4 ms·

I find it reasonable to assume that for thought processes to occur, there must be some processing, so rocks are out. A more interesting example would be, say,

by cygx 7y ago

I find it reasonable to assume that for thought processes to occur, there must be some processing, so rocks are out. A more interesting example would be, say, trees, which do communicate with each other to some degree. Is that accompanied by some form of awareness? What's the lower bound of complexity at which consciousness kicks in? What about supraorganisms such as insect hives? Human cities? An ecosystem as a whole?

gear54rus 7y ago
Information could be encoded in molecule oscillations and therefore 2 rocks of different temperatures would communicate it when touching (heat transfer would occur). Can this be considered processing?
- cygx 7y ago
  It's doubtful you'll be able to make such a scheme work: Information processing tends to involve storage and retrieval of data. That's not really possible in the thermodynamic regime. You'd have to do all your processing before the information gets lost to thermalization.
  - ghjyui 7y ago
    Image recognition needs no storage or retrieval of data: it's a single pass thru a series of matrix multiplications. Yet, image recognition is the very definition of data processing.
    - cygx 7y ago
      How do you propose to perform calculations without data storage (think registers)?
      - adrianN 7y ago
        Like this for example https://hackaday.com/2019/07/16/neural-network-in-glass-requires-no-power-recognizes-numbers/ https://hackaday.com/2019/07/16/neural-network-in-glass-requ... though I'm not sure that you can really count this as calculation. But I guess that's the point of this discussion.
- hackinthebochs 7y ago
  Heat is entropy, but entropy is basically a measure of disorder/randomness. So when two substances transfer heat, they necessarily are not communicating information, except perhaps that there exists a temperature differential.
skissane 7y ago
See this paper: “If Materialism is True, the United States is Probably Conscious” https://faculty.ucr.edu/~eschwitz/SchwitzAbs/USAconscious.htm https://faculty.ucr.edu/~eschwitz/SchwitzAbs/USAconscious.ht...
- mannykannot 7y ago
  This builds on a basic fallacy: 1. According to materialism, consciousness is purely a result of physical processes. 2. Our consciousness is complex. 3. Therefore, materialism says that any physical entity with a complexity comparable to us is probably conscious. At best, going from the two premises to the conclusion involves a tendentious and useless definition of 'consciousness' that equivocates over the specific features of the consciousness of genus Homo that make it distinctly interesting. Tononi's IIT makes the same mistake. https://www.scottaaronson.com/blog/?p=1823 https://www.scottaaronson.com/blog/?p=1823
  - skissane 7y ago
    So: Consciousness = Complexity + X X = what? And how can we be sure that, whatever the X is, that the United States does not have it?
    - mannykannot 7y ago
      Firstly, consciousness is probably not well-modeled as arbitrary complexity plus X, as the complexity is likely inherent to X, and have a quite specific form. For example, it seems unlikely, on basic information-theoretic grounds, that the notoriously complex turbulent flow of a fluid in a pipe is materially closer to consciousness than laminar flow in that pipe. Secondly, the burden of proof (or just of showing evidence) rests squarely on the person making the claim that the United States probably has it (whatever 'it' happens to be.) It is hard to do that whan you cannot say, with any specificity at all, what X is. If your latest post is intended to be a defense of Schwitzgebel's paper, I think you are missing his point. I think he regards the consciousness of the United States counterfactually, and is attempting to show that materialism leads to an implausible conclusion, but he fails to do so, partly on account of the problem I raised in my original post.
cr0sh 7y ago
> What's the lower bound of complexity at which consciousness kicks in? What about supraorganisms such as insect hives? Human cities? An ecosystem as a whole? There's also the question of the nodes and edges of the graph; does the complexity (or processing capability) of these have any effect on the complexity bounds? That is - compare a neuron as the node (and the dendrites as the "edges") in the larger graph of the neural network that is the human brain. Each node and edge are relatively simple compared to the entire brain. Now - what happens when that brain becomes the node (via a human) and the edges of the graph of interaction in society becomes the internet (social networking, email, etc)? Both are much more complex than the former - so do we need fewer nodes/edges to "kickstart consciousness"? The problem of course is how can we (as individual nodes) perceive this consciousness? How could we ever query it? And - if we could query it - and it perceived us querying it - what would it do? If for instance - you perceived one or more of your neurons suddenly say "Hey - I'm a neuron - can you talk to me? I want to know what you are thinking? Helloooo?" - what would you do? Well - more than a few of us would probably head to the doctor in an attempt to either shut that neuron up or remove it or kill it or...you get the idea. So - if we were to do the same thing...?

