3 ms·

I think you are suggesting (but let me know if I'm wrong) that consciousness is not a testable or measurable phenomena, and therefore we cannot make scientific

I think you are suggesting (but let me know if I'm wrong) that consciousness is not a testable or measurable phenomena, and therefore we cannot make scientific claims about who/what experiences it. All you can know for certain is that you have it (if you have it). Ok, so you have consciousness; you know it, but I can't be certain. Nevertheless, I hook you up to an EEG machine, and embed electrode arrays in various brain regions. We have a conversation, show you some images, run through some tasks, etc. and then drip some sedative into your IV. Soon you drift into a deep and dreamless sleep. Run the same tasks and indeed you are completely unresponsive. Later you wake up, and I show you how your neural nets respond and oscillate at particular frequencies when u are awake and then show you the marked difference right after you fall asleep. We do this over a bunch of sessions until you yourself can recognize the neural signatures of your conscious awake brain vs your unconsious asleep brain. Would you say in that experiment we did something scientific to study consciousness?

xamuel 8y ago
>I think you are suggesting ... consciousness is not testable Yes, that's right. (Edit: I mean, we don't know how to test it yet.) Re: your experiment: In principle, I could design a dumb machine that you can plug your EEG machine up to, and my machine will play recordings of human brain readings to it. I'll even make my machine detect when sedatives are injected into it, and respond by tapering off transmitting those recordings for awhile, eventually resuming them after awhile, just like a human. Surely you're not going to propose that this bizarre contraption I've just described is conscious :) Here's a way to think of it: you want to design a new Captcha based on EEGs. Now, every computer comes with EEG sensors for use on the new Captchas. How will you prevent spammers from attaching the sensors to a device that produces pre-recorded EEGs.
- subroutine 8y ago
  A key point here is that I'm not talking about any ol' random human. I am talking about you specifically. That is, this experiment is science-couture, just for you. Why is that important? Well, if you indeed have consciousness then you must be willing to admit the experiments we conducted are probing and characterizing bonafide consciousness. No? I have an if-you-concede-that-much-is-true follow-up; but do you?... wrt. a machine designed to mimic the signals output by your brain during awake states vs sleep states (at least as they are interpreted by my electrodes) is irrelevant, i think. We are not performing a turing test here; you know you have consciousness, so you know the data we have gathered on a conscious vs unconsious state is legitimate. Building a device to trick a sensor is trivial for any number of things we can study scientifically, or have engineered to detect and diagnose a known correlate to some phenomena. I feel this is a bit like saying "i can design a machine that tricks your tire pressure sensors so therefore we cant know anything about tire pressure". I'd admit that would be true if we didn't know, when we designed the sensor, if we were working with a tire with the capacity to hold different air pressures.
  - subroutine 8y ago
    Just wanted to tack-on a final thought, related to another HN post from yesterday titled *The Map Is Not the Territory* >The map of reality is not reality. Even the best maps are imperfect. That’s because they are reductions of what they represent. If a map were to represent the territory with perfect fidelity, it would no longer be a reduction [it would be the territory] I think this is relevant when it comes to managing expectations with regard to a description of consciousness. That is, consciousness seems to be a gestalt experience that arises in only the largest, most complex and dynamic biological organs on Earth. So a full empirically-derived description of consciousness, en total, could require a map that is damn nearly the territory. But like the article points out, maps that large are hardly useful. Basically the TLDR is that if you have consciousness, and we systematically probe your brain while you are conscious vs. unconscious, and find some striking differences between the two states, I'd take that as revealing some small, but empirical detail about how conscious vs unconscious brains function. If you collect enough details independently confirmed, we can begin to build a theory about when and how consciousness arises. After that we can examine your brain and my brain and my dog's brain against these theories and arrive at some probability these brains are conscious (at least to what degree they is similar to general human consciousness). So I guess in some ways you are right - we can never know for certain. But that is basically how all of science works. (eg we 'proved' the Higgs boson exists, but only with a certain probability). https://fs.blog/2015/11/map-and-territory/ https://fs.blog/2015/11/map-and-territory/
    - xamuel 8y ago
      We can only speculate at this point. Maybe consciousness is a gestalt experience whose full description would require a map that is nearly the territory. Or maybe there's something simpler to it that we just don't know about yet. Imagine if man had access to nuclear bombs from earliest history. For an extremely long time, nuclear bombs would have seemed just utterly incomprehensible, basically magical. Until we figured out the science behind them, after which point, suddenly they become predictable applications of science.
      - subroutine 8y ago

