You wake up in a hospital, eyes open but completely unable to move. You can’t even blink. How’s anyone to know you’re conscious, let alone aware?
While that’s an extreme case, there’s a real-worldneed to understand whether patients with more common disorders of consciousness, such as those in a vegetative state, are alive and thinking. Now, researchers report they’ve taken a step in that direction by comparing patterns of electrical activity in the brains of healthy adults with those of patients suffering from a consciousness disorder.
Inspired by research indicating a small number of patients might be at least marginally aware and able to control their thoughts despite showing no outward signs of consciousness, Srivas Chennu and an international team of neuroscientists decided to see whether they could identify consciousness using electroencephalography, or EEG, which tracks oscillating electrical signals from the brain and measured on the scalp. The team collected 10 minutes of EEG data from 91 points on the heads of 32 patients in a vegetative or minimally conscious state as well as a control group of 26 healthy men and women. Next, they broke the data down according to the electrical signals’ frequency bands, commonly known as delta (0-4 Hertz, or cycles per second), theta (4–8 Hz), and alpha (8–13 Hz).
A small number of patients might be at least marginally aware and able to control their thoughts despite showing no outward signs of consciousness.
On the first pass through the data, the team noticed that their patients tended to have stronger delta-band and weaker alpha-band signals compared to those of healthy people, but really getting a handle on the data required a more sophisticated approach. First, the researchers computed correlations between delta, theta, and alpha-band signals from each pair of the 91 EEG measurement points. From those correlations, they next built a connectivity network, a graph showing the strongest correlations—hence the strongest connections—between different parts of the brain. Finally, they compared graphs from the patients to those of healthy people using measures such as clustering, which describes how dense the connections are between a subset of points in the brain, and modularity, a measure of how easily one could break the graph down into smaller components by cutting individual links.
Healthy subjects, the neuroscientists found, had more clustered and less modular alpha-band networks than patients, and alpha networks also spanned a greater physical distance in healthy people than in patients. (Though it’s a bit counter-intuitive, clustering and modularity don’t actually take physical distance into account.) Much the opposite was true of delta- and theta-band networks: These were more clustered and less modular in patients than in healthy controls, although they didn’t extend as far across the brain as alpha networks did in control subjects. Finally, behavioral and EEG data combined suggested that the more responsive a patient was, the more that patient’s alpha-band connectivity resembled a healthy person’s.
Those last two points could be key, the authors argue today in PLoS Computational Biology. The shift to delta- and theta-band networks in patients with consciousness disorders doesn’t bring quite the same pattern of connectivity as healthy alpha-band networks, suggesting that it’s the long-range connections that underlie consciousness.
