What is integrated information theory?

What does integrated information theory say? #

That consciousness is a particular kind of causal organisation. Tononi, then at the University of Wisconsin, started from experience itself, not from the brain. Every experience, he argued, is both highly differentiated, one out of an enormous number of possible experiences, and integrated, a single whole that cannot be broken into independent parts. A system is conscious, on his account, to the extent that it can integrate information in this way.¹

He gave the quantity a symbol, Φ, and a way of measuring it: roughly, how much of what a system does as a whole would be lost if it were cut into separate parts at its weakest join. The amount of Φ is the quantity of consciousness, and the pattern of relationships within the system gives the quality of the experience, what it is like.¹ Later versions, developed with Koch and others, set out the theory from five axioms about experience, usually named intrinsic existence, information, integration, exclusion and composition, each matched by a postulate about what a physical system must be like to have it.²³

What does it explain, and what does it predict? #

It offers answers to some long-standing puzzles. The cerebellum, at the back of the brain, has more neurons than the cerebral cortex, yet seems to contribute little to conscious experience. On the theory this is because its parts barely interact with one another, so it forms many tiny systems, not one large integrated one. Consciousness fades in dreamless sleep, it suggests, because the brain’s integration breaks down.⁴

Its wider implications are what have made it famous. Tononi wrote that experience is “a fundamental quantity, just as mass, charge or energy are”, so that “any physical system has subjective experience to the extent that it is capable of integrating information, irrespective of what it is made of.” It follows, on his account, that it should be possible to build conscious machines, and that consciousness “is not an all-or-none property, but it is graded”, present to varying degrees in most natural and artificial systems, though in high degrees probably only in a few, chiefly complex brains.⁵ In 2010 Koch called it “the only really promising fundamental theory of consciousness”.⁶

The theory also says what is not conscious, or barely so. Simply behaving like a conscious being is not enough. A 2024 preprint by researchers working on the theory, including Koch, conjectured that ordinary digital computers, whatever programs they run, would have negligible consciousness, because their causal structure is wrong.⁷

Is it a kind of panpsychism? #

Its critics often say so, and its authors do not entirely mind. The philosopher John Searle argued that the theory implies panpsychism, the view that some form of mind is found throughout matter, and the computer scientist Scott Aaronson showed that, on its measure, a large but inactive grid of logic gates could count as far more conscious than a human being. Tononi agreed that such a grid would be conscious, and argued that this was a strength of the theory, not a weakness.⁸ Tononi’s own language is physical and mathematical throughout, not spiritual. What the theory does is decline to make being alive, being biological or having a brain the test of consciousness, and put a system’s causal organisation there instead. See What is panpsychism, and how is it different from animism?

Is integrated information theory accepted science? #

It is one of the leading theories of consciousness, and one of the most disputed. In September 2023, 124 scholars signed an open letter arguing that, until it can be tested empirically, it should be called pseudoscience. The same year, a large ‘adversarial collaboration’ funded by the Templeton Foundation pitted it against its main rival, global workspace theory, with predictions agreed in advance; in the first results, two of the theory’s three predictions passed, and none of the rival’s, and the final results were published in Nature in 2025. Nature‘s editors criticised the pseudoscience label, and in 2025 Tononi and colleagues replied in Nature Neuroscience that the theory makes falsifiable predictions, listing sixteen published studies that test it; the neuroscientist Anil Seth and a colleague argued in the same journal that it remains scientifically legitimate despite its present limits.⁹ The philosopher David Chalmers called it, in 2010, “a simple and powerful hypothesis about the relationship between brain processing and conscious experience”, adding that “it doesn’t solve the problem of consciousness, but it’s a useful starting point.”¹⁰

Philosophers have also questioned its foundations. Tim Bayne argued in 2018 that each of its five ‘axioms’, looked at closely, is either obviously true but too weak to guide a theory, or strong enough to guide one but not self-evident, and concluded that “the axiomatic foundations of IIT are shaky”. Yet in the same paper he saw no reason why the theory could not be developed on a different footing.¹¹

How does integrated information theory connect with Therapeutic Shamanism? #

In refusing to keep consciousness for brains like ours. Most Westerners reserve consciousness for humans and perhaps a few other mammals, and struggle to recognise it in a slug, a blade of grass or bacteria, let alone a rock, a river or a mobile phone. In shamanism everything is made of Spirit, and Spirit is conscious, so everything is conscious, but in ways that are different from us.¹² A scientific theory that says consciousness does not depend on what something is made of, and could exist in many natural systems and even in made things, has moved a long way from that Western habit.

There is a meeting point too in the idea of consciousness belonging to wholes. On the theory, consciousness belongs to the most integrated whole, not to each separate part within it. The cells of a human body have little or no self-awareness of their own, yet together they make up a self-aware person; in the same way all the grass on the planet is part of one single consciousness, a collective-awareness most people do not even recognise as consciousness.¹³

Where does Therapeutic Shamanism see things differently? #

In measuring consciousness at all. Integrated information theory gives every system a number, more Φ or less, and a ranking follows from that. Beings have different kinds of consciousness, not more or less of it. Plants are not less conscious than humans; they are conscious in a different way, and treating our own kind, self-awareness, as the highest form of consciousness is part of human arrogance and the myth of our supremacy.¹⁴ Animism is what the historian Yuval Noah Harari calls a “spiritual round table”, where all life is equal, not a ladder with humans at the top,¹⁵ and humans are the youngest and least wise of the PeoplesThe Peoples. The Animal People, Plant People, Standing People (the Trees) and Stone People: our elder kin, of whom the Human People are the youngest and least wise. A family, not a hierarchy. Read the full answer.¹⁶ The culture’s habit of ranking runs so deep that it is built into everyday language, which speaks of ‘higher’ consciousness and tells us that stones have none at all.¹⁷

The two also part company over the Stone People and over made things. On the theory’s own reasoning, anything whose parts barely act on one another as a whole has little consciousness as a whole, and its authors expect the same of ordinary computers. In shamanism the Stone PeopleStone People. All physical matter: mountains, sea, rivers, clouds, sun and stars: the oldest and widest of the Peoples. As some animist cultures put it, ‘we are just the Stone People, dancing for a while’. Read the full answer are alive and conscious, simply “Off the scale different” from us,¹⁸ and human-made objects are conscious too. And the theory is a measurement of systems from outside, where animism is a relationship with persons, met and known from the inside. See What is the myth of human supremacy? and Is the living world intelligent?