I think is sad that the documentary could have been made with the state of conventional neuroscience in the 1980s. David needs PCT but he doesn’t know it. I tried emailing him a while ago but no response…
inadvertently–holding the Ctrl key instead of Shift,
evidently, when I meant to paste Rupert’s timestamp. (And of
course a conversation here intervened.) Trying again,
formatted as intended. The indentation now indicates where
the extended quote ends.
Rupert
Young (2016.01.30 13.00) –
How is the reorganization system organized?
The simplest mechanism that I can think of is just
like the mechanism of memory. From a paper in
preparation:
Memory is a recreated neural signal. Memories are
constructed locally at each synapse (Powers
2005.207–230, Kandel 2006, Rudy 2008). Short-term
memory is due to neurochemicals at a synapse
sustaining the neural signal. To establish a long-term
memory, repeated firing of the synapse alters gene
expression in the cell nucleus of the neuron,
affecting protein synthesis mediated by membrane
transport proteins, such as RNA. The growth and
maintenance of new branches and synaptic terminals,
which is the general means for amplifying a neural
signal, is also how memory is further strengthened and
made more persistent (Kandel 2006). Ascending
perceptual input signals and descending
output/reference signals evoke and strengthen memory
at synapses all along the neural pathways in which the
remembered perceptual signal originated. Interruption
of this process of ‘memory consolidation’ may prevent,
long-term memory formation. Conversely,
reconsolidation can change an established memory
(Loftus 1998, 1999).
Efforts have been made to localize certain kinds of
memory in certain parts of the brain. Given the
distributed, synapse-local character of memory, this
appears to reflect the location of systems for
recognizing and controlling the given kind of
perception, or of important contributors to it.
Establishing and strengthening memories appears also to
involve associative links between different parts of the
brain. For example, a memory that includes strong
emotion involves signals to and from the amygdala. These
links are the basis of Pavlovian conditioning (Mirolli
et al. 2009).
What happens at the synapses that construct a
comparator when the error output remains high? A simple
mechanism for reorganization could depend upon
neurochemical signals there and at distal synapses to
which the connected neurons project, signals that
stimulate random changes in connections and (more slowly)
branching. These changes might disturb control at other
connected comparators, spreading the scope of
reorganization, and something must oppose this and keep it
in check; and obviously there are many problems of detail.
But it is a simple scheme, it answers how reorganization
is targeted yet global (can happen anywhere needed), and
it rests on a fundamental question that must be addressed
regarding the nervous system and indeed regarding any
multicelled organism: what’s in it for the cell?
Reorganization is the mechanism of development in
embryology. We can’t posit a pre-existing reorganization
system for that. Unless we go to Sheldrake’s morphogenic
fields, maybe. IMO, if such fields are effective, they are
effective as constraints on a random process as above.
General-purpose problem-solving systems.
What I mean is Program-level and perhaps higher control
systems (or control loops, or elementary control systems,
if you like) within the hierarchy. They are
general-purpose systems because of the variety of
perceptual inputs that can be recognized as a complex
relationship perception that we think of as a
category/symbol perception. These serve as the
‘argument’ symbols of a program perception. They are
general-purpose control systems because the same sequence
or program logic can apply to a variety of specific
problems. A lot of human practical logic relies on analogy
and pattern recognition, not what computer programs are
best at. Could be that a principle-level perception maybe
something like “try it and see if it works” controls a
Sequence or Program whose initial inputs are matched, and
moves on to another if that one fails. It’s only when the
problem-solving routines fail and error persists that
reorganization kicks in.
Two questions beckon here, one about attention, the
other about what happens in MoL. In MoL, there are two
instigators of an internal conflict, and it appears to me
that the process brings both of them into a single
relationship perception. I found an interesting clue about
attention in Henry Yin’s beautiful paper about the basal
ganglia, in the first paragraph of subsection 8.1–of
course in context of the rest of the paper. I don’t see
any criterion as to what constitutes salience. One would
guess salience = error, e.g. in an animal error in systems
controlling for a quiet and familiar environment, in
systems controlling for food or prey if hungry, etc.
Qualia.
It's all well and good to say that the taste of lemon,
the color teal, the opening melody of the Franck violin
sonata in A, are all no more than rates of firing in my
nerves, or voltages in your robot. But I know I do not
experience rates of firing, and there is no way to
convince me that your robot experiences lemon, teal, or
Franck’s melody. This is the “But I’m not a solipsist,
really!” problem that stumped Bill; he asked more than
once if anyone knew a way out of it.
I accept that you do experience lemon, teal, and
Franck’s melody because I perceive you (imagine you) to be
a human like me, and at Principle and System Concept
levels I control perceptions of commonality of my
experience with that of fellow humans. Those perceptions
are repeatedly tested in the course of interactions every
day. Doubtless, there’s a lot of imagination involved, but
I am able to control perceptions that matter to me and get
along with living, and that’s the general criterion as to
the veridicality of the whole universe of constructed
perceptions, right down the the intensity at a taste bud
for sourness or at the cochlear hair cells that resonate
most strongly at 440 Hz.
The rates of firing of nerves at all levels support a
unitary experience of sitting in a comfortable
teal-colored chair sipping lemonade listening to the
Franck violin sonata in A. The experience arises from the
multitude of neural firings, but is none of them. (This is
a completely imagined perception, by the way–I’m not on
vacation in some warm place.)
(Bruce Nevin (2016.02.02.20:36 ET)
Not sure if you were looking for a response to anything, or
just adding some points.
Anyway, I watched this science program which may be of interest to
the list, The Brain with David Eagleman - 3. Who Is in Control?
()
Not sure if it’s available outside UK (though would be through VPN
such as ), though presenter is
American so it might be in US.
There are some interesting points in it some of which are relevant
to PCT consciousness and PCT in general.
7:50 Still seems to suggest a computational, anti-PCT approach where
says we estimate weight (etc) of coffee cup when lifting it.
9:30 onwards, subject (Ian Waterman) lost ability to manage body
automatically (learn?) and so has to make conscious effort for
everyday tasks such as walking. This would fit with the PCT idea
that consciousness is associated with systems that require
reorganisation.
12:30 onwards, shows an expert at cup stacking (cup stacking?,
what’s that all about?). It shows a brain scan of the expert and a
novice, and shows that the expert actually has very little brain
activity as opposed to the novice. This would fit with the idea that
systems which have improved/optimised the quality of control have
very little error, and, therefore, lack associated consciousness.
And the activity that is seen in the brain of the novice is error.
Eagleman goes on to describe the brain state of the expert as flow,
or being in the zone.
31:50 He also asks the question why aren’t we just unconscious
beings? He concludes that when we come across something unexpected
our unconscious mind is called into action. Although he says it is
the job of consciousness to make sense of the situation of the
violation of expectations, in PCT terms this would be error
introduced into the system as a result of the mis-match with the
reference “expectations”. This might suggest that any time there is
this error then the quality control system comes into play with
consciousness and reorganisation.
33:20 Mentions internal conflict, though talks about consciousness
making an “executive” decision, so sort of going up a level, but a
bit vague.
Regards,
Rupert
inadvertently–holding the Ctrl key instead of Shift,
evidently, when I meant to paste Rupert’s timestamp. (And of
course a conversation here intervened.) Trying again,
formatted as intended. The indentation now indicates where
the extended quote ends.
Rupert
Young (2016.01.30 13.00) –
How is the reorganization system organized?
The simplest mechanism that I can think of is just
like the mechanism of memory. From a paper in
preparation:
Memory is a recreated neural signal. Memories are
constructed locally at each synapse (Powers
2005.207–230, Kandel 2006, Rudy 2008). Short-term
memory is due to neurochemicals at a synapse
sustaining the neural signal. To establish a long-term
memory, repeated firing of the synapse alters gene
expression in the cell nucleus of the neuron,
affecting protein synthesis mediated by membrane
transport proteins, such as  RNA. The growth and
maintenance of new branches and synaptic terminals,
which is the general means for amplifying a neural
signal, is also how memory is further strengthened and
made more persistent (Kandel 2006). Ascending
perceptual input signals and descending
output/reference signals evoke and strengthen memory
at synapses all along the neural pathways in which the
remembered perceptual signal originated. Interruption
of this process of ‘memory consolidation’ may prevent,
long-term memory formation. Conversely,
reconsolidation can change an established memory
(Loftus 1998, 1999).Â
Efforts have been made to localize certain kinds of
memory in certain parts of the brain. Given the
distributed, synapse-local character of memory, this
appears to reflect the location of systems for
recognizing and controlling the given kind of
perception, or of important contributors to it.
