For much of the twentieth century, many scientists treated consciousness as too vague to study. That changed around 1990, when the British biologist Francis Crick, famous for his work on DNA, and the American neuroscientist Christof Koch began arguing that it was time to look for consciousness in the brain. Today it is a busy field with rival theories, large experiments and sharp public quarrels.
Hunting for correlates
Crick and Koch proposed a modest first goal: find the neural correlates of consciousness, the minimal brain activity that is enough for a particular conscious experience. A correlate is not an explanation. It tells you where and when, not why. But finding it would narrow the search.
The key method is contrast: keep the stimulus the same while experience changes, and see what in the brain changes with it. In binocular rivalry, each eye is shown a different image, say a face to one eye and a house to the other. People do not see a blend. They see the face, then the house, then the face, switching every few seconds while the input stays constant. Researchers can then ask which brain activity tracks what the person reports seeing, as opposed to what the eyes receive.
Global workspace theory
Proposed by the psychologist Bernard Baars in 1988 and developed by the French neuroscientist Stanislas Dehaene and colleagues, global workspace theory holds that information becomes conscious when it is broadcast widely across the brain, so that many specialized processes, such as memory, planning and speech, can use it at once. Most processing happens locally and unconsciously. When a signal is strong enough, networks spanning the front and sides of the brain suddenly light up together, a sudden spread Dehaene calls ignition, and the content becomes available everywhere. The theory fits well with access consciousness, and critics say that is its limit: it explains availability, not feeling.
Integrated information theory
The Italian American neuroscientist Giulio Tononi proposed integrated information theory in 2004. It starts from features that experience seems to have, such as being unified and highly specific, and claims that consciousness is integrated information: information generated by a system as a whole, over and above its parts. The amount is measured by a quantity called phi. The theory has striking consequences. Some simple but highly interconnected networks would have a little consciousness, while a conventional digital computer would have very little, whatever program it runs, because of how its circuits are wired. Supporters see a rare attempt to tackle experience head on. Critics say phi is nearly impossible to calculate for real brains and that the theory's strange implications count against it.
Higher-order theories
A third family, defended by the American philosopher David Rosenthal among others, holds that a state is conscious when the mind represents itself as being in that state. Seeing red becomes a conscious experience when you are, in a suitable way, aware of seeing red. These theories link consciousness to the front of the brain and to self-monitoring, and they raise questions about animals and infants whose capacity for such self-representation may be limited.
Putting theories to the test
In an unusual experiment, supporters of global workspace theory and integrated information theory agreed in advance on predictions that would favor one theory over the other. Independent labs then ran the tests. This approach is called an adversarial collaboration. Early results, announced in 2023 and published in full in 2025, challenged key predictions of both theories and crowned neither. Later in 2023, more than a hundred researchers signed a public letter calling integrated information theory pseudoscience, and many others objected that the label was unfair. The episode showed both how far the field has come and how unsettled it remains.
Note what this research does and does not do. It ties particular experiences to particular brain activity, which matters greatly in medicine. It does not by itself answer the hard problem, since a correlation does not explain why the activity feels like anything.
Worked example: imagining tennis
Some patients survive severe brain injury and open their eyes, yet show no sign of awareness. Doctors may diagnose a vegetative state. In 2006 a team led by the British neuroscientist Adrian Owen placed such a patient, a young woman injured in a traffic accident, in a brain scanner and gave her spoken instructions: imagine playing tennis; imagine walking through the rooms of your home. Her brain activity matched that of healthy volunteers doing the same tasks, in the regions linked to movement for tennis and to navigation for the house.
The result suggested that some patients judged unaware may be conscious but unable to move. Later studies used the same method to let a few patients answer yes or no questions. Apply the level's ideas. The scan shows following instructions, a function linked to access. Most researchers take it as strong evidence of experience too, because it is hard to see how someone could follow such instructions without it. Yet the case also shows the stakes: decisions about care, pain relief and the end of life depend on knowing whether anyone is there.
