a resolution. Seth and Bayne call for more precise and more comprehensive theories, and for measures of consciousness that do not assume its contents must be reportable [9]. Preregistration offers one test of progress: rival theorists agree in advance which results would count against them [14].
Two neighbouring questions have their own pages. Whether explaining these processes would also explain why they are accompanied by experience at all is the hard problem of consciousness; some theories address it directly, while others focus on the functions of consciousness [9]. And because the contents of experience are "bound together as components of a single conscious scene" [9], the question overlaps with the binding problem.
What is known
Feedforward and recurrent activity (2000). Victor Lamme and Pieter Roelfsema argued that the first, feedforward sweep of activity through the visual system "is probably incapable of yielding visual awareness", and that in many cases recurrent processing, through horizontal connections and feedback from higher areas, is needed before a stimulus can enter consciousness [1]. In Lamme's later local recurrency theory, recurrent processing within perceptual cortex suffices for experience, given enabling factors such as brainstem arousal, while parietal and frontal regions may be needed to report it [9].
Signatures of conscious access (2011–2020). Stanislas Dehaene and Jean-Pierre Changeux identified three signatures of conscious access: late amplification of sensory activity, long-distance synchronisation at beta and gamma frequencies, and "ignition" of a large prefronto-parietal network [2]. A later review by George Mashour, Roelfsema, Changeux and Dehaene describes a sudden divergence between seen and unseen trials about 200 to 300 milliseconds after a stimulus, and, in monkeys, sudden and sustained frontal activity for faint stimuli that were reported but not for those that were missed [7].
Front or back of the brain (2015–2022). Christof Koch, Marcello Massimini, Melanie Boly and Giulio Tononi concluded in 2016 that the anatomical correlates of consciousness lie mainly in a "posterior cortical hot zone" that includes sensory areas, not in a fronto-parietal network involved in task monitoring and reporting [5]. Other studies link prefrontal activity to conscious perception, while no-report studies have tended to find less prefrontal engagement [4, 9]. Seth and Bayne see the core of the dispute as whether the ability to report a content is a fair proxy for experiencing it [9].
Measuring the level of consciousness (2013). Adenauer Casali and colleagues measured how complex the cortex's electrical response to magnetic stimulation was. Their index separated levels of consciousness in single individuals during wakefulness, sleep and anaesthesia, and in patients who had emerged from coma [3]; see How general anaesthesia works. It was derived from integrated information theory, but its results are also consistent with the global workspace account [7].
An adversarial test (2025). In the Cogitate collaboration, proponents of integrated information theory (IIT) and of global neuronal workspace theory (GNWT), with a theory-neutral consortium, preregistered divergent predictions together with pass or fail criteria [14]. 256 participants viewed clearly visible faces, objects, letters and false fonts for 0.5 to 1.5 seconds while their brain activity was recorded with fMRI, MEG or, in patients with epilepsy, intracranial electrodes [14]. The consortium found information about conscious content in visual, ventrotemporal and inferior frontal cortex, and reports results that "align with some predictions of IIT and GNWT, while substantially challenging key tenets of both theories" [14]. For IIT, sustained synchronisation within posterior cortex was absent [14]. For GNWT, there was a general lack of ignition when the stimulus disappeared, and prefrontal cortex represented a stimulus's category but not its identity, and its orientation only in MEG [14]. The consortium gave no verdict: "we invite readers to weigh the evidence themselves" [14]. In the same paper, Boly, Koch and Tononi write that the results "corroborate IIT's overall claim that posterior cortical areas are sufficient for consciousness", and that the failed synchrony prediction is "quite possibly the result of sparse electrode coverage" [14]. Dehaene writes that the challenges to GNWT "do not seem insurmountable given experimental limitations": content-specific prefrontal ignition appeared at stimulus onset, IIT's mathematical core was not tested, and participants may have turned their thoughts elsewhere before each image disappeared [14].
