I was wondering if anyone else had stumbled across this article while searching the internet...
http://www.riffkiddy.com/documents/FourReasonsForBeing-SecondDraft.pdf
There seem to be many audio overtones and a lot of discussion of feedback that means it is really not off-topic at all! Mind-blowing stuff or just nonsense?
http://www.riffkiddy.com/documents/FourReasonsForBeing-SecondDraft.pdf
There seem to be many audio overtones and a lot of discussion of feedback that means it is really not off-topic at all! Mind-blowing stuff or just nonsense?
I have tried twice to read the article, and find a lot of trendy wording wrapped around a few facts. I never made it past the 3rd or 4th principle either time.
Fact, the human body is a complex multi loop feedback system. Most humans have the same systems, although the tuning can vary from person to person. Example, the human body's internal temperature regulator runs at an average of 98.6 F (37C). Mine usually runs around 97 F, but can vary a half a degree or so over several days.
Fact, humans have a higher level "overlay" on top of that system capable of instinct, intelligence, and emotions. Everyone has a different weighting and capability for each overlay function.
Humans can and do adapt as a species over time responding to their environment and culture.
Fact, the human body is a complex multi loop feedback system. Most humans have the same systems, although the tuning can vary from person to person. Example, the human body's internal temperature regulator runs at an average of 98.6 F (37C). Mine usually runs around 97 F, but can vary a half a degree or so over several days.
Fact, humans have a higher level "overlay" on top of that system capable of instinct, intelligence, and emotions. Everyone has a different weighting and capability for each overlay function.
Humans can and do adapt as a species over time responding to their environment and culture.
It's a pity that this thread has not attracted more comment 
Maybe I am the only one who has not been able to dismiss the article as "hogwash", although some justification for doing so would have been very welcome. I don't agree that the article is presumptuous and even if it is, surely it is deserving of qualified rebuke rather than discounting it because of some other presumption?
I also had to read the article a number of times to "understand" it too (and TBH I still don't completely). But I do have my own cognitive bias to admit here as references 7 (Damasio) and 9 (Hawkins) are IMHO the most insightful in the field and I was therefore very amenable to read an article that attempts their unification.
The article also reads more like pseudo-technical sales blurb than it does a scientific paper. Considering it appears to concern undisclosed intellectual property on machined sentience rather than human consciousness, some obfuscation is probably to be expected. But I did persevere and in the hope of stimulating some conversation, I have added some comments here:
The systematic presentation of the central nervous system encompasses all the variations mentioned in the previous posts. It does not matters whether the neural feedback is multi-level, "re-entrant" or whatever. This article is the first of which I am aware that has started from a "presumption" that our conscious manifestation must have a relatively simple cause if it occurs in us all despite our differences. The evolutionary steps follow in a straightforward manner also, but the "subsuming" of part of the emotive apparatus in the later-evolved cognitive network is again novel as far as I am aware. The generation of whole sensory perception discussed in many other texts follows on neatly from here.
The next parts of the article are where I think the 'real' novelty lies. The cognitive apparatus is theorised as incorporating a model based on directly sensed mechanical impedances in the world immediately around us (the "reciprocating" senses). All the other senses (the "non-reciprocating" and interoceptive senses) are related to this rational basis, so that every change in our metabolic state is theorised to be equivalent to an "energetic" change in the physical world or other sensory mode and vice versa. And where a "back projection" of the interoceptive model part is correlated with a change sensed in the other modes, we get the perception of an "embodied self" grounded in our physical, external world.
Despite the circular language and logical contradictions, the most striking idea to me was the assertion that without the interoceptive correlations, there is no conscious state or even a self. Instead we are unconscious of all we do in between the instants of conscious "Gestalt". This makes a lot of sense and the explanation of consciousness/sentience fits neatly in to the existing body of knowledge about perception and the delays involved. It also fits in with our reality of a single conscious thread set against a background of continuous, parallel unconscious activities. The article suggests a rational explanation that subjectivity and feeling are cast in the past (that "we think we were" - my rephrasing!) and that are a part of the same energetic based model.
Since the article seems to have been gathered from studies into producing sentient machines (presumably also theory ATM?), I have now been left with the question of whether this rational "machine sentience" is genuine or just emulated. And if it is emulated, what then does it say about our own consciousness? Maybe it can only ever said to be as real as the reality around us? Which gets us nowhere. But I certainly don't yet subscribe to the view the article is hogwash, although I would like it to be!
Maybe I am the only one who has not been able to dismiss the article as "hogwash", although some justification for doing so would have been very welcome. I don't agree that the article is presumptuous and even if it is, surely it is deserving of qualified rebuke rather than discounting it because of some other presumption?I also had to read the article a number of times to "understand" it too (and TBH I still don't completely). But I do have my own cognitive bias to admit here as references 7 (Damasio) and 9 (Hawkins) are IMHO the most insightful in the field and I was therefore very amenable to read an article that attempts their unification.
The article also reads more like pseudo-technical sales blurb than it does a scientific paper. Considering it appears to concern undisclosed intellectual property on machined sentience rather than human consciousness, some obfuscation is probably to be expected. But I did persevere and in the hope of stimulating some conversation, I have added some comments here:
The systematic presentation of the central nervous system encompasses all the variations mentioned in the previous posts. It does not matters whether the neural feedback is multi-level, "re-entrant" or whatever. This article is the first of which I am aware that has started from a "presumption" that our conscious manifestation must have a relatively simple cause if it occurs in us all despite our differences. The evolutionary steps follow in a straightforward manner also, but the "subsuming" of part of the emotive apparatus in the later-evolved cognitive network is again novel as far as I am aware. The generation of whole sensory perception discussed in many other texts follows on neatly from here.
The next parts of the article are where I think the 'real' novelty lies. The cognitive apparatus is theorised as incorporating a model based on directly sensed mechanical impedances in the world immediately around us (the "reciprocating" senses). All the other senses (the "non-reciprocating" and interoceptive senses) are related to this rational basis, so that every change in our metabolic state is theorised to be equivalent to an "energetic" change in the physical world or other sensory mode and vice versa. And where a "back projection" of the interoceptive model part is correlated with a change sensed in the other modes, we get the perception of an "embodied self" grounded in our physical, external world.
Despite the circular language and logical contradictions, the most striking idea to me was the assertion that without the interoceptive correlations, there is no conscious state or even a self. Instead we are unconscious of all we do in between the instants of conscious "Gestalt". This makes a lot of sense and the explanation of consciousness/sentience fits neatly in to the existing body of knowledge about perception and the delays involved. It also fits in with our reality of a single conscious thread set against a background of continuous, parallel unconscious activities. The article suggests a rational explanation that subjectivity and feeling are cast in the past (that "we think we were" - my rephrasing!) and that are a part of the same energetic based model.
Since the article seems to have been gathered from studies into producing sentient machines (presumably also theory ATM?), I have now been left with the question of whether this rational "machine sentience" is genuine or just emulated. And if it is emulated, what then does it say about our own consciousness? Maybe it can only ever said to be as real as the reality around us? Which gets us nowhere. But I certainly don't yet subscribe to the view the article is hogwash, although I would like it to be!
To me, it is not at all clear what "sentient" actually means; we are an irrational and illogical species that believes in invisible magic men in the sky, an invisible thing called "luck" that is apparently dispensed by some other invisible magic being, and non-existent connections between completely unrelated events ("I won $5 at the Lotto on Wednesday the 9th, so Wednesday is my lucky day, and 9 is my lucky number.")
Read "Incognito" ( Amazon.com: Incognito: The Secret Lives of the Brain (9780307389923): David Eagleman: Books ) and "Subliminal" ( Subliminal: How Your Unconscious Mind Rules Your Behavior: Leonard Mlodinow: 9780307472250: Amazon.com: Books ), and it is difficult to attribute much sentience to our species.
Perhaps we humans, as an entire species, have far too high an opinion of our own abilities. A sort of global case of the Dunning-Kruger effect. ( Dunning–Kruger effect - Wikipedia )
With that as background, the initial premise of the entire paper sounds like hogwash to me: the "fundamental feeling of being" that the paper refers to, and connects to Descartes' famous mistake, "“I think therefore I am.”
Think about Descarte's statement for a second: an earthworm almost certainly does not think; yet it "am", i.e., it exists. A rock, or a planet, or a drop of water, most certainly does not think; all of them exist.
These things have objective existence in any number of objective ways that do not involve human senses: they have mass, they have inertia, they have temperature, they have a density different from their surroundings, they have their own gravitational field, et cetera, et cetera.
I offer the hypothesis that the "fundamental feeling of being" we humans think we experience, is actually nothing more than a combination of two things: the stream of sensory information constantly being fed into the brain by the nervous system and processed by your sub-conscious mind, and the more-or-less conscious stream of thoughts that occupy most of us almost every waking minute.
I have no idea what intelligence is (no surprise, since I'm basically a dumb ape, slightly more evolved than the other sub-species of ape on this planet). But a fair degree of intelligence seems to have evolved in numerous species (for instance, we now know that various birds, monkeys, apes, and humans all use tools.)
So it's a plausible hypothesis that intelligence is one of the routes that evolution has blindly and repeatedly found, in the constant search for ways to increase survival rates and reproductive success.
Interestingly, the only successful approaches to AI seem to be ones that use somewhat the same strategy that nature does; provide some initial information to the organism, then let it "learn" to cope, until it either dies, or learns enough to survive long enough to reproduce.
-Gnobuddy
Read "Incognito" ( Amazon.com: Incognito: The Secret Lives of the Brain (9780307389923): David Eagleman: Books ) and "Subliminal" ( Subliminal: How Your Unconscious Mind Rules Your Behavior: Leonard Mlodinow: 9780307472250: Amazon.com: Books ), and it is difficult to attribute much sentience to our species.
Perhaps we humans, as an entire species, have far too high an opinion of our own abilities. A sort of global case of the Dunning-Kruger effect. ( Dunning–Kruger effect - Wikipedia )
With that as background, the initial premise of the entire paper sounds like hogwash to me: the "fundamental feeling of being" that the paper refers to, and connects to Descartes' famous mistake, "“I think therefore I am.”
Think about Descarte's statement for a second: an earthworm almost certainly does not think; yet it "am", i.e., it exists. A rock, or a planet, or a drop of water, most certainly does not think; all of them exist.
These things have objective existence in any number of objective ways that do not involve human senses: they have mass, they have inertia, they have temperature, they have a density different from their surroundings, they have their own gravitational field, et cetera, et cetera.
I offer the hypothesis that the "fundamental feeling of being" we humans think we experience, is actually nothing more than a combination of two things: the stream of sensory information constantly being fed into the brain by the nervous system and processed by your sub-conscious mind, and the more-or-less conscious stream of thoughts that occupy most of us almost every waking minute.
I have no idea what intelligence is (no surprise, since I'm basically a dumb ape, slightly more evolved than the other sub-species of ape on this planet). But a fair degree of intelligence seems to have evolved in numerous species (for instance, we now know that various birds, monkeys, apes, and humans all use tools.)
So it's a plausible hypothesis that intelligence is one of the routes that evolution has blindly and repeatedly found, in the constant search for ways to increase survival rates and reproductive success.
Interestingly, the only successful approaches to AI seem to be ones that use somewhat the same strategy that nature does; provide some initial information to the organism, then let it "learn" to cope, until it either dies, or learns enough to survive long enough to reproduce.
-Gnobuddy
In my reading of the article, I formed the conclusion that it refuted/adjusted Descartes "I think therefore I am" rather than supported it. Its conclusion is more of the form "I think therefore I was" as it accounts for feeling and sentience as retrospective cognitively-inferred perceptions.
The "fundamental feeling of being" I think refers to knowing (or believing) of ones own existence rather than an objective account of the existence of some other thing - which probably requires reference to ones own existence in any case. It is always a logical minefield, however, where language and even tenses (see above) can lead to significant misunderstandings.
I am not sure Gnobuddy's hypothesis offers any hypothesis at all, although anyone is more than welcome to correct me... But it is worthwhile to note that IMO the article distances itself from the idea that intelligence, sentience and consciousness develop from purely cognitive process - and that which is the basis of AI, "super AI" and whatever else.
Instead the article's systematic approach makes a lot of sense both from an evolutionary point of view and in providing an objective account of our subjective experience. And in those two aspects I still cannot write it off as "hogwash" and a step forward in accounting for Chalmers' hard problem.
The "fundamental feeling of being" I think refers to knowing (or believing) of ones own existence rather than an objective account of the existence of some other thing - which probably requires reference to ones own existence in any case. It is always a logical minefield, however, where language and even tenses (see above) can lead to significant misunderstandings.
I am not sure Gnobuddy's hypothesis offers any hypothesis at all, although anyone is more than welcome to correct me... But it is worthwhile to note that IMO the article distances itself from the idea that intelligence, sentience and consciousness develop from purely cognitive process - and that which is the basis of AI, "super AI" and whatever else.
Instead the article's systematic approach makes a lot of sense both from an evolutionary point of view and in providing an objective account of our subjective experience. And in those two aspects I still cannot write it off as "hogwash" and a step forward in accounting for Chalmers' hard problem.
Fundamental concepts are always like that. We all know what these things are, yet we cannot explain them in terms of other concepts. The apparent failure to explain is not because we don't know what they are, but because they are fundamental. Most attempts at explaining them usually end up as merely playing philosophers' games with the meanings of words, or saying what they are not rather than what they are.gpauk said:
'Existence' is another one. I know I exist, yet I cannot explain existence without using the term existence which I am trying to explain. Existence is thus a fundamental concept. For me a much more interesting question is "Why do I exist?", although many folk would regard that as a meaningless question because their assumptions do not allow for it.
Surely jumping to your question "Why do I exist?" assumes the fundamental nature of your existence? In the article there is a clear inference that your 'self' (that which recognises its own existence in its reality) is an emergent, learned property developed by the interaction of discrete elements within the central nervous system. Such an interaction implies our existence is not a fundamental concept (if we accept our reality), but capable of rational explanation. As discussed above, the question "Why do I exist?" also presumes existence in the present tense.
It's a pity that this thread has not attracted more comment
I also had to read the article a number of times to "understand" it too (and TBH I still don't completely). But I do have my own cognitive bias to admit here as references 7 (Damasio) and 9 (Hawkins) are IMHO the most insightful in the field and I was therefore very amenable to read an article that attempts their unification.
The article also reads more like pseudo-technical sales blurb than it does a scientific paper.
As far as I can see, the article closely reflects the current state of our understanding. It is indeed not as scientific as some are, but it is clearly aimed at the lay person to understand what would otherwise be a lot of jargon.
I always chuckle when people react: 'it's a bunch of hogwash - let me tell you how it really is'. Yeah. Really.
Jan
Yes.soundbloke said:
People infer 'emergent' properties when they see a property which they cannot explain from a reductionist point of view. Self-awareness certainly exists. To infer that this is an 'emergent' property from the central nervous system mean that other possible explanations have already been discarded in the initial assumptions made. The result is usually a circular argument:
I have a self.
The only explanation must come from the central nervous system only, because there isn't anything else.
Therefore 'self' 'emerges' from the central nervous system.
Having 'explained' 'self', I have no need for anything other than the central nervous system.
The conclusion is thus contained in the initial assumptions.
Yes, that is exactly what my hypothesis addresses. If a blindfolded person were in a car and noticed certain types of vibration, occasional G-forces, certain types of sounds, she would conclude that the car was being driven along city streets.
My hypothesis is that, in much the same way, if your brain (totally blind inside your skull) receives the "noise" of a normally operating nervous system, it thinks "Stuff is happening inside this body, I can feel my fingers and toes, I can hear and touch and smell stuff, so I exist". In other words, there is nothing profound about the belief "I exist" at all; it is no more profound than the belief "this car is running along city streets now".
And this is why I don't usually waste my time on philosophical discussions. Because of the nature of the subject, I can only offer an idea, not a testable hypothesis. You immediately shoot it down, but you cannot prove me wrong, any more than I can prove I'm right, because neither my idea, nor your opinion of it, is testable in any objective way.
So now we're all ready to waste another five hundred years arguing about how many angels can dance upon the head of a pin...and, at the end of it, everyone walks away, no wiser than before.
That's what makes philosophy hogwash, the mental equivalent of empty calories. Everyone has an opinion or belief, but none of them are testable, and therefore, all of them are ultimately worthless.
-Gnobuddy
Now if we can also train the cat to play dead if a tree falls in the forest, I guess we can put that conundrum to bed as well.
-Gnobuddy
"Because of the nature of the subject, I can only offer an idea, not a testable hypothesis. You immediately shoot it down, but you cannot prove me wrong, any more than I can prove I'm right, because neither my idea, nor your opinion of it, is testable in any objective way."
My intention was not to "shoot it down" but to point out that describing a flow of sensory information is not sufficient to explain the emergence of a self or of a subjective experience:
"Stuff is happening inside this body, I can feel my fingers and toes, I can hear and touch and smell stuff, so I exist"
This presumes an "I" exists already and does not account for the nature of subjective feeling. Feeling and sensing are not the same thing. What is the difference between this notion and assuming my laptop is feeling the sound pressure changes incident at its microphone diaphragm?
The article makes a link (I believe for the first time?) between a sensed "energetic" reality and a model component of our emotively respondent metabolic state. It is not proposing a spontaneous or miraculous appearance of a self as per the normal discussions of AI, "super AI" (as it has now been dubbed) or as described by Gnobuddy.
Instead (I think the article says):
- It proposes a cognitive model of the world in which (because of the "reciprocating" sensory information) an energetic model component representing the emotive centre exists in a like energetic model of its reality and so is inherently part of it.
- In that model there is an inherent inference of a causal agent (just like from other external causal agents such as where apples fall to the ground) which becomes the basis of the "self" but from which a whole perception is generated around.
- And when generating a whole perception, the model infers the self having 'witnessed' this model of reality, which is a restrospective 'view' inferred from what we have learned to expect the self would view, if only in the conscious instants punctuating our unconscious 'normality'.
I agree the article is "fluffy" although not as fluffy as what I have just written! Hopefully somebody will put it into words better than I have here. But I do see a hint of rationality that would move this from philosophical conjecture. (And I really like the idea that colour really does exist even if it is actually not part of the E/M spectrum).
"Everyone has an opinion or belief, but none of them are testable, and therefore, all of them are ultimately worthless."
The article claims that because perception is not necessarily private in a machine, then machine sentience is testable (?). As I first replied in this thread, I think it is whether the claimed machine sentience (if it is proven) is emulated sentience or of the same basis as our own subjective feeling of being in the world around us.
"I guess we can put that conundrum to bed as well"
It just might be the next step to doing exactly that, but hopefully not towards developing a fridge that can feel if its light is on and imagine what the cat is experiencing.
My intention was not to "shoot it down" but to point out that describing a flow of sensory information is not sufficient to explain the emergence of a self or of a subjective experience:
"Stuff is happening inside this body, I can feel my fingers and toes, I can hear and touch and smell stuff, so I exist"
This presumes an "I" exists already and does not account for the nature of subjective feeling. Feeling and sensing are not the same thing. What is the difference between this notion and assuming my laptop is feeling the sound pressure changes incident at its microphone diaphragm?
The article makes a link (I believe for the first time?) between a sensed "energetic" reality and a model component of our emotively respondent metabolic state. It is not proposing a spontaneous or miraculous appearance of a self as per the normal discussions of AI, "super AI" (as it has now been dubbed) or as described by Gnobuddy.
Instead (I think the article says):
- It proposes a cognitive model of the world in which (because of the "reciprocating" sensory information) an energetic model component representing the emotive centre exists in a like energetic model of its reality and so is inherently part of it.
- In that model there is an inherent inference of a causal agent (just like from other external causal agents such as where apples fall to the ground) which becomes the basis of the "self" but from which a whole perception is generated around.
- And when generating a whole perception, the model infers the self having 'witnessed' this model of reality, which is a restrospective 'view' inferred from what we have learned to expect the self would view, if only in the conscious instants punctuating our unconscious 'normality'.
I agree the article is "fluffy" although not as fluffy as what I have just written! Hopefully somebody will put it into words better than I have here. But I do see a hint of rationality that would move this from philosophical conjecture. (And I really like the idea that colour really does exist even if it is actually not part of the E/M spectrum).
"Everyone has an opinion or belief, but none of them are testable, and therefore, all of them are ultimately worthless."
The article claims that because perception is not necessarily private in a machine, then machine sentience is testable (?). As I first replied in this thread, I think it is whether the claimed machine sentience (if it is proven) is emulated sentience or of the same basis as our own subjective feeling of being in the world around us.
"I guess we can put that conundrum to bed as well"
It just might be the next step to doing exactly that, but hopefully not towards developing a fridge that can feel if its light is on and imagine what the cat is experiencing.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.