Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!cs.utexas.edu!uunet!tdatirv!sarima From: sarima@tdatirv.UUCP (Stanley Friesen) Newsgroups: comp.ai.philosophy Subject: Re: Testing for machine consciousness Message-ID: <21@tdatirv.UUCP> Date: 9 Oct 90 16:16:42 GMT References: <3499@media-lab.MEDIA.MIT.EDU> <1990Oct4.154655.23004@canon.co.uk> <7@tdatirv.UUCP> <1990Oct8.120927.8648@canon.co.uk> Reply-To: sarima@tdatirv.UUCP (Stanley Friesen) Organization: Teradata Corp., Irvine Lines: 59 In article <1990Oct8.120927.8648@canon.co.uk> rjf@canon.co.uk writes: >In article <7@tdatirv.UUCP> I (Stanley Friesen) wrote: >>The counter-argument is simple. This is also true of the human brain! >>No individual neural mechanism in the brain is conscious, nor is any >>individual subsystem in the brain conscious. ... > >What I meant was, we associate consciousness particularly with >short-term memory, for instance, but it would (I guess) be relatively >easy to implement a machine with short-term memory which functioned >just like ours, though the machine was not conscious. The same >argument applies to any other function. So, if not by its functioning, >how else can we tell whether a machine is conscious? If it functions just like our brain it *is* conscious! Or do you mean that it has short-term memory just like ours, but none of our other functionality? In that case the issue of emergence comes back in. Consciousness probably is an emergent property of the sum total of all of our functionality (or at least a very large subset thereof). Thus taking in one sub-component, like short-term memory, and expecting it to be 'conscious' is silly. Consciousness appears to be based on a complex interaction amoung: internal world models, self-monitoring, decision-making processes, spontaneous learning, abstraction, and perhaps other things. Short-term memory may well be a critical component, giving rise to the sense of continuity necessary for a sense of self, but it is scarcely a defining feature. >>Because we do not agree that no current machine is conscious - we all agree >>that the human machine is indeed conscious. > >No we don't. Because (a) some people would argue that maybe we're not >really conscious, we just think we are (I personally do not think this >position worth dealing with) and (b) you have just redefined >"machine". In such discussions we have to distinguish between natural >human beings on one hand and artefacts made by them on the other, in >order to compare their qualities. O.K., I will make the same point in another way. I know of no mechanism within the human brain that is not strictly physical in nature. Thus an exact copy of a human could, in theory, be constructed (a machine). Since this exact copy is indistinguishable in any way from a naturally born human, we can, as a shortcut, say that humans are 'machine-like' in construction. Thus, if we are conscious, and that consciousness is based in our machine-like body functioning, then a machine may be conscious. [I base this on years of study as a biologist]. In short, I think the distinction you are making between 'machine' and 'human' is largely artificial, it is based on a false dualism. >BTW, I'd still be interested in hearing whether anyone has a test for >machine consciousness.. Now, as for testing a machine for consciousness, that is harder. It is quite a different question than whether a machine can be conscious. [We were denying that blacks were conscious(had a soul) for many years, despite the contrary evidence, so how are we going to be objective about a *machine*!] -- --------------- uunet!tdatirv!sarima (Stanley Friesen)