Duel between two theories of consciousness leaves without a winner – 05/07/2023 – Science

On a sultry June night in New York City, more than 800 neuroscientists, philosophers and onlookers packed an auditorium. They were looking for the first results of an ambitious investigation into a profound question: what is consciousness?

To start, two friends —philosopher David Chalmers and neuroscientist Christof Koch— took the stage to recall an old bet. In June 1998 they went to a conference in Bremen, Germany, and one night ended up talking late in a bar about the nature of consciousness.

For years, Koch collaborated with Francis Crick, a biologist who shared the Nobel Prize for discovering the structure of DNA, in the search for what they called “the neural correlate of consciousness.” They believed that every conscious experience we have—looking at a painting, for example—is associated with the activity of certain neurons essential to the consciousness that goes with it.

Chalmers liked the concept, but was skeptical that such a neural marker could be found anytime soon. Scientists still had a lot to learn about consciousness and the brain, he thought, before they could have a reasonable hope of finding it.

Koch gambled that scientists would find a neural correlate of consciousness within 25 years. Chalmers accepted the bet. The prize would be a few bottles of good wine.

Recalling the gamble on the auditorium stage, Koch admitted it was spurred on by drinks and enthusiasm.

“When you’re young, you have to believe that things will be simple,” he said.

Much has happened over the subsequent quarter century. Neuroscientists and engineers invented powerful new tools to probe the brain, leading to an explosion of eye-opening experiments on consciousness. Some scientists have used brain scans to detect signs of consciousness in people diagnosed as being in a vegetative state, for example, while others have used brain waves to determine when people become unconscious under anesthesia.

These experiments sparked an explosion of new theories. To narrow them down, the Templeton World Charity Foundation has begun supporting large-scale studies that put different pairs of theories to a head-to-head test, a process called adversarial collaboration.

And last month researchers at the event in New York revealed the results of the foundation’s first test, a combination of two of the most prominent theories.

The first, known as Global Workspace Theory, holds that consciousness is a by-product of the way we process information. Neuroscientists have known for a long time that most of the signals that come from our senses do not reach our consciousness. Experiments led by Stanislas Dehaene, a cognitive neuroscientist at the Collège de France in Paris, suggest that we become aware only of signals that reach the prefrontal cortex, a region at the front of the brain. Dehaene argued that a special set of neurons can rapidly transmit information throughout much of the brain, generating consciousness.

“Awareness is the total availability of information,” Dehaene said.

Melanie Boly, a neurologist at the University of Wisconsin, took the stage to explain the other competitor: Integrated Information Theory.

What makes consciousness special, Boly argued, is the way it manages to appear both rich and unified over time. Brains can produce this phenomenon thanks to the way neurons are organized, she said. Groups of them can process information in specific ways—identifying the colors or outlines of an image, for example. But the long-range links between these groups also allow them to pass on information.

In 2017, Koch, then working at the Allen Institute in Seattle, invited a dozen experts to design experiments that would test the two theories against each other. Chalmers also came from New York University to provide philosophical rigor. They agreed in advance on what the results of each experiment would mean for each theory. And the experiments would be conducted by independent scientists who didn’t push for any of the theories.

The Cogitate Consortium, as the team called itself, took two years to prepare the experiment, but it was interrupted by the coronavirus pandemic. In May 2022, researchers were able to start collecting data.

They asked 256 volunteers to look at a series of faces, letters and shapes and then press a button under certain conditions — if the image was a face, for example, or the face of a certain person.

Some of the volunteers performed tasks on an fMRI brain scanner, which measures the flow of oxygenated blood in the brain. Others were seen with magnetoencephalography, which reads magnetic fields in the brain. The researchers also found volunteers preparing to undergo brain surgery for epilepsy. They were tested with implants inserted directly into their brains.

The researchers looked for common brain patterns that emerged whenever volunteers had the conscious experience of seeing an object — regardless of what they saw, what their task was or what technology recorded their activity.

The two theories made different predictions about what patterns scientists would see. According to Global Workspace Theory, the clearest signal would come from the prefrontal cortex because it relays information throughout the brain. Integrated Information Theory, on the other hand, predicted that the regions with the most complex connections — those at the back of the brain — would be the most active.

The timing of the activity could also point to one theory or another. The Global Workspace Theory predicted that the prefrontal cortex would send only short bursts of information: one when an image first appeared and another when it disappeared. But Integrated Information Theory predicted that the back of the brain would be continuously active the entire time volunteers were perceiving an object.

Lucia Melloni, a neuroscientist at the Max Planck Institute for Empirical Aesthetics in Germany, who helped conduct the experiments, took the stage to present the results, with images of brains splashed in red, blue and green projected onto a giant screen.

Melloni explained that in some tests there was a clear winner and a clear loser. The activity in the back of the brain lasted as long as the volunteers viewed an object, for example. A point for the Theory of Integrated Information. But in other tests the predictions of the Global Workspace Theory were confirmed.

“The current experiment is enough to show that neither theory is currently sufficient,” said Anil Seth, a neuroscientist at the University of Sussex in England.

But the 25-year wager, at least, has been settled: no one has found a clear neural correlate of consciousness. Koch ended the evening by bringing a wooden box full of wine onto the stage. He picked up a bottle of 1978 Madeira and handed it to Chalmers.

Then he challenged his friend to a new bet, this time double or nothing: a brain marker of consciousness until 2048.

Chalmers instantly accepted the challenge, despite the questionable odds that the two are still alive to see the result.

“I hope to lose,” he said. “But I suspect I will win.”

^{Translated by Luiz Roberto M. Gonçalves}