Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Test measures spark of consciousness

Researchers distinguish between vegetative and minimally conscious states.

It can be difficult for doctors to differentiate between vegetative and minimally conscious states. Credit: MASSIMO BREGA, THE LIGHTHOUSE/SCIENCE PHOTO LIBRARY

Consciousness is not all or nothing: some people languish for months or years in a middling state. A person could have some sense of and interaction with the world, or could be completely unconscious of their surroundings. But today in Science, researchers report a test that can distinguish between states of consciousness using a simple electroencephalogram (EEG) and some mathematics1.

The test suggests that the key difference between minimally conscious and totally unconscious non-coma states is communication between the frontal cortex — the planning, thinking part of the brain — and the temporal cortex, where sounds and words are processed.

The researchers say that the technique could help doctors to make accurate diagnoses about consciousness, and better predict how a patient will recover. Having this extra information might help families to make difficult decisions about a loved one's care or end of life in cases such as that of Terri Schiavo, who was removed from life support in 2005 after several years in a vegetative state.

A person in a vegetative state will open their eyes spontaneously and make reflexive movements, but has no cognitive function and likely does not feel pain. Recovery is possible, but the chances of improvement are greatly diminished after a year.

Someone in a minimally conscious state, by contrast, has intentional, non-reflexive but inconsistent responses to stimuli. They might speak a few words or track their image in a mirror, and they feel pain. The category has only been an official diagnosis since 2002, so the likelihood of recovery is not established, but again it decreases after one year in the state.

"The line between the two is quite complicated to draw," says Marta Garrido, a neuroscientist at University College London and a co-author of the study. Even when a team of doctors agree to diagnose a vegetative state, they have been shown to be wrong more than 40% of the time compared to the a standardized test, the Coma Recovery Scale2.

Complementing the gold standard

The Coma Recovery Scale is the current 'gold standard', says Melanie Boly, a neuroscientist at the University of Liège in Belgium and another co-author of the latest study. The scale scores patients on the basis of whether they are capable of conscious behaviours such as speech and directed gaze.

Boly and her colleagues sought a biological tool to complement the behavioural scale. They used EEG to measure electrical signals from the brains of 8 people in vegetative states, 13 in minimally conscious states and 22 healthy participants. The subjects were played a series of tones, which occasionally changed in pitch. The differing tone constituted a surprising event in the environment — something that the frontal cortex has to consider, so in all subjects the temporal cortex would send the frontal cortex a message.

In minimally conscious and healthy people the frontal cortex would then send a message back to the temporal cortex. The reason for this is uncertain; it may be to let the temporal cortex know what to expect in the future. But for people in a vegetative state, the communication was one-way: signals passed from the temporal to frontal area, but not back.

Researchers have tried to study consciousness using raw EEG data before, without much luck, says Nicholas Schiff, a neurologist at Weill Cornell Medical College in New York City. EEG data are measured through the scalp, so the technique provides at best a fuzzy approximation of what is going on inside the skull.

Garrido and her colleagues took their analysis a step further. They used dynamic causal modelling, a mathematical method developed by several members of the team, to create a computer model showing the best guess of what inner-brain activity could have yielded the EEG. The researchers have spent several years validating the technique, but it is only a model. A more direct experiment would be to use electrodes implanted in the brain, perhaps in animal studies, Boly suggested.

But Schiff is sceptical about the usefulness of the test as diagnostic tool. With so few subjects, it is impossible to know how well the test truly works, he says. Garrido said the EEG test will not replace the Coma Recovery Scale, but could be a useful supplement.

Schiff is most interested in what the test results say about consciousness. They do not necessarily imply that the connection between the frontal and temporal areas is the seat of consciousness itself; the signals that Boly and her team observed could simply be a proxy marker for whatever truly distinguishes the conscious mind. However, the research does imply that this and similar kinds of in-brain communication could be crucial for consciousness.


  1. Boly, M., et al. Science 332, 858-862 (2011).

    ADS  Article  Google Scholar 

  2. Schnakers, C., et al. BMC Neurol. 9, 35 (2009).

    Article  Google Scholar 

Download references


Related links

Related links

Related links in Nature Research

Nature Neuroscience Gateway

Related external links

US National Institutes of Health coma definition

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Dance, A. Test measures spark of consciousness. Nature (2011).

Download citation

  • Published:

  • DOI:


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing