"NMF" <nm_fournier at ns.sympatico.ca> wrote in message
news:84jYb.5353$Cd6.334665 at news20.bellglobal.com...
> > > The effect acts both mechanically [altering K+
> > concentration within glia alters glial Geometry,
> > which alters neural Geometry, including synaptic
> > 'pressures', etc.], and conductance-dynamically
> > [altering passive spreading, action potential
> > thresholds, etc.],
>> Yep. That is all well established. There is considerable evidence
> suggesting that oligodendroglia exhibit rhythmic pulsatile movements that
> may be responsible for regulating the infraslow potentials (i.e. 2-5 min)
> generated by the brain. Theories implementing glial cells in learning,
etc.,
> have been presented in the 30's and 40's.
That wasn't what I found when I researched
it in the mid-1970s. I didn't find anything that
coupled glial fx to learning and memory.
Except for the research that's cited in AoK,
I worked out NDT's position by tracing the
necessity of it back from TD E/I-minimization.
Since AoK and other papers and letters that I
circulated, glial research took-off. It's, now,
quite active.
It's 'hard' to experience such, BTW.
> It's a fascinating area of
> research and I agree with you that perhaps more active roles are involved
> with these cells.
It's a virtual 'hydraulic'-shape-shifting world in-there.
> Read some of Jim Roberston's work on the subject. He has an interesting
> paper published in Journal of Physiology a few years ago (2002) on the
> subject. And Perea and Araque published an extremely interesting paper on
> the potential dynamic communication between neurons and astrocytes. It was
> also published in J. Physiology and in the same issue as well. (An issue
> that as you expected reviewed glial cells and their involvement in the
> nervous system.) (more below).
>> > And this position is Testable.
> >
>> > The necessary experimental design has to present
> > experimental subjects with the opportunity to
> > acquire, and switch between, widely-differentiated,
> > robustly-established memory 'states'.
> >
> > Train to task, then train to widely-differentiated
> > task. Then, present one task or the other, and
> > sacrifice at various stages of 'switching' [which can
> > be monitored by measures that quantify completeness
> > of 'memory state' switching.
>> Yes. There are a many paradigms that can be done to test this aspect of
> "switching". There are many hippocampal pattern recognition tasks could
be
> used.
>>> >> If NDT's position is Correct [it is :-], then there'll be
> > observable glial anion differentiation that's correlated
> > with 'memory state'.
> > What happens is [I expect] that glial anion 'conforma-
> > tions' are tunable, and this affects both glial structural-
> > 'conformation' and glial K+ conductance in a way that's
> > correlated to 'memory state', and which can vary
> > Profoundly, with respect to different neurons, none of
> > which does the "Nernst Equation" see, or allow for.
>> I don't buy the K+ component you are advocating simply because this has
> already been well studied and the results aren't all that impressive.
> Simply put, the range of encoding that would accompany both K+ influx and
> efflux is not sufficiently dynamic and broad enough to serve as
> communicative template to provide any meaningful information for a neuron.
It's not K+ coding. All that has to happen is
very-sleight K+ concentration variation, and
profound alterations in neural dynamics can
occur - because it'd alter conductivity thres-
holds =locally=, which alters the way that
both passive spread and action potentials
permute - which is everything.
So, forgive me, please, but I stand on what
I've posted, thinking that you've not yet
considered what's in-it, fully. [It's relatively-
easy for a computer programmer to see it,
because the Criticality of a bit's being on or
off with respect to other bits being on or off
is a problem that the programmer is always
dealing with.]
> I doubt you would find a correlated memory state based upon the
> intracellular levels of potassium. K+ doesn't elevate as much as what you
> are think it does in the normal brain. Your best bet would be to
> concentrate your theory on calcium oscillations. Carmagnto(?) has shown
the
> importance of these processes within astrocytes. Moreover, calcium
> oscillations have been shown to regulate the release of glial-related
> proteins, that could serve as modulatory in synaptic efficacy and
important
> for the regulation of long-lasting potentation.
In my view, it all 'revolves' around K+, because
glia are 'exclusively' permeable to K+ - they are
like K+ 'electrodes, and, through glia-glia K+ flow,
all sorts of wonderful 'hydraulics' occurs, all of
which acts back upon passive spread and actuon
potential thresholding, as above.
If there's one ion that asserts this functionality,
there's just no need for other ions with respect
to this functionality.
Note, I'm =not= saying that other ions aren't
involved in =other= ways. Of course they are.
I'm saying that glial K+ conductances function
as I've discussed them. It's easiest to see all of
this diagrammatically.
My view is that folks got 'lulled into sleep' with
respect to K+ conductances, when their role in
the action potential was worked-out. [I also see
analogous [but functionally-different] dynamics
with respect to all ion varieties, BTW.] But there's
the need for only one 'hydraulic' mediator. Every-
thing else 'falls into place' as a function of the one
'hydraulic' mediator.
This's fairly-easy to Test [at least crudely] via
localized manipulations of K+ concentration [via
localized K+ 'clamps'].
> If, glial cells play a role, it would be a dynamic and modulatory role on
> neuronal functioning.
I disagree.
> Glial cells DO NOT (or shall say, HAVE NOT BEEN
> SHOWN) to play a direct role in memory consolidation.
I disagree, as I've been discussing.
> Their role is simply
> to modulate neuronal transmission.
