Where in the brain does consciousness originate? Theories abound, but neuroscientists still haven’t coalesced around one explanation, largely because it’s such a hard question to probe with the scientific method. Unlike other phenomena studied by science, consciousness cannot be observed externally. “I observe your behavior. I observe your brain, if I do an intracranial EEG [electroencephalography] study. But I don’t ever observe your experience,” says Robert Chis-Ciure, a postdoctoral researcher studying consciousness at the University of Sussex in England.
Scientists have landed on two leading theories to explain how consciousness emerges: integrated information theory, or IIT, and global neuronal workspace theory, or GNWT. These frameworks couldn’t be more different—they rest on different assumptions, draw from different fields of science and may even define consciousness in different ways, explains Anil K. Seth, a consciousness researcher at the University of Sussex.
To compare them directly, researchers organized a group of 12 laboratories called the Cogitate Consortium to test the theories’ predictions against each other in a large brain-imaging study. The result, published in full on Wednesday in Nature, was effectively a draw and raised far more questions than it answered. The preliminary findings were posted to the preprint server bioRxiv in 2023. And only a few months later, a group of scholars publicly called IIT “pseudoscience” and attempted to excise it from the field. As the dust settles, leading consciousness researchers say that the Cogitate results point to a way forward for understanding how consciousness arises—no matter what theory eventually comes out on top.
On supporting science journalism
If you’re enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
“We all are very good at constructing castles in the sky” with abstract ideas, says Chis-Ciure, who was not involved in the new study. “But with data, you make those more grounded.”
What Are the Two Consciousness Theories?
The contenders in this face-off are, in some ways, direct inverses of each other. “The two theories are very different creatures,” says Christof Koch, a cognitive scientist at the Allen Institute in Seattle and a co-author of the Cogitate results. Global neuronal workspace theory takes what psychologists have learned about the brain and cognition to suggest that consciousness is a sort of stage. When a particular stimulus enters our conscious awareness, such as an annoying buzzing sound or a bright color, it gets thrust onto the stage and spotlighted.
Integrated information theory, Koch’s preferred framework, instead starts by defining what consciousness is more abstractly and then imagines what properties a system would need to have in order to experience it. Consciousness, IIT argues, arises from processing information—the more information, the more conscious a system can be, roughly speaking.
While these explanations are both quite abstract, the theories can be used to make testable predictions about what happens in the brain when someone consciously perceives something. For example, GNWT says that frontal regions of the brain such as the prefrontal cortex turn the “spotlight” on information when one first becomes conscious of it, a phenomenon called ignition that should be detectable in an imaging study when an individual has a conscious experience. And IIT says that conscious experience would emerge more toward the back of the brain, where nearby networks of neurons are more closely connected, Koch explains.
The Cogitate Consortium project, which began in 2018, set out to pit the predictions against each other at multiple labs run by theory-neutral teams. “These were people that did not have skin in the game,” explains Chis-Ciure. They used three different brain-imaging techniques to observe the brains of 256 participants—a very large sample size for a neuroscience study—while those participants completed the same visual tasks, which involved rotating faces and letters. These types of tasks require conscious experience.
The results challenge both theories because neither’s predictions were fully borne out by the data. For instance, GNWT predicted that electrodes in the prefrontal cortex would detect a signal when a stimulus disappeared and was removed from the mental stage of consciousness, but that was largely absent in the findings.* “The fact that you didn’t see that … is something that I think is a significant challenge,” says Seth, who wasn’t involved in the new study. IIT, on the other hand, predicted that a sustained synchrony of networks of neurons would occur at the back of the brain, which also wasn’t observed.
The researchers expected that the findings would be nuanced. “It was always understood … that a single experiment [wasn’t] going to refute a specific theory,” Seth says. That’s extremely rare in science, where knowledge builds incrementally. “You’re probably not going to change the minds of the proponents of each theory, but it’s more allowing the community as a whole to sort of alter their consensus about what’s going on.”
Will Scientists Ever Find the Source of Consciousness?
The results were first presented at a conference in June 2023. Koch and philosopher David Chalmers used the inconclusiveness of the findings to settle a long-running bet: in 1998 Koch bet Chalmers that neuroscientists could determine how consciousness arises in the brain within the next 25 years. Koch graciously conceded his loss at the 2023 conference.
In September of that year, an open letter that called IIT’s status into question first circulated online. The letter publicly raised the label of “pseudoscience,” saying that IIT was unscientific because its core tenets were not falsifiable, meaning they couldn’t be disproven with current technology. More than 100 authors ultimately signed on. Indeed, critics have linked IIT with panpsychism, the philosophy that consciousness perfuses everything, even nonliving entities. If consciousness arises from complex systems processing information, could a computer chip or even the whole universe be conscious?
The attempted “scientific excommunication,” to quote Seth, was ultimately hashed out in the pages of Nature Neuroscience last month. IIT is a bold theory, and “when it comes to consciousness, we have the right to be wrong and perhaps even the duty to be bold,” Seth wrote in a commentary in the journal.
Seth doesn’t think, as Koch does, that the results of the Cogitate study inspired the open letter. He sees it more as the result of a field at a crossroads. “Everyone has their own theory. And that’s not a great state of affairs. So I think there’s this feeling that, indeed, the theories need to become a bit more precise.”
What’s Next?
During the total solar eclipse of May 1919, two competing theories of gravity faced off against each other: Isaac Newton’s classical explanation versus Albert Einstein’s then new general theory of relativity. Scientists aimed their telescopes at the eclipsed sun to test Einstein’s prediction that our home star’s gravity would bend the light of distant stars. It did, vindicating Einstein.
Nearly 100 years later, Koch and others involved in the Cogitate Consortium wanted to test consciousness theories in a similar way. This kind of adversarial collaboration, as it’s called, can be productive and instructive because it requires both camps to meet and test predictions on neutral ground, with carefully controlled and reproducible findings. Other adversarial studies on consciousness are ongoing. Seth, for example, is on the steering committee of another adversarial collaboration between IIT and two other theories of consciousness.
This type of research “will encourage a new way of doing studies, which is to design experiments that have the best chance of distinguishing between theories rather than finding evidence for or against one specific theory,” Seth says.
For a field with such a lofty goal as explaining the phenomenon of consciousness, “it’s often easy to get lost in the existential fog,” Seth notes. But “even if we don’t have an adequate theory, we’re making progress in really important, practical applications,” such as understanding consciousness in cases of brain damage and coma or during general anesthesia.
“There are people that are taken away from life-support intervention because they are deemed as not being ‘there,’” Chis-Ciure says. “The stakes are too high to not tackle the problem head-on.”
*Editor’s Note (4/30/25): This sentence was edited after posting to correct the description of the GNWT prediction that was largely absent in the findings.