I’ve just clocked a stunning experiment, shortly to be published in the Journal of Cognitive Neuroscience, that recorded brain activity from patients who had part of their skull surgically removed for several months and had only flaps of skin between their brain and the outside world.
The operation is called a hemicraniectomy and is often used when the brain swells or the pressure builds up inside the skull to the point where it is damaging the brain.
Neurosurgeons will sometimes remove a portion of the skull (see the scans on the left) and just leave the scalp protecting the brain until the swelling subsides, before replacing the skull flap some months later.
As an aside, sometimes the surgeons will surgically insert the piece of skull into the abdomen so the bone marrow doesn’t die and it can be replaced ‘alive’ when the time comes. There’s a great description of this here.
The patients normally wear helmets, for obvious reasons, but they are unique in having such a thin covering of the brain.
A team of researchers, led neuroscientist Bradley Voytek, realised this provided a unique opportunity to examine the exactly how the skull affects EEG, one of the most common techniques for measuring the electrical activity of the brain.
EEG records brain activity from electrodes on the skull, but the signal gets ‘smeared’ as the electrical charge passes through the bone and so the source of the activity can’t be located very precisely to specific brain areas.
By working with the hemicraniectomy patients, the researchers could compare electrical activity on one side of the brain – recorded through just the skin, and the other, where recordings were made normally – through electrodes on the skull.
The researchers found that the non-skull signals were richer, were less subject to interference, were more closely tied to specific tasks and could be better linked to specific brain areas.
On the right is a comparison of the signal coming from a listening task, where participants are suddenly presented with an ‘oddball’ noise in the midst of a bunch of otherwise identical sounds. The brain reacts strongly to the change and this is reliably reflected in a positive spike in the electrical activity at about 50 milliseconds (consequently, the wave is called the ‘P50’ signal).
You can see that the activity on the craniectomy side is much stronger, tighter and cleaner whereas on the skull side it is quite indistinct. The team found similar results in several other tasks.
This not only helps us better understand EEG results on people with intact skulls, but it also meshes with brain activity recordings that are taken from electrodes implanted directly in the brains of patients undergoing neurosurgery.
Link to PubMed entry for study.
pdf of scientific article.
Link to Bradley Voytek’s blog post about his work.