bascule Posted January 29, 2009 Posted January 29, 2009 Okay Sione, clearly we aren't getting anywhere. Let's start over. What do you make of this argument? Consciousness results from brain activity The brain is a physical system The behavior of physical systems can be modeled using the laws of physics Therefore, it should be possible to model a physical system which results in consciousness Can you perhaps point out which of these three statements you agree/disagree with before giving me your thoughts on the conclusion?
Sione Posted January 29, 2009 Posted January 29, 2009 (edited) Okay Sione' date=' clearly we aren't getting anywhere. Let's start over. [/quote'] Sure, and don't be shy. What do you make of this argument? 1. Consciousness results from brain activity 2. The brain is a physical system 3. The behavior of physical systems can be modeled using the laws of physics Therefore' date=' it should be possible to model a physical system which results in consciousness [/quote'] 2. The brain is a physical system Physical system can be classical and quantum. It can be deterministic, chaotic or random. You think it is not random and perhaps that there is no such thing as random, while I suggest everything is random and truly non-deterministic, how wonderful! This should not be too hard to resolve, after all we are talking about everyday life and these positions are diametrically opposite, someone must be completely wrong! I bet $25 it is not me. 3. The behavior of physical systems can be modeled using the laws of physics I like the idea, but how do you model quantum uncertainty? Even if it is deterministic and in the scope of classical physics, how do you model something you can not define? I suggest thought is not something you can capture in a freeze-frame and draw it as a geometrical representation, but something that is inherently dynamic in nature and connected to time variable. So to know the mechanics (electronics) of it is far away from knowing the structure in which it occurs. Brain is like a computer and thoughts are what you see on screen, only that brain output feeds recursively back into brain again, as input + current feelings, and so on. If we could differentiate input from output, only then we can hope to be able to catch it's head and tail, only then we could begin to understand mechanics of it. We simply do not know how to recognize the "information" (thoughts, emotions, consciousness), we do not know the "form" in which it exists and so we can not really define it, therefore we can not simulate it. If we can achieve to "illustrate thoughts", to project dreams as some motion picture, that would be good enough definition I think, can we do anything like that? Can we collect human memory in some meaningful way or reproduce some "photos" from photographic memory? Do we know the location of memory, can we remove some parts of person's memory? Have you considered the peculiarity of brain's holographic qualities, like when you remove quarter of the brain and person goes on to live like nothing happened? You can simulate structure as much as you can define them and you say you can define them well, that's great. Imagine now emotion or consciousness somehow emerged inside such computation, how would you know? We would not know how to test it without definition and we would not be able to recognize it. What would you look for? If you can not name it, you are blind to it. It could be right on your nose and you would not see it. My turn... 1.) Do you think chemistry is deterministic? 2.) Do you think the "information" is within chemical structure or within electromagnetic fields? 3.) Do following life forms have emotions or instincts? a. Virus b. Bacteria c. Squid d. Bird e. Dog f. Dolphin Edited January 29, 2009 by Sione
npts2020 Posted January 29, 2009 Posted January 29, 2009 Sione; The rest of the last post that you cut off in your response explains in as perfect wording as I can think of why I think even chemistry is deterministic. Just because nobody is smart enough to figure it out doesn't make it not true. IMO in order for the world to be non-deterministic you would need to provide an example of something that exists that violates some underlying physical principle.
Sione Posted January 30, 2009 Posted January 30, 2009 (edited) Sione; The rest of the last post that you cut off in your response explains in as perfect wording as I can think of why I think even chemistry is deterministic. I have no idea what are you referring to' date=' other people probably know even less, so can you please be more specific, and could you use quotes? Just because nobody is smart enough to figure it out doesn't make it not true. I'm smart, more than enough to make it true... thought, I really have no idea what are you talking about, what is "it"? IMO in order for the world to be non-deterministic you would need to provide an example of something that exists that violates some underlying physical principle. deterministic -adjective 1. Describes a system whose time evolution can be predicted exactly. Contrast probabilistic. 2. Describes an algorithm in which the correct next step depends only on the current state. I can give you examples, I say EVERYTHING is non-deterministic, so pick your favorite example, anything in the whole wide world and beyond, anything at all... and tell me just how much is deterministic? However, note that non-deterministic DOES NOT MEAN "violating physical principles", please acknowledge yourself with the subject: http://en.wikipedia.org/wiki/Quantum_mechanics The main question now is this: - Brain is Electrical Quantum machine or Mechanical Turing machine? We know brain is some physical system. There are many ways to model dynamical systems in physics and before we start we need to know if our system is based on Classical or Quantum physics, whether it is determinist or not ...alternatively, you can doubt the uncertainty of Quantum Mechanics, which I, of course, can prove, most certainly. Edited January 30, 2009 by Sione
mooeypoo Posted January 30, 2009 Posted January 30, 2009 Sione, lose the attitude, you're here to discuss not to lecture. Whatever you may think of others' intelligence or your own IQ, it has absolutely no relevance to any discussion. The claims on both sides are weighed and examined by the merit of the claims themselves, and not by whatever it is you think about the other person - or however higher in the evolutionary chain you see yourself. Lose the ad hominem attack and get your behind off that high tree you put it on so this discussion can actually get somewhere.
