Does consciousness—our awareness that we are perceiving something—arise from a special region in the brain, or from the coherent workings of multiple regions? Analyzing data from electrodes implanted in the brains of epilepsy patients, French researchers suggest the latter, although their results, published March 17 in the online journal PloS Biology, also point to a role for special, consciousness-related circuits in the prefrontal cortex.
Lionel Naccache, senior author on the paper and a researcher at Pitié-Salpêtrière, a teaching hospital in Paris, says he found that ordinary nonconscious visual perception was reflected in a quick sweep of activity from the primary visual cortex at the back of the brain to the prefrontal cortex. Such activity was liable to fade away just as quickly, but above a certain threshold, it evoked a sustained “long-distance coherent communication” between prefrontal areas and other areas of the cortex, a phenomenon that corresponded to conscious perception in his research participants.
“It’s a nice paper,” says Christof Koch, a neuroscientist and consciousness researcher at the California Institute of Technology. “If one can generalize this result [in other humans] and maybe do it in monkeys, it could be useful as a signature of consciousness. So it’s definitely a step forward.”
The phenomenon of consciousness lacks any basis in current standard theories of physics or biology. Cognitive scientists therefore have not had any objective method for measuring it and have had to content themselves with a search for its “neural correlates”—the specific neuronal activity that is necessary and sufficient for people to be aware of whatever their brain perceives. But even this quest has proved difficult.
To isolate these correlates, explains Gabriel Kreiman, a specialist on the neural workings of vision at Harvard Medical School, “we need to try to dissociate conscious from unconscious processing under situations where all the other variables remain as constant as possible.”
Such a task has never been easy. It requires a careful experimental design as well as fine-grained measurement of the geography of brain activity and its timing patterns—measurement that generally lies beyond the capacity of magnetic resonance and other noninvasive imaging technologies.
Yet a research team led by Naccache has now tried to meet these criteria, using electrodes implanted in participants’ brains and a procedure that can momentarily hide images from consciousness.
Surgeons implanted electrodes deep into the brain tissue of 10 epilepsy patients being prepared for surgery. The electrodes guided them to the precise brain region responsible for the patients’ seizures. Only in circumstances such as these is it considered ethical to make use of implanted brain electrodes for scientific purposes, but the findings based on such experiments are increasing.
Participants were shown a random sequence of images featuring either a word or a blank white space, each of which flashed on a screen for 29 milliseconds. For a randomly chosen selection of these images, the researchers flashed a “mask” of “&” symbols for 400 milliseconds after an image disappeared.
The masking technique is standard in perceptual research, yet as Naccache and his colleagues acknowledge, the presence of the mask was likely to complicate the data by activating brain areas on its own. However, by measuring the difference between the brain state induced by masked blank images and that induced by masked word images, the researchers were able, in principle, to isolate the state induced by word images. To confirm whether a participant had perceived an image, consciously or not, the researchers used words that were either “threatening” or “nonthreatening” and had participants press a button to indicate which of these came to mind.
Using the electrodes, the researchers eavesdropped on the brain’s electrical activity following each image presentation. Over several hundred trials for each participant, they were able to map what appeared to be key differences in brain-activity patterns between conscious and nonconscious word perception.
Essentially, they found that during nonconscious perception, activity occurred in multiple areas of the cortex, yet never became coherent—firing in sync—over large distances; this nonconscious activity also dissipated relatively quickly.
By contrast, during conscious perception the activity was able to “ignite” into much longer-term, self-reinforcing, interconnected activity across widely separated cortical areas. This coherent activity included areas of the prefrontal cortex and appeared to be concentrated in the “gamma wave” range of frequencies, which previous research has linked to attention and consciousness.
Naccache says that his group, which includes cognitive scientists at the French national research institute INSERM, now plans to do confirmatory experiments with a slightly different experimental design, using a technique known as “attentional blinking”—which does not weaken the intensity of the stimulus image in the way the masking technique does.
Other labs, including Kreiman’s, are working on similar experiments using a technique known as “binocular rivalry,” in which conscious perception is made to shift from an image in the left visual field to one in the right, or vice versa. “We are just at the very beginning of trying to formulate the relevant questions about consciousness,” says Kreiman.
Koch, too, notes that even when research is able to identify a region such as the prefrontal cortex as a key player in the circuits of consciousness, it needs to go further still: “We have to ask, what’s different about prefrontal areas? What’s the difference between prefrontal and parietal such that activity in prefrontal cortex can give rise to consciousness, but not activity in parietal cortex? There has to be some more abstract, general explanation for that.”
Among possible explanations, Koch tentatively favors the “Information Integration” theory of consciousness put forward by neuroscientist Giulio Tononi at the University of Wisconsin. Tononi proposes that consciousness is a fundamental property arising from any system that uses interdependent, information-exchanging parts. By this logic, the most powerful consciousness-generating networks of the brain would be those that integrate the largest amount of neural activity—as the results from Naccache and colleagues also suggest.
The theory implies, however, that consciousness is not limited to highly evolved animals or even to biological brains. As Koch puts it, “Whether it’s my iPhone or the flatworm C. elegans or the human brain, it would differ only in the amount of consciousness. But all would be conscious.”