Neuroscience Unit


New Research Clarifies Human Understanding
Provides A Solution to the Elusive "Binding" Problem

Birmingham, October 1996 -- It is one of the great and enduring scientific mysteries: how the three pounds of protoplasm -- the nerve cells and molecules that make up the brain -- generate the state we call consciousness.

Now, an Anglo-American team of scientists from the University of Birmingham, IBM Research Center at Yorktown Heights New York and Imperial College School of Medicine at St. Mary's may have provided a major step in establishing a cellular basis for consciousness. In the 17th October, 1996 issue of the international journal Nature, they report a potential solution to a famous quandary known as the "binding problem" - how the brain "binds" together activity from spatially separated areas into a single coherent picture.

The research suggests new experimental approaches to studying the mind-body problem, while at a practical level it has implications for dementing and psychiatric illnesses such as schizophrenia, that somehow disrupt the process of useful thought.

For years, researchers have suspected that the binding task is accomplished by nerve cells in distinct areas of the brain communicating between themselves by oscillating in phase -- like two different chorus lines kicking to the same beat even though they're dancing in different theatres. These oscillations have been detected in everything from the olfactory bulb of rabbits to the visual cortex of cats and even conscious humans. IBM, Birmingham and Saint Mary's researchers now believe they have explained not only how the oscillations come about but how the oscillatory rhythm is communicated from one area of the brain to another. These two findings are critical to understanding how the complex electric signals of large numbers of nerve cells generate awareness and perhaps even consciousness.

At the heart of the binding problem is how the brain processes its perception of objects. "The brain uses widely distributed parallel processing mechanisms, that are best understood in the case of vision," explains John Jefferys, Professor of Neuroscience in Birmingham. "Most people have no difficulty identifying objects such as a tennis ball flying through the air. The way our brains link the different areas that process colour, shape and motion so that we see the ball as a single object is so effective that we simply are not aware of the complex computations involved".

As long ago as the 1950s, researchers observed a distinct 40 cycles per second (Hertz) rhythm in the brains of animals while they were discriminating between smells or visual stimuli. In 1989, Gray and Singer in Germany detected the rhythm in the visual cortex of cats, and noticed that not only were these oscillations tightly synchronized over a large part of the brain, but they appeared strongest when the cat was looking at a single object, and weakest or non-existent when the cat was looking at different and unrelated objects. No less a biological luminary than the Nobel Laureate Francis Crick has suggested that this synchronized firing of nerve cells was used by the brain not only as a means of focusing attention, but as the general mechanism of consciousness.

But how were these 40 Hertz oscillations generated? More than thirty years later, in the 16 February, 1995 issue of Nature, Miles Whittington, Roger Traub and John Jefferys reported that the oscillations seem to emerge naturally from a network of neurons known as interneurons, or inhibitory neurons, that work to suppress and control the firing rate of the bulk of the brain's neurons, which are known as pyramidal cells. What's more, the researchers showed that repetitive stimulation to the interneuron network would prompt the elicit 40 Hertz oscillations. This finding arose from laboratory work on fresh hippocampal brain tissue done by Jefferys and Whittington and computer simulations of networks of neurons constructed on an IBM SP supercomputer by Traub, a research staff member at IBM in New York.

Their proposal is that the inhibitory network receives a steady or slow excitatory drive which makes it oscillate, and that provides a clock which determines when pyramidal cells can fire. The fact that the inhibitory network can, by itself, sustain this oscillatory rhythm, separates the synchronizing control, or clock', from the specific neuronal processing of information the central processing unit'. If the role of these so-called gamma rhythms is indeed to mediate binding, then mechanisms must exist for the selective coupling of areas involved in processing common entities.

In their latest Nature paper, Traub, Whittington, Stanford and Jefferys report that they may have found this coupling mechanism; they show how these 40 Hertz oscillations can achieve a tight synchronization across different areas of the brain, thus providing a possible solution to the famous binding problem. In the latest work, Traub added pyramidal cells to his computer simulation of the network of inhibitory cells. With the pyramidal cells included, Traub found that the interneurons began firing twice in quick succession -- in doublets, as Traub calls it.

This double beat then serves as the time keeper that keeps neurons synchronized over long distances. The time lag of the second beat matches up with how long it takes a signal to travel to the next ensemble of neurons that are paying attention to the same object. And that beat keeps the oscillation synchronized. Not only do the doublets seem to explain the tight synchronization needed by the binding problem using "very simple circuitry,'" says Traub, but the model also led to three firm predictions about the oscillations in live cells that could be tested in the laboratory. When Whittington, Stanford and Jefferys tested the predictions in the lab, the results on all three predictions matched the model exactly.

"This is how theoretical and experimental science should work together. Roger Traub's computer models are built firmly on real experimental data, and make predictions that can be tested by experiment" says Jefferys. "In this case it has worked out better than we dared hope".

For more information on this visit our web sites at
http://medweb.bham.ac.uk/neuroscience/home
http://www.research.ibm.com

Research Team members:-

  • Prof John Jefferys, Professor of Neuroscience, University of Birmingham
  • Dr Ian Stanford, Research Associate, University of Birmingham
  • Dr Miles Whittington, Lecturer, Imperial College School of Medicine at St Mary's (London)
  • Prof Roger Traub, Staff member, IBM Research Center, Yorktown Heights, New York and Columbia University Medical Center, New York.


 

Home | Contact Us  | Products Pricing | How To Order | Product Specifications | Links & Additional Technical Information |

 Copyright 1995 - 2013 Intelegen Inc. All rights reserved