The Problem of Continuity of Conciousness
When talking about the future and transhumanism to people, specifically on the topic of life extension and the possibilities available to us, one of the most interesting debates I always get into is on the topic of “uploading”.
When we talk about the potential in the future of radically extending human life there are 3 major options.
- Extension of biological life, via therapies or processes designed to halt or reverse the aging process allowing a human to live young and healthy indefinitely.
- “Re-Life”, where a complete record of your brain, post death, is imprinted on a younger genetic clone of yourself.
- Uploading, where your mind and consciousness are transferred out of the organic substrate of your brain into an artificial substrate such as a computer, robot or distributed network.
With the last two options discussion always falls to the problem of “continuity of consciousness”. This is the situation where many people feel that an uploaded consciousness would only be a “copy” or a “clone” of their consciousness rather than a continuation of their self. In this post I will discuss this debate and posit my possible solution to the problem.
The main issue facing the concept of uploading is that most proponents would prefer the process to occur pre physical death. e.g they would like to take their current consciousness and move it into a more resilient or robust substrate.
It’s important to note that, at this time, this is of course a total thought experiment as a technology for copying human consciousness into a machine and sustaining that consciousness in a coherent form does not currently exist. Having said that there are many technologies currently already in use or being trialed that allow for the direct reading of neural signals via electrodes implanted in the brain. At the moment these technologies benefit people with brain injuries and other disorders but one day, with higher concentrations of electrodes and greater computing power, these technologies may allow us to read every signal travelling through a healthy adults brain. Allowing us to effectively copy and simulate their running mind.
It has already been predicted by some technologists that by 2050 we will have computers powerful enough to simulate every neuron in a human brain and build and maintain a stable simulation of a working human mind. That mind, in essence, will be “alive” while it’s running. But what if we combine these two technologies. The reading out of a living humans brain and insert that output into a running brain simulation. The simulation “becomes” the person the mind is read from and gains a consciousness of it’s own.
This raises a whole set of moral questions of it’s own; is the new mind a different person? Is it murder to turn it off? Does it diverge from the original donor? Who gets to sleep with their wife and is there a USB attachment for that? These are all questions we may actually have to answer in the not too distant future. However this technology almost certainly will become available, and at that time what do we do about people who wish to undergo the procedure to move their mind into a more robust and essentially immortal format?
The problem then becomes; What about the organic source? If nothing is done, all you’ve really succeeded in doing is cloning your mind. “You” would continue on and eventually die in your organic form. It may be comforting to know that “you” would live on in some other form but for many it would not be the “continuation” they would hope for.
One option is destruction of the organic original as the mind is read out. This has a high “ick” factor in that to most people you would be making a copy and then murdering the original. It’s generally an unsatisfying solution for those wishing to upload one day.
Here we have the core of the problem of continuity of consciousness. How to move your mind from one substrate to another without “interrupting” your consciousness, and is the resulting non-organic consciousness really “alive”? Really “you”?
At this point I would like to present a thought experiment I have heard presented a few times that, I think, presents the “cut off point” of the problem quite well.
Lets theorise that there is a person who has sustained an injury requiring the amputation of their leg and it’s replacement with an artificial prosthetic. Is this person still human? The majority of people I have spoken with would say “of course”… Now lets theorise that they have sustained more severe injuries. They now need an artificial heart and all limbs replaced with prosthetics.
Are they still human?
I would still expect most people to reply in the affirmative to this question.
Now lets theorise that the subject has received a major brain injury that stops signals from their motor cortex getting to the rest of their body. Lets also theorise that an artificial “signalling circuit” exists that we can insert into their brain to allow the upper brain to communicate with the rest of their body. Essentially replacing the broken motor portion of the subjects brain. Are they still human?
Part of their brain is now artificial and not organic at all. But it’s only the part that controls signals to his body. I would now also suspect that most people would reply “yes” that they are still human.
Now lets assume The subject has taken a very unlikely injury to their frontal cortex. They essentially still have their base memories and most of their brain is unharmed but there is no part of the mind left to process these memories or interpret the outside world. Now lets say we can replace that portion of the brain with a prosthetic as well, giving their mind something to “run on” again. Are they still human? Has the original person died and now a machine operates in this persons place with access to their memories and motivations? This is where I usually find opinion divides. There is a “break” in consciousness. That which existed before was destroyed, ended, and a new consciousness was raised in it’s place. A person interacting with this subject would not be able to tell the difference between pre-accident and post accident personality but there was a definite break in the flow of mind.
It’s a tricky question but at this point the subject is not much different to someone who had their mind read out then the original mind destroyed. The order is simply different.
I think, however, that this thought experiment actually presents the solution to the problem in it’s form. It’s clear that when the change is gradual and incremental the line between what was before and what is now blurs. Arguably there should be a granularity at which point the process becomes invisible to the subject and continuity is maintained.
My proposal for the solution to this problem is that for true acceptance of an “uploading” procedure or a migration to a non organic brain we would have to wait for the development of a nano-scale “artificial neuron” that has all the properties and functionality of an organic neuron with the advantage of being electronic based rather than organic.
