There's been a small, but stunning, step forward in the quest to help people who have suffered consciousness-impairing injuries. Scientists inserted electrodes into the brain of a man in a minimally conscious state. They used the electrodes to stimulate parts of the brain believed to be crucial for binding together the brain into an aware state. As the doctors hoped, the stimulation made the man more responsive. He was able to name objects and to hold a cup to his lips. The details of the experiment appear in this week's issue of Nature.
One experiment on a single subject is a far cry from a clinical trial. But I was still very impressed by it. Perhaps that's because I spent some time with these particular scientists a few years ago, back when they were just trying to figure out how to understand what was going on in the heads of their patients. I wrote an article about the experience for the New York Times Magazine in 2003, which I've posted here. If I remember correctly, it was the first account of their research to appear in the press. Consider it the prequel.
[ Image credit: from animation created by the Cleveland Clinic]
Comments
Kind of a new way of looking at "artificial intelligence". Are we possibly on the road to "hybrid intelligence"?
Posted by: John B. | August 1, 2007 3:07 PM
This appears to be an impressive finding. It certainly is food for thought.
I'm a neuroscientist, but this general area is not my area, so I'm mostly an amateur here.
What I like about this finding is its implications on our thoughts about recovery of function. Our simple mental model of the brain is that a brain cell is either dead, or injured or healthy. If healthy, it will do the best it can do. When a brain is damaged some sets of cells die and others are injured. According to the simple model, recovery of function involves getting injured cells healthy. Among other things, this experiment demonstrates that this simple model is wrong. There is no room in this model for electrical stimulation of the brain inducing spontaneous and rapid recovery. For it is impossible for brain stimulation to instantaneously repair damaged neurons.
My thinking about this stems from a personal anecdote. My mother in law had a stroke which produced a visual field deficit. Almost half of what she saw on the left side of each eye was, according to her "cloudy". After thorough neurological workup it was determined that she had a stroke that affected parts of visual cortex in her right hemisphere. Presumably, this region of cortex contained neurons that were either dead or injured. Standard neurological advice is that, after such an incident there may or may not be recovery. If there is to be recovery, it is likely to be within a period of weeks to months.
About a month after the stroke, while we were walking thru a mall, my mother in law said something like "wow! I can see again. Just this instant, it was like a curtain was drawn back and the cloudy area became sharp" (this was actually in Chinese). While this recovery is within the timeframe predicted by neurologists, the suddenness was a shock. Moreover, the extremely sharp recovery does not fit the simple model of injured neurons slowly recovering. What I imagine is that her visual cortex was damaged. The injured neurons did recover, but recovery did not lead to incremental improvement. Rather, the visual cortex is a system of neurons that can be in several states. The system can be jump-started, or induced to change from one state to another. These state changes have rather sharp temporal boundaries.
Since my mother-in-law's recovery I've questioned neurologists whenever possible to see if her recovery pattern was common or exceptional. In general, they wouldn't commit.
The brain-stimulation induced recovery seems to fit with the brain-state model of cortical function. This makes it similar to the heart, where electrical (or other) stimulation can change functional patterns.
As an side note, I'm suspicious about the rigorous scientific validity of categories such "vegetative state". I don't mean to suggest that they are not empirically valid categories. But since we have only a vague sense of what the physiology of a "vegetative state" is, it is imprudent to make firm rules or predictions about what is possible or impossible.
Posted by: john | August 2, 2007 3:39 PM
picky comment:
In the brain image the brain is too small for the skull. frontal lobes should press the front of the forehead; orbital surface of the frontal lobes should be directly above the eyes; caudal medulla (which I think is visible) should fit into the foramen magnum which is at the base of the skull.
Reason for comment: 3-d brain images are plentiful. They are beautiful and popular in things like tv documentaries. It is extremely annoying how many of them have either terrible errors or misleading appearance. The difficulties probably come from difficult communication between neuroanatomists and illustrators.
Posted by: johnk | August 2, 2007 6:14 PM
John said: "As an side note, I'm suspicious about the rigorous scientific validity of categories such "vegetative state". I don't mean to suggest that they are not empirically valid categories. But since we have only a vague sense of what the physiology of a "vegetative state" is, it is imprudent to make firm rules or predictions about what is possible or impossible."
What do FMRI results show on people who have been given the general tag "vegetative"?
Posted by: John B. | August 3, 2007 2:08 PM
John B,,
I don't know the answer about fMRI and vegetative state. Good question. But I don't think standard fMRI answers questions about vegetative state. fMRI measures regional blood flow, so it is an indirect measure. Moreover, the standard practice is to measure blood flow relative to blood flow in the same region in a control situation. "No change" is what you expect in vegetative state, but it doesn't say much about the health of the tissue.
Posted by: john | August 7, 2007 1:54 AM