| | |
Summary: nature neuroscience · volume 3 no 9 · september 2000 919
Hebbian learning, the development of neural circuits based on
correlated activity, relies on two critical mechanisms. The best
known of these is activity-dependent synaptic modification along
the lines proposed by Hebb1. Equally important is a mechanism
that forces different synapses to compete with one another so
that when some synapses to a given postsynaptic neuron are
strengthened, others are weakened2,3. For example, correlation-
based rules of synaptic modification can provide a reasonable
account of many aspects of development in visual cortex, but
only when they are combined with constraints introduced to
ensure competition4. Although Hebbian synaptic modification
has received support from experiments on long-term potentia-
tion and depression5, much less is known about the mechanisms
that generate competition between synapses.
At first, it might seem that any mechanism that imposes com-
petition among synapses must involve a global intracellular sig-
nal that reflects the state of many synapses. The constraints used
in many models of Hebbian learning6, although not biophysi-
cally realistic, are based on this idea. Typically these constraints
|
