anhinga_anhinga (![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png)
anhinga_anhinga) wrote2010-03-28 11:30 pm
anhinga_anhinga) wrote2010-03-28 11:30 pm
Entry tags:
"Hybrid Consciousness": an object for thought experiments
Imagine a hypothetical situation where a brain and an artificial device are coupled in such a way that a single consciousness (a single subjective reality) is associated with them. Let's call this a "hybrid consciousness" (perhaps, a better name can be suggested for this).
One can do various interesting thought experiments with such an object. The contexts where this might be useful include the abstract topics such as "Hard Problem", technological singularity, or uploading, and more practical topics, for example, something like "if I could change a way I am [reading mathematical papers/writing computer code/...], how would I want to do that, and what would I want to feel in the process of doing that".
1) In context of the "Hard Problem", I think the ability to achieve a situation like this would be the best test of our understanding of the "Hard Problem" (and would make the subjective realm much more amenable to tests and experiments).
Here we must assume that a sufficiently non-trivial part of a given subjective personality is associated with the artificial device (and, for example, when the body gets sleepy, we are likely to observe the shift of the subjective towards what is mostly running on the artificial substrate; and if the coupling remains during sleep, we might get some version of lucid dreams as part of the overall activity).
2) In context of mind uploading, this seems to provide a much saner approach that what people usually discuss; an example of an approach along these lines is here:
http://www.aleph.se/Trans/Global/Uploading/gupload.html
3) In more practical contexts, the question what would I want to feel in the process of doing [A/B/C/writing computer code/reading mathematical papers/...] makes sense on its own and might be useful to consider.
4) We can also ignore the fact that we don't know whether one can use an ordinary computer as a device coupled in this way to a human brain, and even if the answer is positive, that we don't know what would it take to achieve a coupling like this (both depend on what the answer to the "Hard Problem" really is), and ignoring our lack of knowledge here we can try to meditate on how we might want to program the computer to do this or that, if such-and-such layer of the processing were conscious. (Ignore the fact that without further progress on the "Hard Problem" we don't know how to distinguish between conscious and unconscious processing; and decide that we have some freedom is designating some views into the running system as conscious.)
5) One also has some freedom to run things slower or faster in the artificial device, although if one wants to preserve coupling with the biological there are probably some constraints. When I start to think/introspect about accelerating some parts of the train of conscious thought relative to other parts, what I mostly feel is cognitive dissonance, so I am not sure whether this direction of thinking is fruitful (в этом месте, оно несколько сносит крышу).
I am not sure how much of the above makes sense, but I was thinking about this topic during the previous week and have not lost interest, so I decided to share this.
One can do various interesting thought experiments with such an object. The contexts where this might be useful include the abstract topics such as "Hard Problem", technological singularity, or uploading, and more practical topics, for example, something like "if I could change a way I am [reading mathematical papers/writing computer code/...], how would I want to do that, and what would I want to feel in the process of doing that".
1) In context of the "Hard Problem", I think the ability to achieve a situation like this would be the best test of our understanding of the "Hard Problem" (and would make the subjective realm much more amenable to tests and experiments).
Here we must assume that a sufficiently non-trivial part of a given subjective personality is associated with the artificial device (and, for example, when the body gets sleepy, we are likely to observe the shift of the subjective towards what is mostly running on the artificial substrate; and if the coupling remains during sleep, we might get some version of lucid dreams as part of the overall activity).
2) In context of mind uploading, this seems to provide a much saner approach that what people usually discuss; an example of an approach along these lines is here:
http://www.aleph.se/Trans/Global/Uploading/gupload.html
3) In more practical contexts, the question what would I want to feel in the process of doing [A/B/C/writing computer code/reading mathematical papers/...] makes sense on its own and might be useful to consider.
4) We can also ignore the fact that we don't know whether one can use an ordinary computer as a device coupled in this way to a human brain, and even if the answer is positive, that we don't know what would it take to achieve a coupling like this (both depend on what the answer to the "Hard Problem" really is), and ignoring our lack of knowledge here we can try to meditate on how we might want to program the computer to do this or that, if such-and-such layer of the processing were conscious. (Ignore the fact that without further progress on the "Hard Problem" we don't know how to distinguish between conscious and unconscious processing; and decide that we have some freedom is designating some views into the running system as conscious.)
5) One also has some freedom to run things slower or faster in the artificial device, although if one wants to preserve coupling with the biological there are probably some constraints. When I start to think/introspect about accelerating some parts of the train of conscious thought relative to other parts, what I mostly feel is cognitive dissonance, so I am not sure whether this direction of thinking is fruitful (в этом месте, оно несколько сносит крышу).
I am not sure how much of the above makes sense, but I was thinking about this topic during the previous week and have not lost interest, so I decided to share this.
no subject
Photoreceptors --- can we place THAT many?
For animals, I do not see interesting experiments. The only value of these efforts is developing a harmless technique that could be applied to humans... Thus I would not expect this to happen in less than 50 years...
no subject
Yes, this is done via genetic engineering. At least the recent talks at MIT were sounding very upbeat. They placed enough to control the animals by turning on small lights inside there skulls.
> It is more interesting to connect two mature persons
>
> The only value of these efforts is developing a harmless technique that could be applied to humans
Then you surely want to do it via an interface to an electronic device, or something like that. "Harmless" first of all implies "reversible". The last thing you want, in this sense, is to have physical synapses between their neurons.
no subject
Surprisingly, it turns out not to be always the case. It seems that these Siamese twins are sufficiently joined to share each other sensations, but remain separate persons:
http://www.nytimes.com/2011/05/29/magazine/could-conjoined-twins-share-a-mind.html?nl=todaysheadlines&emc=tha210&pagewanted=all
"[...]Their brain images reveal what looks like an attenuated line stretching between the two organs, a piece of anatomy their neurosurgeon, Douglas Cochrane of British Columbia Children’s Hospital, has called a thalamic bridge, because he believes it links the thalamus of one girl to the thalamus of her sister. The thalamus is a kind of switchboard, a two-lobed organ that filters most sensory input and has long been thought to be essential in the neural loops that create consciousness. Because the thalamus functions as a relay station, the girls’ doctors believe it is entirely possible that the sensory input that one girl receives could somehow cross that bridge into the brain of the other. One girl drinks, another girl feels it. [...]"
no subject
Since time passed [since our communication] I had also another thought about all this. While natural sources of information (eyes, ears,...) is what we are doing for years, a different type of device (say, external memory or a direct connection to someone else's brain) may be more difficult to handle. Probably genetic evolution is not absolutely necessary; the "cultural" development of concepts around such connections can replace it. However, we may still need several generations to learn how to use and even feel it properly, because proper hormones have to be released, etc.
On the other hand, if two mature humans are connected so strongly that they couldn't distinguish whose memory it is and whose feeling it is, it will almost certainly create fatal madness --- just because we are incapable of handling this in our "culture", no experience, no learning how to handle it from the childhood, etc.
All this makes real experiments even more difficult.