American Institute of Physics
SEARCH AIP
home contact us sitemap
Physics News Update
Number 551 #2, August 8, 2001 by Phil Schewe, James Riordon, and Ben Stein

Evidence for chaos in the neocortex

Evidence for chaos in the neocortex, the most complex brain structure specific to humans and other mammals, has been obtained in a model by researchers in Australia (David Liley, Swinburne University of Technology, 011-61-3-9214-8812, dliley@swin.edu.au).

Chaos in the brain would manifest itself as unpredictable and seemingly random electrical activity in a population of nerve cells, or neurons. Chaos may have an important neurological function: it could provide, as researchers have speculated, a flexible and rapid means for the brain to discriminate between different sounds, odors, and other perceptual stimuli.

Electroencephalograms (EEGs) record electrical activity in the cerebral cortex, but they, and all other current experimental techniques, may never be able to detect clear and unequivocal signs of chaos, since the cortex also emits a very large amount of obscuring "noise" or random electrical activity.

Using realistic models of brain physiology, many researchers are trying to devise models which reproduce the output of EEGs yet also offer new insights into the brain's inner workings. However, previous models either do not allow for chaos to appear, or have been unable to demonstrate that chaos can occur under the conditions imposed by the structure of the brain.

In the present work, the researchers model the behavior of two large populations of neurons: excitatory (which bring other neurons closer to firing) and inhibitory (which make it more difficult for other neurons to fire). Specifically, they look at the "mean soma membrane potential," the electric potential between the outside and inside of the neuron's cell body (higher potential means more frequent firing).

Varying the rate of external electrical impulses to each neuron population, they found the mean electrical activity was irregular and noise-like (it looked like noise but really wasn't) for a wide range of external inputs. Quantitatively such behavior is associated with a positive Lyapunov exponent, a hallmark of chaos. The existence of chaos, the researchers say, would provide a means for the brain to change its response rapidly to even slightly different stimuli. (Dafilis et al., Chaos, September 2001.)