Energy transfer "into" nervous system

From Tom Bourbon [931203.0919]

[From Bill Powers (931202.1715 MST)]

Too much to comment on right now, but this is worth a few words:

Martin Taylor (931202.1030) --

Maybe the analogy is flawed, but you haven't hit the flaw, and
I don't see one yet. Do you think light quanta don't carry
energy? The whole business of receptor chemistry is to use that
energy to modulate a greater flow (amplify it), and so on up
the line through the nervous system. Same applies for acoustic
energy, or its analogue, the pressure waves corresponding to
touch. The energy cost you are talking about is that of the
amplifier.

I raised this point some time ago in connection with the
entropy/information metaphor and got nowhere. Let's have another
brief round.

My contention is that neural signals do not carry energy from one
place to another, nor do sensory receptors accept energy and pass
it on into the nervous system. This is probably a fairly
important assertion in relation to certain theoretical
structures.

Huzzah! It *is* an important assertion. Receptors do not convert, transform,
or transduce one form of energy into another. Neither do they pass energy,
transmit energy, or communicate energy into the nervous system. Nor do
neurons carry, transmit, or convey energy from point to point within
the nervous system. The operations are all local and they all depend on
momentary perturbations of high-energy stores which are established and
maintained by the expenditure of large amponts of energy -- that was your
point in the remainder of your post.

Perhaps this round will progress beyond the previous one.

Until later,

Tom

[Martin Taylor 931203 10:45]
(Bill Powers 931202.1715)

On the relation between self-organized structures and other
negative feedback systems:

Maybe the analogy is flawed, but you haven't hit the flaw, and
I don't see one yet. Do you think light quanta don't carry
energy? The whole business of receptor chemistry is to use that
energy to modulate a greater flow (amplify it), and so on up
the line through the nervous system. Same applies for acoustic
energy, or its analogue, the pressure waves corresponding to

touch. The energy cost you are talking about is that of the
amplifier.

I raised this point some time ago in connection with the
entropy/information metaphor and got nowhere. Let's have another
brief round.

It has to be brief. I smell tarbaby again. I don't have time or
inclination to pick it up until I have the wherewithal for a good wash.

My contention is that neural signals do not carry energy from one
place to another, nor do sensory receptors accept energy and pass
it on into the nervous system. This is probably a fairly
important assertion in relation to certain theoretical
structures.

They don't carry energy to power the neurons, no. But energy does pass,
in a modulated fashion. It is the modulation that matters, and you
can't have that without having the energy flow itself.

What is an amplifier? It isn't just something that increases the energy
coming from a power source. It is something that modulates a larger energy
flow at its output than the energy flow at its input. The patterning of
the input has to be recoverable from the patterning of the output, and
in the case of a perfect amplifier, this recovery is perfect, though
the actual form of the patterning may be vastly different. There is
no loss of information when a signal is perfectly amplified (and no gain,
either). (Remember, information is always "about" something. In this case,
it is "about" the original pattener of the lower energy flow--what we often
call the modulation source). The important thing about an amplifier is
that it reduces the vulnerability of the information to disturbance in
an uncontrolled environment, by increasing the energy in the signal
degrees of freedom as compared to the energy in other environmental
degrees of freedom.

Every stage in a control loop is itself uncontrolled. Every stage of
the nervous system has amplifiers, and if you go a little deeper, the
primary business of those amplifers is to compensate for transmission
losses at the synaptic junctions and elsewhere in the neural mechanism.

A literal -- i.e., physical -- analysis of systems like these
shows that the concept of a "power amplifier" is only a metaphor.
All signal-handling systems of this kind dissipate energy at
every point in their operations. This precludes analyzing signal
transmissions in this kind of system in terms of any literal
application of physical laws concerning energy transfer.

All amplifiers, of whatever kind, dissipate energy at every point in
their operations. It's a necessary aspect of being in a physical
universe. Neural systems are in that universe. They dissipate energy.
How are they exempt from "any literal application of physical laws
concerning energy trasnfer?"

