Consciousness: here, there and everywhere?

The scientific study of consciousness has traditionally relied on identifying behavioral and neural correlates of experience. However, these methods struggle to explain why certain brain regions, like the cerebral cortex, generate consciousness while others, like the cerebellum, do not, despite their high complexity. Furthermore, relying on behavioral reports makes it difficult to assess consciousness in non-verbal subjects, such as infants, brain-damaged patients, animals, or sophisticated machines. Tononi and Koch argue that we need a theory that defines what experience is and what physical properties are required to support it.

IIT takes a different approach by starting from the essential phenomenological properties of consciousness itself—intrinsic existence, composition, information, integration, and exclusion. From these axioms, the theory derives postulates about the physical mechanisms required for consciousness. It posits that consciousness is identical to a conceptual structure that is maximally irreducible intrinsically. The quantity of consciousness is measured by 'big phi' (Φ), which quantifies the system's integrated information, while the quality of the experience is defined by the specific shape of this conceptual structure in a high-dimensional cause-effect space.

IIT offers several counterintuitive predictions. It suggests that consciousness is graded and can exist in simple systems, but it explicitly denies consciousness to purely feed-forward networks, even if they are functionally equivalent to conscious systems. Crucially, the theory implies that digital computers, even those running faithful simulations of the human brain, would lack consciousness because they lack the necessary intrinsic cause-effect power. The theory has been used to develop the Perturbational Complexity Index (PCI), a clinical tool that measures the brain's capacity for information integration to assess consciousness in non-responsive patients.