Professor
12-31-2006, 05:45 PM
http://www.chicagotribune.com/news/local/chi-0612310318dec31,1,6578594.story?coll=chi-news-hed
They want to read our minds
At UIC, a powerful new MRI machine may offer the clearest view yet of the human mind working at the speed of thought
By Ronald Kotulak
Tribune science reporter
December 31, 2006
At 40 tons, the new functional magnetic resonance imaging machine at the University of Illinois at Chicago is one of the world's largest and most powerful.
But what really sets it apart, scientists say, is its ability to track the firing of individual neurons in the brain--that is, to watch thoughts form in real time.
That capability could be a boon to diagnosing and treating strokes, Alzheimer's disease, autism, schizophrenia and other mental disorders. It also could help people become smarter by discovering how the brain learns best. And it could aid scientists in building intelligent computers that function more along the lines of the brain.
The ultimate goal, however, said Dr. Keith Thulborn, director of the Center for MR Research at UIC, which houses the magnet, is to be able to find out what people are thinking.
"We'd like to get to the stage of reading thoughts," he said. "At the moment all we can do is look at the response of the brain to a particular stimulus. The next step, which we'd all like to get to, is what actually has been processed. Is it the time of day? Is he actually reading a sentence? Which word in the sentence is he understanding?"
Magnetic resonance machines have revolutionized brain imaging in the last 20 years because of their ability to provide exquisite pictures of the brain's structure and broad function. The biggest MRI machines typically used in medical research generate a field of up to 3 tesla, a measure of magnetic strength.
To look at how the brain works in fine detail, from neuron to neuron, a huge magnet is needed. The one at UIC is 9.4 tesla.
At 100,000 times the strength of the Earth's magnetic field, the UIC magnet is so strong it pulls on the steel shanks in the shoes of people who come near it. If it weren't for a 520-ton steel shield encasing the magnet, people would be able to feel its fringe magnetic field a couple of blocks away.
The targets sit in the magnet's hollow core, and they are much smaller: the sodium ions in the brain that generate the electrical charges powering our thoughts.
Following the sodium
The UIC machine is the first to be able to follow the flow of sodium, allowing it to see the firing of individual brain cells instantaneously. Other brain imaging machines follow water--the main ingredient in blood--and that means a time lag.
Neurons will register a flash of light in 100 milliseconds, for instance, but the flow of blood that accompanies this activity is comparatively slow, taking two to five seconds.
Thulborn was the first person to have his brain imaged in the new machine four months ago. "It was very satisfying," he said. "What people said couldn't be done was actually done." He could see that billions and billions of neurons were working busily in his brain. A very reassuring image: "My brain is tightly packed."
Since then, he and his team have learned to slide test subjects into the machine very slowly. The magnetic field is so great that a fast or jerky movement can cause an unwanted stimulation of the brain cells, making people see flashing lights, sense a metallic taste or feel dizzy.
Thulborn spent 10 years designing the ambitious machine, which costs more than $10 million and was funded by the state of Illinois, the National Institutes of Health and GE Healthcare. The big superconducting magnet and the advanced imaging system were custom-built to his specifications by companies that have since been acquired by GE.
To prevent vibrations from outside traffic or construction from interfering with delicate measurements, the magnet sits on a thick free-standing concrete slab supported by eight caissons sunk 60 feet into clay.
Unlocking the thought process
Now that the machine is running and going through safety tests, Thulborn anticipates that in time, it will reveal some of the mind's deepest secrets: how we make decisions, how we perceive things, how we process language.
Tracking precisely what happens in the brain as people experience sights, sounds, smells, tastes and touch could reveal how memories get started, where they go to be analyzed for their emotional significance, where they are stored and how they are recalled.
"That obviously has implications for an understanding of what it is to be a human being, what it is about our brains that makes us the way we are," said Michael Rugg, director of the Center for the Neurobiology of Learning and Memory at the University of California, Irvine, who uses a 3-tesla machine to study memory. "And it obviously has important implications for an understanding of why different kinds of disease processes give rise to impairments in these everyday functions."
