IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP

[Neuroscience] Re: Philosophical Foundations of Neuroscience

John H. via neur-sci%40net.bio.net (by j_hasenkam At yahoo.com.au)
Mon Dec 4 18:09:35 EST 2006


As luck would have it Michael's suggestion reawakened in me both
Freeman's ideas and R and W Blue's "correlational opponent process". It
opens a window with a nice view and sheds some light on my problem. I
did some reviewing today, it provided some justification for my
scepticism regarding the wholly materialist and Neuron Doctrine
approach to understanding complex brain functions. In particular, the
paper by Freeman,
Neural Networks, Vol. 10, No. 7, pp. 1175-1183, 1997

1997 SPECIAL ISSUE
Three Centuries of Category Errors in Studies of the Neural
Basis of Consciousness and Intentionality
Walter J. Freeman

adds weight to my argument that if we keep digging in the same
materialist hole we may be looking the wrong way.

Matt said this:

"This is serious stuff, right? What's at stake here is whether
neuroscience is incoherent, and will never go anywhere, because we are
pursuing a line of thinking that is utterly flawed. "

That is wrong and this is why:

The bods tend to get upset because they believe that such assertions
invalidate their life's endeavours. It is like saying that Einstein
made Newton redundant, when in point of fact Newtonian physics is still
far more widely applicable than Einstein or QM. The simple fact is that
the mountain of data created (bit like the tower of Babel) is not made
erroneous by abandoning the Neuron Doctrine or a strict materialist
approach to the problem. The data, or much of it,  is still good and
has value, what changes is the frame of reference for interpreting that
data. In any event, akin to the what happened in physics, much of that
data and its interpretation will still be valid and useful. So whatever
Bennet and Hacker think about the current state of Neuroscience, I fail
to see how it imperils the value of the work or the validity of the
interpretation. It most certainly cannot deny the clinical advances
provided by the same. It would be just as silly to abandon the existing
framework as it would to abandon all use of Newtonian physics. If
anything, what we will be pleased to witness is an increased richness
in the interpretation of the existing data. Crisis? What Crisis?

So IF Bennet and Hacker are asserting that neuroscience is doomed they
are being silly, they need to learn a few lessons from the history of
science. Yes, their ideas do have serious implications for cog neuro
but that doesn't really surprise me. I really don't know what the fuss
is about because other authors have been making similiar assertions for
years. Cog Neuro will have to undergo some sort of revolution, that
much is obvious without reference to this text. However the rest of
Neuroscience will remain relatively unscathed and will provide a wealth
of data up for reinterpretation.

My understanding is that AI is doing quite well by instantiating rather
basic ideas drawn from neuroscience, I think they are enjoying more
success there than trying to utilise concepts drawn from models of
"higher order processing" from cog. neuro. Not sure, Michael knows
better. One interesting comment I heard on a cap discussion a few years
back was that the best AI models "cheated" ie. they introduced "reward"
and "punishment" to enable the device to learn. I fail to see why this
is "cheating" because that is what happens in the world. I presume the
poster was operating under the assumption that intelligence is embedded
in the organism. I reject this, intelligence occurs through the
organism responding to its environment. The organism becoms intelligent
because of its environment. Now that is spooky, it is essentially non
materialistic, non Neuron Doctrine, it appears to be a step backwards
towards dualism. Hah, this is one of my criticisms of modern
psychiatry: it is too reductive. People think I'm being spooky!

In regard to my Great Quandry, this statement from the Blues' paper is
relevant:

"For human neuro function all that is
necessary is that there be a consistent relationship or association
with
neuro wavelets and external and internal events.

I do not find this explanation at all satisfactory, I had already
considered it and dismissed it  as being too simplistic and avoiding
the central issue. It's a cop out. But if some people are happy with it
then perhaps I should be.

The incredible thing is that Freeman's 1997 paper basically makes the
same argument against neuroscience in general and cog sci in
particular. Has it made any difference? Have all the papers addressing
these issues ever made any difference? When will neuro reach the point
of criticality? Does this point even exist? Damn  Kuhn, I don't like
his ideas but then ... .

Strange as it may seem, one reason I am looking at these ideas is
because in my own area of interest there are clinical findings that
bear no relationship to the original trauma, at least in terms of our
current understanding of neuropathology. The three key mysterious
symptoms that occur are: depression, emotional lability, and
attentional issues(even in mild brain injury). For example, one
reference I state indicates that post injury there is a very widespread
reduction in GABA receptors.  In depression without brain injury, there
is very good evidence to show a reduction in GABAergic transmission in
the hippocampus, il-1 appearing to be a key culprit here. These
experimental findings receive additional support from a clever MRI
study showing that a positive response to SSRIs corresponds with a
reduction in hippocampal activity, while the non-responders maintained
high hippocampal activity. Additionally, I have seen one reference
indicating a decline in GABAergic transmission in the occipital cortex
during depression (lost the ref dammit!) and in tinnitus one finds a
sharp reduction in GABA in the MGN, in my view a compensatory mechanism
because of damaged inner ear cells that goes too far. There is a lot
more going on here that relates to the linkages I mentioned earlier,
I'm still working on that, and working on that, and working on that ...
. Hyperexcitability is a key problem post trauma. Antidepressants are
known to modulate immune function, cort levels, and increase
neurogenesis in the DG, now I wonder what types of functional neurons
those progenitor cells become, a chronically inhibited neuron is a
dying neuron(ADNF???) ... .  What was that stuff about Lashley's "mass
action" idea??? Reaching this far shows how desperate I have become in
trying to establish the causes of these TBI symptoms.

PS: I have no way of critiquing either wavelets or chaos theory as
viable enterprises. There is one quasi problem: simple nervous systems,
which work quite well, may be "linear" in their function. At some
point, perhaps for example the bee, which has a tiny "proto frontal
cortex", these other dynamics come into play, though even ganglia
should be able to do the job.

Thanks Michael,



John..


Michael Olea wrote:
> John H. wrote:
>
> > I still maintain that Glen's position that what can be safely said is
> > that memory is "change" is the more rigourous position to hold. It is
> > empirically verifiable, "storage" is not. Unfortunately it confronts
> > one with a despair because at present it is so lacking in informational
> > value. Perhaps this is one reason why people prefer to think of
> > "storage": at least you can play with that concept. "Change" just damns
> > my intelligence.
>
> Don't dispair. See:
>
> CHAOS IN THE CNS: THEORY AND PRACTICE
> Walter J. Freeman
> http://sulcus.berkeley.edu/FLM/MS/WJF_man2.html
> 
> I hope that helps.
> 
> -- Michael



More information about the Neur-sci mailing list

Send comments to us at biosci-help [At] net.bio.net