"Ron Blue" wrote in message <089e01c23339$ac568920$ce02030a at rblue>...
>Hutcheon, B. and Yarom, Y. (2000). Resonance, oscillation and the
intrinsic
>frequency preferences of neurons. Trends in Neuroscience, vol 23, No.5,
>216-222.
i won't quote from the paper, mainly because i've little to say that's kind'
to its stuff. folks who want read it can find it on the net. i found it by
doing a Google on the complete title.
the paper contains occasional 'bridges' to physiological data, but it's is
written from a simulation perspective, and its emphasis is on facilitation
of such simulation. a worthy goal, but i found the article's discussion to
be almost-completely 'disconnected' from biological nervous systems, which
are the stuff correlated to my prior comments, in this thread, re'
"oscillations".
the 'problem' is that the paper attributes too much decision-making power to
properties intrinsic to discrete 'neurons', almost completely disregarding
Global-Network architecture.
for instance, it claims that 'eigenfrequencies', more or less intrinsic to
individual 'neurons', determine when, if, if not, a 'neuron' will
communicate with other 'neurons'.
there is a parenthetical statement in which it's said that the
'eigenfrequencies' can change, but no mechanism is provided with respect to
such, and without such a mechanism, the information-processing capacity of a
'network' comprised of such 'neurons' is extremely-limited, because, without
a mechanism for varying the eigenfrequency, a 'network' of such 'neurons'
can only exist as a =relatively= static entity. it's 'neurons' can only
communicate when the "eigenfrequency's" 'bell' is rung
but, if a mechanism for varying the 'eignefrequencies' is incorporated, the
'eigenfrequencies' cease existence, the allowed-to-vary thing becoming,
instead, stuff that's determined at the level of the global network.
what about physiologically-occurring neuronal-spike variation?
can't such be termed "oscillation"?
Nope.
Why not?
because whatever appears to be 'oscillating' exists in the midst of
continuous-variation-'pressure'. even when there's an appearance of this or
that neuron "state's" being 'steady, it's being actively 'held'-there
through the application of ionic-conductance force.
the 'problem' inherent in the paper, with respect to such
physiologically-real ionic force, is that it's discussion resorts to
'time'-coding of information, when information is encoded within the
energy-flow in which the ionic force derives. no 'time', so no
'time'-coding. [it's 'ok' in a simulation, but it just isn't-it with respect
to physiologically-real nervous systems, 'cause there's no such thing as
'time' in-there.]
as i said in my earlier post in this thread, in physiologically-real nervous
systems, what has the appearance of "being 'oscillation'" is dynamically
determined through the actions of the TD E/I-minimization mechanisms.
for instance, the "septal rythms [that] 'call cadence' for [the] alternation
[of stochastic and specific inputs to the hippocampus], do not constitute an
"oscillation", at least not in any rigorous sense, because the "rythm's"
'state' is being continuously tuned.
the way that such continuous tunning-'pressure' gains it's existence occurs
at the 'level' of the global network.
a simple example is with respect to the setting of the lengths of "loop
circuits", which exist all over the place within the nervous system.
the frequency at which any "loop" will fire occurs as a function of its
'momentary' loop-length.
in cortex, for instance, loop-lengths are converged-upon as a function of TD
E/I-minimization-delimited injection of stochastic activation provided by
"protopathic" projections to the cortex [from the reticular formation,
intralaminar thalamic nuclei, thalamic reticular nuclei, etc. [further
discussion in AoK, Ap3.]]
the stochastic activation is injected. the TD E/I-minimizations eliminate
what's extraneous, under continuous feed-back from inputs from the
environment, until convergence upon relatively-minimized loop-circuit
lengths occurs.
everything taps-into such loop-circuits to gain it's tuning information.
for instance, loop-circuits that're converged-upon via TD E/I-minimization,
as is briefly discussed above, determine the activation of the
motor-pryamidal cells which, then empower the musculature in the
manifestation of behavior.
everything that happens within the nervous system happens like this.
it's how, as i've discussed in msgs posted in other threads, everything is
'translated' into directionalities within the One Internal Frame of
Reference Geometry 'language', which 'translation' into a single
directionality-'language', via TD E/I-mim=nimization, maps everything
right-into WDB2T. the result being that all information is simultaneously
available with respect to all other information, because it all exists in
the one 'language'.
get-it?
the only 'time' in which this generalized-information-availability
'disappears' is when, as i've discussed in other threads, system-outputs
must be tuned with respect to manifesting specifically-directed external
force.
but the loo-pcircuits are not 'oscillators', and their activation does not
constitute 'oscillation'.
there're dynamically-tuned periodisities in the loop-circuits' activation,
but there's nothing that's "oscillating".
what there are is dynamically-varying circuit-lengths.
such isn't 'oscillation', is it?
Nope.
all of this stuff has been in AoK all along, although i wanted to point
folks to "loop circuits" in the ms., but was disappointed to find that there
was no Index entry for "loop circuits", and i don't have the 'time' to
reread Ap5, where i think "loop circuits" are discussed, within the
discussion of the TD E/I-minimization mechanisms.
anyway, this global-network way is completely-Substantiated in the proven
experimental results.
its information-processing capacity is Awesomely-Immense because
=everything= within it is continuously tuned via TD E/I-minimization.
it's more of why i've been unable to locate any 'boundaries' with respect to
the "brain's" [nervous system's] information-processing capacity.
there's a huge quantity of continuously-varying stuff in-there, but there're
no "oscilations" in-there.
[unless one wants to call relay stuff 'oscillation', which, coming from a
background in Physics, is a bit too-far-s-stretch for me. 'atoms'
"oscillate". "loop-circuits", and all of their correlated neuronal dynamics,
'get-in-line' and 'run-relays'. =huge= difference.]
something i found 'interesting' in the paper was it's 'side discussion' with
respect to "multiplexing". i expect that the mechanism proposed doesn't
occur within physiological nervous systems because, if it did, it would
greatly-diminish nervous-system information-processing capacity [because
only 'neurons' having the necessary intrinsic [but
varying-without-a-mechanism to tune such variance] frequency could enter
into loop-circuits. the way the brain does it suffers no such
restrictedness.
anyway, if 'you' Pray, please Pray for folks in the Middle East. don't 'wait
'til tomorrow.
k. p. collins
