GaryGaulin
Posts: 5385
Joined: Oct. 2012
|
The scientific method is actually very simple (unless of course when it interferes with the objectives of political activists who are promoting their ideology/beliefs).
First, one proposes a testable theory. This is the introduction to what I proposed (which political activists have been claiming they cannot understand a word of):
From the Theory of Intelligent Design (by Gary Gaulin)
| Quote | Introduction – Intelligence, Intelligent Cause
The theory of intelligent design holds that certain features of the universe and of living things are best explained by an intelligent cause, a nonrandom force guided self-assembly process whereby an intelligent entity is emergent from another intelligent entity in levels of increasingly complex organization producing self-similar entities systematically in their own image, likeness. As in a fractal, multiple designs are produced by an algorithm producing emergent fractal-similar designs at the next size scale (atom -> molecule -> cell -> multicellular).
Large arrows show this emergent causative pathway from behavior of matter (a Behavioral Cause) and intelligence from intelligence (an Intelligent Cause). The last arrow to Multicellular Intelligence indicates a predicted sudden event scientifically witnessed by the fossil record known as the Cambrian Explosion which will be covered in a section of its own. Shown in the lower half of the illustration is a simplified block-flow diagram of the same cognitive/intelligence system that is at each level of the progression shown above it.
Successful designs remain in the biosphere’s interconnected collective (RNA/DNA) memory to help keep going the billions year old cycle of life. We are the result of a molecular learning process that keeps itself going through time by replicating previous contents of genetic memory along with good (better than random) guesses what may work better in the next replication, children. Resulting cladogram shows a progression of adapting designs evidenced by the fossil record where never once was there not a predecessor of similar design (which can at times lead to entirely new function) present in memory for the descendant design to have come from.
Behavior of matter is produced by electromagnetic force created atomic bonds and intermolecular interactions (covalent, polar covalent, van der Waals polar force, ionic, metallic, hydrogen) and follows the “laws of physics”. This is covered by Atomic Theory, which describes the atoms in the model’s particle system environment. Behavior of matter can only respond to exteroceptive stimuli one way, such as bonding with another molecule or not, therefore has two of four requirements for intelligence (but does not by itself qualify as intelligence). It is not possible to rule out intelligence at this behavior level, but with no scientific evidence existing for this the behavior of matter is assumed to not require intelligence to produce intelligence (origin of intelligent life).
As in physics algorithms, there is a Time Step. Each particle/entity in the virtual environment is something to CONTROL that is moved a small amount each time according to surrounding forces/conditions acting upon it. What response to take in a given condition is stored in a memory that is addressed by sensory that produces a unique action response for each environmental situation the particle can encounter. Memory can be here thought of as a binary digital RAM or analog neural network that has in it a truth table to produce the behavior for each kind of atom.
For modeling purposes where a “Behavior” produces an emergent intelligence the behavior that created it can be thought of as being “all knowing” in the sense that the behavior is inherent, does not have to learn its responses. A computer model then starts off with this behavior already in RAM or ROM and has no GUESS or CONFIDENCE included in the algorithm, as does intelligence. Memory contents then never changes, in this model only a GUESS writes data to MEMORY.
Intelligence is not a lifeless mass responding to physical forces by drifting downstream, intelligence can do such things as decide to swim upstream instead. In a complete physics model where all matter obeys physical laws, intelligence is an emergent deterministic internal force inside (then living) things that “at will” becomes an outside force causing change in motion to matter around it.
Intelligent behavior results in an entity with the ability to self-learn. The flowchart becomes:
Intelligence is here operationally defined by how it works: Intelligence is an autonomous sensory-feedback (confidence) guided sensory addressed memory system that through trial-and-error learns new successful actions to be taken in response to environmental conditions. In addition to something to CONTROL and MEMORY there must be one or more CONFIDENCE levels gauging failure or success of its motor actions towards reaching the goal and a way to GUESS motor actions when a successful response does not yet exist. A good-guess is based upon existing knowledge. A random-guess is the last resort and only has to be "random" to the intelligence. For example where one must produce "random numbers" for another to guess they may use their phone number, which to them is not a random string of numbers, but to the other person who does not know their phone number it is a random string of numbers. What is most important for something to be "random" is that the intelligence perceives it as such.
Confidence gauges whether it is getting closer to its goal or not. In0/Out0 is typically a 4 state (count of 0-3) analog signal that recalls confidence level which increases each time the action worked, decreases when it failed. Upon reaching 0 a Guess is taken. In a most simple chemotaxis system Guess and Motor are combined, changing motor direction produces a tumble/guess where to go next. Only a single on/off memory location is then required to control its motor(s).
We know when we need to take a guess, or have an action response we are confident will work. To a newborn baby, almost everything is a new experience. No memory at all of what to do is then sensed by Out0 being 0 which then causes a guess to be taken. Responses that work are stored with increasing confidence, for as long as it keeps working, but confidence level does not need to increase past 3 for a good model. In bacteria the interoceptive sensors would simply be metabolic pathway molecules reporting motor condition back to the sensory end of the system to provide time delay that through Confidence being restored by that action switches motor back to swimming after tumble has been completed.
There should always be an easily recognizable circuit where each part works with others in a certain way. This includes motors/muscles where there are expected to be two connections to the memory/brain. The input connects to the data action outputs of a Random Access Memory controlling it. The output is a sensory feedback signal to RAM addressing that adds (usually subconscious) awareness of the muscle action. This sensory output can be from other sensors not directly connected to, such as touch sensors on skin that “feel” muscles moving or eye sensing travel direction. Without at least indirect sensory feedback of motor actions addressing RAM the system has no way to know whether the motor has in turn produced the expected action, or not.
Although not a circuit requirement (as in the four above) there should be the production of regular detectable synchronized cycles, as the algorithm/system keeps repeating the one thought at a time process. Where these cycles are no longer present then the intelligence is nonfunctional.
Where a system is missing one or more requirements we have a system which may appear to be intelligent but only qualifies as a protointelligent behavior. This is true where the sensor(s) connect directly to the motors in a way that keeps the system on course but does not learn how to control itself. There must be a memory system between sensors and motors being controlled. An example of cellular intelligence so simple it is almost like cheating is the E.coli chemotaxis system where chemoreceptors address a single memory location that increases or decreases according to the amount of chemical being sensed. When going the wrong way it tumbles to try another direction.
Being self-learning, given enough time, intelligence will produce the next emergent level of intelligence when it learns how to achieve it. Large numbers of rudimentary intelligences are predicted to have a tendency to spontaneously produce easily detectable and measurable emergent intelligence at the next level. No extra computer code is needed, entities learn how to on their own. Demonstrating this intelligent cause/causation would require many intelligent entities with rudimentary intelligence which in time self-assemble (at higher complexity is also called self-organize) to produce an emergent intelligence, much the same way a molecular genome produces a living cell, or living cells produced us.
Because of atom by atom computing being too memory intensive to computer model a large volume of matter, macromolecules can be approximated in the next level above atoms where it is no longer an atom by atom particle system physics problem, there are instead (combination of atoms) molecules each with unique behavior that can be summed up as a unique single entity (in the same way as classic Argon particle system describes all argon atoms but instead as molecular binding/reaction site dynamics of all its atoms combined). Macromolecules next self-assemble to form cells, which likewise can also be modeled at the next level by just modeling the cellular detail where a muscle cells are a regular spherical shape that shortens in length during contraction. And that can next be taken another step as is demonstrated by the Intelligence Design Lab where the behavior of many sensors and neurons/synapse is summed up to form the brain that connects to muscle cells that control muscles that can also send dirt particles flying or (as is the case in the Lab) simply propel it on a flat surface without disturbing anything. It is not necessary to start at the atomic level we only need to properly sum up one level to the next to produce a representative model. We must also keep in mind that with a computer it is easy to model a perfect memory that never forgets. In some ways the intelligence may be too perfect to be biologically possible, but at least it is easy to achieve that perfection, here made possible by the reliability of computer RAM to hold data
Reciprocal causation brings all of our complex intelligence related behaviors back to the behavior of matter where it's basic physics, that begins with common particle systems such as for modeling Argon (and other) atoms on a parallel processing GPU. Here all argon atoms are alike, as well as helium, carbon and all of the other elements and their isotopes. This regular atomic structure becomes the fundamental starting point for models like this where atoms combine to form molecules, molecules combine to form cells, and cells combine to produce multicellular organisms.
The reciprocal causation pathway goes from one level to the next but not directly from brain to matter, because just thinking about digging a hole in the ground does not propel virtual soil particles through air. There is first a neural connection between brain and muscle cells, then a neural feedback connection from the muscle cells back to the brain. After the muscle cells successfully receive this signal to contract it next has to convert that signal into a pulling "force" by powering its internal motor protein molecules, and like any other motor it needs energy to make it move on command and where that has run out there will be no digging either because it will then be too weak to move. There is no force applied to the digging limb until the molecular level systems have actually produced muscle force, to apply force to the limb accelerating soil particles into the air to dig a hole in an otherwise perfectly flat environment.
For sake of theory, “consciousness” is considered to be in addition to intelligence, otherwise the most rudimentary forms of intelligence and even simple algorithm generated computer models of intelligent processes would have to be expected to be conscious of their existing inside of a personal computer. It is not possible to rule-out electronic or algorithmic consciousness existing, therefore even though it is not expected to exist in a computer model it is still possible that any functioning intelligence system is somehow conscious of their existence. In either case, consciousness is not a requirement for intelligence, and here must be considered to be in addition to intelligence. |
The scientific method now requires critics in this forum to test the above (as I did in sections that follow this Introduction) and show scientific evidence that its predictions are impossible.
Claiming they cannot understand any of it is just another political tactic to get them out of having to follow the scientific method. A qualified science teacher knows this, and US education laws and ethics forbid them from siding with political activists at Panda's Thumb who cannot even follow the scientific method.
This theory is legal to teach in any US public school classroom. Excuses found in this forum are not legal to teach as science, they are now examples of how the scientific method is routinely ignored/distorted by political activists who have no legal right to force their unscientific opinions into public school classrooms.
The trashing of the scientific method does not happen in science education forums for serious science teachers. This only happens in forums where political activists are trying to control what is taught in the science classroom.
--------------
The theory of intelligent design holds that certain features of the universe and of living things are best explained by an intelligent cause, not an undirected process such as natural selection.
|
