'U' See?

Kenneth Collins k.p.collins at worldnet.att.net
Sat Aug 10 23:17:38 EST 2002


Of course, the 'pawl' is not like a marlin spike, but consists of a
locally-focused [short-range] network, within this or that TD
E/I-minimization mechanism [AoK, "Short Paper", Ap3, 5, 6, 7], which
receives stochastic inputs from the ipsilaterally-aligned protopathic
system and, typically, widespread inputs from, and sends outputs to,
the epicritic system.

All TD E/I-minimization mechanisms converge upon supersystem
configurations in which information is represented in terms of
relative-direction within the Internal Frame of Reference Geometry.
[IFR; See the "neural topology" thread I started this evening. See,
also, the discussion of "supersystem configuration" AoK, Ap5, and
"volition", AoK, Ap7.]

Anyway, TD E/I-minimization procedes within globally-integrated
'arrays' as the locally-focused 'array-elements' "whittle" [AoK, Ap5]
their 'momentary' circuit lengths down to minima.

As is discussed in AoK, the locally-focused circuits 'blindly' and
automatically 'eliminate' all activation that can be eliminated until
such results in the occurrence of TD E/I(up), which constitutes the
'over-shoot' condition, which is what triggers the 'pawl'.

As is discussed in AoK, Ap5 my analysis is that the 'cycling' occurs
via the alternation of epicritic and stochastic inputs.

But, see?

The 'oscillation' is not "oscillation", but only an artifact of the
alternation of epicritic and stochastic inputs within the TD
E/I-minimization dynamics - TD E/I-minimization's 'shadow', and is,
itself, non-information-containing. Within the TD E/I-minimization
dynamics, it correlates only to the 'latching' that occurs at
epictitic\stochastic input-phase 'over-shoot' during TD
E/I-minimization, upon which the 'other' phase is triggered. The
"oscillation" is a 'shadow' of TD E/I-minimization.

Information exists in energy-flow directionality set-up by, and
inherent in, the TD E/I-minimized circuits that are 'momentarily'
converged upon, =not= in the artifact-stuff of the alternation of
epicritic and stochastic inputs through which TD E/I-minimization
occurs. That stuff is 'only' the 'grinding of the gears' of TD
E/I-minimization - 'machinery' - a TD E/I-minimization
servo-mechanism - all 'blindly'-automated.

In the converged-upon TD E/I-minimized 'state', there's no
"oscillation", but 'only' the loop-circuits that are converged upon,
and which are, ideally, the minimum-length-allowed by the 'momentary'
epicritic inputs, with respect to previously-established "biological
mass", which "biological mass", is, itself, tuned-in during the
course of TD E/I-minimization, which tuning enters into determination
of the TD E/I-minimization 'over-shoot'.

The IFR energy-flow directionalities that are converged upon via TD
E/I-minimization constitute the stuff that precipitates, maps, and
guides the by-production of formerly-non-existent "microscopic
trophic modifications" ["micro mods"; "biological mass"] which,
subsequently, encode[s] information, enabling 'memory' storage,
retrieval, and cross-correlation, via subsequent TD E/I-minimization.
[There's a =lot= going-on simultaneously - everything that's
discussed in AoK.]

[Don't ask me to provide circuit designs. I worked this stuff through
more than 20 years ago, worked-out simple prototype circuits, then
[preserved in my notes [somehwhere in my boxed archives]], and
haven't returned to it until "oscillations" came up here in
bionet.neuroscience. The circuits are not 'difficult'. They just have
to preserve the IFR Geometry relative to their locus within IFR,
while receiving epicritic and stochastic inputs within a 'local'
network that's rich enough in neurons to allow full-spectrum
loop-circuit tuning, via TD E/I-minimization, with respect to the
functional range of =each= epicritic input. There has to be
sufficient stochastic input to allow the local network optimal
convergence [loop-circuit-configuration] opportunity. The only other
requirement is that the locally-focused network must 'latch' within a
narrow range of relatively-low TD E/I(up), and receive 'governing'
inputs [examples: AoK, Ap5; amygdala re septo-hippocampal axis;
substantia nigra re basal ganglia, inferior olive re cerebellum,
prefrontal cortex re the rest of the brain ['recursively', re
lower-'level' TD E/I-minimization mechanisms] that dynamically set
the 'momentarily'-acceptable TD E/I limits.

This seems like 'a lot', but Because of the stochastic inputs,
integration of i/o with respect to diverse functionality is not
difficult.