Recent claims. In September 2023, after preliminary results of an adversarial collaboration testing IIT had been presented at a public event and reported in the press, a letter posted on PsyArXiv under the collective name IIT-Concerned, with 124 named signatories, argued that the studies tested predictions "not really logically related to the core ideas of IIT", and that until the theory as a whole is empirically testable "the pseudoscience label should indeed apply" [10]. On 10 March 2025 Nature Neuroscience published three Comment articles. IIT-Concerned argued that the theory "is indeed unscientific because its core claims are untestable even in principle" [11]. Tononi and colleagues replied that IIT accounts for consciousness "in objective, testable terms", and that the attempt to label it pseudoscientific exposes "a crisis in the dominant computational-functionalist paradigm" [12]. Alex Gomez-Marin and Seth assessed the charge critically and offered "thoughts to elevate the clash into positive lessons for our field" [13].
Main approaches
Global workspace. Information becomes conscious when it is broadcast, through ignition, across a workspace of neurons with long-range connections, with particular weight on prefrontal and parietal cortex [2, 7]. Strongest evidence: late, non-linear ignition separating seen from unseen stimuli across paradigms and modalities [2, 7], and transient, category-specific prefrontal responses at stimulus onset in the Cogitate data, which Dehaene reads as ignition [14]. Main difficulty: no-report studies find less prefrontal engagement [9], and in Cogitate there was no ignition at stimulus offset and no prefrontal representation of identity [14].
Integrated information. Starting from the properties of experience, IIT holds that the physical substrate of consciousness is a maximum of intrinsic cause–effect power, which it places mainly in posterior cortex [6, 9]. Strongest evidence: content decoding and duration-tracking in posterior cortex in Cogitate [14], and a complexity index derived from the theory that tracks the level of consciousness [3, 7]. Main difficulty: the predicted sustained synchrony was not found [14]; some predictions, such as consciousness in simple non-biological systems, cannot be evaluated without methods for detecting consciousness in such systems [9]; and its scientific status is disputed [10, 11, 12, 13].
Recurrent processing. Feedback and horizontal activity within sensory cortex makes contents conscious, and frontal areas serve report [1, 9]. Strongest evidence: feedforward activity alone is "probably incapable of yielding visual awareness" [1], and in Cogitate conscious content was decodable from posterior cortex, a prediction the consortium notes is shared by recurrent processing theory [14]. Main difficulty: showing that posterior activity suffices without anterior contribution relies on report, which itself needs anterior cortex, and the theory must still specify which recurrent processes are distinctive of consciousness [9].
Higher-order representation. A state is conscious when it is the object of a suitable meta-representation, which most versions place in prefrontal cortex [9]. Strongest evidence: lesion, stimulation and imaging evidence that advocates read as implicating prefrontal cortex in consciousness [9]. Main difficulty: the Cogitate consortium writes that its failures to confirm prefrontal content representation also challenge higher-order theories in which prefrontal cortex supplies the content of visual experience [14].
Open sub-questions
- Offset ignition. If a stimulus's disappearance is made meaningful, for example as an object hiding behind a screen, does prefrontal cortex ignite at offset, as Dehaene proposes [14]?
- Synchrony and binding. Does content-specific synchrony between category-selective cortex and early visual areas appear at lower frequencies (2–25 Hz), or with denser sampling of early visual cortex, as IIT's proponents suggest [14]? See the binding problem.
- Seen and unseen in one design. With conscious and non-conscious trials contrasted, as Dehaene says a second Cogitate experiment will do, is posterior duration-tracking specific to conscious perception [14]?
- Inactive or inactivated. IIT predicts that inactive neurons in visual cortex can contribute to experience while inactivated ones cannot; can optogenetic experiments in animals test this [9]?
- Early or late. Is perceptual consciousness marked by an early negativity at 120–200 milliseconds, or by later signatures at about 250–400 milliseconds [9]?
- Contents without report. Can a perturbational complexity measure [3] be extended from the level of consciousness to its contents in no-report designs [4]?
Work on this in Papex
If you have a testable answer to one of the sub-questions above, you can draft it as a private hypothesis in Papex, or prepare the draft from Claude or ChatGPT with the Papex connector. You review the exact version, sign it and decide whether to publish it; a public record that names this problem is listed on this page. A Papex record shows who signed which version and when. It is not peer review and does not mean the hypothesis is correct.
