Why does anything feel like anything?

Particular lesions remove particular kinds of experience: a lesion outside the striate cortex can take visual motion away and leave the rest of vision working. Anaesthetics suspend consciousness and return it, reliably enough to operate on people, though which of their effects removes experience rather than only the report of it is still argued. Stimulating specific cortical regions produces specific sensations. The dependency is not vague. It is detailed enough to be used in surgery.

Using binocular rivalry, masking and related paradigms, researchers can hold the stimulus constant while what the subject reports experiencing changes, and watch which brain activity tracks the report. Several of these markers recur across studies, though some may track the act of reporting rather than the experience, and newer designs try to do without the report.

These are correlates. Knowing that a state accompanies an experience is not the same as knowing why it produces one.

Thomas Nagel argued in 1974 that an organism is conscious if there is something it is like to be that organism, and that this subjective character is not captured by the reductive analyses on offer. He stopped short of calling physicalism false: his position was that we have no conception of how it could be true. David Chalmers later separated the “easy problems” (attention, reportability, integration, all hard in practice but conceptually tractable) from the hard problem: why physical processing is accompanied by experience at all.

Global workspace theories hold that consciousness is what happens when information is broadcast widely enough across the brain to become available to many systems at once. Integrated information theory holds that consciousness is identical to a particular kind of causal structure, and assigns it a quantity. These are not variations on a theme; they make different predictions and imply different things about which systems are conscious.

A large adversarial collaboration has tested them directly against each other, and a second is under way in monkeys and mice. The results so far constrain both without settling between them.

There is no agreed test for whether a system is conscious. This is not an academic gap: it affects clinical decisions about unresponsive patients, questions about which animals can suffer, and increasingly, questions about artificial systems. In one study of 241 brain-injured patients with no observable response to commands, a quarter performed a cognitive task detectable on fMRI or EEG.

Nothing in physics, chemistry or neuroscience predicts that any arrangement of matter should feel like something from the inside. A complete functional account of the brain (every signal traced, every computation described) would still, on Chalmers’ argument, leave this unexplained.

That is what makes it hard. It is not that the answer is complicated. It is that nobody agrees on what an answer of the right shape would look like.

We infer consciousness in other humans from similarity. That inference weakens smoothly as we move to other mammals, to birds, to octopuses, to insects, and it provides little guidance for systems that are behaviourally sophisticated but built differently.

Integrated information theory proposes that experience is what a system with a particular causal structure is, from its own perspective. Its authors say it makes testable predictions; a group of critics argued in 2025 that its core claims cannot be tested even in principle. It has the unusual property of implying that consciousness is widespread and graded rather than rare and binary.

On global workspace accounts, there is no additional fact to explain: being conscious of something just is having it globally available. Chalmers replies that this shows promise as an account of the easy problems and offers no explanation of the hard one.

Some philosophers argue that if experience cannot be derived from physical description, it may belong in the inventory of basic features of reality rather than being built from them. This is a serious position in the literature and it is not established. It also trades one mystery for another: why this experience, in this arrangement?