When a sensory receptor fires, it can do so only because
metabolic sources in the organism have pumped it full of energy,
which the incoming stimulus then causes to be dumped. This
process causes a chain reaction, in which the input part of a
pumped-up axon is caused to dump its stored energy, which
triggers the dumping of energy from the next part, and so on down
the fiber to the following synapse. To maintain repeated firings,
metabolic sources must continually run their energy pumps,
resupplying the energy dissipated by succeeding pulses. Continual
stimulation of sensory nerves causes a continual loss of energy
from the nervous system.

Yes, like that. Quite ordinary physical amplifiers.

In an equilibrium system, the energy is partitioned equally (within
a Gaussian distribution) among all degrees of freedom in the system.
Entropy is maximum. In a closed system, you can't reduce this entropy.
You can't have "signals." What, after all, IS a signal? It is a difference
between the energy in some particular degrees of freedom and the majority
(which are roughly equipartitioned). It is a reduction in the entropy
of the system as compared with the entropy that would be found if the
energy were partitioned equitably. The greater the difference between
the energy per degree of freedom of the heat in the system and the
energy in the "signal" degrees of freedom, the more precisely can
the signal be specified. You have to have a non-equilibrium system
in order to have a signal, and that means a dissipative energy flow.
An amplifier takes the signal in a weak energy flow, and imposes it on
a stronger one.

Back to light quanta. The important point about them, for vision, is
that they don't arrive randomly in space and time. Each one brings
some energy, but if they did come randomly, all we would see would be
a white blur. It is the fact that from some directions and at some times
they come faster than from other directions and at other times that
allows us to see. This is "amplitude modulation," in both space and
time, of a relatively weak energy flow. It's the modulation that has
to be amplified. And in the end, it is the integrity of the signal,
in the sense of the degree to which one could recover the original
modulation, that determines the operation of the entire system, whether
it be open-loop amplification or control.

All this becomes important when we come back to deal with the informational
structure of control systems, but as I said, I don't want to get back
into that now, particularly with Rick in pit-bull mode. Right now, it
is only an interesting continuation of the chase after the red herring
that was dragged across the trail a few days ago.

By the way, you said a few weeks ago that you didn't like red herrings
--a.k.a. kippers. I suspect that this thread of discussion has a similar
scent. Personally, I like both kinds of kipper, but only if carefully
cooked and not overdone.

Martin

[From Bill Powers (931202.1715 MST)]

Too much to comment on right now, but this is worth a few words:

Martin Taylor (931202.1030) --

Maybe the analogy is flawed, but you haven't hit the flaw, and
I don't see one yet. Do you think light quanta don't carry
energy? The whole business of receptor chemistry is to use that
energy to modulate a greater flow (amplify it), and so on up
the line through the nervous system. Same applies for acoustic
energy, or its analogue, the pressure waves corresponding to
touch. The energy cost you are talking about is that of the
amplifier.

I raised this point some time ago in connection with the
entropy/information metaphor and got nowhere. Let's have another
brief round.

My contention is that neural signals do not carry energy from one
place to another, nor do sensory receptors accept energy and pass
it on into the nervous system. This is probably a fairly
important assertion in relation to certain theoretical
structures.

When a sensory receptor fires, it can do so only because
metabolic sources in the organism have pumped it full of energy,
which the incoming stimulus then causes to be dumped. This
process causes a chain reaction, in which the input part of a
pumped-up axon is caused to dump its stored energy, which
triggers the dumping of energy from the next part, and so on down
the fiber to the following synapse. To maintain repeated firings,
metabolic sources must continually run their energy pumps,
resupplying the energy dissipated by succeeding pulses. Continual
stimulation of sensory nerves causes a continual loss of energy
from the nervous system.

A literal -- i.e., physical -- analysis of systems like these
shows that the concept of a "power amplifier" is only a metaphor.
All signal-handling systems of this kind dissipate energy at
every point in their operations. This precludes analyzing signal
transmissions in this kind of system in terms of any literal
application of physical laws concerning energy transfer.

ยทยทยท

----------------------------------------------------------------
Best,

Bill P.