by cygx 7y ago

gear54rus 7y ago

Information could be encoded in molecule oscillations and therefore 2 rocks of different temperatures would communicate it when touching (heat transfer would occur). Can this be considered processing?

cygx 7y ago
It's doubtful you'll be able to make such a scheme work: Information processing tends to involve storage and retrieval of data. That's not really possible in the thermodynamic regime. You'd have to do all your processing before the information gets lost to thermalization.
- ghjyui 7y ago
  Image recognition needs no storage or retrieval of data: it's a single pass thru a series of matrix multiplications. Yet, image recognition is the very definition of data processing.
  - cygx 7y ago
    How do you propose to perform calculations without data storage (think registers)?
    - adrianN 7y ago
      Like this for example https://hackaday.com/2019/07/16/neural-network-in-glass-requires-no-power-recognizes-numbers/ https://hackaday.com/2019/07/16/neural-network-in-glass-requ... though I'm not sure that you can really count this as calculation. But I guess that's the point of this discussion.
hackinthebochs 7y ago
Heat is entropy, but entropy is basically a measure of disorder/randomness. So when two substances transfer heat, they necessarily are not communicating information, except perhaps that there exists a temperature differential.

skissane 7y ago

See this paper: “If Materialism is True, the United States is Probably Conscious” https://faculty.ucr.edu/~eschwitz/SchwitzAbs/USAconscious.htm https://faculty.ucr.edu/~eschwitz/SchwitzAbs/USAconscious.ht...

mannykannot 7y ago
This builds on a basic fallacy: 1. According to materialism, consciousness is purely a result of physical processes. 2. Our consciousness is complex. 3. Therefore, materialism says that any physical entity with a complexity comparable to us is probably conscious. At best, going from the two premises to the conclusion involves a tendentious and useless definition of 'consciousness' that equivocates over the specific features of the consciousness of genus Homo that make it distinctly interesting. Tononi's IIT makes the same mistake. https://www.scottaaronson.com/blog/?p=1823 https://www.scottaaronson.com/blog/?p=1823
- skissane 7y ago
  So: Consciousness = Complexity + X X = what? And how can we be sure that, whatever the X is, that the United States does not have it?
  - mannykannot 7y ago
    Firstly, consciousness is probably not well-modeled as arbitrary complexity plus X, as the complexity is likely inherent to X, and have a quite specific form. For example, it seems unlikely, on basic information-theoretic grounds, that the notoriously complex turbulent flow of a fluid in a pipe is materially closer to consciousness than laminar flow in that pipe. Secondly, the burden of proof (or just of showing evidence) rests squarely on the person making the claim that the United States probably has it (whatever 'it' happens to be.) It is hard to do that whan you cannot say, with any specificity at all, what X is. If your latest post is intended to be a defense of Schwitzgebel's paper, I think you are missing his point. I think he regards the consciousness of the United States counterfactually, and is attempting to show that materialism leads to an implausible conclusion, but he fails to do so, partly on account of the problem I raised in my original post.

cr0sh 7y ago

> What's the lower bound of complexity at which consciousness kicks in? What about supraorganisms such as insect hives? Human cities? An ecosystem as a whole? There's also the question of the nodes and edges of the graph; does the complexity (or processing capability) of these have any effect on the complexity bounds? That is - compare a neuron as the node (and the dendrites as the "edges") in the larger graph of the neural network that is the human brain. Each node and edge are relatively simple compared to the entire brain. Now - what happens when that brain becomes the node (via a human) and the edges of the graph of interaction in society becomes the internet (social networking, email, etc)? Both are much more complex than the former - so do we need fewer nodes/edges to "kickstart consciousness"? The problem of course is how can we (as individual nodes) perceive this consciousness? How could we ever query it? And - if we could query it - and it perceived us querying it - what would it do? If for instance - you perceived one or more of your neurons suddenly say "Hey - I'm a neuron - can you talk to me? I want to know what you are thinking? Helloooo?" - what would you do? Well - more than a few of us would probably head to the doctor in an attempt to either shut that neuron up or remove it or kill it or...you get the idea. So - if we were to do the same thing...?