What Is Integrated Information Theory?
Integrated Information Theory (IIT) is one of the most ambitious attempts to solve the problem of consciousness with mathematical precision. Developed by neuroscientist Giulio Tononi at the University of Wisconsin-Madison beginning in 2004, IIT starts not from the brain but from the phenomenology of experience itself and works backward to determine what physical properties a system must have to be conscious.
The theory's central claim is deceptively simple: consciousness is integrated information. More precisely, consciousness is identical to a specific structure of cause-effect relationships within a system, quantified by a measure called Phi (Φ, pronounced "fee").
The Core Framework
IIT begins with five axioms — properties of experience that are self-evidently true from the first-person perspective. Every experience exists (intrinsic existence), is structured (it has parts and relationships), is specific (it is this experience and not another), is unified (it cannot be divided into independent components), and is definite (it has borders). From these axioms, IIT derives corresponding postulates about the physical substrate that could give rise to such experience.
The key postulate is integration. A conscious system must process information in a way that is irreducible — the whole must do more than the sum of its parts. Phi (Φ) measures exactly this: how much the system's cause-effect structure would be lost if you partitioned it into its most independent components.
Who Proposed It
Giulio Tononi first published the foundations of IIT in 2004, with major revisions in IIT 2.0 (2008) and IIT 3.0 (2014). Christof Koch, former president of the Allen Institute for Brain Science, has been IIT's most prominent advocate and collaborator. Together, they have argued that IIT provides a principled, non-arbitrary answer to which physical systems are conscious and to what degree.
Key Evidence
IIT's predictions have found support in clinical neuroscience. The perturbational complexity index (PCI), derived from IIT's principles, has proven remarkably effective at detecting consciousness in unresponsive patients. By delivering a magnetic pulse to the brain and measuring the complexity of the resulting electrical response, researchers can distinguish vegetative states from minimally conscious states with over 95% accuracy. This clinical utility represents one of the strongest empirical validations of any theory of consciousness.
Additionally, IIT correctly predicts that consciousness fades during dreamless sleep and under general anesthesia — states where the brain's ability to integrate information measurably decreases — while persisting during dreaming, when integration remains high.
Key Objections
IIT faces several significant challenges. The most practical is that calculating Φ for any system larger than a handful of elements is computationally intractable. Critics like Scott Aaronson have pointed out that certain simple grid-like systems would have very high Φ under IIT, potentially implying that inactive structures like a wall of logic gates could be highly conscious — a reductio ad absurdum.
The "hard problem" objection persists as well: even if Φ perfectly tracks consciousness, IIT may not explain why integrated information feels like something. The theory identifies consciousness with a mathematical structure, but whether this constitutes a genuine explanation or merely a correlation remains debated.
Why It Matters
IIT matters because it offers a concrete, falsifiable framework in a field often dominated by vague theorizing. It makes specific predictions about which systems are conscious, providing a potential basis for assessing consciousness in brain-injured patients, animals, infants, and eventually artificial systems. The Adversarial Collaboration between IIT and Global Workspace Theory, results of which were published in 2023, represents a landmark attempt to empirically test competing theories of consciousness — bringing the field closer to genuine scientific progress.
Whether IIT ultimately proves correct or not, it has raised the bar for what a scientific theory of consciousness should look like: precise, mathematically rigorous, and empirically testable.
