MIT model explains how the brain can learn novel tasks while still remembering what it has already learned.
The experiment that made it all happen for the 1998 Nobel Prize in physics took place in 1982 at the Francis Bitter Magnet Laboratory at MIT.
Although the prize did not go to an MIT faculty member, "this is very much an MIT Nobel Prize," said Robert Birgeneau, dean of science and professor of physics, who recalled one of the three recipients, Stanford University's Robert B. Laughlin, who received the PhD in physics from MIT in 1979, as "a brilliant and charismatic graduate student, very self-confident, strong-willed and outspoken, even at the student stage" of his career. "He was very clear in his own mind that he wanted to pursue fundamental issues in physics.
"His approach was remarkably creative and intuitive. It wasn't that he sat down and slogged away at calculations for five years. He thought deeply about the problem and had a very different way of approaching it," Professor Birgeneau said. MIT Professor of Physics Peter A. Wolff said of Dr. Laughlin's theory, "The minute you saw it, you knew it had to be right, but it was a surprise. It was a very profound theoretical thing, but it was based on a very physical phenomenon."
The experiment that led to Dr. Laughlin's theory on the fractional quantum Hall effect took place in the magnet lab 26 years ago, but it was not an isolated experiment for researchers Horst L. Stï¿½ï¿½ï¿½ï¿½ï¿½ï¿½rmer and Daniel C. Tsui, who shared the 1998 Nobel Prize with Dr. Laughlin. The pair, well-known to a handful of MIT physicists and graduate students as colleagues and friends, have enjoyed an unusual, enduring partnership that made them frequent visitors to the Institute.
The physics prize, which is accompanied by $978,000 to be split among the recipients, is shared by Dr. Laughlin; Dr. Stï¿½ï¿½ï¿½ï¿½ï¿½ï¿½rmer, who works at Columbia University and Bell Laboratories in Murray Hill, NJ; and Dr. Tsui of Princeton University. All three were at Bell Labs when they began the research that led to last week's prize for discovering how electrons can change behavior. Professor Laughlin subsequently did the theoretical basis for the work at Lawrence Livermore National Laboratory in Livermore, CA.
Professor Laughlin attributed his comprehensive understanding of physical phenomena to MIT's strong emphasis on engineering. It was a "confluence of things from engineering that prepared me for understanding the fractional quantum Hall effect and coming up with an explanation," he said during a TV interview at Stanford.
The trio's work is based on a 19th-century discovery by American Edwin H. Hall. Hall applied a magnetic field perpendicularly to a thin metal plate, causing an electric potential to appear that is perpendicular to both the plate and the magnetic field.
A German physicist, Klaus von Klitzing, delving deeper into the Hall effect in 1980, discovered that when the strength of the magnetic field applied to the plate increased smoothly, the change in electrical potential occurred in steps proportional to integer numbers -- as might be predicted by quantum mechanics. Dr. Klitzing won the 1985 Nobel in physics for this phenomenon, called the integer quantum Hall effect.
Taking the Klitzing discovery yet another step, Professors Stï¿½ï¿½ï¿½ï¿½ï¿½ï¿½rmer and Tsui prepared at Bell Labs the experiment that was ultimately carried out at MIT. They sandwiched extremely thin layers of gallium and arsenic together to create a transistor in which electron movement was restricted to two dimensions instead of three.
"They did an experiment in a novel physical regime that had not been studied, and they observed some very unexpected, unusual results," said Professor Laughlin's thesis adviser, John D. Joannopoulos, the Francis Wright Davis Professor of Physics at MIT's Research Laboratory of Electronics. "Then Bob constructed a novel and creative theory to explain this behavior."
To make an experiment like this work, "you have to be a good physicist and you need artists with materials behind you," Professor Wolff said.
The three sponsoring institutions -- the magnet lab, Princeton University and Bell Labs -- each contributed to the cost, but Professor Wolff, as head of the magnet lab for several years in the 1980s, had to fight for Professors Stï¿½ï¿½ï¿½ï¿½ï¿½ï¿½rmer and Tsui to get the extra time they needed. In this unusual partnership, the pair continued to do experiments at the lab until around 1995, according to Professor Robert Griffin, current director of the magnet lab.
