Share This Page
Can the key to consciousness be found in the folds of the cerebrum? Can the simple unfettered state of “being conscious” be localized in the brain, its properties deconstructed to precisely timed patterns of neural firing? Finding the answers is the goal of a $20 million international research program to search for the neural footprints of consciousness.
The broad, multi-year initiative—termed Accelerating Research in Consciousness (ARC)—is being funded by the Templeton World Charity Foundation. In the first phase, representing $5 million, two leading brain theories of consciousness with diametrically opposed assumptions will face off to test their hypotheses. ARC pits the Integrated Information Theory (IIT) and the Global Neuronal Workplace (GNW) theory directly against one another, in what Templeton calls “adversarial collaboration,” to settle some fundamental questions about how, when, and where the brain processes subjective awareness of ourselves and the world around us.
The two theoretical models are in stark contrast to one another: their definitions and assumptions of what constitutes consciousness differ and their whole approach to the subject is fundamentally different. What they have in common is that they both study the neural correlates of consciousness.
IIT is the brainchild of Giulio Tononi, a professor and director of the Wisconsin Institute for Sleep and Consciousness at the University of Wisconsin. GNW has been elaborated by Stanislas Dehaene of INSERM/Unicog, in concert with Lionel Naccache of Sorbonne/INSERM, Jean-Pierre Changeux of Institut Pasteur, and others. These two theories were selected by Christof Koch, a leading consciousness researcher who is serving as an advisor to the Templeton project, because each has an established following among scientists and a “preponderance of evidence” backing them, says Koch, who now heads the Allen Institute for Brain Science.
Why This Approach?
ARC is audacious not only in its approach and its subject matter, but also in its commitment to model best practices in open science. The underlying premise is that meaningful progress on big questions like consciousness requires focused, structured collaboration beyond what any isolated research group can do.
“The days of the lone genius scientist, the chap in the lab who solves the big problem, are pretty much over,” says Dawid Potgieter, senior program director at Templeton, which is bankrolling the project on the premise that siloed science is the enemy of progress. “There’s a need to do science differently,” notes Potgieter.
“What’s happened over the last 50 years, in biomedical sciences in general, is that you never have a single experiment that tests two competing theories,” says Koch. “[Adversarial collaboration] requires people to work together in a productive way so disagreements can be tested.” He points to the experiments of 1919 that directly tested Einstein’s then recently introduced Theory of General Relativity against the prevailing Newtonian view of the universe, largely settling the argument in favor of Einstein. In contemporary science, Koch says: “This is very rarely done.”
Naccache, a French neurologist and neuroscientist who is a co-architect of GNW, says the nature of the consciousness question calls for a bold approach. “We don’t yet have a full theory of consciousness. We have only sketches—‘esquisses’ in French,” he says. “The best way to go beyond our current knowledge is to provoke collisions between these theories in order to test their respective core ideas, and to go forward with new ideas,” he adds.
ARC provokes collisions by bringing the leaders of opposing views to the same table—literally, over the course of a multi-day workshop—to hash out and agree to “killer ideas” that could disprove the other’s theory. Then it turns to six top-notch independent laboratories to empirically test the predictions and find out who’s got it right—or at least, who might be closer to right. The scientific protocol is designed to ensure scientific objectivity and rigor with maximal transparency, with an eye toward moving the field forward. A team of independent investigators will direct the scientific program, led by principal investigator Lucia Melloni of Max Planck Institute and New York University and co-principal investigators Liad Mudrik of Tel Aviv University and Michael Pitts of Reed College.
The Templeton initiative reflects the level of maturity of a field that used to be a no-go for young scientists. Tononi, who became intrigued by the topic as an adolescent contemplating ethical dilemmas, recalls that young people were strongly dissuaded from going into this area of investigation. “I asked neuroscientists at the time, and I was literally told ‘Shut up. Don’t even ask. Hush. Go away.’” It was considered an occupation, he says, “for aging Nobel prize winners,” a reference to Francis Crick and Gerald Edelman, who both took on consciousness only as senior, world-recognized scientists.
“Crick could do it because he was a half-god,” deadpans Koch, who collaborated with Crick on a seminal 1990 paper proposing a roadmap for rigorous investigation of the neural correlates of consciousness. “Retired people could do it, but reasonable working scientists didn’t work on consciousness. It was career-killing,” says Koch.
Melloni came to the field as a Ph.D. a decade after Koch and Crick’s influential paper, which she said re-energized thinking about consciousness and laid out concrete steps for its investigation. “From that moment on, somehow consciousness went back to business,” she says.
Scientific theories about consciousness are now ubiquitous and proliferating. “We’ve gone from ‘you couldn’t talk about it’ to ‘everyone can talk about it,’” Tononi says, pointing to the plethora of books on the subject before adding wryly: “It’s not necessarily making things easier.”
