In short
Physicist Max Tegmark proposed treating consciousness as a state of matter, which he called perceptronium, defined by how it stores and integrates information. The idea draws on integrated information theory, is highly speculative and contested, and is a rich source of questions for science fiction.
- Perceptronium is a proposal, not an established finding
- IIT measures integration of information with a quantity called phi
- SF uses the idea to ask what counts as a mind
Solids, liquids and gases are defined by how their particles are arranged and how they behave, not by what they are made of. Water and iron can both be solid. In a paper first circulated in 2014, the physicist Max Tegmark asked whether consciousness could be approached the same way: as a particular arrangement of matter with particular information-processing properties, whatever the matter happens to be. He gave this hypothetical state a name, perceptronium. This article explains the idea carefully, separates it from established science, and looks at why it is so useful to science fiction writers.
The starting point: substrate independence
Many researchers who think about minds suspect that consciousness depends on organisation rather than on a specific material. If that is true, a brain made of neurons and a hypothetical system made of silicon, or something stranger, could in principle both be conscious, as long as they are organised in the right way. This is called substrate independence. It is a widely discussed assumption, not a demonstrated fact.
Tegmark's proposal takes this assumption seriously and asks a physicist's question: if consciousness depends on organisation, can we describe that organisation in terms of physical properties, the way we describe phases of matter? Calling the result a state of matter is partly a metaphor and partly a research programme.
What perceptronium is supposed to be
In Tegmark's framing, perceptronium is the most general substance that feels subjectively self-aware. He suggested several principles such matter would need to satisfy. Stated in plain language, they include the following.
- Information: the system must be able to store a large amount of information.
- Integration: that information must form a unified whole rather than independent parts.
- Independence: the system must have substantial independence from the rest of the world, so its state is not simply dictated from outside.
- Dynamics: it must be able to process information, not just hold it.
- Utility: the information it holds should be useful to the system, for example for making predictions.
These are proposed conditions, not experimental results. Tegmark himself presented the work as an exploration of what a physical theory of consciousness would require, and noted unresolved difficulties, including problems that arise when such principles are applied to quantum descriptions of physical systems.
Integrated information theory, briefly
The integration principle draws on integrated information theory, or IIT, developed by the neuroscientist Giulio Tononi and colleagues. IIT starts from features of experience, such as the fact that each experience is unified and specific, and tries to derive what kind of physical system could have them. Its central quantity, written with the Greek letter phi, is meant to measure how much a system's whole carries information over and above its parts.
In IIT, a system with high phi is highly integrated: cutting it into pieces destroys information that only exists in the whole. A system with near-zero phi, however complex its behaviour, would not be conscious in IIT's sense. This leads to counterintuitive conclusions. Some simple but highly integrated networks could score above zero, while some powerful computers built with feed-forward architectures might score very low.
| Idea | Who proposed it | Status |
|---|---|---|
| Substrate independence | Many philosophers and scientists over decades | Widely discussed assumption, not proven |
| Integrated information theory (phi) | Giulio Tononi and colleagues | Influential, actively tested and strongly disputed |
| Perceptronium / consciousness as a state of matter | Max Tegmark | Speculative theoretical proposal |
Why the idea is contested
IIT is one of the most debated theories in consciousness research. Critics point out that phi is extremely hard or impossible to compute for realistic systems, that some of the theory's predictions seem to attribute consciousness to very simple structures, and that its core axioms are not testable in the usual way. In 2023 a large group of researchers publicly criticised it in strong terms, and its defenders responded just as strongly. Experimental programmes have compared IIT with rival theories, such as global workspace theory, with mixed results that both camps interpret in their favour.
The responsible summary in 2026 is simple: nobody knows whether consciousness can be defined by physical organisation, and if it can, nobody has agreed which organisation. Perceptronium is a stimulating way to frame the question, not an answer.
How science fiction uses the idea
Science fiction has always been interested in what counts as a mind. Ideas like perceptronium give writers a vocabulary for questions the genre already asks.
- Non-human minds: if consciousness is a state of organised matter, it could arise in a gas cloud, an ocean, a planetary network or a machine. The question becomes how to recognise it.
- Measuring minds: a society that could compute something like phi would face ethical decisions about which systems deserve rights. Fiction can explore what happens when that measure is wrong.
- Uploading and copying: if a mind is a pattern, copying it raises the problem of which copy is you.
- Watchers and guardians: stories about beings that observe or guide humanity often hinge on whether those beings experience anything at all, or merely process.
The Daimones Trilogy works in this territory of questions: it deals with humanity's survival and with intelligences that are not human, and it invites readers to ask what a mind is and what responsibility it carries. Readers interested in the theories above will find them a useful lens for that kind of story, without needing the novels to take a position on any particular theory.
A good speculative premise does not need to be true. It needs to be clear enough that readers can argue with it.
There is also a structural advantage. A story built around the question of whether something is conscious has a natural mystery engine: evidence accumulates, interpretations compete, and the answer, or the refusal to give one, can carry the ending.
Writing with the idea responsibly
For writers, the useful parts of these theories are their questions, not their equations. Use the core intuition, that a mind might be a pattern of integrated information rather than a particular kind of tissue, and let the plot test its edges. Avoid presenting phi or perceptronium as settled science, avoid invented precision, and remember that readers who know the field will notice when a contested theory is presented as consensus.
A practical approach is to decide in advance which version of the idea is true inside your story, and keep it private. Characters can argue, measure and get things wrong, but the world itself stays consistent. Readers sense when an author knows the rules even if the book never states them.
Handled that way, consciousness as a state of matter becomes what good speculative ideas should be: a lens that makes familiar questions strange again.
Questions
Is perceptronium a real substance?
No substance has been identified as perceptronium. It is a theoretical label Max Tegmark proposed for whatever arrangement of matter would be conscious, defined by information-processing principles.
Is integrated information theory accepted by scientists?
It is influential but strongly disputed. Researchers disagree about its axioms, its testability and its predictions, and experimental comparisons with other theories have not settled the debate.
Why does science fiction use these ideas?
They give a concrete vocabulary for old questions: whether non-human systems can be conscious, how we would recognise or measure a mind, and what copying a mind would mean.
Independent reader guides, science notes and writing craft.
