7 ms·
Not unless we can reduce self-awareness to an algorithm, which I don't think we can. We can, however, reduce the appearance of self-awareness down to an algorit
by hliyan 8y ago
Not unless we can reduce self-awareness to an algorithm, which I don't think we can. We can, however, reduce the appearance of self-awareness down to an algorithm.
- red75prime 8y agoI think there's nothing in self-awareness beyond specifically structured information processing, and such processing is self-awareness. But this position is hard to argue for or against, so I'll leave it at that.
- kazinator 8y agoThe fascinating thing is that complex information processing in the brain doesn't cease when you're unconscious.
- red75prime 8y agoIt changes though, and significantly so. Even EEG (which isn't neuron level precise) shows the changes.
- Mahn 8y agoThere's no evidence that we cannot reduce self-awareness to an algorithm.
- naasking 8y ago> Not unless we can reduce self-awareness to an algorithm, which I don't think we can. By the Bekenstein Bound, the human brain is at best a finite state automoton. Unless you wish to appeal to some supernatural element in consciousness, like dualism, consciousness is reducible to an algorithm.
- hliyan 8y agoa. I don't think the Bekenstein Bound is relevant here. b. The commenter below has a point. We don't know whether consciousness is theoretically reducible to an algorithm or not. All we know is that we can't at the moment, and we don't even know where to begin. So the answer is unknown.
- naasking 8y ago> a. I don't think the Bekenstein Bound is relevant here. Yes it is. A system with finite volume can only contain finite information. A system with finite information can be encoded as a finite state automoton. Our brain is contained within a finite volume, ergo, our brain is at best a finite state automoton. > We don't know whether consciousness is theoretically reducible to an algorithm or not. We do unless you want to rewrite physics. Which some people do with panpsychism. There seems little compelling reason to though.
- simsla 8y ago> by Bekenstein Bound Can you explain why this would be the case? (Or point me in the right direction.)
- red75prime 8y agoGiven that a brain can be perfectly described by a limited amount of information, you can write an algorithm which perfectly simulates a brain. Consciousness needs to be supernatural to not be present in this simulation. ETA: Algorithm will require quantum computations to be feasible though.
- aninhumer 8y agoThis only proves that behaving identically to a human is reducible to an algorithm. The actual experience of consciousness is something secondary to behaviour. It's possible that an algorithm which behaves as a human experiences consciousness identically to a human, as some fundamental property of reality, or it's possible that it does not. We cannot know, because consciousness is not something that we can observe.
- fvdessen 8y agoHow is the experience of consciousness secondary to behaviour when the concept of consciousness emerged from and influenced our behaviour ? It seems as strange to me to imagine that we could have two identically behaving beings, one with consciousness, and one without; as imagining that we could have identically behaving beings, one with eyesight and one without.
- majewsky 8y agoWhen I think of my mother (e.g. how she would react when I tell her something), that's basically a simulation of her running in my head. Is this simulation self-aware? If not, then that's one way to imagine two identically behaving beings, one with consciousness (my actual mother) and one without (my brain's simulation of her). (For the sake of argument, I assume that I know my mother so well that her behavior will be indistinguishable for all intents and purposes from my simulation of her in my head.)
- fvdessen 8y agoOr maybe you substitute her consciousness with your own.
- red75prime 8y agoIf your model of your mother is aware of her own existence as a model in your brain, wouldn't she beg you to continue modelling her, or mention the fact at least? > I assume that I know my mother so well that her behavior will be indistinguishable for all intents and purposes from my simulation of her in my head. Then you are running perfect simulation of your mother and some parts of her environment (such that she'll not freak out finding herself in nothingness) in your brain, so yes, she'll be conscious.
- simsla 8y agoTuring-completeness might be a sufficient framework for consciousness. (Meaning it would be possible to distill it into an algorithm, even if we have not sufficiently advanced to do so, and even if the slowness of current hardware might make it slow beyond belief.) It might also prove wholly insufficient some day; too constrained to express what we'd call consciousness. The problem, I think, is that we're still so unbelievably far removed from that point (of being able to model and completely probe apart consciousness) that assuming either is just random conjecture.
- vertexFarm 8y agoWe don't even know if the appearance of self-awareness is all that other humans have. We only know for sure of our own self-awareness, since that is precisely the definition of self-awareness. For all we can prove, the appearance of self-awareness is all that is necessary for consciousness. I could be that if you point a webcam at a computer, then that computer gains some rudimentary self-awareness like an insect might have, or not have, we can't really prove it either way. Sentience is not necessarily sapience. It could also be that computer sentience is absurd and rendered impossible by some as of yet unknown mechanic. In either case we have no reason to make assumptions about it.
- mannykannot 8y agoPresented with an example of an algorithm having the appearance of self-awareness, how can you tell that it is not self-aware, yet when you apply the same criteria to humans, you conclude they are self-aware? If you are thinking of Chalmers' p-zombie argument, that does not do the job, because it is, at most, a claim that Chalmers sees the logical possibility of p-zombies as plausible, which, even if one agrees, is a long way from an argument that apparently self-aware algorithms must therefore be faking it. Not that I have an actual example to try this on, which is why the issue is a source of endless debate...
- fvdessen 8y ago> how can you tell that it is not self-aware ? Let the algorithm run for a while and see if it starts asking itself existential questions on the nature of consciousness.
- majewsky 8y agoSo a dog is not self-aware because it's not pondering its own existence? (For all we know.)
- fvdessen 8y agoIt is hard to know for dogs since they lack the language necessary to ponder. But if we have an intelligence with the necessary language to ponder, and we see that it doesn't, then I think it is reasonable to assume that it doesn't. There's an interesting book that uses this argument to advance the idea that consciousness as we understand it now emerged 3000 years ago, as it was only then that humans started to make artworks about pondering on consciousness. (The Origin of Consciousness in the Breakdown of the Bicameral Mind)
- asfasgasg 8y agoP-zombies do appear to ponder, though. They are in every way externally indistinguishable from normal people. That's the whole point of the thought experiment.
- mr_toad 8y ago> which I don't think we can This is also begging the question.