The search for neural correlates
In 1990 Francis Crick and Christof Koch made a bold move: instead of solving consciousness philosophically, look for its neural correlates, the minimal brain activity that reliably goes with a given conscious experience. This reframing launched a whole research program. Find the switch that, when flipped, turns a specific experience on or off, and you have a foothold, even if you do not yet know why the switch feels like anything.
Decades of work show consciousness is not housed in one spot but depends on the thalamocortical system, the looping traffic between the thalamus and the cortex. The cerebellum, packed with most of the brain's neurons, can be lost without abolishing consciousness, while small strokes in the brainstem or thalamus can end it. So the question is less about a place and more about which networks, doing what, are jointly sufficient for experience.
Front versus back: the hot-zone debate
A sharp dispute concerns whether the front or the back of the cortex matters most. Koch, Tononi and colleagues argue for a posterior 'hot zone' in parietal, occipital and temporal regions, where the content of an experience, the actual face or color, appears to be constituted. On this view the prefrontal cortex is doing reasoning, reporting and monitoring around the experience rather than generating it.
The rival view, global workspace theory associated with Stanislas Dehaene and Bernard Baars, gives the prefrontal cortex a starring role: an experience becomes conscious when information is 'ignited' and broadcast widely, especially through frontal networks, making it available across the brain. Part of the disagreement is methodological, because tasks that require a report naturally light up frontal regions, which can inflate their apparent importance.
The 2025 adversarial test, and why 'where' isn't the whole story
In 2025 a large adversarial collaboration nicknamed Cogitate put these two theories head to head, publishing in Nature. Rival proponents agreed in advance on predictions, then a neutral consortium tested 256 people using fMRI, MEG and intracranial recordings. The results bruised both camps: the posterior cortex did not show the sustained connectivity integrated information theory predicted, and the workspace theory struggled with the absence of the frontal 'ignition' it expected when a stimulus ended.
The lesson is that 'where' is only part of the question. Locating the circuits is real progress and clinically valuable, for instance in assessing awareness in unresponsive patients. But knowing which networks light up does not tell us why that activity is accompanied by felt experience. A famous 25-year wager between Koch and philosopher David Chalmers ended in 2023 with the honest verdict that the deeper puzzle is not yet solved.