Mr Skeptic Posted January 30, 2009 Posted January 30, 2009 (edited) If you think that this thread has become too long and unreadable, clicking this link should fix that. Sorry if this was inappropriate, but I think its the only thing that can be done. Edit: Crossposted with mooeypoo. I thought the mods had abandoned us Edited January 30, 2009 by Mr Skeptic
Sione Posted January 30, 2009 Posted January 30, 2009 (edited) Sione' date=' lose the attitude, you're here to discuss not to lecture. Whatever you may think of others' intelligence or your own IQ, it has absolutely no relevance to any discussion. [/quote'] I have no idea what are you referring to, perhaps you misunderstood some joke? The claims on both sides are weighed and examined by the merit of the claims themselves, and not by whatever it is you think about the other person - or however higher in the evolutionary chain you see yourself. It is not what I think, but the ARGUMENTS I provide, like this one: - "The Copenhagen interpretation, is the interpretation of quantum mechanics most widely accepted amongst physicists. Quantum mechanics does not simply reflect our limited knowledge, the results are in fact probabilistic because the physical universe is itself probabilistic rather than deterministic. ...what can I say, most physicists and 'Copenhagen interpretation' agrees with me, what do we do about that? Lose the ad hominem attack and get your behind off that high tree you put it on so this discussion can actually get somewhere. You misinterpreting something, and this discussion would move if people would at least have some general knowledge, at least glance over Wikipedia or search the Internet for quotes and real arguments. I do not comment people IQ, but laziness to support nonsense claims. If you think that this thread has become too long and unreadable' date=' clicking this link should fix that. Sorry if this was inappropriate, but I think its the only thing that can be done. [/quote'] If it is unreadable then try to increase the size of the font, or articulate what confused you. What is it you mind? Are you worried about 'space on the internet'? Do you think knowledge and dialog presented here hurts anyone in any way? If it is not your concern, if you are not hurt in any way, than why do you not just mind your business and save some space, you made the thread even longer by your unnecessary comment. Can you rationalize the purpose and why did you HAVE to say that? Do you think your behavior could be simulated? Do you think it is rational, emotional, instinctive, determinist, random or chaotic? Are you helping someone, doing some good? Helping justice? Perhaps, helping censor knowledge? Why do you not let npts2020, bascule and me have a conversation, since you are not interested? That's what public forums are for, right? Perhaps, there is such thing as having too much of discussion, if so, then just leave, no one is forcing you... sheesh! Edited January 30, 2009 by Sione
Mokele Posted January 30, 2009 Posted January 30, 2009 It's very simple - while it is very complex, the brain is deterministic. You can determine the behavior of any neuronal network if you simply know the membrane properties and neurotransmitters involved. Yes, there are a lot of factors affecting membrane properties, but it's still predictable. All this stuff about quantum influences in neurons is simply bullshit. Why? Yes, everything is quantum probability, but why don't baseballs behave like quantum particles? Because they've got so many atoms that all the probabilities just cancel out. The throw of a pair of dice is random, but if you roll 100,000,000 pairs of dice, the average result is going to be 3.5. Your average cell has 10^14 atoms in it. And neurons are actually very, very big cells. We're talking *quadrillions* of atoms. Any quantum effects are going to be averaged out easily. Mokele
bascule Posted January 30, 2009 Posted January 30, 2009 My turn... 1.) Do you think chemistry is deterministic? Yes, very much so. If you start from properly measured ingredients, you can follow a chemical formula and get the same results every time. Biology performs some absurdly complex chemical reactions, such as unfolding the DNA helix, unzipping it, reading off its nucleotides, and feeding that into a giant protein factory. Once the proteins emerge from this factory, they fold the same way every time. Cells are filled with little precisely engineered machines that have been carrying out their tasks with amazing accuracy for billions of years. Do you think there's some sort of chemical indeterminacy? What part of chemistry do you find non-deterministic. 2.) Do you think the "information" is within chemical structure or within electromagnetic fields? Both. Given your preoccupation with electromagnetic fields, I'm going to guess you believe there's some sort of quantum effect which affects electromagnetic fields in your brain, and that's a gateway to magical fantasy land where free will resides. Or if that's not it, can you perhaps describe what quantum majigger you think is going on in the brain? Like specifically? Penrose certainly believes there's a quantum effect going on. He hypothesizes quantum waveform collapse affects microtubules. Do you have something that specific to propose? 3.) Do following life forms have emotions or instincts? a. Virus b. Bacteria c. Squid d. Bird e. Dog f. Dolphin I'm going to ignore your question and substitute my own: which ones do I think are conscious? That's the much more interesting question. Virus and bacteria, no, duh. I really don't know enough about squid to answer, but one of their relatives the cuttlefish I believe to be conscious. Birds I believe evolved a structure which is functionally similar to the neocortex and I do believe they possess a limited degree of consciousness. I think it varies a lot depending on the species. As for dogs and dolphins, they're both mammals and pretty much all mammals are conscious in my book.
npts2020 Posted January 30, 2009 Posted January 30, 2009 Sione; Well, the Webster's New World Dictionary sitting on my desk says de-ter-min-ism; The doctrine that everything, including ones choice of action, is the necessary result of a sequence of causes. Deterministic is listed as an adjective, with no other definitions, where did your definition come from? I see nothing about predictability but an implication that the same causes will have the same outcome every single time. IMO the definition you got is poorly worded and you are confusing can with being able to predict an outcome. Just because something can be done does not mean any particular entity is able to do it (A rocket ship can be launched but who had the ability to do it 100 years ago?). I do not discount the possibility that the universe is non-deterministic but I have seen no evidence of it thus far in my life. That is why I view things through the eyes of determinism, yes even chemistry.
