Would you like to make this site your homepage? It's fast and easy...
Yes, Please make this my home page!
MIND AND BRAIN
Information Processing
NOTE: by "matter" we understand a metalanguage shortcut to
the physical concept of discretized, integrated high density
areas of Gravity/Inertia Field. Inverted commas stress its
contrast to the common "materialistic" usage of the term,
which denotes a meaningless abstraction void of physical
sense and absent from physical vocabulary.
Information Processing may be schematized with help of three
entities: Processor, Input and Output. Input, a set of
Receptor devices, feeds the Processor which generates Output,
a set of action Trigger devices, fed partially back to the
Input as the cybernetic feed-back..
Within this schema and in an arbitrary - not preprogrammed
or memorized - situation, an action has to start by relating
Input with Output, i.e. by establishing all Receptor/Trigger
couples, in order to select those pertinent to the particular
act.
Now, "material" Processor cannot exceed 2*10e47 operations
per gram and second (Bremermann's limit[*] (BL)). Enormous
as it may seem, it turns out to be negligible. 2*10e47 is a
small number when compared with processing rates that one
would need to carry out certain search processes. For example,
Minsky gives the
number of all possible move sequences in chess as about
10**120.
Conservatively estimated Input and Output of an advanced
organism contains 10e6 devices each. Supposing simplest 2
state devices, relating Input with Output requires
R=2**(2*10e6) operations. Its Log=602060, indicates R to be
1 with 602060 zeros.
On the other hand BL for a kg (roughly the mass of the brain)
is about 2*1e50 or 1 with 50 zeros.
Dividing R by BL we get about 1 with 602010 zeros seconds
as the lowest limit of time required by "material" brain
to relate conservatively estimated input and output of an
organism. The practical "synaptic" limit[**] is by about 34
orders longer, because the neuron structure of the brain
is much courser than the atomic granularity implied by
the BL.
The real neuronic brain would need over 1 with 90 zeros
seconds to Relate Input with Output.
Thus, Morphogenesis, behavior of living Organisms, Human
complex scientific and social Projects clearly require some
additional agent different and incomparably faster than the
raw "corpuscular Matter", let alone the real neural brain.
MIND, BRAIN AND THEIR INTERFACE
Mind
The chapter STRUCTURE OF MIND postulates
a Structure homomorphic with the "Three Worlds" of Ref,
encompassing three domains: Sensorium, Imagery and Symbolics,
accounting for all Mind's functions, such as perception,
imagination, cognition, symbolizing, inference, reflection
and understanding
Brain
Brain's structure pertinent to information processing
encompasses specialized cells called Neurons and their
interfaces called Synapses. A neuron consists roughly of
several dendrites, of a cell body and an axon. Dendrites are
receptors of electric impulses which travel through them to
the cell body and make it "fire", sending electric impulse
further down the axon till axon terminal.
Axon terminal lies close to dendrites of neighboring neurons,
separated from them by synapses.
A synapse consists of:
1. a presynaptic ending that contains neurotransmitters,
2. a postsynaptic ending that contains receptor sites for
neurotransmitters and,
3. a synaptic cleft or space between 1 and 2.
Neurotransmitter is a chemical released by impulse reaching
the presynaptic ending at the axon terminal and travelling
through the cleft to the postsynaptic ending in vicinity of
a neighboring neuron's dendrite. Neurotransmitter reaches the
dendrite and triggers an electric impulse that travels down
to the cell body of the neighboring neuron which in turn may
fire.
This network structure of neurons-vertices and synapses-edges
allows millions of neurons to "talk" with each other.
However, it would imply so far chain reactions in which a
single impulse will fire the whole network rather than create
any local pattern.
That's why we have two sorts of neurotransmitters: activating
and inhibiting. The receiving neuron "counts the votes" and
decides to fire in case of sufficient "activating majority"
or else to stay idle.
While freely stretching an arm, the triceps contracting
neurons will get activated, while those of the biceps
inhibited in order to relax the opposite bending force. In
controlled stretching both activations and inhibitions will
be precisely measured out to achieve the desired action.
Neurons are connected so that when they fire together, they
"wire" together, to form stronger synaptic connections. This
is the Brain's dialectic counterpart of Mind's Learning
faculty defined above.
Brain-Mind interface
We have seen that "material", discrete, corpuscular processor
of the size of human brain would need about 1e600000 seconds
to process the Information involved by a simplest action of
an advanced organism. This is a hypothetical limit of what
the evolution may reach in another billion years.
The practical limit of brain's neural network by itself is
about 1e900000 seconds.
