X-Message-Number: 8324
From: Stephen Bogner <>
Subject: RE: Consciousness Defined
Date: Tue, 17 Jun 1997 12:10:42 -0600

> From: Hara Ra <>
>

> -------------------------
> Definition: Consciousness
>
> A system is conscious if:
>
* 1. It is capable of lingusitic communication (it can talk)
*
In some respects, the issue of communication (or equivalently,
observable behavior) has more to do with our ability to decide whether
or not an entity is conscious than it does with whether or not the
entity is actually conscious.  No doubt an uncommunicative consciousness
would be of a rather primitive form, and if it had no observable
behavior then by definition it cannot be measured in any case, so we can
safely ignore this (hypothetical) case.

> 	2. It can discuss its own existence (it knows it exists)
>
* -------------------------

An Alternate Definition (drawn somewhat by my work on conceptual mapping
for autonomous systems):

A system is conscious if (and only if) it has all of the following 4
features:
1. it has the capacity to sense it's environment (it can perform "Data
Acquisition").
2. It has a capacity for  "situational awareness" (it can perform "Data
Interpretation")
3. It has the capacity for "anticipation" (it can perform "Data
Extrapolation and Inference")
4. It is capable of "behavior" (it can act upon the "Reality State" from
within it)

Terminology can be a frustrating source of confusion.  I will attempt to
supply clear definitions of my terminology so interested parties can
match the definitions to their own terminology without loss or
distortion of the information (meaning and truth) being conveyed.

My terminology is defined as follows:

1. Reality State - the observer independent state of reality as defined
by the Critical Realist philosophical point of view:  "There is a
physical world which exists objectively and independently of our minds. 
Our sense data can be used to examine the world itself."
2. Sensor - a physical system incorporating a measuring transduction
structure and a transmission structure to output results of the system.
3. Sensor Capabilities - the limiting characteristics of the sensor
system.  In general these are technical limitations resulting from the
imperfect state of the art and the laws of science.  These limitations
may also be intentional so that undesired behavior can be avoided.
4. Data Acquisition - the activity of obtaining sense data, and
transforming that data into quantitative information about the measured
parameter.
5. Sensor Report - a data structure containing, directly or indirectly,
quantitative information on the state of measured parameters.  In
practice, the Sensor Report takes the form of a mapping of the
independent variables of the Physical Model to their measured or
inferred values.
6. Intelligence - a mechanism characterized by its intrinsic cognitive
ability to resolve problems in complex reasoning by manipulating
abstractions in an algorithmic and/or intuitive manner.
7. Algorithmic Intellect - a processing mechanism that uses mathematical
laws and algorithmic procedures to interpret or reduce data.  In
practice, the Algorithmic Intellect creates quantitative descriptions of
the dependent variables using the independent variables from the Sensor
Report, according to the deterministic equations of the Physical Model.
8. Intuitive Intellect - a processing mechanism that rejects the
universal validity of the logical law known as the "Law of the Excluded
Middle" (a fundamental law of the Algorithmic Intellect), and allows the
probabilistic resolution of that class of propositions that are
undecidable in the absence of adequate proof.  In other words, this is
the intellect that resolves ambiguities and generates creative original
solutions by supplying theories and hypothesis (which may be tested by
other means) when complete information is unavailable.
9. Reality Model - a conceptual map of the idea of reality itself.  In
particular, this conceptual map will contain a model of the physical (ie
sensible or measurable) aspect (a model of being) and a model of the
operational (ie. Behavioral ) aspect (a model of becoming).
10. Physical Model - a deterministic set of relationships between
measurable parameters based upon physical science.  In practice, this
model will take the form of a more or less fully defined set of
mathematical equations describing the relationships between measurable
or inferrible parameters.
11. Operational Model - a probabilistic set of relationships between
transformational entities based upon social and operational science.  In
effect it is a probabilistic model of the steps (operations) required to
transform one Reality State into another.
12. Data Interpretation - the activity that transforms Sensor Reports
into Situational Awareness, through the application of Intelligence,
within the constraints of the Reality Model.  In practice, this involves
first solving the equations of the Physical Model so that the Reality
State of being becomes more or less fully defined, and then assigning
qualitative and contextual significance to that Reality State of being
(ie. The solution matrix of the Physical Model) such that the driving
factors in the Operational Model become defined.
13. Situational Awareness - a data structure containing quantitative,
qualitative, and contextual information on the independent and dependent
parameters of the Physical Model, with tentative evaluations of the
undecidable propositions resolved by the Intuitive Intellect.  All
driving factors (ie. Independent parameters) of the Operational Model
are declared or inferred in the Situational Awareness.
14. Anticipatory Control - the activity that generates a goal oriented
Instruction Set on the basis of an intelligent evaluation of the
situation.  This evaluation is constrained by the Reality Model, which
identifies goals, and the set of operations needed to achieve them, on
the basis of its Operational Model.
15. Instruction Set - a data structure that contains a quantified
description of intentions that are ultimately reducible to sets of
controller impulses to mechanisms.
16. Entity - a coherent assembly of physical systems that acts upon the
Reality State from within it.
17. Entity Capabilities - the limiting characteristics of the Entity. 
In general these are technical limitations which result from the
imperfect state of the art and from the laws of science.  These
limitations may also be intentional so that undesired behavior is
avoided.
18. Mission Execution - the activity of engaging the entity to perform
tasks that alter the Reality State so that it corresponds to the goal
intended by the Instruction Set.


As a simple test, we can try to assign a very (very) minor level of
consciousness to a simple feedback mechanism such as an automatic
thermostat:
1. It senses temperature.
2. It forms a situational awareness (from its extremely simple
"thermostat level Reality Model" and "thermostat level Algorithmic
Intellect") about whether or not the temperature is below its set-point.
3. It decides to switch on (or not) based upon that temperature,
4. It demonstrates "observable behavior" by turning on the furnace or
the air conditioner.

Some will doubtless argue that at 3 there is no "anticipation"; however
I would argue that the anticipation in inherent in the simple
deterministic Reality Model of such a system.  Others will argue at 3
that there is no opportunity for "wrong" choices, and so there is no
consciousness because the outcome is pre-determined.  I would argue
firstly that this supposes that an absence of determinism is a
pre-requisite for consciousness and there is no evidence for such an
assumption (for example, evidence that any system in the human organism
functions in a non-deterministic manner), and secondly that the
appearance of strict pre-determination is an illusion based upon the
macroscopic scale of the entity - if we consider a "nano scale
thermostat" then the possiblity of "wrong" choices becomes more
apparent.

My position is that the simple "feedback loop" discussed here represents
a possible "quantum" of conciousness, and that the high level
conciousness demonstrated by advanced entities (such as ourselves) is an
emergent "observable" created by the synergy of countless millions of
such quantums.  The implication of this view for cryonics is that the
information concept of identity could theoretically be valid if it can
accurately simulate the function and synergy of these countless
quantums, many of which loop through the environment as well as the
entity, but that it is difficult to see how this might be done in a
practical sense without duplicating the entity physically at the scale
at which these quantums occur.  Hence, I am inclined toward the concept
of reanimation via physical nanoscale repair, which I would consider
both necessary and sufficient to re-establish both consciousness and
identity.


Steve.

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=8324