My scenario would involve the regular injection into the brain or bloodstream of millions of dormant artificial neurons that remain in the body for a period of time. In the brain, as a neuron dies an artificial neuron moves into the space occupied by that neuron and takes over it’s function in the brain. This could even be facilitated or accelerated by the artificial neurons (at a reasonable pace) moving in on living neurons, making the same connections as it then killing the neuron off itself. Over many years as natural cells die off, or months in the facilitated model, the brain would move from an organic system to an electronic one. The subject themselves would be completely unaware of the process as their thoughts and thought processes would remain totally un-interrupted during the transition.
You,as the person undergoing thus procedure, would be totally un-aware of when the first or last neuron was replaced. One day a scan would simply show that you were now 100% artificial. At that point an interface unit could be implanted that allows the now wholly electronic signal to be read out as a series of information, shutting down the artificial neurons as they are read out. The process is similar to the reading of an organic mind and it’s destruction but for some reason an electronic to electronic transfer lacks the “yuck factor” of “destroying” an organic brain as you copy it. We understand data transfers, we are comfortable with that. Destroying a brain however seems like murder.
Or, if this does give the newly inorganic mind the feeling of being simply “copied” the new artificial mind can simply be expanded and enhanced until it’s un-recognisable as related to it’s original organic cousin.
Anyway. This is my solution to the continuity of consciousness problem. I don’t claim it to be original or ground breaking but rather a thought experiment into the potential for making the movement of human minds from the current very fragile, non redundant and ultimately terminal organic substrate to something much more resilient and long lasting. A form of existence where backups and redundancy are a realistic route to virtual immortality and the possibilities for the direct meeting of human minds on a scale never before realised is possible.
The repercussions in the forms of interacting with the physical world, either via machines or android “avatars” is something for discussion on another day, but also very worth thinking about.
Thanks for taking the time to read this entry and I hope it has given you something to think about and I’d love to hear other peoples opinions on the topic and it’s related issues.
AM
That`s probably the best solution I`ve read about regarding the uploading of our mind`s.
I like continuity of consciousness. If my consciousness could be transferred to an artificial brain, and especially if I could be aware of both my birth and artificial brain at the same time, but then later the body is stored via cryogenics until it can be restored (while my consciousness is in an artificial brain, or voluntarily dormant, like sleep) , that seems pretty good to me.
My consciousness simply moved around.
Also if I’m merely revived from a cryogenic state, and my consciousness is intact…that seems a little iffy, but not too bad. It is you, but you were dormant for a while.
Someone making a copy of me later seems pretty much useless to me. If you make an exact duplicate of me RIGHT NOW, and it lives on the other side of the world, and I never here of it again…how is that helping me?
I don’t understand people who get excited over cloning. Some woman cloned a dog and acted like they brought her dog back to life, and it knew her and it was the same dog. Ugh.
If there were ten genetic duplicates of me at birth, and they all go to different, unassociated lives in different places…you just have ten different people who have nothing to do with each other, except they have the same DNA. Right? They’re not me, they just have the same DNA code.
Hi, I’m a journalist based in Sydney doing a video story on the transhumanist movement. If you are in Sydney anytime over the next couple of weeks, could you email me so I could possibly get in touch and ask you some questions?
Cheers
Thankyou for this. I hve podered this question since seeing the “star trek” transporter and you have the most rational consideration of the question that I have read. To take the thought experiment one step further though, how could you objectively test consciousness continuity in a scientific manner? To the outside observer no change may be apparent, to the resulting entity with cloned memories continuity would be also be apparent. Would you volunteer for early technology knowing you could be facing destruction of consciousness even if an observer or the cloned inhabitant of your meat would not know that the original was gone?
Hey
It’s an interesting idea but I’m not sure I understand how the procedure makes the process of ‘reading out’ a brain any easier. The artificial neurons are an artificially created simulacrum for a human neuron. So they are components with the same properties as natural neurons.
A brain after replacement with artificial neurons will thus have the same characteristics as a brain before the process. What’s changed to suddenly make it easier to read it out? I’m skeptical that artificial neurons that participate as exact simulacrums in what must be a fairly tricky electrochemical network (assuming it is purely electrochemical) can also have the ability to participate in some sort of side-network to allow reading out of structure and state.
Your idea reminds me of a great Greg Egan short story ‘Jewel’ from the collection Axiomatic. In that story people have a device inserted into the brain that learns their neural structure and syncs up, the idea being that you can do a cut over with no perceived discontinuity.
I think the general idea was that once you have nanomachines that can replicate the network and performance of a neuron (not necessarily with the same physical structure) then these can re-arrange themselves into an analogous network in a completely different form with no loss of continuity. i.e they can “collapse” into a completely different processing substrate without changing the “state” of your mind. So once the cells are all replaced it can re-arrange itself into a more computationally dense structure that is independent of the meat sack and self powered and can simply be removed or expanded or whatever turns you on now you’re running on nanomachines instead of meat. The important part is the continuity of consciousness. But you make a valid point. I probably should have been clearer on that at the end. I might edit the article.