Slashdot Mirror
Intelligence Map Made From Brain Injury Data
An anonymous reader writes with this news out of the University of Illinois:
"Scientists report that they have mapped the physical architecture of intelligence in the brain. Theirs is one of the largest and most comprehensive analyses so far of the brain structures vital to general intelligence and to specific aspects of intellectual functioning, such as verbal comprehension and working memory. Their study, published in Brain: A Journal of Neurology (abstract), is unique in that it enlisted an extraordinary pool of volunteer participants: 182 Vietnam veterans with highly localized brain damage from penetrating head injuries. ... The researchers took CT scans of the participants’ brains and administered an extensive battery of cognitive tests. They pooled the CT data to produce a collective map of the cortex, which they divided into more than 3,000 three-dimensional units called voxels. By analyzing multiple patients with damage to a particular voxel or cluster of voxels and comparing their cognitive abilities with those of patients in whom the same structures were intact, the researchers were able to identify brain regions essential to specific cognitive functions, and those structures that contribute significantly to intelligence."
I believe mine is currently functioning as intended.
This is one of those fine moments when I wish scientific journals posted online weren't pay-walled. Kinda kills the dissemination of knowledge to the masses when one has to pay $32 to view a single article once, and makes it economically infeasable for an individual to read and verify the information they hear from primary sources.
A voxel is a 3D (volumetric) pixel.
Floating in the black seas of infinity without a paddle.
There's the map of The Brain; is anyone working on a map of Pinky?
Sheesh, evil *and* a jerk. -- Jade
In Soviet Russia, Brain Voxels divide YOU into 3,000 three dimensional units.
Why did the chicken cross the road? Because Elon Musk put an AI chip in its head.
While this is undoubtedly an important study, their findings are going to have to be replicated somehow in a larger, more diverse set of subjects. They're looking at just 182 people and, while it's not mentioned explicitly in the article, it appears they're all men. We know from other studies that there are anatomical differences in men's brains compared to women's brains, and even between left handed and right handed men. It would be very interesting to see, for example, a FMRI study to see if the structures play the same role in all patients.
There's no point in questioning authority if you aren't going to listen to the answers.
brain mapping.
Maybe phrenology was onto something after all!
Occasionally living proof of the Ballmer peak.
From the video, I understand that damage to the right frontal and temporal lobe have no effect on general intelligence. I wonder what they are used for then? What kind of mental processes are run there? Something that the tests did not measure, or are they simply useless? Seems to be a waste of resources.
Let's see if they can explain this guy.
TFA did not specify any pre-injury base-line for intelligence.
Did they all take intelligence tests before enlisting?
Seems unlikely.
Did they have any other way to check cognitive function prior to the injury so they had some sort of a useful base-line?
Is it possible that a majority of the differences, especially in general intelligence, were less related to the injuries and more related to nature/nurture?
How about compensation?
Humans are great at adapting.
Did they check their results with people who had more recent injuries?
Might be a good starting point, but it sounds like there is a lot that could affect the things they were testing for that were not isolated or otherwise accounted for.
This study seems to be making the assumption that we all put the same brain functions in precisely the same places. But each individual has different intellectual strengths, weaknesses, and talents. Although I wouldn't say they shouldn't do this study, I fail to see how it would give us more than the coarsest understanding, biased based on the individual personalities of those tested.
182 people were investigated and the results compared. This means only the similarities show up, whereas the differences you mention largely average away.
If the assumption (that we all put the same brain functions in more or less the same places) was false, an experiment with a big enough sample size would show there'd be no correlation. Heck, a too small sample size would on average even be more likely to show an anecdotal lack of correlation!
This is how science works: There is an initial assumption (a. k. a. hypothesis), then you try to disprove it. If you fail, you've won.
This coarse understanding of the brain is a step-up from the object-level understanding we have know. From there, we can refine the model in gradual steps.
It's definitely a study with flaws, but it's also a study that advances the current state of the art.
Patience, grass hopper. The neural interfaces to allow for Total Recall to become reality are still a few coarse and flawed studies away.
Those who can, do. Those who can't, sue.
AFAIK, most people will lose the same function if they lose the same part of their brain. I'd wager that anyone who didn't had already lost function in that part of their brain in their childhood, and their brain compensated.
Still, it's a crap shoot, but more data is better than none. If you have a brain tumor, and the surgeon has to choose to sacrifice one part of your brain to remove the tumor, this data helps to guide that choice.
All ideas^H^H^H^H^Hprocesses in this post are Patent Pending. (as well as the process of patenting all postings)
People also have different physical strengths and weaknesses, but we still have the same muscles in the same areas. It would be reasonable to assume the brain is the same way until we have evidence that suggests otherwise.
We hope your rules and wisdom choke you / Now we are one in everlasting peace
I dont care what your confidence is!
They took just 182 people, all of similar age, all of similar education, all gone through military training, all with penetrating head wounds and trauma to their brain, and used that to "map the brain". Sounds solid.