Consequently, the "material", discrete Brain, including its
neural network structure cannot account alone for human and
living organisms' behavior. We must look for some continuous
aspect of Brain as Interface supporting covariant maps to
such Mind's functions as Reflection and Inference.
Field suggests itself as the most fundamental Continuum.
Brain's neural network clearly "generates" electromagnetic
field and many Neurologists see it as Mind itself, or at
least as Brain's interface to some other, as yet undefined,
but clearly observable, thus physical "Mind". This question
exceeds the scope of our essay. We stop at the assertion
that human behavior posits a polar continuous counterpart
of the discrete Brain and that the EM field satisfies this
requirement.
Another, more complex and rather exciting theory proposes
Field generated hologram as interface to Mind, or Mind
itself. Hologram has many striking features, starting with
Totality being in each Part. Some Neurologists claim that
memories are not stored in any specific location in the
brain, but appear to be distributed over the whole brain,
much as any portion of a holographic image is distributed
over the whole structure. Suppressing a part does not change
the anything in the image, else than making it fainter.
The dynamic TimeSpace of the holographic image is homomorphic
with our Imagery structure and could eventually support its
patterns. Again, like in the case of Field, the question of
hologram being "Mind", or an interface to some as yet
undefined "Mind", exceeds the scope of our essay. We shall
stop at saying that hologram may satisfy the requirement of
a polar continuous counterpart of the discrete Brain.
We shall just mention here a rather poetic theory of
pan-hologram based on "holographic paradigm". A small group
of researchers sees "real universe" as a holographic blur of
frequencies and the brain also as a hologram which selects
some of this blur's frequencies and transforms them into
sensory perceptions. In this "ocean of consciousness" we
would be the "weary rivers that wind somewhere safe to sea".
There is a striking affinity of this paradigm with religions
of the East which have long upheld, the physical world is
Maya, an illusion, and although we may think we are physical
existants moving through a physical world, this too is an
illusion. Most, if not all protagonists of the paradigm are
attired by eastern beliefs, include the paranormal in their
inquiries and some report about shamans making instantly
disappear and reappear entire groves of trees.
Let's note that so far we have justified the need of some
covariant continuum with consideration of the processing
speed. Another necessary justification is provided by the
Morphogenesis. Whether we consider it from the point of
view of Haken's Synergetics or Thom's Catastrophes, it
consists in ordering a continuum.
Final note
Brain and Mind structures described above are Models
satisfying all Model requirements, including falsifiability.
Brain-Mind Interface, perhaps the most exciting area of
inquiry, produced so far only Theories, i.e. deductively
consistent structures factual falsification.
References:
Ref: Popper and Eccles "The Self and Its Brain"
(Springer International) Page 327, Fig E5-7, etc.
[*]BREMERMANN'S LIMIT
No "material" system whether artificial or living can compute
more than 2 x 10e47 bits per second and per gram of its mass.
This limit refers to a self-contained system where the power
supply is included in the "matter" and where "computation" is
defined as the transmission of information over one or more
channels within the system. Recognizing that computation
requires energy for changing physical markers and for
recognizing such changes in subsequent steps, the limit is
obtained from Einstein's relation between energy and matter,
E=mc2, and Heisenberg's uncertainty principle, which specifies
the inaccuracy or noise in the measurement of energy.
Current computers as well as the brain are far from
approximating this limit because they don't utilize atomic
events for the storage and transmission of information.
Ref: Bremermann's paper
OPTIMIZATION THROUGH EVOLUTION AND RECOMBINATION*
H. J. Bremermann
Part I. Limitations On Data Processing Arising From Quantum
Theory
Part II. Evolution Processes and Optimization
Part I is pertinent to the derivation of the limit and to
its implications for the Mind-Brain issue.
[**] BRAIN'S COMPUTATIONAL POWER
The total computational power of the brain is limited
by several factors, including the ability to propagate
nerve impulses from one place in the brain to another.
Propagating a nerve impulse a distance of 1 millimeter
requires about 5 x 10-15 joules. Because the total
energy dissipated by the brain is about 10 watts,
this means nerve impulses can collectively travel at
most 2 x 10e15 millimeters per second. By estimating
the distance between synapses we can in turn estimate
how many synapse operations per second the brain can do.
This estimate is only slightly smaller than one based
on multiplying the estimated number of synapses by the
average firing rate, and two orders of magnitude
greater than one based on functional estimates of
retinal computational power. It seems reasonable
to conclude that the human brain has a raw computational
power between 10e13 and 10e16 operations per second.
