Re: Can machines be conscious?
Posted: 01 Mar 2010 17:37
Well, to be clear, I don't believe in a 'magic spark'. I think we are very complex, adaptable primates ... the best around ... but in the grand scheme of things not *that* different from the other intelligent species on the planet.
If there is a single adaptation separating us from chimps, orcas, dolphins, etc., to me it is the ability of our brains to ask "what if," to think on a level of abstraction. Although we may ask a computer to consider a billion different scenarios in reaching a calculation, that is not really the "what if" I'm referring to. We cannot tell a computer, today, "well, use your imagination."
"What if" has been a good and bad adaptation for humans. It lets us build tools, adapt to any environment, solve complex problems. It also occasionally makes us psychotic: "what if my wife is sleeping with the mailman ..." or "what if [insert deity] wants me to kill infidels."
But I don't think anything magical brought us to that place. It is a function of our brain's complexity and the survival problems we were faced with. The ability to ask "what if" was a huge adaptive advantage over creatures that couldn't ask it, but it may be just as much of a freak in nature as a giraffe's neck. Ultimately I think a deeper understanding of what it is about our brain, structurally, that lets us both accurately perceive the world around us (important to avoid being eaten by predators) and yet consider completely imaginary scenarios (dangerous if you did nothing but live in that imaginary world) will lead us to the ability to 'virtualize' this process and create a machine with similar capabilities.
But as I said earlier, I think we won't understand it as nearly well as we expected once we achieve that.
If there is a single adaptation separating us from chimps, orcas, dolphins, etc., to me it is the ability of our brains to ask "what if," to think on a level of abstraction. Although we may ask a computer to consider a billion different scenarios in reaching a calculation, that is not really the "what if" I'm referring to. We cannot tell a computer, today, "well, use your imagination."
"What if" has been a good and bad adaptation for humans. It lets us build tools, adapt to any environment, solve complex problems. It also occasionally makes us psychotic: "what if my wife is sleeping with the mailman ..." or "what if [insert deity] wants me to kill infidels."
But I don't think anything magical brought us to that place. It is a function of our brain's complexity and the survival problems we were faced with. The ability to ask "what if" was a huge adaptive advantage over creatures that couldn't ask it, but it may be just as much of a freak in nature as a giraffe's neck. Ultimately I think a deeper understanding of what it is about our brain, structurally, that lets us both accurately perceive the world around us (important to avoid being eaten by predators) and yet consider completely imaginary scenarios (dangerous if you did nothing but live in that imaginary world) will lead us to the ability to 'virtualize' this process and create a machine with similar capabilities.
But as I said earlier, I think we won't understand it as nearly well as we expected once we achieve that.