Key references
- Lamme, V. A. F. & Roelfsema, P. R. (2000). The distinct modes of vision offered by feedforward and recurrent processing. Trends in Neurosciences 23(11), 571–579. 10.1016/S0166-2236(00)01657-X · Checked 14 Oct 2026
- Dehaene, S. & Changeux, J.-P. (2011). Experimental and theoretical approaches to conscious processing. Neuron 70(2), 200–227. 10.1016/j.neuron.2011.03.018 · Checked 14 Oct 2026
- Casali, A. G., Gosseries, O., Rosanova, M., Boly, M., Sarasso, S., Casali, K. R. et al. (2013). A theoretically based index of consciousness independent of sensory processing and behavior. Science Translational Medicine 5(198), 198ra105. 10.1126/scitranslmed.3006294 · Checked 14 Oct 2026
- Tsuchiya, N., Wilke, M., Frässle, S. & Lamme, V. A. F. (2015). No-report paradigms: extracting the true neural correlates of consciousness. Trends in Cognitive Sciences 19(12), 757–770. 10.1016/j.tics.2015.10.002 · Checked 14 Oct 2026
- Koch, C., Massimini, M., Boly, M. & Tononi, G. (2016). Neural correlates of consciousness: progress and problems. Nature Reviews Neuroscience 17(5), 307–321; erratum 17(6), 395 (colour coding of Fig. 3e). 10.1038/nrn.2016.22 · Checked 14 Oct 2026
- Tononi, G., Boly, M., Massimini, M. & Koch, C. (2016). Integrated information theory: from consciousness to its physical substrate. Nature Reviews Neuroscience 17(7), 450–461 (Opinion). 10.1038/nrn.2016.44 · Checked 14 Oct 2026
- Mashour, G. A., Roelfsema, P., Changeux, J.-P. & Dehaene, S. (2020). Conscious processing and the global neuronal workspace hypothesis. Neuron 105(5), 776–798. 10.1016/j.neuron.2020.01.026 · Checked 14 Oct 2026
- Yaron, I., Melloni, L., Pitts, M. & Mudrik, L. (2022). The ConTraSt database for analysing and comparing empirical studies of consciousness theories. Nature Human Behaviour 6(4), 593–604. 10.1038/s41562-021-01284-5 · Checked 14 Oct 2026
- Seth, A. K. & Bayne, T. (2022). Theories of consciousness. Nature Reviews Neuroscience 23(7), 439–452. 10.1038/s41583-022-00587-4 · Checked 14 Oct 2026
- IIT-Concerned, Fleming, S. M., Frith, C. D., Goodale, M. A., Lau, H., LeDoux, J. E., Lee, A. L. F. et al. (2023). The integrated information theory of consciousness as pseudoscience. PsyArXiv preprint, posted 16 September 2023 (15 September in Nature's news report). 10.31234/osf.io/zsr78 · Checked 14 Oct 2026
- IIT-Concerned, Arnold, D. H., Baxter, M. G., Bekinschtein, T. A., Bengio, Y., Bisley, J. W. et al. (2025). What makes a theory of consciousness unscientific? Nature Neuroscience 28(4), 689–693 (Comment; published online 10 March 2025). 10.1038/s41593-025-01881-x · Checked 14 Oct 2026
- Tononi, G., Albantakis, L., Barbosa, L., Boly, M., Cirelli, C., Comolatti, R. et al. (2025). Consciousness or pseudo-consciousness? A clash of two paradigms. Nature Neuroscience 28(4), 694–702 (Comment; published online 10 March 2025). 10.1038/s41593-025-01880-y · Checked 14 Oct 2026
- Gomez-Marin, A. & Seth, A. K. (2025). A science of consciousness beyond pseudo-science and pseudo-consciousness. Nature Neuroscience 28(4), 703–706 (Comment; published online 10 March 2025). 10.1038/s41593-025-01913-6 · Checked 14 Oct 2026
- Cogitate Consortium, Ferrante, O., Gorska-Klimowska, U., Henin, S., Hirschhorn, R., Khalaf, A. et al. (2025). Adversarial testing of global neuronal workspace and integrated information theories of consciousness. Nature 642(8066), 133–142 (published online 30 April 2025). 10.1038/s41586-025-08888-1 · Checked 14 Oct 2026
Source
Open, as listed by Wikipedia · Unsolved problems in neuroscience. Checked 29 Sep 2026.
The status is the source’s. Papex does not decide whether a problem is solved.
Records on this problem
No public record addresses this problem yet.