The brain runs on electricity generated by positively charged sodium ions, and the goal of UIC's new machine is to observe the electrical charges racing through individual brain cells as they communicate.
When a neuron receives information from neighboring neurons and decides to act, it opens pumps on its surface that allow sodium ions to flood in. A tiny jolt of electricity quickly builds up, then zooms down the neuron's long axon, passing the information to other neurons. The neuron's sodium pumps then close, and other pumps clear the sodium out of the cell so it can be ready to fire again.
The massive magnetic field of a powerful MRI machine aligns the nuclei of the atoms in the same direction, much as a magnet pulls a compass needle. When the nuclei are allowed to relax, they emit characteristic radio frequency signals that tell observers what kind of atoms they are and what they are doing.
Advancing technology
Other machines have been attuned to the movement of water molecules in the brain rather than sodium, which means it has been impossible to measure exactly which neurons fire first when a thought arises. Nor can those machines detect their exact timing and communication pathways. Where thoughts start and stop cannot be determined accurately. Responses to stimuli are averaged out, blurring the image of what is really going on.
That's where Thulborn's 9.4-tesla machine comes in. If it lives up to its promise, the machine will spot the first neurons that fire almost instantaneously and track where the communication links go from there.
"What we'd like to understand--and what we're working on--is what happens on a single event, a single thought," Thulborn said. "And we've been able to do this. It turns out that if you look at a single thought versus average thoughts, then in fact you get a different view of how the brain functions."
The only other 9.4-tesla machine is at the University of Minnesota. It too is custom-made, but its main focus is following the flow of blood in the brain.
One target of UIC's machine is the brain's capacity for parallel processing, Thulborn said. The brain does not respond to a series of identical stimuli in the same way; instead, one area mulls over the first stimulus while other areas process subsequent events.
"This is very important," Thulborn said. "In the days when survival was our main goal, if you saw some food sitting on the ground and you're going to run toward it, but you saw a saber tooth tiger standing off on the side, if you could only process the food signal, then you wouldn't have survived. You'd have been eaten by the tiger. You have to be able to respond to multiple stimuli."
Learning just how the brain does that is the task of the 9.4-tesla machine. Instead of looking at the flow of blood as a marker of brain activity, it will look directly at the biochemistry that underlies thought.
It's like the difference between ground-based telescopes that have to peer through the distorting atmosphere and the orbiting Hubble Space Telescope, Thulborn said. The Hubble provides a clear, spectacular view of a previously unseen universe.
"We're looking at a new dimension that wasn't available previously," he said. "We're not looking at a blurred view of brain function through a water signal. We're actually now looking at what we really want to look at, which is the cellular machinery."
The University of California's Rugg agreed. "If we could remove our dependency on blood flow measures ... that would be a big deal," Rugg said. "If we could get to something that responded much more in real time, then that would be a very big deal indeed."
The importance of such a feat would go beyond medical applications, he said. "By understanding better how the brain does things very easily that computers at the moment do very badly or can't do at all, the idea is that people will be able to build more efficient computers."
- - -
`Mind reading' stirs concerns
The ability to decipher the brain's operational manual raises a host of ethical concerns. Do we want our thoughts to be revealed to others? Will it lead to a foolproof lie detector test? Will employers or school officials seek out such neural technology to select certain types of candidates and eliminate others? Will it predict who is prone to violence?
"People are dying for accurate, objective measures of all aspects of human psychology," said Martha Farah, director of the University of Pennsylvania's Center for Cognitive Neuroscience. "When a new technology comes along they're going to grab for it."
Farah worries that the brain imaging technology will be adopted too early.
"The bad outcomes related to this kind of mind-reading imaging are going to come from people thinking the techniques are more accurate and more valid than they really are," she said.
"People may be denied jobs on the basis of the brain scan. People may be thrown in jail for being suspected terrorists on the basis of a brain scan. Some kid may be put in the wrong educational track on the basis of a brain scan."
Any new technology is not inherently good or bad, Farah added--it's how people use it.
