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

TD E/I-minimization & WDB2T

KP-PC k.p.collins at worldnet.att.net%remove%
Thu Apr 3 05:52:24 EST 2003


Thank you for continuing.

----- Original Message -----
From: "BilZ0r" <BilZ0r at TAKETHISOUThotmail.com>
Newsgroups: bionet.neuroscience,sci.med.psychobiology
Sent: Thursday, April 03, 2003 1:35 AM
Subject: Re: Sundowning: severe dementia and bringing on the night
(corticosterone increases aggression in the rat)


>
> > There are neurons that manifest impulse activity. With respect to
> > such, if an impulse generates increased activation, it is
> > "excitatory". If it generates decreased activation, it's
> > "inhobitory".
> >
> > Within the nervous system, neurons are organized in a rigorously
> > topologically-distributed fashion. A tract, for instance, might
> > 'twist' around as it travls into the nervous system so that the
> > neighborhood relationships among the neurons comprising the tract
at
> > the body-environment interface [say the skin] get turned
upside-down
> > en masse. This sort of thing happens commonly within the neural
> > Topology, and is what enables various 'areas' of the nervous
system
> > to interact gracefully [with optimal efficiency] - which is
important
> > because 'graceful' neural activation correlates with minimal
response
> > latencies, which correlates to enhanced survival propensity.
>
> Okay, outside the CNS, nerves don't do that at all. Any of the
> parasympathetic nerves? In fact the whole autonomic system isn't
set up
> how you suggest, that could be a whole lot more efficient as far as
speed
> is concerned.

Even autonomic stuff is Topologically organized. To see this, imagine
'rewiring' the heart - it'd lose its synchronicity. Imagine
'rewiring' the smoothe muscles of the bowel with the nerves that
activate the heart. "Drizzly sh...".

And, as far as autonomic response latencies go, there's no need for
'lightning'-quickness in-there, because everything that's necessary
over the short term is stored in local tissue. 'Lighting'-quick
autonomic response latencies would only induce the system to
'thrash', but, more-importantly, even within nervous systems, it's
good to 'count to ten' before 'going ballistic', which is what the
autonomic nervous system does, and is why it's 'engineered' the way
it is.

And sympathetic is antagonistic to parasympathetic, which is 'just'
more "E/I" stuff.

> > Such Topological structure is, itself, an instance of TD
> > E/I-minimization because, if it didn't happen, there'd be the
need
> > for mor-complex interconnectednessm say, via a set of
interneurons
> > that 'bridge' the non-topologically-aligned structure to create a
> > version of the 'graceful' Topology, but in which all pathways
have
> > increased length.
>
>
> > Increased path length = lengthened response latencies and
reaction
> > 'times', so this 'bridged'-by-interneurons Topology can't compete
> > with the 'gracefully'-aligned neural Topology, because the
'graceful'
> > neural Topology always works faster - which is why the 'graceful'
> > neural Topology became ubiquitous during the course of
evolutionary
> > dynamics.
>
> In the CNS, how about the path the optic tract takes past the LGN?

The fiber=length variation within the optic radiations is necessary
to achieve proper timing of thevisual 'wave front' with respect to
visual cortex topology [this's an example I sight in the "Automation
of Knowing..." ms. [AoK].

> The
> optic radiation isn't as straight as it should be. And the Corpus
> callosum, why is it so small, it should extend all the way to the
back of
> the calcirine sulcus inorder to minamize the distance that the
optic
> cross over fibres take.

I presume you're talking about visual fibers that pass to the
opposite hemisphere via the corpus sallosum?

They must resolve the same problem that the optic radiations
resolve - timing of the neural-activation 'wave front' within target
cortical areas. Their distribution even looks Topologically-like the
optic radiations.

> Oh, and the path the IV and the VIII cranial
> nerves?

IV - trochlear cranial nerve; VIII - vestibular cranial nerve - what,
with respect to them, do you think 'violates' anything I've said?

>
>
> > All of the above applies in the abstract with respect to
excitation
> > and inhibition -minimization of the ratio of excitation to
inhibition
> > reduces response latencies and reaction 'times' allowing a system
> > which performs such minimization to out-compete any system that
does
> > not.
>
> If its all about responce latencies, why do we have so many GPCRs?
LGICs
> are far faster. phase out GABAb, all GABAa eh?

"Functional multiplexing" [AoK, Ap9]. Neurochemistry is a 'toolkit'
that allows the brain to be wired-up in an efficient way. The
existence of the various neurotransmitters creates the equivalent of
'chemically-mediated insulation' which enables neural circuitry to be
used for multiplexed purposes, without there being any 'cross-talk'.