I didn't say the study wasn't worthy or interesting.. I am just saying with 182 subjects, the strong claim presented to me - We've mapped intelligence!) in the title of this article isn't warranted.
Who submits these things. Helper undergrads who want to see the study and their name in the "thanks to" section made famous?
Someone call quality control....
Anything to promote research into three-boobed women...
This somehow reminds me of how much of genetic research is done. If you want to figure out what a gene does, make a cell line where that gene is broken and see what happens. Or for that matter, particle physics: figure out atoms by smashing them together. It is in many ways a brute-force method, but it's effective enough nonetheless.
"Mapping" intelligence is like herding cats. The problem is that a brain is a swarm of neurons, meaning that its function is the sum of all of its parts. Sure some brain areas are "mappable", because they connect to specific peripheral organs, but otherwise intelligence as a function is unmappable. That is why a location for memory has not been found. In effect, the more cerebral matter is surgically removed, the deeper the memory loss is.
Even the humble slime mould (Physarum polycephalum) can navigate mazes to find a food source, using the most optimal (least expenditure of energy) path. Slime moulds have been used to create maps of major metropolitan transportation systems (such as the Tokyo subway system). Likewise, Darwin's famous experiments with earthworms revealed that earthworms use what the environment affords them in order to strengthen their burrows. They accomplish this despite lacking a central nervous system and any of the "big five" sensory organs. Finally, there are parallels between the pattern dynamics of the BZ (Belousov–Zhabotinsky) reaction and the aggregation phase of the slime mould life cycle (in which a chemical signal for starvation pulls distinct amoebas into an aggregate, and each amoeba that sends the starvation signal becomes the center of a circle towards which the other amoebas move).
Examples like suggest that many complex systems, both biological and otherwise, can demonstrate intelligent behavior. The social, cultural, political, biological, and other environmental contexts afford and constrain the kind of intelligence an organism has. Brains, especially human ones, aren't particularly special in this regard.
I volunteer you for the next round, with a statistically meaningful sample size. By the way, they have confidence bounds on the presented data, had you defeated the paywall to RTFA. Yes, the researchers are smarter than you. In the off chance that you happen to be smarter, go ahead and tell me how you'd perform a better experiement to study this.
You wrote this with you fingers, no? Intelligently designed post!
My informed opinion: --- the researchers were presenting, primarily, a MAPPING SYSTEM ("An integrative architecture for general intelligence mapping.") There are many many areas of intelligence that were not represented within this limited group, composed of verbal deficit injuries. There's no big news here, there's just a promising system for conducting future research on intelligence.
I can't help it:
http://prefrontal.org/files/posters/Bennett-Salmon-2009.pdf
When it comes to fMRI studies, I always remember the story of a dead salmon in an fMRI scanner, that was shown a series of photographs depicting human individuals in social situations. The salmon was asked to determine what emotion the individual in the photo must have been experiencing.
Of course, it was a resounding success! And now SCIENCE knows where in the brain of a dead salmon, the mental process to evaluate human emotions occurs.
I had a massive brain injury over six years ago and my intelligance was severly damgaged for a long time. I had a temeoral lobectomy at the end of last year to stop the seizures that were brought on from the injury and it seems like my intelligence has mostly came back. Even the neurosurgeon said that I seemed smarter than when he first started seeing me. Yet, I'm still not sure wether I'm as smart as when I was a computer science major back in 2001 and started reading ./
3000 voxels too beaucoup!
my public library has an interlibrary loan department, they can get journal articles within a week or two through the ILL system. another library will fax your local library with a copy of the article for a $1 fee or something like that.
i can tell by some of the voxels and from having seen a lot of shops in my time (playing Commanche helicopter simulator in 1993)
"I am not deceiving myself about anything"
This study seems to be making the assumption that we all put the same brain functions in precisely the same places.
Within limits, we do. There is normal variation from person to person, and to some degree the brain will attempt to reroute around damage, but on the whole there is a standard layout where the same functions get put in the same locations. These researchers were looking for certain higher cognitive functions. It is not absolutely sure that they have them all precisely located, but they have identified areas of interest.
I recently had a small stroke on the left side of my brain. The only impact I can tell is some very slight speech problems, which are almost gone just a few weeks later. It would be neat to compare my MRI to the areas mapped out by this to see if I can notice any losses. Too bad my neurologist is not interested in talking more than spending five minutes telling me to eat better and take aspirin and a statin from now on.
There is no way in hell your going to get me to do all that stuff to my brain just to get smarter. On the other hand, I'm still pretty young. If I get started now I can take it slow.
So.... are trolls actually missing functional areas, or are they just cognitively uncoordinated?
DNA is a Turing machine. You, however, being dynamic and emergent, are not.
Surprisingly low resolution tbh. Even a decade ago resolutions like 256x256x256 (16,7M voxel) were common. Then again higher resolution you use, more divergence between patients you get...
Just because I always remember the story doesn't mean I discard all fMRI studies outright. But I still think you should be on your toes.
After an extensive search, researchers failed to locate intelligence and concluded that Homo Sapiens is, in fact, a misnomer.