Integrated Information Theory (IIT), developed by neuroscientist Giulio Tononi, proposes that consciousness is identical to integrated information — a property that can be precisely quantified by the mathematical measure Phi (Φ). The theory begins not from neuroscience but from phenomenology: from the undeniable fact that experience exists, and from the essential properties that every experience necessarily has.
IIT identifies five axioms of phenomenal existence. First, intrinsicality: consciousness exists for itself, independently of any external observer. Second, information: every conscious experience is specific — it rules out all other possible experiences simultaneously. Third, integration: experience is unified and irreducible; it cannot be decomposed into independent parts without destroying it. Fourth, exclusion: consciousness is definite, with exact borders determined by the system that maximizes Phi. Fifth, composition: experience has structure — it is made of distinctions that relate to one another in specific ways.
From these axioms, IIT derives physical postulates. Any system that satisfies all five in its causal structure is conscious, to the degree that its integrated information exceeds the sum of its parts. Phi measures this excess. A system with zero Phi — such as a feedforward network where information flows in one direction without looping back — has no consciousness. A system with high Phi — such as the thalamocortical complex in the human brain — has rich, unified, structured experience.
The theory accounts for several puzzling features of consciousness. It explains why the cerebellum, despite containing more neurons than the cerebrum, does not support consciousness: its neurons are densely packed but weakly interconnected, yielding low integration. It explains why computers, despite their computational power, have low Phi: their modular, feedforward architecture prevents information from becoming irreducibly unified. It explains why consciousness diminishes during dreamless sleep and epileptic seizures: both states produce low-information, low-integration activity patterns.
IIT makes a radical prediction: any system with nonzero Phi has some degree of consciousness. This implies a form of panpsychism — that consciousness is not exceptional to biological brains but fundamental to any sufficiently integrated physical system. Even simple arrangements of logic gates, if their causal structure is sufficiently integrated, would have minimal experience. This conclusion follows directly from IIT's axioms, though it contradicts common intuition.
The theory faces serious objections. Philosopher Scott Aaronson demonstrated that certain arrangements of inactive logic gates would, by IIT's own formulation, possess consciousness vastly exceeding that of humans. Tononi accepts this result and considers it a feature rather than a flaw. Neuroscientist Michael Graziano has dismissed IIT as pseudoscience, arguing it offers no mechanistic account and cannot be empirically tested. Philosopher Adam Pautz has argued the theory remains too vague to be properly evaluated. Others challenge the axiomatic foundations themselves, questioning whether the five axioms are truly self-evident or merely assumed.
Supporters include Christof Koch, who has called IIT the only genuinely promising fundamental theory of consciousness, and Anil Seth, who endorses IIT's core insight while cautioning against its panpsychist extrapolations. David Chalmers, who formulated the hard problem of consciousness — the question of why physical processes give rise to subjective experience at all — has expressed measured enthusiasm, calling IIT a development in the right direction regardless of its ultimate correctness.
IIT occupies a unique position in consciousness studies. It takes first-person evidence as its starting point rather than a problem to be explained away. It attempts to bridge phenomenology and physics through a single mathematical framework. It does not eliminate the hard problem but proposes to dissolve it through identity: the structure of experience and the causal structure of its physical substrate are, IIT claims, the same thing described from two vantage points.
Whether Phi is the measure of consciousness or merely correlated with it remains deeply contested. What IIT has accomplished, regardless of its ultimate fate, is to transform the philosophy of consciousness into a domain where precise, falsifiable, mathematical claims can be made and tested against empirical evidence.
Abstract synthesized for experimental use. Source: Mørch, H.H. (2017). The Integrated Information Theory of Consciousness. Philosophy Now, Issue 121. Secondary reference: Albantakis et al. (2023). IIT 4.0. PLoS Computational Biology.