Establishing and strengthening  memories appears also to
involve associative links between different parts of the
brain. For example, a memory that includes strong
emotion involves signals to and from the amygdala. These
links are the basis of Pavlovian conditioning (Mirolli
et al. 2009).Â
What happens at the synapses that construct a
comparator when the error output remains high? A simple
mechanism for reorganization could depend upon
neurochemical signals there and at distal synapses to
which the connected neurons project, signals that
stimulate random changes in connections and (more slowly)
branching. These changes might disturb control at other
connected comparators, spreading the scope of
reorganization, and something must oppose this and keep it
in check; and obviously there are many problems of detail.
But it is a simple scheme, it answers how reorganization
is targeted yet global (can happen anywhere needed), and
it rests on a fundamental question that must be addressed
regarding the nervous system and indeed regarding any
multicelled organism: what’s in it for the cell?
Reorganization is the mechanism of development in
embryology. We can’t posit a pre-existing reorganization
system for that. Unless we go to Sheldrake’s morphogenic
fields, maybe. IMO, if such fields are effective, they are
effective as constraints on a random process as above.
General-purpose problem-solving systems.
What I mean is Program-level and perhaps higher control
systems (or control loops, or elementary control systems,
if you like) within the hierarchy. They are
general-purpose systems because of the variety of
perceptual inputs that can be recognized as a complex
relationship perception that we think of as a
category/symbol perception. These serve as the
‘argument’ symbols of a program perception. They are
general-purpose control systems because the same sequence
or program logic can apply to a variety of specific
problems. A lot of human practical logic relies on analogy
and pattern recognition, not what computer programs are
best at. Could be that a principle-level perception maybe
something like “try it and see if it works” controls a
Sequence or Program whose initial inputs are matched, and
moves on to another if that one fails. It’s only when the
problem-solving routines fail and error persists that
reorganization kicks in.
Two questions beckon here, one about attention, the
other about what happens in MoL. In MoL, there are two
instigators of an internal conflict, and it appears to me
that the process brings both of them into a single
relationship perception. I found an interesting clue about
attention in Henry Yin’s beautiful paper about the basal
ganglia, in the first paragraph of subsection 8.1–of
course in context of the rest of the paper. I don’t see
any criterion as to what constitutes salience. One would
guess salience = error, e.g. in an animal error in systems
controlling for a quiet and familiar environment, in
systems controlling for food or prey if hungry, etc.
Qualia.
It's all well and good to say that the taste of lemon,
the color teal, the opening melody of the Franck violin
sonata in A, are all no more than rates of firing in my
nerves, or voltages in your robot. But I know I do not
experience rates of firing, and there is no way to
convince me that your robot experiences lemon, teal, or
Franck’s melody. This is the “But I’m not a solipsist,
really!” problem that stumped Bill; he asked more than
once if anyone knew a way out of it.Â
I accept that you do experience lemon, teal, and
Franck’s melody because I perceive you (imagine you) to be
a human like me, and at Principle and System Concept
levels I control perceptions of commonality of my
experience with that of fellow humans. Those perceptions
are repeatedly tested in the course of interactions every
day. Doubtless, there’s a lot of imagination involved, but
I am able to control perceptions that matter to me and get
along with living, and that’s the general criterion as to
the veridicality of the whole universe of constructed
perceptions, right down the the intensity at a taste bud
for sourness or at the cochlear hair cells that resonate
most strongly at 440 Hz.Â
The rates of firing of nerves at all levels support a
unitary experience of sitting in a comfortable
teal-colored chair sipping lemonade listening to the
Franck violin sonata in A. The experience arises from the
multitude of neural firings, but is none of them. (This is
a completely imagined perception, by the way–I’m not on
vacation in some warm place.)