This eliminates the need for redundant circuitry, and whole 'levels'
of interneuron connectivity, which greatly minimizes brain size,
which greatly minimizes response latencies, which enables most-robust
survival propensity.

> > Put all of this together and you get Topologically-Distributed
> > minimization of the ratio of Excitation to Inhibition - TD
E/I(min).
>
> If you minimised the RATIO of Excitititory recpetors over
inhibitory
> receptors you'd have a brain full of inhibitory receptors... I
don't see
> how that makes anything faster.

Of course zero activation only happens with death.

There're whole subsystems that have the purpose of injecting
stochastic activation into the system. This, of course, constitutes a
'contradiction' to the TD E/I-minimization principle, but only over
the short-term. Over the long-term, all of this stochastic stuff is,
itself, part and parcel of TD E/I-minimization.

For instance, a simple ambulation algorithm directs its host into a
wall. If the host has sensory receptors that act to inject TD E/I(up)
back within the algorithm, that 'detunes' the algorithm so that it
can 'recognize' that it has to 'try a new direction'. When it does,
the sensory activation ceases, and TD E/I-minimization happens.

Short-term TD E/I(up) signals corrections that are required to
achieve long-term TD E/I(down).

There're =many= instances of analogous dynamics that occur within the
brain, all the way up to the highest 'levels' of cognition and
volition.

> It might minimise the ENERGY used in tranmission, but it would
result in
> a scrambled signal.

Nope. Think about it in this way. If I'm to move my arms, hands and
fingers in a way that dances over my keyboard so that a sentence with
a particular meaning is typed out, all of the activation that would
cause my muscles to be activated in any of the virtually limitless
other ways that they can be activated, must be relatively-inhibited,
or I'll not type out the sequence of letters that comprise the
sentence that I have 'in-mind'.

That's just more TD E/I-minimization - in which an extremely-ordered
set of neural activations are selected from the virtually-infinite
set of all possible neural activations.

Our nervous systems 'cut to the chase' and 'just' do the TD
E/I-minimization thing directly, and behavior, etc. are produced as
by-products of the degree to which TD E/I-minimization is actually
achieved.

A Brilliant solution on the part of evolutionary dynamics, but one
that's inherently ;flawed' because, as a result of the way
'blindly'-automated TD E/I-minimization is converged upon within
nervous systems, anything that's merely repetitively experienced will
become TD E/I-minimized. This yeilds a system that tends to be
'blindly' and automatically prejudiced with respect to that which,
through prior experience, has merely become relatively 'familiar' to
it - instances include the "us vs. them" nationalistic prejudice that
precipitates the Savagery of war. [wanting to understand these
dynamics was the wellspring of the work work I've done in
Neuroscience, BTW.]

The tragic stuff inherent can be transcended by internalizing
understanding of how nervous systems work.

> and if the ratio was important you could then you
> could just add say 50% more receptors of each kind ( Excit. vs
Inhib.) of
> receptors and you wouldn't alter the ratio, so again, I don't see
why
> this ratio is so important.

The added neurons would make the brain larger, increase response
latencies, and increase energy consumption, all of which would render
such a brain relatively uncompetitive, and it would not survive under
evolutionary pressures.

> Now the idea of putting this that work together close together
(esp. in
> the cortex) is accetped, somatopy in the Frontal and prefrontal
cortices,
> retinotoy in V1. But its far from the rule, How about all of the
thalamic
> nuclei? and the SMA?

What about them?

> And again I ask, How does minimizing a ration lead of cognition?

For that, you'll have to get a copy of AoK and study it and the refs
cited in it. [It's non-trivial, but is available in AoK.]

> > There is. The order differential inherent in WDB2T is. itself, an
> > energy-gradient.
>
> Okay, you lost me there. As far as I was awear the 2nd law of
> thermodynamics was simply a statment that "The entroy of an
isolated
> sysmtem always increases" One can use that statement to construct
> mathamatical equations, but there is no gradient. Entropoy can
exsist
> quite happily without an gradients of any kind.

I don't talk about "the 2nd Law of Thermodynamics". I talk about
what's =described= by the 2nd Law of Thermodynamics" [WDB2T], and
that's the one-way flow of energy from order to disorder, which is a
physically-real energy-flow which has its energy-gradient which
permeates all of physical reality.

> > All that stuff that flows from high concentration through a
> > semi-permeable barrier to low concentration does is follow the
WDB2T
> > energy-gradient, until 'equilibrium' is 'reached' ['equilibrium'
is
> > never actually attained because WDB2T is ubiquitous, everywhere
> > within physical reality, 'going' in the one direction -
evaporation,
> > heat radiation, the heat radiation that underpins condensation,
etc.
> > WDB2T is always in-there, everywhere within physical reality..
>
> See, this is where you go wrong. the 2nd law of thermodynamics is
an
> emperical law, its not based on anything but experiments, its like
a
> framework in an argument, all of those examples can be explained
through
> it, but it is not needed for them to happen. You can use the 2nd
law to
> explain a lot of non physical phenomina, like economics and law.

