X-Message-Number: 32932 Date: Wed, 13 Oct 2010 14:41:37 -0700 (PDT) From: 2Arcturus <> Subject: experimental validation of uploading --0-1486200287-1287006097=:3767 Sorry - I meant to respond to this in a more timely fashion, but I couldn't... I think you're right, Daniel. Experiments with neural prosthetics are going to tell us a lot, and provide us with evidence as to whether machines and digital processing can substitute for some parts of the brain. As you noted, the evidence is even already coming in, and it is coming in in the affirmative. Of course, anti-uploaders might back up their argument and argue that the primary visual cortex, while interfacing with the part of the brain that gives rise to conscious, is not the part of the brain that gives rise to consciousness. The trouble with extending the experiment to include more and more of the brain would be that anti-uploaders would say that at some (undefined) point, the crucial part of the brain that supports consciousness would get swapped out with a device that merely supports a "philosophical zombie"/p-zombie. They might say - well, the subject is reporting being consciousness, but that is just the p-zombie glibly misrepresenting itself as being conscious. I am convinced the anti-uploaders' survival/doubling issue is a simple philosophical error, but the issue of whether machines can support consciousness the way the brain does is a little more challenging. Part of the problem is that we have no science of how the brain supports consciousness. I am convinced that someday we will have that, because in theory it should be possible to test the brains of living human subjects at any scale, from brain-area to the molecular level, for example, by temporarily silencing or reversibly-modifying certain areas and seeing what the results are to consciousness, as reported by the subjects. This should bring about a whole new field of science within neuroscience and allow at least some provisional theories about how matter such as the brain relates to the subjective consciousness and how it gives rise to it. This is the kind of information that could be brought to bear on the issue of consciousness "in silicon" - not just the Turing test, but a test that does look inside the box and examines how a purported mind works and what it is doing (structure and dynamics) and how that plausibly relates to inferred consciousness. I have a suspicion that truly simulating verisimilitudinous consciousness (Turing-test level) in a non-conscious system, especially by a sequential formalism, would be much harder than simply building an actually conscious system. Surprise - nature, the blind watchmaker, discovered the shortcut to making a conscious-looking animal - by making a conscious animal! Even if a p-zombie were possible, we might be able to discern that it was a p-zombie, by looking inside its 'brain' and noticing the vast, circumlocutious complexity of its attempts to simulate consciousness, compared to the simpler, more direct way of the brain. Of course, even a science of consciousness, an "objective/subjective" science, wouldn't answer radical (paranoid?) skeptics who would reject all the evidence of the new science by doubting the subjective consciousness of the test subjects, and every other living human being except themselves. After all, the only evidence we have of subjective consciousness is of our own, by definition. The critics would have to decide where to draw the line between suspecting that everyone except themselves was a p-zombie, or believing that anything that looks conscious is conscious (the 'hard AI'/Turing test approach that Searle rails against). And this is why I don't think it's 'even money' whether machine consciousness is possible. Believing that it is possible is simply extending belief that what the brain is doing to make consciousness is something procedurally possible in the real world and that it can be understood and that it does not bear some sort of unique relationship to some unknown characteristic of the brain, supported in some sort of unknown area of the brain. Conscious animals evolved from inanimate matter, in the stuff of inanimate matter. Occam's razor brings the burden of proof to those who would suppose, without warrant, that there is something unknown and magical, some tertium quid/quintessence, that lets a brain make consciousness and that keeps anything else, including a machine, from making it. What is it and why should we think so? Since I can't think of anything else or a reason for it, I don't suppose that, although of course I admit it there is a lot left to learn. >>> Message #32895 From: Daniel Crevier <> References: <> Subject: experimental validation of uploading Date: Fri, 01 Oct 2010 12:28:37 -0400 I'd like to come back to my posting No. 32877 of september 28. It didn't get any reaction, so let me try to add some zest to it. I think this posting shows that the theory that uploading preserves consciousness is falsifiable : it can be proven or disproven experimentally whether the uploaded person is still conscious, and not a zombie. The experiment I suggest is the following. Do only a partial uploading of a subject : only replace the primary visual cortex by a digital circuit. It has been shown that loss of this area leaves stroke victims consciously unaware of visual information, even if some visual processing seems to be still occuring in other parts of the brain. So if there is a short list of brain areas intimately related to consciousness, then the primary visual cortex is very much in it. Note also that loss of this brain area leaves all other mental abilities intact: the subjects can still think, talk, and report on their internal states. If this brain area is digitized, then one of two things could happen. First, the subjects could report that they still have normal vision. According to uploaders, this is the expected outcome, since we assume that the digitized part of the brain will interface with the rest in exactly the same way as the pre-existing biological one. Alternatively, the subject could report that he or she has become blind. This should be the outcome expected by anti uploaders, since according to them the digital circuitry lacks whatever magic is required to induce consciousness. Right now, this can't be done in practice, but we'll get there. The evidence so far, though, is pretty much in favor of uploaders: for example, retinas are made of neurons, and electronic replicas have been made; subjects were quite aware of their inputs. So, after all, belief in uploading may not be a matter of personal choice or values. It may be objectively verifiable. Any comments? Daniel Crevier Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=32895 --0-1486200287-1287006097=:3767 Content-Type: text/html; charset=iso-8859-1 [ AUTOMATICALLY SKIPPING HTML ENCODING! ] Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=32932