IIT and GNW take fundamentally different approaches to trying to understand consciousness. GNW, according to Naccache, is inspired by knowledge of the psychological properties of being conscious. It says that as soon as you are conscious of something—a face, a sound, a memory, a feeling—you can not only self-report it (e.g., I see X, I hear X, I remember X, I feel X) but you can also apply all your cognitive abilities to it – you can think about it, you can remember something related to it, or make some plan of action as a result of it. This “cognitive availability,” a term coined by psychologist Bernard Baars, who proposed the Global Workplace theory, is at the core of the model. As such, GNW takes a functionalist approach. It says consciousness is a function of the brain, so let’s look for where in the brain this function is orchestrated.
In contrast, IIT starts from consciousness itself, the subjective experience of being as opposed to doing. Rather than viewing consciousness as a particular brain function and searching for the neural correlates, it describes the essential properties of consciousness, including the core concept of how information is integrated to engender awareness of the world and self. Then, IIT theory postulates that the physical substrate of consciousness will share those essential properties, providing a sort of treasure map to the kind of neural environment likely to support conscious being.
Back or Front?
The IIT approach has led Tononi and others to look at cortical tissue in the back of the brain, where primary sensory areas form a dense “supergrid” of neuronal connections that are integrated both vertically and laterally. This particular part of the brain, marked by extraordinary complexity, [fits] the essential properties of consciousness as enumerated in IIT.
“If we could unfold the structure of the back of the brain, the complexity there is astronomical,” Tononi says. “It boggles the mind, and fills you with a certain humility and awe…,” he notes.
In contrast, GNW theory postulates that a network engaging both the front and the back of the brain is where the action is. According to Naccache, the cognitive availability that characterizes consciousness should be associated with a neural availability, a particular functional architecture in the brain. GNW theory predicts this neural signature is long-distance, coherent, and complex, a kind of sustained, complex conversation between brain regions engaged in high-level neural processing, located mostly in frontal and parietal regions of the neocortex, he says.
“This is where the theories disagree,” Koch says. This split is of particular interest because there is no universal agreement as to whether the essential wiring for being conscious is—roughly speaking—in the back of the cortex or the front.
The models also diverge in how they envision the timing of neural processing. IIT says as long as one remains conscious of something, a neural correlate representing that “something” will be evident in the brain. GNW postulates that the moment of “conscious access” will be associated with a particular neural representation that is separate and distinct from activity surrounding that moment. A central question of GNW is how to disentangle the actual neural signature of conscious access from events just before and after.
The experimental protocols and methods selected by the investigative teams are designed to tease out these two issues of localization and timing. Study participants will perform tasks designed to pinpoint the moment of conscious recognition and researchers will look at the data for the corresponding neural signatures. Three different neuroscience techniques will be used: functional magnetic resonance imaging (fMRI), which tracks patterns of activity in the brain with high spatial resolution; magneto-electroencephalography (MEG/EEG), a method for recording electrical activity in groups of neurons noninvasively and with high temporal resolution; and electrocorticography (ECoG), direct recordings from implanted electrodes. The latter procedure, which offers the best spatial and temporal resolution, is done in patients with intractable epilepsy as part of treatment to localize seizure initiation to determine where to intervene.
Data collection and analysis for this initial phase of the broader Templeton-funded program is expected to take three years. Subsequent phases, which have already been initiated, will match up other theories to face off and run complementary animal studies in rodents and nonhuman primates to examine questions that cannot be answered with human subjects. At the conclusion of each phase, all data will be made publicly available in an open-science protocol intended to propel progress toward a more comprehensive view of how the brain does consciousness.
The search for the neural underpinnings of consciousness is more than academic; the implications are broad and diverse if not immediately obvious. Improving the detection of conscious awareness in unresponsive patients is one clear application for this type of research, and Tononi’s work has already yielded promising results in developing an objective measure of consciousness. More far-fetched but entirely plausible applications include managing chronic pain or treating mental illness such as depression; the idea being that if we understand how the brain processes subjective awareness, it may be possible to tweak the system to alter one’s experience of pain or illness. Outside the realms of medicine, a current hot topic is when and whether sophisticated machine learning systems that mimic brain processing might actually be “conscious” to some degree, and what that means for the burgeoning field of Artificial Intelligence.
“We see consciousness as a big concept that needs to be solved before one can expect to solve other big problems in society,” says Templeton’s Potgieter.
For his part, Koch sees consciousness as a question that science cannot afford to ignore. “If science cannot explain how my conscious mind comes into the world, then it’s leaving a gigantic hole in the center of our everyday existence,” he says. He goes on to note that “It’s a simple challenge for science. Science so far has not done it. It’s time that we do.”