left-right reversal of CNS: Why?

Jerry Larson jlarson at hsc.usc.edu
Sat Sep 14 22:19:31 EST 1996

> dantso at cris.com wrote:
> >       You are confusing image inversion with visual field representation.
> >         The optics of the eye have nothing to do with crossed CNS
> > representations.
> >         Humans are among the few species that have an appreciable binocular
> > overlap, perhaps to provide stereopsis, or whatever.
> >         Crossed CNS representations are ancient, many species having very
> > poor or no vision, or no single retinal image. Many species in which other
> > senses far dominate vision. The visual *field* (not image) representation
> > crossing followed the previous cross of the rest of the CNS. It did not
> > drive it.
Sounds right to me.  I think there's some confusion of levels in this
whole thread.  

This whole business about the topological equivalences and deformations of
the various homunculi (and generic animal-unculi), I just don't get it.
The brain doesn't know its own topological layout.  If you could take and
twist it around through space, it would still work; in fact just that has
happened.  I'm inclined to think that there was some good reason for
crossed control circuits, but it's likely to have predated the elaborate
encephalization of mammals, especially human beings.

btw, I remarked earlier that many cranial nerves have ipsilateral or
bilateral representation; one person agreed, another disagreed, saying
that they map onto the same, contralateral, cortical homunculus.

  I'll admit that I was thinking in large part of hearing (bilateral),
smell (bilateral or nonlateralized), and vision, which works by crossed
fields rather than crossed eyes, as I said.  But for many of the other
cranial nerves, yes, it's true that the _cortical representation is in the
opposite hemisphere.  However, the first (most rostral) nucleus concerned
with that nerve is generally on the ipsilateral side of the brainstem,
whereas the somatic nerves, having already decussated at the level of the
medulla, find their first nucleus on the contralateral side of brainstem
or thalamus. So there are crossed fibers, but they're postsynaptic;
they're crossing after they get to the brain, whereas the spinal nerves
cross just before, or just as, they get to the brain.  And then, of course
there are vision, hearing and smell, but maybe they don't count, since
they're not directly concerned with the CNS representation of the body.

This does all seem to make some kind of functional sense, with the somatic
sense and motor control for one side of the body, plus the the
corresponding visual field, all in one hemisphere (and not necessarily so
for smell, hearing, and the face).  Why it needs to be the contralateral
hemisphere, that I still don't see.  I don't find the business about
avoidance movements convincing, since the crossing phenomenon, and the
avoidance movements, probably predated advanced cerebral and cerebellar
development, but obviously, if you have legs or fins, or even paired
muscles as in snakes, and you want them to work in a coordinated way,
there has to be cross-talk of some kind.  Many quadrupeds, e.g. horses,
can stand or even "walk" by spinal reflex when decerebrated, showing there
must be crossed connections at the spinal level.

Obviously the issue arises with bilateral symmetry.  As soon as you have a
left and right, and a front and a back, you can have direction; you can
put eating, fighting and sensing equipment in the front, and you'll want a
lot of nervous system at that end to control all that stuff; and you'll
want more nervous system running from that end clear to down to the other,
and you'll want locomotor equipment (legs, fins, wings, whatever) on both
sides so you don't just go in circles, and you'll need some kind of way to
coordinate the two sides.  

Basically, if you bump into something on the left and you want to avoid
it, the right limbs have to do something.  Maybe what they have to do is
relax, so the left ones can push you away.  Of course, you also need a way
to do the opposite, so you can go after things you want, and that sort of
decision, whether to go for it or avoid it, that is just the sort of thing
a brain would come in handy for.  But even just to go in a straight line,
you need some sort of feedback mechanism.  If you haven't upgraded your
nervous system to a brain-based one yet, some reflexes that take
information from the left and do something on the right would be just the
ticket.  In other words, crossed circuits,i.e.decussated fibers!

If you want to build a robot with a complex brain, you can probably make
it work by ipsilateral control.  I don't know if it can compete with
contralaterally-controlled robots or not, but it will work fine. 
But if you have really simple control circuits, some of them are going to
have to be crossed, so that you can, for example, have the left rear and
right front leg work as a functional unit.

See, when we ask the question why there is "left-right reversal of CNS",
we (at least those of us who aren't experimental neuroscientists) are
thinking about "the left side of the brain controls the right" as Paul
Simon says, i.e. contralateral _cortical representation.  But cerebral and
cerebellar cortex came long after bilateral symmetry.  The conclusion I'm
coming to here is that the crossed representation was there _before the
encephalization, in fact it was necessary exactly because there wasn't
much central control yet.

Does anybody know offhand, what is the lowest thing on the evolutionary
ladder that demonstrates contralateral representation?  Do insects have
it?  I would guess they do, because they don't have much of a brain, so
there have to be "subcortical" control circuits that coordinate the two
sides of the body.

Do all vertebrates have it, even critters like snakes without legs or
fins? (I would guess they do, because they USED to have feet or fins in
their evolutionary derivation). How about planaria?  (Planaria, i.e.
flatworms, have a pair of nerve trunks, right and left, running lengthwise
down the body.  I believe they do have some sort of rudimentary brain, but
how do the two sides work together-- are they controlled by the brain,if
any,  or do they talk directly to each other?)


More information about the Neur-sci mailing list