91av

THE background chatter between brain cells isn鈥檛 just random noise but
structured chaos, says a team of scientists who have modelled brain activity.
They say this could explain how subtle changes in our surroundings rapidly
trigger complex mental images.

Chaotic activity may be unpredictable, but it is orderly and follows rules,
unlike random noise. Its hallmark is the butterfly effect: a tiny change in the
initial conditions of a chaotic system will have a dramatic impact on the
outcome. The existence of chaos in the brain could explain how we make sense of
an ever changing world.

A popular alternative theory is that the brain acts like a computer, with
individual neurons signalling to each other in a stepwise fashion. But this may
not explain how we form an impression of, say, a familiar smell without having
to think about it first.

But researchers have had trouble finding clear evidence for chaos in the
brain. In the latest attempt, neurobiologist David Liley at Swinburne University
of Technology in Melbourne and his colleagues built a mathematical model of the
cortex, the part of the brain that deals with memory, language and perception,
and contains billions of neurons. The model is based on a set of equations that
describe the average electrical potential between the outside and inside of the
neurons under different conditions.

Liley and his team found that the model not only generated chaos, but also
produced a pattern very similar to the 鈥渁lpha rhythm鈥� that typically appears
when someone closes their eyes and relaxes. He argues that the spontaneous
appearance of the pattern suggests the model is mimicking what happens in the
brain. 鈥淭he alpha wave is the sanity check,鈥� he says. 鈥淚t suggests that the
chaos we get is also plausible.鈥�

鈥淎 plus of [the team鈥檚] model is that it attempts to include real
physiological quantities,鈥� says neuroscientist Scott Kelso of Florida Atlantic
University in Boca Raton. But he remains sceptical about what it demonstrates.
鈥淭he model produces chaos, there鈥檚 no doubt about that. But does that mean that
chaos is the major dynamical feature of the nervous system? That鈥檚 still an open
辩耻别蝉迟颈辞苍.鈥�

It鈥檚 likely to be a tough question to answer. The simulated brain activity
produced by the model was very sensitive to change, becoming more or less
chaotic depending on the input to the neurons. So it may not be possible to
detect signs of chaos in actual brain recordings because researchers could never
completely control all the inputs in a real person鈥檚 brain.

But the existence of chaos in the brain could be tested indirectly by using
the model to predict how electrical activity in the brain responds to drugs.
That could then be tested directly in living animals, says Liley.