Sione Posted January 30, 2009 Posted January 30, 2009 (edited) It's very simple - while it is very complex' date=' the brain is deterministic. You can determine the behavior of any neuronal network if you simply know the membrane properties and neurotransmitters involved. Yes, there are a lot of factors affecting membrane properties, but it's still predictable. All this stuff about quantum influences in neurons is simply bullshit. Why? Yes, everything is quantum probability, but why don't baseballs behave like quantum particles? Because they've got so many atoms that all the probabilities just cancel out. The throw of a pair of dice is random, but if you roll 100,000,000 pairs of dice, the average result is going to be 3.5. Your average cell has 10^14 atoms in it. And neurons are actually very, very big cells. We're talking *quadrillions* of atoms. Any quantum effects are going to be averaged out easily. [/quote'] I agree, but when you average you lose information. Also, the system is obviously _dynamic_ and three-dimensional, it is not just about position, static snapshot can not tell you anything about linear and angular momentum. Anyway, how do we test it? How do we find out for sure? Consider: - Imagine emotion or consciousness somehow emerged inside such deterministic computation, how would you know? What would you look for? 1.) Do you think chemistry is deterministic? 2.) Do you think the "information" (thoughts/consciousness/feelings) is within chemical structure or within electromagnetic fields? 3.) Do following life forms have emotions or instincts? a. Virus b. Bacteria c. Tree d. Squid e. Bird f. Dog g. Dolphin I'm going to ignore your question and substitute my own: which ones do I think are conscious? That's the much more interesting question. Hold on' date=' you can not really do that, it is not "fair-play" discussion, especially since this is very essence of my point. The question in OP is about EMOTIONS. I say you DO NOT KNOW DEFINITION of emotions nor consciousness nor feelings nor thoughts and therefore you can not simulate it. To prove my point I am trying to demonstrate how you can not recognize emotions/instincts/feeling in real-world WORKING natural "AI", then how would you recognize it in artificial AI that is not even certain to work properly? Do you see how you struggle to define where consciousness is and where is not, you can not even know if a person in coma have any thoughts, feelings or consciousness, you would not know what to look for, or would you? How would you know then your AI has consciousness, what would you look for in BlueBrain project? Given your preoccupation with electromagnetic fields, I'm going to guess you believe there's some sort of quantum effect which affects electromagnetic fields in your brain, and that's a gateway to magical fantasy land where free will resides. Or if that's not it, can you perhaps describe what quantum majigger you think is going on in the brain? Like specifically? Penrose certainly believes there's a quantum effect going on. He hypothesizes quantum waveform collapse affects microtubules. Do you have something that specific to propose? Yes we understand each other, we only need to find a way to confirm theory with reality. I have no problem accepting what you suggest, I only want some evidence for it, because all the evidence I am aware of point to otherwise. I am not proposing to prove either until we settle what is this thread about - POSSIBILITY of EMOTIONAL AI. Quantum/Classic DOES NOT MATTER for that question, this is only thing that matters: - Imagine emotion or consciousness somehow emerged inside such computation, how would you know? What would you look for? I will ignore the rest of your questions too then, until you decide to play fair... boo, booo! Sione; Well' date=' the Webster's New World Dictionary sitting on my desk says de-ter-min-ism; The doctrine that everything, including ones choice of action, is the necessary result of a sequence of causes. Deterministic is listed as an adjective, with no other definitions, where did your definition come from? [/quote'] Please, lets not discuss some definitions that are very essence of concepts and basic terminology in this scientific field. Your definition is general, my definition is SCIENTIFIC, more precise and logically applicable. http://dictionary.reference.com/browse/deterministic http://en.wikipedia.org/wiki/Quantum_mechanics http://en.wikipedia.org/wiki/Determinism It is not fair that I need to argue something that is general knowledge and just a click away. Edited January 30, 2009 by Sione
bascule Posted January 30, 2009 Posted January 30, 2009 Sione, The nature of consciousness is not a question for science, it is one for philosophy. Consciousness is metaphysical, whereas science is a tool for explaining the physical world. Cognitive science has come up with certain approaches to studying consciousness using a scientific methodology and has come up with a number of easily repeatable experiments (most of these take the form of "optical illusions" or other such sensory phenomena which can be reliably repeated on any given test subject) I think since Kant most people, particularly of a scientific mindset, respect the separation between experiential consciousness and the brain activity underlying it. You seem to want to conflate the two. I think this is the source of your confusion. All that said: you did not disagree with the first question I asked, which was "Consciousness results from brain activity." If you believe this, then any questions relating to things like emotions, feelings, thoughts, and what exactly defines "consciousness" are irrelevant. If you believe that consciousness results from brain activity, then you should also believe that if we build a physical system similar enough to the brain, consciousness should result from it despite what knowledge or lack thereof the creator of the physical system may have about emotions, feelings, etc. These are secondary effects which are emergent from the behavior of the physical system itself. To build a copy of a brain inside a computer, we don't need to understand the great mysteries of the brain, the "whys" behind our genes constructing the brain in the way they do. All we need to understand is how the brain is structured, which merely requires we build a map of its structure. One of the most important things to gather from modern neuroscience is that the neocortex, the structure at the heart of consciousness, thought, reasoning, experience, etc, is immensely self-similar with only minor variation. These minor variations take the form of some specific tuning to help the neocortex organize itself into specific functional areas for different behaviors, however these areas are in no way fixed and the neocortex will automagically reorganize itself in the event of brain damage. This makes the task of understanding the operation of the neocortex considerably simpler, and there are comprehensive theories of the combined operation of the neocortex, thalamus, and hippocampus. 1
