From: Jim Fehlinger (fehlinger@home.com)
Date: Mon Mar 26 2001 - 20:57:46 MST
I picked this up at the local Barnes & Noble over the
weekend:
http://www.amazon.com/exec/obidos/ASIN/0880642629
It's some candy to distract me from _IBM and the Holocaust_,
which has become a bit of a slog. The author is a science
journalist (author of _The Myth of Irrationality: The Science
of the Mind From Plato to Star Trek_, which strikes me as
an auspicious title ;->), and the tone is breezy and slightly
irreverent, which should prove entertaining.
I went straight to the index and found the following bit
of gossip about Gerald M. Edelman (Chapter 8, "The Needs
that Shape the Brain", pp. 166-170):
--------------------------
Good timing is the secret to a lot of things. Late in the 1980s,
a Nobel prize-winning immunologist trying to revolutionize the
mind sciences with a more dynamic approach to consciousness
managed to attract an awful lot of bad press. In a gathering of
neuroscientists, the mere mention of Gerald Edelman's name was
enough to send eyes rolling heavenward.
A New Yorker, famous for his intense manner, patrician bearing,
and highflown tastes (he once trained to be a concert violinist[*]),
Edelman won a Nobel prize for medical research in 1972. Then,
like Francis Crick, Edelman decided to branch out and tackle the
bigger question of human consciousness. But whereas Crick spent
most of his time visiting other people's labs, learning about
what they were doing and helping to tease out the implications of
their research, Edelman wanted to have his own set-up and pursue
his own vision of the brain.
A charismatic figure, Edelman proved adept at drumming up
sponsorship. Eventually, enough money was raised to found the
Neurosciences Institute, a $16 million 'monastery of science'
built into a hillside at the Scripps Research Institute in
southern California. The cash paid for wet labs to study
neurology and dry labs for computer simulations of brain
circuits. Edelman could also afford thirty full-time staff. For
other neuroscientists, it was bad enough that a complete outsider
was getting his own lab, free of any of the usual teaching
responsibilities or funding constraints. But what really raised
hackles was that Edelman was talking as if he had already
discovered the secret of consciousness. Interviewed in the _New
Yorker_ magazine, Edelman said that the mainstream of mind
science -- dominated as it was by the computational model -- was
so far off the mark as to be 'not even wrong'. His own big idea
was that brain processing was based on neural competition.
States of consciousness evolved through exactly the same kind of
classical Darwinian mechanisms that drove the evolution of life.
In a trilogy of densely argued books published between 1987 and
1989 [_Neural Darwinism: The Theory of Neuronal Group Selection_
(1987), _Topobiology: An Introduction to Molecular Embryology_
(1988), _The Remembered Present: A Biological Theory of
Consciousness_ (1989)], Edelman laid out his manifesto for Neural
Darwinism, or TNGS, the theory of neuronal group selection.
Edelman wrote that just as animals compete for food and living
space in the struggle for life, so sensations had to compete for
space in the mapping surfaces of the brain. Every moment would
begin with a battle in which some networks of activity would
blossom, gaining the neural territory needed to become conscious-
level percepts, while other, weaker, nerve-ensembles withered
away. Crucially, one of the factors determining the success or
failure of a new sensation was the support it received from
higher levels of the brain. If a sensation was anticipated, or
deemed important in some other way, positive feedback from higher
areas would help swell the mapping activity, elevating it above
the general clamour. Edelman called this feedback between
levels of processing a 're-entrant circuit'.
Another essential point that Edelman felt the rest of
neuroscience was missing was that it was quite wrong to think of
the brain as a lump of fixed hardware. He saw that it was
incredibly plastic -- plastic on every timescale, from the
fleeting instant of a neural firing pattern, through the
minute-by-minute and day-by-day rewiring changes that create
memory traces, right out to the developmental competition needed
to wire the brain's pathways during childhood, and even the
genetic timescale changes of ordinary evolution. The brain might
look to have a stable design, but -- as Merzenich's work with
finger maps in monkeys had showed -- it is in a constant state of
flux, forever changing its connection patterns as a result of
internal selection and competition. In short, brain circuitry
flowed to meet the processing demands being placed upon it.
This struck Edelman as a beautiful idea. It meant that there was
a continuum tying consciousness back to biology. Exactly the
same selectionist principles ruled at every level of the brain's
operation and only the pace of the adaptation varied.
Genetic-level adjustments in the organisation of the brain
happened with glacial slowness, while the neural competition to
map the contents of a moment would flare and collapse in the
blink of an eye. Yet the identical mechanism of 'processing
through adaptation' applied.
This hierarchy in adaptive levels made consciousness seem a lot
easier to explain. The story of the animal brain would be one of
a gradual increase in the speed of its competition-based
adjustments. Simple animals, like worms and jellyfish, could
only adapt their nervous systems over generations. Slightly
smarter animals, like slugs and snails, could make the nerve
junction growth changes that gave them a rudimentary ability to
remember and learn -- a sort of reflex-level awareness. Then,
further up the evolutionary scale, animals become capable of
creating fleeting mental maps of the world. In under a second
they could evolve a change in brain state, a disturbance to a
memory landscape in response to a fresh wave of sensory input.
