Ray Scanlon rscanlon at wsg.net
Wed Oct 28 09:30:33 EST 1998

Neil Rickert wrote in message <715iua$8ur at ux.cs.niu.edu>...
>"Ray Scanlon" <rscanlon at wsg.net> writes:
>If N neurons have to meet up accurately with N locations, there are
>N! (that's a factorial) ways of doing this.  Given that N is very
>large, N! exceeds the total storage capacity of the known universe.
>Therefore the DNA could not possibly encode such a high degree of

I don't know how this "coding capacity" argument go started but I have seen
it again and again. It is a strawman. It is allied to the view of the brain
as a bowl of porridge with all the structure of oatmeal.

We are not talking about the possible permutations of n things taken n at a
time. We are talking about the human (as a vertebrate) brain with between
100 and 1000 billion neurons all precisely connected according to rules.

Rule number one: Remember thy neighbor.

Again and again and again in the brain, topology is preserved. There are
about one million ganglion cells in the retina, their axons make up the
optic nerve. The axons from one eye are joined at the optic chiasm by the
axons from the other eye. The axons then split with the axons from the left
half of both retinas continuing to the left brain and the right halves going
to the right brain. the axons are completely scrambled in the nerve but upon
arriving at the lateral geniculate nucleus (LGN), rule number one takes
over. The axons split into six groups each having its destination in a layer
of the LGN, three for the left eye and three for the right eye. The axons
sort themselves out and restore the topology, neighbor by neighbor. The
axons terminate on cells in the LGN.  Each layer contains a topological
mapping of the retina.

The axons from the cells of the LGN proceed (with laterals to the reticular
nucleus of the thalamus) to area seventeen of the cortex where the axons of
the four upper layers terminate in layer 4Cbeta and the two lower layers go
to layer 4Calpha of the cortex. Again rule number one determines that
topology is preserved.

How is the rule implemented? There are three major possibilities: mechanical
guidance, following chemical gradients, or homing in on some complementary
molecule. This is an active area of research with results pointing toward
more than one mechanism.

The important thing is that this beautifully precise wiring is accomplished
with only a few rules. One to preserve neighborhood, and several to
determine destination. This is NOT n factorial. Mathematical arguments from
the "bowl of porridge" are hopelessly naive.

Other classes of sensory input (somatic, auditory) similarly follow rule
number one.

When the axons that originate in the cortex proceed to association areas,
again topology is preserved.

>>My disagreement is with your statement that the brain is not limited to
>>signal energy flowing through the sensory neurons and information
>>encoded in DNA. What is this other source?
>For example, the chemical constituents of the air, the food, other
>parts of the environment.

Any molecule from the environment that desires to get to the brain must pass
the blood-brain barrier. This is not easy and is a basic pharmacological
problem when attempting to get medication to the brain.

Don't forget that I have never downplayed the environment. The DNA
constructs the brain, the milieu impinges on it. The effect of the milieu on
the brain is called "learning". The important thing is that the milieu
affects the brain through the sensory neurons and, from the past, through
the DNA.

You say I am preaching dogma but I say I am merely reciting basic

Those interested in how the brain works might look at

More information about the Neur-sci mailing list