I 'dropped' the prototype-circuit development, back then, because I
thought it'd be better done in an Academic setting, which opportunity
I expected my early work to win for me, which 'expectation' took me
away from my work in order to communicate the work, which is what
I've been doing for the last 20 years :-]

Born 'naive', still 'naive', and will die 'naive' with respect to
'the way things are supposed to be". Please Forgive me for 'lobbying'
against the waste, inherent. [I'm not really 'whining' on my own
behalf, but on behalf of doing what needs to be done in, and through,
Neuroscience.]

ken [k. p. collins]

Kenneth Collins wrote in message ...
>Kenneth Collins wrote in message ...
>>Sorry, I left out some necessary details.
>>
>>'down-ness' and 'up-ness' are just relative TD E/I 'momentary'
>>directionality, in this exercize, artificially mapped in 2-D
because
>>that's how the data were recorded.
>>
>>It's not 'illegal' Maths. It's just some of how Maths actually
>>happens within nervous systems.
>>
>>k. p. collins
>>
>>Kenneth Collins wrote in message ...
>>>Kenneth Collins wrote in message ...
>>>>[...]
>>>>There is convergence, within a "U"-shaped 'well', and that can be
>>>>termed "oscillation", but it's better described as
damped-harmonic
>>>>'motion' [with respect to TD E/I, not like a tuning fork dunked
in
>>>>a  viscous liquid],
>
>because TD E/I is not 'analogous to the =relatively= 'uniform'
action
>of a viscous fluid.
>
>[which is, again, why the "oscillation" is only an illusion - a
>'shadow' of TD E/I-minimization - an artifact of TD
E/I-minimization.
>
>At the ionic [3-D energy-flow] 'level' in which TD E/I-minimization
>derives, everything's 'just' =continuous= variation.
>
>The mactoscopic "oscillation" is an illusion at-scale.
>
>Same as when a strong wind shakes a tree. The tree might
"oscillate",
>but look-closer, and everything's going directionally willy-nilly.
>
>In nervous systems all of the ionic stuff is, more or less,
>rigorously-mapped in a topologically-distributed fashion with
respect
>to both internal and external WDB2T.
>
>Within this 'storm' of ionic-flow directionalities, the TD
>E/I-minimization mechanisms 'latch' TD E/I(down), which enables
>convergence upon relative TD E/I(min).
>
>But the "oscillation" is just a macroscopic illusion that's
>correlated with the 'pawling'-latencies inherent in the TD
>E/I-minimization mechanisms' functioning.
>
>TD E/I 'overshoots' before the TD E/I-minimization mechanisms can
>'stick-the-pawl' in-there.
>
>The "oscilation" is just an artifact of TD E/I-minimization that
>derives in this 'latching' ["pawling"] over-shoot.
>
>If the macroscopic illusion of "oscillation" is attended-to,
>=everything= that actually matters is ignored.
>
>the other thing is that, within the 'brain', TD E/I-minimization
>incorporates topologically-mapped 'arrays':
>
>"U" "U" "U" "U" "U" "U" "U" "U" "U" "U"
>  "U" "U" "U" "U" "U" "U" "U" "U" "U" "U"
>    "U" "U" "U" "U" "U" "U" "U" "U" "U" "U"
>     "U" "U" "U" "U" "U" "U" "U" "U" "U" "U"
>      "U" "U" "U" "U" "U" "U" "U" "U" "U" "U"
>        "U" "U" "U" "U" "U" "U" "U" "U" "U" "U"
>         "U" "U" "U" "U" "U" "U" "U" "U" "U" "U"
>           "U" "U" "U" "U" "U" "U" "U" "U" "U" "U"
>            "U" "U" "U" "U" "U" "U" "U" "U" "U" "U"
>              "U" "U" "U" "U" "U" "U" "U" "U" "U" "U"
>
>As is discussed in AoK, Ap6, the 'columnar' distributions of 'local'
>TD E/I-minimization 'zones' are set-up as a function of 'normal'
>sensory-inputs, mapped with respect to one another, 'hovering' about
>a cantral 'focus' [that correlates with the front-center orientation
>dynamics that are discussed in AoK, Ap3 [and disclosed in the
>Chronister, et al ref]. See the discussion of "sliding fields" in
>AoK, Ap6 [and run the "RELMO.bas ["relative motion"] QBasic code I
>posted before logging-off the last 'time'.]
>
>You, you, you... See?
>
>k. p. collins
>
>k. p. collins
>
>>>>so the harmonics just demonstrate that TD
>>>>E/I-minimization is happening.
>>>
>>>This needs to be discussed in more detail.
>>>
>>>First the convergence is probably via a high-eccentricity
>>>'paraboloid, so calling it a cross between a "U" and a "V" is
>>better.
>>>
>>>But, "Where is the curve?"
>>>
>>>It's in any of the traces that depict TD E/I-minimization [as in
>the
>>>Elazar and Addey ref that i've cited elsewhere on the board [and
in
>>>AoK], and it's in the Traub, et al 'challenge' paper.
>>>
>>>The 'jaggies' that occur within a trace that shows a downward
trend
>>>over 'time' are just the 'momentary' TD E/I-minimization dynamics
>>>'hunting', back and forth, with respect to the intermingling of
>>>stochastic and epicritic 'chance' fluctuations [scraps of
>>>'creativity'], 'latching' [in AoK, "pawling"] when TD E/I(up)
>>follows
>>>TD E/I(down).
>>>
>>>So, when these "jaggies" are mapped onto 2-D Cartesian
coordinates,
>>>vertical axis: TD E/I, horizontal axis: 'down-ness' --> 'up-ness'
>>>[distributed evenly along the axis], one gets the
high-eccentricity
>>>'paraboloid', which discloses the TD E/I-minimization 'well'.
>>>
>>>"U" See?
>>>
>>>k. p. collins
>>>
>>>>[...]





More information about the Neur-sci mailing list