Because the WDB2T energy-gradient permeates physical reality as a
physically-real energy-flow, one can use WDB2T to explain everything
within physical reality. Anything in which Humans are involved
reduces directly to TD E/I-minimization, which reduces directly to a
'climbing' of the WDB2T energy-gradient.

'Forget' statistical 2nd Thermo. See the physically-real
energy-gradient that exists because WDB2T is a one-way flow of energy
from order to disorder. See that, be-cause WDB2T permeates physical
reality, the WDB2T energy-gradient maps Truth within physical
reality. Sure, it's often the case that there're local energydynamic
minima [like a volcano's crator], but all one has to do with respect
to them is 'wander' in an ordered way, and one will discover that
there's greater Truth than that which is restricted to the local
energydynamics minima.

> >
> >|The brain could theoretically work with
> >| almost no entropy change.
> >
> > Not True.
>
> Yes it is true, it is absolutly and unevquivocable true. Imagine if
VSSC
> and VSCC were sensitive to a change from 0mv to 0.1mv, we can
construct
> diodes that are sensitive to millions of times less change, so why
not a
> channel? Then we can have the brain resting at ionic equilibrium,
and
> instead of a depolaristion of 25mv, one of 0.1mv would be all that
was
> neccassary. Then a AP could be a transient change of 0.5mv. After
that
> the brain would work exactly the same, except Na/K ATPase activity
would
> have to be turned down a whole lot. There we've reduced the entropy
> change massivly.

I don't know, offhand what VSSC  and VSCC  are.

The brain doesn't rest at ionic equilibrium, the "resting state" is
actually maintained, against the WDB2T energy-gradient [energy-flow]
through active [energy-consuming] ionic pumps.

The neural dynamics are as they are with respect to the signalling
work that each neuron must do. Weaken any neuron's energydynamics, as
you've proposed, and it'll not have sufficient signal strength to get
it's signal to it's target. And that'd create disordered activation -
which is TD E/I(up).

I've hammered on this stuff for 31 years. Trust me [or not], there's
nothing in it that'll 'break'.

Your statement is not True.

> >| Life dosn't move toward anything, and it
> >| certainly dosn't move toward energy abundance, if anything life
> > lowers
> >| the amount of free energy in a system.
> >
> > That Life 'climbs' the WDB2T energy-gradient doesn't mean that
Life
> > doesn't 'consume' useable energy. Life does, which is why Life
=must=
> > 'move toward' increasing energy abundance [relative to energy
within
> > itself] - if it doesn't, Life ceases via energy-depletion.
> >
> > If you doubt, stop eating [don't really :-]
> If we looked at me during my post adolencent life. I have lost
total
> energy if anything. I consume it, use it, and then loss it, there
is no
> accumlation leading to abundance. And if we're going to go back to
> entropy, which I wish we had never touched on in the first place,
my
> entropy over my life time decreases

I didn't say what you say I said. I didn't say anything about
"accumulation". I discussed how Life seeks increased energy abundance
[because, in the presence of increased energy abundance, survival
propensity is enhanced.]

I've traced these dynamics all the way down to bacteria. For
instance, in bactreial chemotaxis, flagelites move their flagella
[little 'hairy tails' that extend from the surface of bacteria] in an
ordered fashion when nutrient concentration gradient is increasing,
but in a disordered fashion when the nutrient concentration is
decreasing. The former imparts directed motion, which 'clibs' up the
nutrient concentration gradient. The latter imparts 'tumbling' motion
which is, essentially, stochastic, and which enables bactreia to
'turn-around' so as to 'climb' the nutrient concentration gradient in
another direction, or to 'hunt' for a nutrient concentration gradient
to 'climb'.

'TD E/I-minimization' without neurons - 'just' bacteria 'climbing'
the WDB2T energy-gradient.

> >| and stop citing your own unpublished works, its unproffesional.
> >
> > I'm an Amateur, anyway. My work is 'correct'. It's not published.
> > What would you have me do?
>
> Cite material published in peer review journals. Theres millions of
> neural network experiments published that you could use to support
your
> ideas (or hopefully refine them)

I do cite many, many excellent, hard-won experimental results in AoK,
and make that ref available to folks, gratis.

Since it's available to anyone who wants it, I don't 'rewrite' AoK
online.

Online, I just discuss stuff that needs clarification from the
perspective of folks familiar experience.

Your posts have been very helpful in this regard, and I'm Grateful to
you for the work you put into them.

Thank you.

ken collins






More information about the Neur-sci mailing list

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