Sione Posted January 31, 2009 Posted January 31, 2009 (edited) bascule, Yes, "consciousness results from brain activity", it was obvious I agreed to that, in pretty detailed fashion even. The problem is that you can not define BRAIN ACTIVITY, Ok? You think you can define BRAIN STRUCTURES, and that is not the same as defining dynamics of system, do you agree? I'm trying to be very simple and concrete, trying to focus and discuss what is this thread about. I do not want to make it philosophical discussion, I want to make it very obvious and down to essence, answer the question with pure logic and practicability of mathematics, as in "computer science". If you would like to talk with me, then please respond properly and directly to question: I am not proposing to prove either until we settle what is this thread about - POSSIBILITY of EMOTIONAL AI. Quantum/Classic DOES NOT MATTER for that question, this is only thing that matters: - Imagine emotion or consciousness somehow emerged inside such computation, how would you know? What would you look for? Do you see how you struggle to define where consciousness is and where is not, you can not even know if a person in a coma have any thoughts, feelings or consciousness, you would not know what to look for, or would you? How would you know then your maybe working AI has consciousness, what would you look for in BlueBrain project? Edited January 31, 2009 by Sione
npts2020 Posted January 31, 2009 Posted January 31, 2009 Sione; Your references for defining determinism are not helping your argument. The last definition given in your first reference is the one you chose to work with (yes the one that is IMO poorly worded), all the other definitions are either verbatim or variations of the one I gave. The reason the definition you used is different is not because of being any better scientific use of the term but because that is how it is used in computer programming, which is somewhat different from how the term is used generally in philosophical discussion. If you wish to say that we can't now build an artificial system to perfectly model the current universe in every detail, I would agree with that. However, that is not saying that the universe does not follow an underlying set of rules that do not change and may one day be discernable by humans to the point of being able to predict anything.
Sione Posted January 31, 2009 Posted January 31, 2009 (edited) Do you not understand how hard is to talk to you if you do not use quotes? It is hard, it is not polite because you are making people waste time to reference what you're talking about. I can not argue with you about definitions if you are lazy to learn about the subject. Determinism in COMPUTER SCIENCE, LOGIC, MATHEMATICS, STATISTICS, PHILOSOPHY, QUANTUM PHYSICS and other sciences is what I told you it is, find your favorite SCIENTIFIC dictionary or encyclopedia and learn about it, that's all. Stop jumping over me. If you wish to say that we can't now build an artificial system to perfectly model the current universe in every detail, I would agree with that. If I say? ...was I not clear enough? Did I not repeat 24 times exactly what I mean? Maybe not, so let me repeat it again: -Imagine emotion or consciousness emerged inside such computation, how would you know? WHAT WOULD YOU LOOK FOR? I do not insist on EVERY DETAIL, but I just want you to tell me what is emotion, what is thought, what is feeling? I do not care how much detail, I want to know if person in a coma have any consciousness, emotions or thoughts? If you can not recognize it in a real working machine, then how in a world you expect to know some maybe working AI has anything like "it"? Edited January 31, 2009 by Sione
Mokele Posted January 31, 2009 Posted January 31, 2009 I agree, but when you average you lose information. And? Biological systems are *very* sloppy, and neurons have a one-bit output, on or off. Think of a neuron like this: imagine a bucket tied to a springy light switch. The bucket has water always flowing in, but has a hole in the bottom. When other neurons fire at it, they add an extra set quantity of water. Sometimes it's enough to fill the bucket and flip the switch, other times it takes several additions, and sometimes it takes several neurons all at once to make another fire. Other processes can either change the rate of the constant water flowing in or change the size of the hole. What you're talking about amounts to tracking individual water molecules in the buckets - it doesn't matter and cannot substantially affect the output. Remember, biology isn't optimal. Most systems are fairly sloppy, and some are so badly cobbled together it's amazing they even work at all. Also, the system is obviously _dynamic_ and three-dimensional, it is not just about position, static snapshot can not tell you anything about linear and angular momentum. Um, what? What does momentum have to do with neurons. Nothing. Anyway, how do we test it? How do we find out for sure? With animals. We've been sticking wires into animal's brains for decades now, and we know a LOT. We can actually *read* the visual stimuli a monkey sees directly from its brain. We've mapped the entire nervous system of several simpler animals. Also, humans have a knack for doing stupid shit that causes brain damage, which in turn allows us to actually observe the effects of damage to particular clusters of neurons. Imagine emotion or consciousness somehow emerged inside such deterministic computation, how would you know? What would you look for? 'Consciousness' is a tricky term. Emotions, however, are simple. They're literally nothing more than cognitive reflexes, no more special or complex than withdrawing your hand from a hot stove. Adding emotions to a computer is simple - simply program it to bias its responses towards a particular form upon receiving a certain input. If I make a random number generate only spit out even numbers, I've basically given it emotion. 1.) Do you think chemistry is deterministic? Obviously. Compound A + Compound B = Compound C. 2.) Do you think the "information" (thoughts/consciousness/feelings) is within chemical structure or within electromagnetic fields? Neither. It's been well established via animal experiments that information is encoded into the nervous system by physical rearrangement of the connections between neurons and by alterations of the properties of individual neurons (see the leaky bucket analogy - basically permanently changing hole size or input flow). EM fields in the majority of animals are byproducts of cellular activity, with no usefulness to the animal itself (the exceptions obviously being electric fish such as knifefish, etc.) 3.) Do following life forms have emotions or instincts? a. Virus b. Bacteria c. Tree d. Squid e. Bird f. Dog g. Dolphin It depends how you define the terms, but the first three, definitely not, as they lack a nervous system. Squid are iffy - some cephalopods are smarter than rats and mice. Birds, dogs and dolphins have been empirically observed displaying behavioral tendencies which suggest cognitive processes similar to human emotions. To prove my point I am trying to demonstrate how you can not recognize emotions/instincts/feeling in real-world WORKING natural "AI", then how would you recognize it in artificial AI that is not even certain to work properly? That assumes those are necessary for consciousness and/or intelligence. That's like saying that a beak is necessary for flight because birds have one and they fly - just because humans have certain cognitive traits doesn't mean they're necessary or even useful to an AI. he problem is that you can not define BRAIN ACTIVITY, Ok? We can, have, and have been studying it in detail for close to 30 years. Brain activity is the pattern of firing of neurons. We can directly observe it via FMI and PET scans, the latter of which are becoming higher and higher resolution, allowing us closer and closer to a neuron-by-neuron map of brain activity. We can also use electromyography, but we don't on humans for ethical reasons (it requires implantation of wire electrodes and, usually, termination and dissection to confirm placement). How would you know then your maybe working AI has consciousness How do you know we have it? How do you know it isn't illusory? We tend to imagine that there's a little 'free will' module somewhere in the brain that's somehow consciously making decisions, but there is a growing body of evidence from the imaging methods described above that we don't even make decisions - we just react, then justify it later (albeit only a few dozen milliseconds later). In short, it's possible that even *humans* are merely highly complex learning neuronal networks that spit out info based on what we're given, without any *actual* "man behind the curtain" so to speak. Perhaps the impression of 'making a decision' is just a cognitive illusion. As unflattering as it may be, it's a possibility rooted in empirical studies, and one we have to consider. Mokele
Sione Posted January 31, 2009 Posted January 31, 2009 (edited) Remember' date=' biology isn't optimal. Most systems are fairly sloppy, and some are so badly cobbled together it's amazing they even work at all. [/quote'] I said I agree, but we still must be sure if the system we are talking about is indeed biological as opposed to electrical quantum machine. I accept all your theories, no problem. We just need to talk about experiments and observations from now on. Um, what? What does momentum have to do with neurons. Nothing. Look, we agreed we do not know for sure. I accept your theory, thought if it turns out brain is quantum machine, then direction, velocity and spin of charged particles provide much more information than only spatial position in 3D, it would make for a better computer. We've been sticking wires into animal's brains for decades now, and we know a LOT. We can actually *read* the visual stimuli a monkey sees directly from its brain. We've mapped the entire nervous system of several simpler animals. Beautiful, that is very interesting, can you give some links? However, 'visual stimuli' is INPUT, what we need to know is how to recognize brain "OUTPUT". It is hard to know because the information feeds recursively back into brain, there is no screen to display output. In any case, that is the way to go. 'Consciousness' is a tricky term. Emotions, however, are simple. They're literally nothing more than cognitive reflexes, no more special or complex than withdrawing your hand from a hot stove. Adding emotions to a computer is simple - simply program it to bias its responses towards a particular form upon receiving a certain input. If I make a random number generate only spit out even numbers, I've basically given it emotion. Fantastic! I absolutely agree. Obviously. Compound A + Compound B = Compound C. Yeah, but I heard of this as well: A+B+C = AB, C A+B+C = AC, B Can we test it? Neither. It's been well established via animal experiments that information is encoded into the nervous system by physical rearrangement of the connections between neurons and by alterations of the properties of individual neurons (see the leaky bucket analogy - basically permanently changing hole size or input flow). What you are referring to is what I call "chemical structure", physical as opposed to e/m fields that are SPIRITUAL, like "spiritual mind". How do you call it? You do not really know what is "information" and how it looks inside the brain. You can only say that thought process is SOMEHOW CONNECTED to physical rearrangements and chemical structures and reactions, but unless you can read thoughts you do not know in what form that information really exists. EM fields in the majority of animals are byproducts of cellular activity, with no usefulness to the animal itself (the exceptions obviously being electric fish such as knifefish, etc.) That's fine, now lets talk about pictures of monkey thoughts. It depends how you define the terms, but the first three, definitely not, as they lack a nervous system. Squid are iffy - some cephalopods are smarter than rats and mice. Birds, dogs and dolphins have been empirically observed displaying behavioral tendencies which suggest cognitive processes similar to human emotions. Let me define: Feeling = brain input Thought = brain output Emotion = memorized feeling Instinct = emotion based reaction Intelligence = thought based reaction Consciousness = production of brain output I think all of them have emotions and instincts, even Virus. And here is my argument, which is your argument from above: - "Emotions, however, are simple. They're literally nothing more than cognitive reflexes... Adding emotions to a computer is simple - simply program it to bias its responses towards a particular form upon receiving a certain input. If I make a random number generate only spit out even numbers, I've basically given it emotion." Feeling=Sensation/Perception:INPUT = (Thought=Cognition/Intention:OUTPUT) + (Emotion=Feeling Acquired/Genetic:MEMORY) We can, have, and have been studying it in detail for close to 30 years. Brain activity is the pattern of firing of neurons. We can directly observe it via FMI and PET scans, the latter of which are becoming higher and higher resolution, allowing us closer and closer to a neuron-by-neuron map of brain activity. We can also use electromyography, but we don't on humans for ethical reasons (it requires implantation of wire electrodes and, usually, termination and dissection to confirm placement). So, tell me, according to all the technology mentioned - does a person in a coma feels anything, dream, think or have consciousness? How do you know we have it? How do you know it isn't illusory? That is my question, I asked first. We do not know even if we have it, then how can we talk about simulating it? We tend to imagine that there's a little 'free will' module somewhere in the brain that's somehow consciously making decisions, but there is a growing body of evidence from the imaging methods described above that we don't even make decisions - we just react, then justify it later (albeit only a few dozen milliseconds later). In short, it's possible that even *humans* are merely highly complex learning neuronal networks that spit out info based on what we're given, without any *actual* "man behind the curtain" so to speak. Perhaps the impression of 'making a decision' is just a cognitive illusion. As unflattering as it may be, it's a possibility rooted in empirical studies, and one we have to consider. This is great! I love it, shame that I have to argue against it. "we just react, then justify it later" - this would explain a lot about human behavior, wouldn't it? - how much sense words like "self-awareness" make with such automatic justification? Do you find it possible deterministic interplay of physics forces had this very moment of you reading this line of text destine to happen, guaranteed based only on physics and positions of particle at the time of Big-Bang? According to entropy and efficiency principle, this seem opposite of what should happen. Edited January 31, 2009 by Sione Consecutive posts merged.
Mokele Posted January 31, 2009 Posted January 31, 2009 thought if it turns out brain is quantum machine, then direction, velocity and spin of charged particles provide much more information than only spatial position in 3D, it would make for a better computer. I guarantee there is zero chance of that happening. Biological systems are simply too large to take advantage of quantum effects without it all getting averaged out. Beautiful, that is very interesting, can you give some links? The first thing that cropped up in google: http://cercor.oxfordjournals.org/cgi/content/abstract/13/4/422 Reading visual cortex of the human brain we need to know is how to recognize brain "OUTPUT". It is hard to know because the information feeds recursively back into brain, there is no screen to display output. In any case, that is the way to go. Well, there's simple motor output - We can train monkeys to use a robot arm via implants in the motor cortex However, what you're talking about, inferring higher-level cognitive function, is only really possible via indirect methods such as observing behavior. Possibly in an AI, it would be simpler, because you could essentially watch all the parts move as they do some complex task, but in a living brain, we can only correlate due to technological and ethical barriers. What you are referring to is what I call "chemical structure", physical as opposed to e/m fields that are SPIRITUAL, like "spiritual mind". How do you call it? You do not really know what is "information" and how it looks inside the brain. You can only say that thought process is SOMEHOW CONNECTED to physical rearrangements and chemical structures and reactions, but unless you can read thoughts you do not know in what form that information really exists. There is no evidence whatsoever for any sort of 'spiritual brain', nor any evidence for any role of EM fields in the brain as information storage. Also, the experiments that discovered the basis of learning have been in organisms with very simple nervous systems. We can read their thoughts because they only have a few hundred neurons total. Feeling = brain inputThought = brain output Emotion = memorized feeling Instinct = emotion based reaction Intelligence = thought based reaction Consciousness = production of brain output I don't think your definitons are very good. Emotion, for instance, is a cognitive bias in output in response to a particular input, and has nothing to do with memory - amnesiacs still have emotions. Instincts also have nothing to do with emotions - they're simply automatic behavior patterns and pre-programmed drives. Your definition of intelligence is useless, because it requires a definition of thought. And you define consciousness so broadly that any motor command is evidence of it, and even jellyfish are conscious. think all of them have emotions and instincts, even Virus. A virus isn't alive. And neither bacteria nor trees even have nervous systems, meaning your brain-based definitions *automatically* leave them out. does a person in a coma feels anything, dream, think or have consciousness? It depends on the nature of the injury that put them in that coma. Some are nothing but shells, totally brain dead. Others have rudimentary or chaotic brain activity. And some have the truly horrifying experience of being fully aware and trapped inside a body that doesn't respond. I think the point Sione is trying to raise is an interesting one: How do you *know* if an AI is intelligent or conscious? How can you test it? Especially if you're trying to evaluate someone else's claim to have made one? The Turing test only gets you so far, after all. Mokele
bascule Posted February 1, 2009 Posted February 1, 2009 physical as opposed to e/m fields that are SPIRITUAL, like "spiritual mind". There's nothing spiritual about electromagnetic fields. They are just one of the four fundamental forces. They are very much a part of the physical world, especially with their role of holding together all matter as we know it. You do not really know what is "information" and how it looks inside the brain. You can only say that thought process is SOMEHOW CONNECTED to physical rearrangements and chemical structures and reactions, but unless you can read thoughts you do not know in what form that information really exists. Thing is, we've done that: http://www.newscientist.com/article/dn16267-mindreading-software-could-record-your-dreams.html
ydoaPs Posted February 1, 2009 Posted February 1, 2009 I think the point Sione is trying to raise is an interesting one: How do you *know* if an AI is intelligent or conscious? How can you test it? Especially if you're trying to evaluate someone else's claim to have made one? The Turing test only gets you so far, after all. Mokele How do we know Sione is intelligent or conscious? How can we test it?
Sione Posted February 1, 2009 Posted February 1, 2009 (edited) ydoaPs, What part confused you? Perhaps you lack the ability to process? Anyway, to answer your question, the proof of my intelligence is the DEFINITION that we finally have here. Maybe this definition does not make everyone happy, but it works and everyone is welcome to provide better definition. With definition we can talk about simulation, without definition you can only dream about it. There's nothing spiritual about electromagnetic fields. Please' date=' have you not noticed that I'm the DEFINITION MASTER? Electromagnetic fields are spiritual (incorporeal), as opposed to material. "Spiritual", relates to e/m fields quite equally as to mind. Please forget religious implications, it simply means "not made out of matter". Mind over matter, free your mind! [b']spir⋅it⋅u⋅al[/b]–adjective 1. of, pertaining to, or consisting of spirit; incorporeal. 2. of or pertaining to the spirit or soul, as distinguished from the physical nature: a spiritual approach to life. 1. of, relating to, consisting of, or having the nature of spirit; not tangible or material. See Synonyms at immaterial. 5. characterized by or suggesting predominance of the spirit; ethereal or delicately refined: She is more of a spiritual type... Homer: - There's your giraffe, little girl. Ralph Wiggum: - I'm a boy. Homer: - That's the spirit. Never give up. I think the point Sione is trying to raise is an interesting one: How do you *know* if an AI is intelligent or conscious? How can you test it? Yes' date=' yes, yes... thank Goddess Chance! That is my point, as I said in my 1st post - HOW TO TEST IT? To test it we need to be able to DEFINE it, just so we know WHAT to look for. I guarantee there is zero chance of that happening. Biological systems are simply too large to take advantage of quantum effects without it all getting averaged out. There is no evidence whatsoever for any sort of 'spiritual brain', nor any evidence for any role of EM fields in the brain as information storage. Guarantee sound very assuring, but I guarantee opposite. I also have evidence for it, but this is not important for simulation, it is only a matter of determinism and processing power. What's important is the DEFINITION, because it will apply in either case. Well, there's simple motor output - We can train monkeys to use a robot arm via implants in the motor cortex Fantastic! That's EXACTLY the type of thing I was talking about, great! Soon, we'll be reading your minds! http://scienceblogs.com/pharyngula/2008/12/soon_well_be_reading_your_mind.php - "No, not really, but this is still a cool result: investigators have used an MRI to read images off the visual cortex. They presented subjects with some simple symbols and letters, scanned their brains, and read off the image from the data — and it was even legible! This is an interesting piece of work, but it has some serious limitations." As far as I am concerned, it does not need to be monkey, it can be the simplest nervous system there is, as long as you can "intercept" information and actually "read/predict" reaction before 'organism' reacted, i.e. "read the mind". Feeling = brain input Thought = brain output Emotion = memorized feeling Instinct = emotion based reaction Intelligence = thought based reaction Consciousness = production of brain output I don't think your definitons are very good. My definitions are very much perfect, they work! ...especially considering the number of symbols they use, they are kind of general and not fully defined. Again, my point is that "you" do not have definitions at all, so you could not know what is what, and therefore you can not really judge. If you would like to judge, provide your definitions and we will take better one or modify until agreed upon, but we must have them EXACTLY defined in order to answer these questions properly. I want to work together to make operating definition, so that it applies to all the cases and that we can uniquely agree on all the results our definition produces. We must define ALL of the terms and they have to interrelate so it makes sense and compares with general understanding of these terms. Does this sound fair? Emotion, for instance, is a cognitive bias in output in response to a particular input, and has nothing to do with memory - amnesiacs still have emotions. The correct objection would be if you asked me to define "brain" and "memory". I was just trying to keep it simple and obvious. I can, of course, explain it in any terms as long as we define them. I do not insist on terminology, we can change words as you like, but those are the TERMS we need to define, whatever we call them. I can demonstrate how those definition actually work very well indeed, in practice. They might not give prediction that are in accordance to your understanding, but they can be applied EXACTLY, and according to those definitions I can simulate all of those terms, sure not everyone will agree, but if you think about it, you may realize that those definitions do model life-like 'information processing entities'. Not too bright, not very emotional perhaps and maybe without purpose, but they will move, they will be animated and they will "feel", "remember" and "react" to external input, produce output accordingly to what they previously "felt" or was built into their "genes". As I said, my definitions work, they do model life-like behavior. -> OUTPUT=INPUT+MEMORY -> INPUT=MEMORY+OUTPUT -> MEMORY=INPUT+OUTPUT && -> e.g. Feeling=Sensation/Perception:INPUT = Thought=Cognition/Intention:OUTPUT + Emotion=Feeling Acquired/Genetic:MEMORY Instincts also have nothing to do with emotions - they're simply automatic behavior patterns and pre-programmed drives. And you define consciousness so broadly that any motor command is evidence of it, and even jellyfish are conscious. Ok, so instead of to argue personal opinions about meaning, I want us to try and make definitions together, the shorter the better, according to Occam's razor, and so it can be applied to these cases: a. Virus b. Bacteria c. Tree d. Squid e. Bird f. Dog g. Dolphin h. human embryo, 2 days i. human fetus, 21 weeks j. human born, 2 months k. human coma, 21 years l. BlueBrain AI OUTPUT (Thought), is abstract term, not to be confused with physical reaction/response. Anyway, in order to simulate OUTPUT and be able to TEST IT, for all of these we MUST, very EXACTLY, be able to address questions about these attributes of life forms: Life = ? Memory = ? Feeling = ? Thought = ? Emotion = ? Instinct = ? Intelligence = ? Consciousness = ? --------------------- Processing System/Responding System/Brain/Nervous System = ? These words need to be defined and we do not need to be imperative they are strictly in accord to our general understanding, as long as we know what we are talking about and as long as it relates and computes. I'm happy with any wording, but above terms must be defined in some meaningful relation, they must apply to real life scenario, if we are to start simulating. A virus isn't alive. And neither bacteria nor trees even have nervous systems, meaning your brain-based definitions *automatically* leave them out. The fact is that we do not have EXACT DEFINITION of life, so the scientific community is actually divided in opinion... ah, definitions again. According to most simple definition, Life = self-replicating molecule. Again, I do not care how we define it, but it must precisely describe the real world, and I do not mind definition that says virus is not life, as long as it works mathematically and logically and can be applied in practice. Edited February 1, 2009 by Sione
Mr Skeptic Posted February 1, 2009 Posted February 1, 2009 How do we know Sione is intelligent or conscious? How can we test it? I wonder if he could give some evidence of it? Merged post follows: Consecutive posts mergedPlease, have you not noticed that I'm the DEFINITION MASTER? Electromagnetic fields are spiritual (incorporeal), as opposed to material. "Spiritual", relates to e/m fields quite equally as to mind. Please forget religious implications, it simply means "not made out of matter". spir⋅it⋅u⋅al–adjective 1. of, pertaining to, or consisting of spirit; incorporeal. 2. of or pertaining to the spirit or soul, as distinguished from the physical nature: a spiritual approach to life. 1. of, relating to, consisting of, or having the nature of spirit; not tangible or material. See Synonyms at immaterial. 5. characterized by or suggesting predominance of the spirit; ethereal or delicately refined: She is more of a spiritual type... I never knew that computers were spiritual. But since they are based on electromagnetism and you say electromagnetism is spiritual, surely computers are spiritual too
Xittenn Posted February 1, 2009 Posted February 1, 2009 So if we pass Sione through a giant inductive coil and he induces an electrical current we know he has a soul?
Sione Posted February 1, 2009 Posted February 1, 2009 (edited) I wonder if he could give some evidence of it? We are talking about simulating AI, what are you talking about... me? I'm flattered. Now, would you like to participate in the discussion and talk about what this thread is about? Don't make me repeat, I said my DEFINITION is evidence, Ok? Now, we need to see what evidence you got, can you talk? Can you answer these, so we at least know what is your position on all this: a. You do not understand definition? b. You understand definition and agree? c. You understand and disagree? (provide your definition) Thanks in advance for your input that is related to topic and constructive, insults are welcome as well, they are amusing, taa. I never knew that computers were spiritual. But since they are based on electromagnetism and you say electromagnetism is spiritual, surely computers are spiritual too [hitting its head] Stop hitting yourself! Your conclusion is correct, but your refusal to accept definitions from the dictionary and encyclopedia will make you hurt yourself like that often. If you could understand that everything is defined by definitions, everything. So everything is true or false ACCORDING to some definition. And according to some of those definitions, yes they are "spiritual". "Spiritual", Synonyms: Immaterial, http://dictionary.reference.com/browse/spiritual - what part do you not understand? You know what "synonym" and "immaterial" mean, right? So if we pass Sione through a giant inductive coil and he induces an electrical current we know he has a soul? Ok, I know I'm great, but lets talk about SIMULATING AI and definitions we need to be able to achieve that. Can you offer better definitions or articulate what part you do not agree with so we can improve it, please? ======================= Variables & Units... Life = living status, Yes/No Memory = max capacity, Byte Feeling = information input, Byte Thought = information output, Byte Emotion = memorized input(felt), Byte Consciousness = produced output, Byte/Sec Instinct = emotion biased physical reaction/purpose, (+/- %)/Sec Intelligence = thought biased physical reaction/purpose, (+/- %)/Sec --------------------------------------------------------------------- Responding System/Nervous System = whatever, say physical volume, mm^3 Edited February 1, 2009 by Sione
addictor Posted February 1, 2009 Posted February 1, 2009 Reply to OP. I reckon in the future yeah, when we are able to program them to appreciate what we feel and program them to make choices based on their emotionals. Anything we can program them to do, they will be able to perform it. If we can't, they can't.
Recommended Posts