What is consciousness? Why do we have subjective experiences? Why does it feel like something to see the color blue or feel pain? These questions constitute what philosopher David Chalmers calls the “hard problem” of consciousness—and one theory offers a mathematical framework for tackling it: Integrated Information Theory (IIT).
The Hard Problem of Consciousness
Physical science can explain how the brain processes visual information, controls movement, and maintains bodily functions. What it cannot explain is why these processes are accompanied by subjective experience. Why isn’t consciousness just sophisticated information processing without any “inner life”?
Chalmers named this the “hard problem”—distinguishing it from the “easy problems” of explaining cognitive functions. Solving the easy problems would explain how the brain does what it does. The hard problem asks why there is something it is like to be you at all.
Integrated Information Theory: Starting from Experience
IIT, proposed by neuroscientist Giulio Tononi in 2004, takes a radical approach. Rather than starting from physical mechanisms and trying to derive consciousness, IIT “starts with consciousness”—accepting the existence of subjective experience as certain and reasoning backward to discover what physical properties such experience requires.
This approach grounds the theory in five axioms derived from the structure of experience:
- Intrinsicality: Experience exists for itself—it has intrinsic existence
- Information: Experience is specific—it differs from other possible experiences
- Integration: Experience is unitary—it cannot be reduced to independent components
- Exclusion: Experience is definite—it has a definite composition and boundary
- Composition: Experience is structured—it has parts and relationships
Each axiom maps to physical postulates about systems that can generate consciousness. For experience to be unified, for instance, the physical substrate must be integrated—causing effects as a whole rather than as independent parts.
The Mathematical Formalism
IIT quantifies consciousness using a mathematical measure called phi (Φ). The theory proposes that a system’s quantity of consciousness equals its integrated information—the amount of information generated above and beyond its parts.
The calculation involves examining causal relationships: how the system’s current state constrains its past and future states. Systems with high phi possess irreducible cause-effect structure—their parts influence each other in ways that cannot be decomposed without losing information.
A simple feedforward network, where signals flow only forward, has low phi because information doesn’t integrate across the system. A recurrent network where parts mutually constrain each other can have high phi—and according to IIT, more consciousness.
Applications: Assessing Consciousness in Patients
One of IIT’s most practical applications is assessing consciousness in patients who cannot communicate—those in vegetative states, coma, or locked-in syndrome. Traditional methods rely on behavioral responses, but some conscious patients cannot move or speak.
Researchers developed the Perturbational Complexity Index (PCI), which applies transcranial magnetic stimulation while measuring brain-wide electrical responses. Systems generating simple, predictable responses have low phi and likely lack consciousness. Those generating complex, integrated responses have high phi—and may be conscious despite behavioral evidence suggesting otherwise.
Controversies and Criticisms
IIT remains controversial. A 2023 group of scholars characterized it as “unfalsifiable pseudoscience,” citing insufficient empirical support. A 2025 Nature Neuroscience commentary reiterated these concerns. Critics argue that IIT makes untestable predictions and that phi doesn’t reliably correlate with consciousness.
The theory’s panpsychist implications also trouble some. If integrated information generates consciousness, many systems—including some computers and perhaps the universe itself—might possess some degree of experience. Most researchers find this conclusion uncomfortable or incoherent.
Two Levels of IIT
A 2024 paper proposed distinguishing two levels within IIT. The first level addresses autonomous systems that achieve critical complexity—theory applicable to robots, AI, and biological systems. The second level addresses actual consciousness.
The authors argue that critics confuse these levels, leading to inappropriate skepticism about the first. By adding necessary elements distinguishing genuine consciousness from mere complexity, IIT might evolve toward a complete theory.
The Future of Consciousness Research
Despite controversies, IIT has focused attention on integration as crucial to understanding consciousness. The theory’s mathematical rigor offers a framework for empirical testing that purely philosophical approaches lack.
Whether IIT ultimately proves correct remains uncertain. What seems clear is that understanding consciousness requires new conceptual tools—frameworks that honor both the subjective nature of experience and the objective methods of science. IIT represents one ambitious attempt at such synthesis.