So consciousness turned out to be just a further twist in an old
evolutionary tale. It was the result of speeding up the adaptive
abilities of the brain until its circuitry became able to adjust
to the events of a single moment.
The evolutionary angle on consciousness came naturally to Edelman
because of his own work in immunology. Edelman's Nobel prize had
come from helping prove that immune cells are produced through a
selectionist competition. The problem for the body is that it
cannot predict what kinds of bugs or viruses it might face in
life, so its solution is to manufacture a huge and rather random
variety of immune cell types. All these cells float in the
bloodstream. Then, when there is an invader, whichever immune
cell happens to have the right response characteristics will be
stimulated to reproduce itself in great numbers, rapidly swamping
the disease.
The principle of selection seemed so elegant and powerful that
when Edelman decided to give the problem of consciousness a go,
it appeared obvious that as a biological system, the same sort of
logic must be at work in the brain. So Edelman was rather
shocked when he found that the Darwinian model did not seem to
feature at all in the computer-obsessed thinking of psychology
and neuroscience. With typical vigour, Edelman tried hard to
point this out. But he got little thanks. In a snub that
typified the general response, Crick coolly remarked that Edelman
was an enthusiast notable more for the exuberance of his ideas
than for their clarity. Backs were turned on him, and
researchers resumed their search for the neural code or
topographical mapping principles by which the brain must process
information.
It was true, of course, that Edelman made some presentational
mistakes. He invented a lot of his own jargon for concepts that
already seemed to exist, using words like 're-entrant' which did
not appear to add a lot to the idea of ordinary feedback. Still
more of a problem was that his use of the Darwinian metaphor made
him sound curiously outdated. By the end of the 1980s, anyone
thinking seriously about neural competition or evolution was
already hurrying to jump aboard the new intellectual vehicle
provided by the sciences of chaos and complexity. So while for
some Edelman was a step too far, for others he already seemed a
step behind. A much broader revolution was in the offing. For
Edelman, the timing was all wrong.
The tale of Gerald Edelman is not important apart from what it
says about the state of the mind sciences at the outset of the
1990s. The history of science is littered with individuals who
seemed to have had a considerable personal understanding of the
dynamic nature of the brain and the way it works. Brilliant
theorists like Wundt, Hebb, and Neisser -- and many others not so
far mentioned, such as the Gestalt psychologist Wolfgang Kohler,
or the Russian physiologist Evgeny Sokolov -- have briefly opened
the door to a way of thinking that might have led the study of
the mind in quite a different direction. Yet the time was never
ripe. To most people, the standard reductionist view of
information processing always seemed to be working well enough --
it was getting more things right than wrong -- so there was no
recognition of a need for change.
However, just a few years into the 1990s, the mind sciences did
begin to see that groundswell shift in opinion. Everywhere they
looked, researchers were discovering troublesome facts. Scanning
showed too much activity in the brain; single-cell recording
showed too much plasticity; the simple models of cognitive
science were just not fitting. And suddenly, the sciences of
chaos and complexity were offering a genuine alternative to
reductionism, a way of thinking about an evolving system that was
backed up by solid maths.
The change was certainly not universal. As is the way with all
revolutions, it was the young researchers with nothing to lose
who made the more fervent converts. And the change did not show
through with great clarity. It was easy enough for academics
like Friston or Kosslyn to get across what they meant in private
conversation, where they could flutter their hands a bit or point
at a nearby pond, but much harder to say the same things in the
formal language of peer-reviewed articles. However, gradually a
more dynamic view filtered into the mainstream of thought so that
by the mid-1990s it had become difficult to find anyone who would
disagree with at least the essence of what Edelman had been
trying to say."
--------------------------
[*] Edelman's interest in music is obvious from some of the
anecdotes scattered through his books, for example:
"Writing about these properties [of memory: association,
inexactness, and generalization], I am reminded of the contrast
between different gifted individuals in their approach to memory
and performance. A story told about Fritz Kreisler, the great
violinist, and Sergei Rachmaninoff, the great pianist, provides
a case in point. In 1930, they met in Berlin to record the
Grieg C Minor Sonata together. Rachmaninoff was a meticulous
worker and wanted to practice right away. Kreisler, who didn't
practice much, was not so assiduous and went out on the town. The
next morning, at Kreisler's insistence and with Rachmaninoff's
reluctance, the recording took place; it went well. (It is
still available, I believe, and is stunning.)
[ http://shopping.yahoo.com/shop?d=c&cf=1&id=245334
Nonetheless, Rachmaninoff was not pleased.
Somewhat later that year, the story goes, the two played together
in New York, and the program included the same sonata.
Somewhere in the course of a movement, Kreisler had a memory
lapse. Being Kreisler, he simply made up some cadences,
probably in the hope of picking up the thread later. When that
didn't happen after a minute or so, he leaned over, still
playing, and asked, 'Sergei, where are we?' Rachmaninoff
looked up from the keyboard and said, 'Carnegie Hall.'"
-- _Bright Air, Brilliant Fire_, Chapter 10 "Memory and
Concepts: Building a Bridge to Consciousness", p. 104
Edelman's daughter, Judith, is herself a professional
musician: http://www.compassrecords.com/edelman.htm
Jim F.
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:06:42 MST