FOR the past few decades, Charles Stevens of the Salk Institute for
Biological Studies in La Jolla, California, has been one of the leaders in
unravelling how synapses function. Without a basic understanding of nerve cell
communications, it will be impossible to understand how the brain processes
information. The challenge is truly formidable.
A single human brain contains 10 billion neurons interconnected by 60
trillion synapses. Each synapse is not a simple switch but changes the strength
of the signal it passes on according to the history of its use.
Stevens is the first to have studied how synaptic strength is modulated
during natural brain activity, rather than by the simple and repetitive
electrical stimuli usually used in the lab. He took records of electrical
activity from individual 鈥減lace cells鈥� in the hippocampus of rats as they
explored a room. Place cells play a part in the learning of location, and the
same cell shows the same pattern of electrical activity when the rat returns to
the same spot. Stevens played this complex pattern of recorded activity into
neurons in a slice of rat hippocampus in the laboratory. The result was
remarkable: synaptic strength varied with high precision for a wide range of
patterns (Neuron, vol 22, p 157). He concludes that the synapse rapidly
responds to the pattern of incoming electrical activity and modifies the message
it passes on.
Advertisement
How is this change brought about? Stevens鈥檚 experiments suggest that changes
in synapse strength are largely due to changes in the probability of tiny
vesicles of neurotransmitter鈥攖he chemical that carries a signal from one
cell to another鈥攂eing released from the pre-synaptic neuron. As impulses
arrive, each vesicle effectively rolls the dice and asks: 鈥淪hould I let my
transmitter go?鈥� The other critical factor influencing strength is the number of
vesicles ready at the active zone of each synapse鈥攏ormally between five
and ten.
As yet, Stevens does not know how that change in probability is controlled,
although he is sure there will be a molecular answer. 鈥淲e know a lot of the
molecules that are involved in synaptic transmission and have cloned many of
them,鈥� he says.
And that is just the start of the bigger adventure of understanding how the
brain鈥檚 60 trillion synapses add up. 鈥淥n a descriptive level we can say what the
synapses are doing, and we know a few vague things about why they might be doing
it that way, but there is not really a theory of computation in the brain,鈥� says
Stevens. But he points out a key difference between the brain and a computer. In
computers, data is stored in structures that are remote to the main processor.
In brains, information is stored right inside the processing circuits.
Understanding this difference may help us to see how the brain works, as well as
leading the way to a new generation of computers.
