Information Technology Reference
In-Depth Information
Neurons are complex in a variety of different ways. Consider their shape, which for no
immediately apparent reason can be very complicated, ranging from a spindly structure
of dendrites growing sparsely away from the soma to thickly entangled dendritic forms
converging on compact blobs in space, and pretty much everything in between. Some
neurons even have the branching structure of the lightning bolt displayed earlier. In
Figure
2.5
the morphology of a representative sample of the kinds of neuron structures
that are observed in the human brain and elsewhere in nature is depicted. The geometric
concept of a fractal dimension is introduced to distinguish among the various forms
shown.
There is little if any difference in structure, chemistry, or function between the neu-
rons and their interconnections in man and those in a squid, snail, or leech. However,
neurons can vary in size, position, shape, pigmentation, firing pattern and the chemical
substances by which they transmit information to other cells. Here, in addition to the
differences in the complex spatial structure, we are also interested in the firing patterns
and the distribution of intervals between spikes for neurons of different morphologies.
As Bassingthwaighte
et al
.[
7
] point out, certain cells are normally “silent” while others
are spontaneously active. Some of the active ones generate regular action potentials, or
nerve impulses, and others fire in recurrent brief bursts or pulse trains. These different
Alpha
Beta
Gamma
Epsilon
Delta
The fractal dimension is useful to characterize the shape or form of neurons. Those shown are
for five different classes of ganglion cell in cat retina. When the neuron is less branched, the
fractal dimension is closer to one, whereas when it is more branched, the fractal dimension is
closer to two. Reproduced with permission from [
33
] Fig. 4.
Figure 2.5.