--Ronald Kotulak
----------
rkotulak@tribune.com
They want to read our minds
At UIC, a powerful new MRI machine may offer the clearest view yet of the human mind working at the speed of thought
By Ronald Kotulak
Tribune science reporter
December 31, 2006
At 40 tons, the new functional magnetic resonance imaging machine at the University of Illinois at Chicago is one of the world's largest and most powerful.
But what really sets it apart, scientists say, is its ability to track the firing of individual neurons in the brain--that is, to watch thoughts form in real time.
That capability could be a boon to diagnosing and treating strokes, Alzheimer's disease, autism, schizophrenia and other mental disorders. It also could help people become smarter by discovering how the brain learns best. And it could aid scientists in building intelligent computers that function more along the lines of the brain.
The ultimate goal, however, said Dr. Keith Thulborn, director of the Center for MR Research at UIC, which houses the magnet, is to be able to find out what people are thinking.
"We'd like to get to the stage of reading thoughts," he said. "At the moment all we can do is look at the response of the brain to a particular stimulus. The next step, which we'd all like to get to, is what actually has been processed. Is it the time of day? Is he actually reading a sentence? Which word in the sentence is he understanding?"
Magnetic resonance machines have revolutionized brain imaging in the last 20 years because of their ability to provide exquisite pictures of the brain's structure and broad function. The biggest MRI machines typically used in medical research generate a field of up to 3 tesla, a measure of magnetic strength.
To look at how the brain works in fine detail, from neuron to neuron, a huge magnet is needed. The one at UIC is 9.4 tesla.
At 100,000 times the strength of the Earth's magnetic field, the UIC magnet is so strong it pulls on the steel shanks in the shoes of people who come near it. If it weren't for a 520-ton steel shield encasing the magnet, people would be able to feel its fringe magnetic field a couple of blocks away.
The targets sit in the magnet's hollow core, and they are much smaller: the sodium ions in the brain that generate the electrical charges powering our thoughts.
Following the sodium
The UIC machine is the first to be able to follow the flow of sodium, allowing it to see the firing of individual brain cells instantaneously. Other brain imaging machines follow water--the main ingredient in blood--and that means a time lag.
Neurons will register a flash of light in 100 milliseconds, for instance, but the flow of blood that accompanies this activity is comparatively slow, taking two to five seconds.
Thulborn was the first person to have his brain imaged in the new machine four months ago. "It was very satisfying," he said. "What people said couldn't be done was actually done." He could see that billions and billions of neurons were working busily in his brain. A very reassuring image: "My brain is tightly packed."
Since then, he and his team have learned to slide test subjects into the machine very slowly. The magnetic field is so great that a fast or jerky movement can cause an unwanted stimulation of the brain cells, making people see flashing lights, sense a metallic taste or feel dizzy.
Thulborn spent 10 years designing the ambitious machine, which costs more than $10 million and was funded by the state of Illinois, the National Institutes of Health and GE Healthcare. The big superconducting magnet and the advanced imaging system were custom-built to his specifications by companies that have since been acquired by GE.
To prevent vibrations from outside traffic or construction from interfering with delicate measurements, the magnet sits on a thick free-standing concrete slab supported by eight caissons sunk 60 feet into clay.
Unlocking the thought process
Now that the machine is running and going through safety tests, Thulborn anticipates that in time, it will reveal some of the mind's deepest secrets: how we make decisions, how we perceive things, how we process language.
Tracking precisely what happens in the brain as people experience sights, sounds, smells, tastes and touch could reveal how memories get started, where they go to be analyzed for their emotional significance, where they are stored and how they are recalled.
"That obviously has implications for an understanding of what it is to be a human being, what it is about our brains that makes us the way we are," said Michael Rugg, director of the Center for the Neurobiology of Learning and Memory at the University of California, Irvine, who uses a 3-tesla machine to study memory. "And it obviously has important implications for an understanding of why different kinds of disease processes give rise to impairments in these everyday functions."
