Scientific reference document
Drafting date: July 9, 2026
This document reflects the state of scientific knowledge at this date. Periodic updating is recommended.
Knowing the brain's structure (see R-08 and R-13). Measuring neuronal activity. Mapping circuits. Yet one question persists: Why is this activity accompanied by lived experience? Why do neurons processing the image of red produce not only behavior but also feel red?
David Chalmers, philosopher and cognitive scientist, named this problem the «hard problem» in 1995. He distinguishes two categories of questions:
This document surveys the main scientific theories of consciousness, distinguishing what is established from what is debated.
Origin: Bernard Baars (cognitive scientist, USA), 1988. Redesigned by Stanislas Dehaene and collaborators (cognitive neuroscience, France).
Principle: Consciousness functions like a theater. Unconscious specialists (neural processes) compete for access to a limited stage (the global workspace). Information that gains access is «broadcast» throughout the brain and becomes conscious.
Neural mechanism: Activation propagates through a fronto-parietal network, with nonlinear ignition at high latency (~300 ms). Observed neural signature: P300 component of event-related potentials and high-frequency gamma synchronization.
Publications: Baars, B. J. (1997). In the Theatre of Consciousness. Dehaene, S. et al. (2014). Conscious Processing and the Global Neuronal Workspace Hypothesis (Neuron).
Validation: Numerous experimental studies, particularly on unconscious vs. conscious vision. Testable predictions verified in hundreds of works.
Limitations: Explains access to information, but not why this accessibility creates subjective experience. Some critics note the model localizes consciousness too much in frontal regions.
Status: Dominant theory in cognitive neuroscience. Experimentally validated.
Origin: Giulio Tononi (psychiatrist, University of Wisconsin-Madison, USA), 2004. Developed with Christof Koch, Melanie Boly, and others.
Principle: Consciousness corresponds to the amount of integrated information in a system. The more integrated information a system generates (measured by Φ, phi), the more conscious it is. The system must satisfy five axioms: existence, composition, information, integration, exclusion.
Mathematical formalization: Φ is calculated as the minimum between integrated causal information (φc) and integrated effecter information (φe). In principle, can be applied to any physical substrate (brain, machine) (see B-07).
Applications: Assessing consciousness in coma patients, under anesthesia. Clinical tests (Perturbation Complexity Index — PCI).
Publications: Tononi, G. et al. (2016). Integrated Information Theory 3.0. Albantakis, L. et al. (2023). Integrated Information Theory 4.0 (arXiv). Ellia, F. et al. (2021). Consciousness and Complexity (Neural Computation).
Scientific validity: Rigorous formal framework, numerous peer-reviewed publications. Empirically testable (PCI).
Criticisms:
Status: Major theory, peer-reviewed. Open debate with GWT.
Origin: Roger Penrose (mathematician, physicist, Oxford, UK) and Stuart Hameroff (anesthesiologist, University of Arizona, USA), 1990s.
Principle: Consciousness arises from quantum computations in neuronal microtubules. Each computation ends with objective reduction (wave function collapse) linked to spacetime geometry. Non-computational, explains free will.
Arguments in favor:
Publications: Penrose, R. & Hameroff, S. (2014). Consciousness in the universe: A review of the 'Orch OR' theory (Physics of Life Reviews). Craddock, T.J. et al. (2015). Anesthetics Act in Quantum Channels in Brain Microtubules (Current Topics in Medicinal Chemistry).
Dominant criticisms (2023–2025):
Current consensus: Intriguing theory, minority. Majority of neuroscientists and physicists consider Orch OR as unverified or even unverifiable.
Status: Speculative hypothesis. Little supported by dominant scientific community.
Classical emergence: Consciousness emerges naturally from the complexity of neural networks. «Default» position of many neuroscientists. Problem: Does not answer the «hard problem», sidesteps the question of subjective experience.
Panpsychism (variants): Consciousness is a fundamental property of matter (see B-07), like mass or charge. Complex systems integrate these elementary consciousnesses. Related to IIT. Criticism: How do micro-consciousnesses aggregate into macro-consciousness?
Eliminative materialism (Dennett, etc.): Consciousness as we conceive it is an illusion. There are only computational processes. Criticism: Denies subjective experience, philosophically unsatisfying.
Hybrid approaches (Consciousness Consortium, 2023): Researchers propose several theories are complementary, each explaining a facet of the phenomenon. Ex: GWT for access, IIT for phenomenal quality.
Publications: Consortium, C. et al. (2023). Mapping the Landscape of Consciousness Theories (Trends in Cognitive Sciences).
| Theory | Type | Validation | Strengths | Weaknesses |
|---|---|---|---|---|
| GWT | Cognitive Neuroscience | Strong (experimental) | Testable, verified predictions | Does not explain subjective experience |
| IIT | Formal/Mathematical | Medium (peer-reviewed) | Rigorous framework, applicable to any system | Φ calculation impractical, controversies |
| Orch OR | Quantum | Weak (minority) | Link to fundamental physics, free will | Rapid decoherence, lack of evidence |
| Emergence | Classical | Dominant | Simple, intuitive | Sidesteps hard problem |
Despite progress, several questions remain open:
Hard problem (Chalmers):
Global Workspace Theory (Baars, Dehaene):
Integrated Information Theory (Tononi):
Orch OR (Penrose & Hameroff):
Hybrid approaches: