4 ms·

I made a comment that touches on this more directly elsewhere in the thread: https://news.ycombinator.com/item?id=15245740 https://news.ycombinator.com/item?id

I made a comment that touches on this more directly elsewhere in the thread: https://news.ycombinator.com/item?id=15245740 https://news.ycombinator.com/item?id=15245740 I'm curious what you think of it.

monktastic1 9y ago
Thanks for sharing that. The trouble I have is that "modeling" can happen in the absence of what I'm calling "experience." Given that one can happen without the other, they cannot be equivalent. So the trouble always seems to come back to giving a formal definition for "experience." I admit that I can't, but I can give a procedure (roughly, meditation) that seems to get other people to a point where they say "oh, that." For a particular reason that I seem unable to communicate, at that point it becomes less meaningful to say that experience is "caused by" physical things as opposed to merely correlated with them. The more clearly I see what it is I'm trying to explain, the more clear it becomes why there's an "explanatory gap." I'm sure this is terribly unsatisfying and sounds like hand-waving woo, and I'm sorry I can't do better right now.
- dwaltrip 9y ago
  > The trouble I have is that "modeling" can happen in the absence of what I'm calling "experience." This is the big claim. I'm highly skeptical. The way I see it, the self-modeling that occurs in biological organisms _is_ the conscious experience. They are one and the same. Each momentary, instantaneous experience is the latest output of the self-modeling process. As another angle, this seems closely related to identity. I see identity as being intertwined with, perhaps even stemming from, the accumulated impact and memory of these models. At least, the portions that refer to us. The boundaries can be quite fuzzy -- we are partially defined by our friends, etc. Each instantaneous model changes the brain a bit, and over time this process constructs one's identity (starting with a tiny kernel at birth) In summary, maybe we can say that "I" is simply the brain's current best guess at what the entity that houses it actually is, and that conscious experience is the brain's best guess at what this "I" is currently experiencing (caveat: these guesses are biased towards survival -- accuracy is not the only concern). > but I can give a procedure (roughly, meditation) that seems to get other people to a point where they say "oh, that." Meditation is very interesting. I've done some, and I'd like to dig into it more. However, I interpret the experience differently. > I'm sure this is terribly unsatisfying and sounds like hand-waving woo, and I'm sorry I can't do better right now. Not at all! I'm enjoying the discussion. Thanks for being frank.
monktastic1 9y ago
Replying here, since I don't have the option on your latest. >> The trouble I have is that "modeling" can happen in the absence of what I'm calling "experience." > This is the big claim. I'm highly skeptical. Sure, it depends on what the word "modeling" means. The way I've always used it, it applies to lots of existing systems, including robots. Are they conscious? Maybe. I don't know. It seems easy to build a very simple system that trivially "models" itself. Does that mean it's conscious? > As another angle, this seems closely related to identity. As it happens, (lack of) identity is a critical realization on the Buddhist path (among others). If we look carefully enough, we can't find one. Instead, we find something like this. Take a perception like sound. Instead of discovering a self that hears it, it's more like the sound itself is made of the sheer fact of subjectivity; i.e., "made of" consciousness. There is no "I" that is conscious (though of course the concept still remains useful). Instead, consciousness is that property which [everything I could possibly call or think of as] the world is "made of." > Meditation is very interesting. I've done some, and I'd like to dig into it more. However, I interpret the experience differently. That's fair. There are different ways and degrees of doing it. Ultimately, I think the distinction comes down to whether we treat "consciousness" as an abstract, objective property (that I assume others have), or this "sheer fact of subjectivity" that seems impossible to communicate (that I can know I have). One of these I'm happy to reduce to other objective properties, and the other I am not.
- dwaltrip 9y ago
  > Sure, it depends on what the word "modeling" means. The way I've always used it, it applies to lots of existing systems, including robots. Are they conscious? Maybe. I don't know. It seems easy to build a very simple system that trivially "models" itself. Does that mean it's conscious? No, it would not be. I think everyone agrees what humans can do is fairly distinct (and many animals, to a much lesser degree). If it wasn't clear, I'm arguing it can be understood as an incredibly sophisticated form of modeling. A toy system can have a model of itself, but it will be a very weak and limited model with essentially no broader context. The models that the human brain constructs are immensely richer, with many orders of magnitude more complexity. I would argue that significant awareness requires this depth and richness. > Take a perception like sound. Instead of discovering a self that hears it, it's more like the sound itself is made of the sheer fact of subjectivity; i.e., "made of" consciousness. We are now getting into territory that feels more difficult to address. I don't know how to meaningfully interpret this claim that sound itself is made of consciousness. Do you mean to say that the phenomenon going on in the human brain is the actually same thing that is occurring in a wave of air molecules? I'd be curious to see what evidence you believe there is for this idea, beyond your personal meditation sessions.

monktastic1 9y ago

Thanks for sharing that. The trouble I have is that "modeling" can happen in the absence of what I'm calling "experience." Given that one can happen without the other, they cannot be equivalent. So the trouble always seems to come back to giving a formal definition for "experience." I admit that I can't, but I can give a procedure (roughly, meditation) that seems to get other people to a point where they say "oh, that." For a particular reason that I seem unable to communicate, at that point it becomes less meaningful to say that experience is "caused by" physical things as opposed to merely correlated with them. The more clearly I see what it is I'm trying to explain, the more clear it becomes why there's an "explanatory gap." I'm sure this is terribly unsatisfying and sounds like hand-waving woo, and I'm sorry I can't do better right now.

dwaltrip 9y ago
> The trouble I have is that "modeling" can happen in the absence of what I'm calling "experience." This is the big claim. I'm highly skeptical. The way I see it, the self-modeling that occurs in biological organisms _is_ the conscious experience. They are one and the same. Each momentary, instantaneous experience is the latest output of the self-modeling process. As another angle, this seems closely related to identity. I see identity as being intertwined with, perhaps even stemming from, the accumulated impact and memory of these models. At least, the portions that refer to us. The boundaries can be quite fuzzy -- we are partially defined by our friends, etc. Each instantaneous model changes the brain a bit, and over time this process constructs one's identity (starting with a tiny kernel at birth) In summary, maybe we can say that "I" is simply the brain's current best guess at what the entity that houses it actually is, and that conscious experience is the brain's best guess at what this "I" is currently experiencing (caveat: these guesses are biased towards survival -- accuracy is not the only concern). > but I can give a procedure (roughly, meditation) that seems to get other people to a point where they say "oh, that." Meditation is very interesting. I've done some, and I'd like to dig into it more. However, I interpret the experience differently. > I'm sure this is terribly unsatisfying and sounds like hand-waving woo, and I'm sorry I can't do better right now. Not at all! I'm enjoying the discussion. Thanks for being frank.

monktastic1 9y ago

Replying here, since I don't have the option on your latest. >> The trouble I have is that "modeling" can happen in the absence of what I'm calling "experience." > This is the big claim. I'm highly skeptical. Sure, it depends on what the word "modeling" means. The way I've always used it, it applies to lots of existing systems, including robots. Are they conscious? Maybe. I don't know. It seems easy to build a very simple system that trivially "models" itself. Does that mean it's conscious? > As another angle, this seems closely related to identity. As it happens, (lack of) identity is a critical realization on the Buddhist path (among others). If we look carefully enough, we can't find one. Instead, we find something like this. Take a perception like sound. Instead of discovering a self that hears it, it's more like the sound itself is made of the sheer fact of subjectivity; i.e., "made of" consciousness. There is no "I" that is conscious (though of course the concept still remains useful). Instead, consciousness is that property which [everything I could possibly call or think of as] the world is "made of." > Meditation is very interesting. I've done some, and I'd like to dig into it more. However, I interpret the experience differently. That's fair. There are different ways and degrees of doing it. Ultimately, I think the distinction comes down to whether we treat "consciousness" as an abstract, objective property (that I assume others have), or this "sheer fact of subjectivity" that seems impossible to communicate (that I can know I have). One of these I'm happy to reduce to other objective properties, and the other I am not.

dwaltrip 9y ago
> Sure, it depends on what the word "modeling" means. The way I've always used it, it applies to lots of existing systems, including robots. Are they conscious? Maybe. I don't know. It seems easy to build a very simple system that trivially "models" itself. Does that mean it's conscious? No, it would not be. I think everyone agrees what humans can do is fairly distinct (and many animals, to a much lesser degree). If it wasn't clear, I'm arguing it can be understood as an incredibly sophisticated form of modeling. A toy system can have a model of itself, but it will be a very weak and limited model with essentially no broader context. The models that the human brain constructs are immensely richer, with many orders of magnitude more complexity. I would argue that significant awareness requires this depth and richness. > Take a perception like sound. Instead of discovering a self that hears it, it's more like the sound itself is made of the sheer fact of subjectivity; i.e., "made of" consciousness. We are now getting into territory that feels more difficult to address. I don't know how to meaningfully interpret this claim that sound itself is made of consciousness. Do you mean to say that the phenomenon going on in the human brain is the actually same thing that is occurring in a wave of air molecules? I'd be curious to see what evidence you believe there is for this idea, beyond your personal meditation sessions.