I see much more, but modulating neuronal
'transmission', through a modulation of pas-
sive spread and action potential thresholds,
selectively tunes everything that's involved
in 'memory'. [This's a realm that's in-vivo-
critical, though, it cannot even be seen in
vitro - be-cause, absent global-integration,
the glial fx just isn't there, and, therefore,
only weakly-observable, so, in vitro, what's
important is just be discarded as 'noise'.
> Two recent studies may be of interest to
> you. One study showed that activation of working memory networks was
> significantly correlated with the extent of glial cell activation and
> metabolites in HIV brain injured patients (which exhibit abnormal glial
> inflammatory effects). They hypothesized that the increased glial
> processing is associated with a decrease in neuronal processing.
If it's as you've described, it's probably 'just'
the glial 'hydraulics', occurring 'abnormally' -
which provides a 'handle' into the glial dyn-
amics that I've been discussing [a differential
that can be analyzed, and cross-correlated in
myriad ways with respect to all aspects of
nervous system function].
> A
> second study (published in PNAS) showed that astrocytes synthesize an
> important calcium binding protein, S-100B. It is believed that astrocytes
> release this protein extracellularly and that it can modulate neuronal
> functioning. Mice devoid of S-100B exhibit greater synaptic plasticity
and
> enhanced learning.
This =might= 'correspond' to the glial anion
'conformational' variation that I referred to
in my prior reply. [I understand that my dis-
cussion has been inadequate. The way I work,
online, is to 'construct' concepts via reiterative
discussion.] In these dynamics, an alteration in
K+ conductance alters ambient energy-gradients
which alters protein-folding dynamics, and, there-
fore, end-'states'(?) ["(?)" = "working-hypothesis".
[This is a very-'thermodynamic' view on protein-
folding that I've been discussing, here in b.n, for
'years' ["3-D energydynamics"].]]
The crucial thing is the resultant "multiplexing" of
physical structure [of the neural Topology]. This's
"crucial" because sprouting, spine-Geometry, etc.
just cannot be sufficiently-dynamic be-cause this
'hard'-structural stuff is too-metabolically-costly.
So a more-energy-efficient means is necessary
to tune =within= already-established structure.
This's in the realm of the glial functionality that
I've been discussing [and with respect to which
I've not, yet, adequately conveyed the inherent
power and functional-extent].
> These points suggest that glial cells are not directly involved with
> learning and memory consolidation per se, but instead regulate the
activity
> of neuronal elements that are important for the long-term maintenance of
> information.
Yeah, 'support' roles are old stuff.
What I'm discussing is much-more than
'support' roles, though. I'm discussing
an acive role for glia in =all= aspects
of 'memory'.
Let's see, what's a useful teaching-analogy?
A parametric "case" statement in computer
code, in which the parameters are, themselves,
learned by the code.
[Gees! My PC is under attack as I'm writing
this! :-[
The "parameterss'" 'states' modify the code's
flow.
> The memory state, as you suggest, would not be associated with
> glial cells directly but instead would reflect the role of glial cells in
> regulating the operations of neurons in a dynamic manner.
Perhaps we're saying the 'same' thing,
but, as far as I'm concerned, if it act-
ively enters into any aspect ot 'memory',
then it =is= "memory" [because, if it wasn't
in-there, "memory" would, at least, different,
and probably not achievable.
In NDT, all of this stuff is of-a-piece [even
though some of the dynamics are in NDT's
"abstract conceptual" perspective, these
point directly to experiments that, when
done, will verify NDT's position, and dis-
close physiological specifics that are not,
yet, explicitly-reified within NDT].
> Considering the
> strong electrotonic coupling that exists between glial cells, the
syncytium
> nature of glial cells could be instrumental in orchestrating perturbation
in
> neuronal firing dynamics in such a matter that profound shifts in
cognitive
> processes may occur.
Yes, that's what I've been discussing :-]
Except that I work solely in conductances,
seeing 'oscillations' as being artifactual [not
what's significant].
> In light of Pribr[a]m concept on the holographic
> representation of memory within the brain, glial cells and glial processes
> may be useful for producing the interference patterns that can be
> superimposed upon the background activity of the brain. Enhancing the
> contrast between electrical patterns of experience and background patterns
> of activity.
Yes, it was Pribram's work that got
me started on NDT's glial hypothesis
back in the mid-1970's. [And it's good
to hear of his work being referred to,
by others, with Respect [which is
relatively-recent. It used to be that
Pribram's work was, rather savagely,
'ridiculed'.]
> (However, there is a cautionary note. Older studies have shown that the
> actual oscillatory shifts exhibited by glial cells are on a time scale
that
> is too large to play any direct or active role in modulating complex
> cognitive processes. Simply put, if glial cells are going to play the
> complex and global role in producing memory state switches, some other
> mechanism besides current models on neuronal signaling must be shown. )
As above, 'oscillations' are artifacts - not
information-relevant.
The stuff I've been discussing acts at all scales
of the neural architecture, from extremely-localized
ways with respect to sub-portions of neurons, to
the nervous system as a whole [through profound
globally-integrated cognitive shifts].
The glial fx that I'm discussing literally tunes neural
fx - like the flow of water in a stream 'tunes' the
swimming of a fish - only, in a globally-integrated,
actively-co-operating way with respect to cognition,
including 'memory' addressing.
It's easier to convey the essence of it diagrammatically.
I had this stuff diagrammed, but the diagrams were
stolen a couple of 'years' ago [shortly after I offered
to post the diagrams here in b.n].
Cheers, Neil, ken [k,p, collins]