The brain runs on electricity generated by positively charged sodium ions, and the goal of UIC's new machine is to observe the electrical charges racing through individual brain cells as they communicate.
When a neuron receives information from neighboring neurons and decides to act, it opens pumps on its surface that allow sodium ions to flood in. A tiny jolt of electricity quickly builds up, then zooms down the neuron's long axon, passing the information to other neurons. The neuron's sodium pumps then close, and other pumps clear the sodium out of the cell so it can be ready to fire again.
The massive magnetic field of a powerful MRI machine aligns the nuclei of the atoms in the same direction, much as a magnet pulls a compass needle. When the nuclei are allowed to relax, they emit characteristic radio frequency signals that tell observers what kind of atoms they are and what they are doing.
Advancing technology
Other machines have been attuned to the movement of water molecules in the brain rather than sodium, which means it has been impossible to measure exactly which neurons fire first when a thought arises. Nor can those machines detect their exact timing and communication pathways. Where thoughts start and stop cannot be determined accurately. Responses to stimuli are averaged out, blurring the image of what is really going on.
That's where Thulborn's 9.4-tesla machine comes in. If it lives up to its promise, the machine will spot the first neurons that fire almost instantaneously and track where the communication links go from there.
"What we'd like to understand--and what we're working on--is what happens on a single event, a single thought," Thulborn said. "And we've been able to do this. It turns out that if you look at a single thought versus average thoughts, then in fact you get a different view of how the brain functions."
The only other 9.4-tesla machine is at the University of Minnesota. It too is custom-made, but its main focus is following the flow of blood in the brain.
One target of UIC's machine is the brain's capacity for parallel processing, Thulborn said. The brain does not respond to a series of identical stimuli in the same way; instead, one area mulls over the first stimulus while other areas process subsequent events.
"This is very important," Thulborn said. "In the days when survival was our main goal, if you saw some food sitting on the ground and you're going to run toward it, but you saw a saber tooth tiger standing off on the side, if you could only process the food signal, then you wouldn't have survived. You'd have been eaten by the tiger. You have to be able to respond to multiple stimuli."
Learning just how the brain does that is the task of the 9.4-tesla machine. Instead of looking at the flow of blood as a marker of brain activity, it will look directly at the biochemistry that underlies thought.
It's like the difference between ground-based telescopes that have to peer through the distorting atmosphere and the orbiting Hubble Space Telescope, Thulborn said. The Hubble provides a clear, spectacular view of a previously unseen universe.
"We're looking at a new dimension that wasn't available previously," he said. "We're not looking at a blurred view of brain function through a water signal. We're actually now looking at what we really want to look at, which is the cellular machinery."
The University of California's Rugg agreed. "If we could remove our dependency on blood flow measures ... that would be a big deal," Rugg said. "If we could get to something that responded much more in real time, then that would be a very big deal indeed."
The importance of such a feat would go beyond medical applications, he said. "By understanding better how the brain does things very easily that computers at the moment do very badly or can't do at all, the idea is that people will be able to build more efficient computers."
- - -
`Mind reading' stirs concerns
The ability to decipher the brain's operational manual raises a host of ethical concerns. Do we want our thoughts to be revealed to others? Will it lead to a foolproof lie detector test? Will employers or school officials seek out such neural technology to select certain types of candidates and eliminate others? Will it predict who is prone to violence?
"People are dying for accurate, objective measures of all aspects of human psychology," said Martha Farah, director of the University of Pennsylvania's Center for Cognitive Neuroscience. "When a new technology comes along they're going to grab for it."
Farah worries that the brain imaging technology will be adopted too early.
"The bad outcomes related to this kind of mind-reading imaging are going to come from people thinking the techniques are more accurate and more valid than they really are," she said.
"People may be denied jobs on the basis of the brain scan. People may be thrown in jail for being suspected terrorists on the basis of a brain scan. Some kid may be put in the wrong educational track on the basis of a brain scan."
Any new technology is not inherently good or bad, Farah added--it's how people use it.
--Ronald Kotulak
----------
rkotulak@tribune.com