xamuel 8y ago

>I think you are suggesting ... consciousness is not testable Yes, that's right. (Edit: I mean, we don't know how to test it yet.) Re: your experiment: In principle, I could design a dumb machine that you can plug your EEG machine up to, and my machine will play recordings of human brain readings to it. I'll even make my machine detect when sedatives are injected into it, and respond by tapering off transmitting those recordings for awhile, eventually resuming them after awhile, just like a human. Surely you're not going to propose that this bizarre contraption I've just described is conscious :) Here's a way to think of it: you want to design a new Captcha based on EEGs. Now, every computer comes with EEG sensors for use on the new Captchas. How will you prevent spammers from attaching the sensors to a device that produces pre-recorded EEGs.

subroutine 8y ago
A key point here is that I'm not talking about any ol' random human. I am talking about you specifically. That is, this experiment is science-couture, just for you. Why is that important? Well, if you indeed have consciousness then you must be willing to admit the experiments we conducted are probing and characterizing bonafide consciousness. No? I have an if-you-concede-that-much-is-true follow-up; but do you?... wrt. a machine designed to mimic the signals output by your brain during awake states vs sleep states (at least as they are interpreted by my electrodes) is irrelevant, i think. We are not performing a turing test here; you know you have consciousness, so you know the data we have gathered on a conscious vs unconsious state is legitimate. Building a device to trick a sensor is trivial for any number of things we can study scientifically, or have engineered to detect and diagnose a known correlate to some phenomena. I feel this is a bit like saying "i can design a machine that tricks your tire pressure sensors so therefore we cant know anything about tire pressure". I'd admit that would be true if we didn't know, when we designed the sensor, if we were working with a tire with the capacity to hold different air pressures.
- subroutine 8y ago
  Just wanted to tack-on a final thought, related to another HN post from yesterday titled *The Map Is Not the Territory* >The map of reality is not reality. Even the best maps are imperfect. That’s because they are reductions of what they represent. If a map were to represent the territory with perfect fidelity, it would no longer be a reduction [it would be the territory] I think this is relevant when it comes to managing expectations with regard to a description of consciousness. That is, consciousness seems to be a gestalt experience that arises in only the largest, most complex and dynamic biological organs on Earth. So a full empirically-derived description of consciousness, en total, could require a map that is damn nearly the territory. But like the article points out, maps that large are hardly useful. Basically the TLDR is that if you have consciousness, and we systematically probe your brain while you are conscious vs. unconscious, and find some striking differences between the two states, I'd take that as revealing some small, but empirical detail about how conscious vs unconscious brains function. If you collect enough details independently confirmed, we can begin to build a theory about when and how consciousness arises. After that we can examine your brain and my brain and my dog's brain against these theories and arrive at some probability these brains are conscious (at least to what degree they is similar to general human consciousness). So I guess in some ways you are right - we can never know for certain. But that is basically how all of science works. (eg we 'proved' the Higgs boson exists, but only with a certain probability). https://fs.blog/2015/11/map-and-territory/ https://fs.blog/2015/11/map-and-territory/
  - xamuel 8y ago
    We can only speculate at this point. Maybe consciousness is a gestalt experience whose full description would require a map that is nearly the territory. Or maybe there's something simpler to it that we just don't know about yet. Imagine if man had access to nuclear bombs from earliest history. For an extremely long time, nuclear bombs would have seemed just utterly incomprehensible, basically magical. Until we figured out the science behind them, after which point, suddenly they become predictable applications of science.
    - subroutine 8y ago