The lab, which operated on the MIT campus with federal funding for 27 years, drew researchers from far beyond MIT to conduct a variety of research needing the world's most powerful magnets. In 1990, the National Science Board designated Florida State University as the site for a new national high-magnetic-field laboratory. The magnet lab on the MIT campus continues to serve MIT and other scientists, but on a more limited scale.
Making the experiment work was an expensive, iffy proposition, Professor Wolff recalled. It necessitated squeezing a large high-field magnet into a tiny space with very low-temperature equipment. "It was really a gamble. There was a real chance it wouldn't work," he said. In fact, at one point, "the whole inside of the magnet blew up and pieces literally hit the ceiling. They thought they had wrecked the equipment, but it survived and they went on."
The variation of the Hall effect occurs at a temperature of almost absolute zero and in a magnetic field almost one million times stronger than the Earth's. Under those conditions, electrons in the transistor appeared to exhibit electrical charges that occur in thirds of a step, seemingly only one-third of what they should have been.
Professor Laughlin, the theorist of the group, immediately framed an explanation that he later said was wrong. It came to him eventually that the electrons were undergoing what has come to be known as the fractional quantum Hall effect. They were interacting as if they were a fluid, and only looked as though they had been broken into thirds. "When you take the magnetic field away," Professor Stï¿½ï¿½ï¿½ï¿½ï¿½ï¿½rmer explained at a news conference at Bell Labs last Tuesday, "the normal electrons reappear and the one-third fractional electrons disappear."
Experiments on this subject are so fundamental to so much of physics that they have, over the years, generated thousands of papers, Professor Wolff said. Related to high-energy physics, the fractional quantum Hall effect may help scientists understand the quantum structure of the vacuum throughout the space-time of the universe.
Of the three, Professor Wolff knows the German-born and -trained Dr. Stï¿½ï¿½ï¿½ï¿½ï¿½ï¿½rmer best, first meeting him when Dr. Stï¿½ï¿½ï¿½ï¿½ï¿½ï¿½rmer was a postdoctoral researcher at Bell Labs in the 1970s.
"His boss called me up and said, 'Could you come meet this guy and give me an appraisal?' I spent an hour with him and was absolutely blown away. He was super-energetic, also tremendously creative." Their relationship continued as at least two of Professor Wolff's students went on to work with Dr. Stï¿½ï¿½ï¿½ï¿½ï¿½ï¿½rmer at Bell Labs. "He's generous, and he was kind of a father to them. When they were taking classes and doing research at Bell Labs one week out of four, they went to his house and were together like a family."
Wolff says Dr. Stï¿½ï¿½ï¿½ï¿½ï¿½ï¿½rmer is outgoing and personable, and recalls Dr. Tsui as profound and the more inward-looking of the pair. Dr. Tsui spent a few months at MIT as a visiting scientist in 1988.
After taking a class with Professor Joannopoulos in the early 1970s, Dr. Laughlin joined his research group from 1975-79, when he completed his thesis on "The Structure and Excitations of Amorphous Solids and Surfaces" -- a topic unrelated to the one for which he wonthe Nobel Prize.
Dr. Laughlin "had spent a few years in the army after graduating from Berkeley, so he came in to graduate school at a more mature level," Professor Joannopoulos said. "He was extremely motivated. My impression is that he had a very strong basic understanding of physics, and that he was very imaginative and creative. So that, combined with his outgoing and outspoken personality, gave me the impression that he was really a unique kind of individual."
From MIT, Dr. Laughlin went as a postdoc to Bell Labs, where Professor Joannopoulos said he became intrigued with the quantum Hall effect. "The first thing he did was try to come up with a theory for the integer quantum Hall effect. Bob then moved to Lawrence Livermore National Laboratory. He did his work on his own -- with a loose connection to MIT in that he would talk to me about it off and on -- in a very systematic way. It was a very creative idea and involved brilliant intuition on his part."
Professor Joannopoulos also remembers Professor Laughlin as "always great fun as a graduate student. The group was always laughing and fooling around." He added that a little-known fact about Professor Laughlin is that he is an accomplished pianist and composer. Dr. Laughlin has played some of his original concertos when visiting Professor Joannopoulos at home.
Professor Laughlin has had an interesting career, Professor Wolff said. "He came at this problem from a different angle than others. It turned out to be clearly and intuitively right."
A full list of Nobel Prize winners with connections to MIT can be found within the News Office web site.
A version of this article appeared in MIT Tech Talk on October 21, 1998.