By 2100 we should be able to scan this info from a frozen brain. If we scan your brain and then build and run a computer simulation of it, someone who remembers being you would wake up and feel alive.
Let's say we are actually able to do that by 2100. Somebody is scanning in patterns and current state of my neural activity. He copies that into a simulation - more than once. Which of the simulations will be me? All of them?
If we keep writing down common sense datums until 2100, we can make computers as smart as people.
Would it make them smart or just filled with hard coded knowledge? I would expect that in this case the goal would be to create a system that is able to acquire common sense knowledge on its own from a given environment and reason from it. What we might consider as common sense in our daily lives might not be fully applicable in different situations.
I will be your resident biologist tonight (not working in cognition/memory but I will try to make a more general point).
Putting aside the philosophical problems for a moment, the notion that the only relation between a brain and a mind is the pattern of neuronal connections is pretty naive.
The formation and maintenance of memory is controlled at multiple levels: single atoms (ionic gradients determining the polarization of neurons), small chemicals and hormones (e.g. neurotransmitters in the synapses and signalling peptides), DNA (epigenetic control, mainly methylation), proteins (constant synthesis and degradation in the process of memory consolidation and long-term potentiation) and finally cells (pattern of connections between neurons).
I highlight these different levels because they are formed by compounds that are very different from a biological and chemical point of view. You can't just snap-freeze them all in their place. And freezing is the easy part -- how do you thaw a brain and preserve the state of all these components (all of which have different thermodynamic properties)?
Assuming you don't want to thaw it but, as the article suggests, 'scan it' you would have to be able to determine the conformation and activation state of each involved protein and compound, along with all their density gradients in intracellular spaces... Not by 2100, sorry.
Biology is hard because it's all relations, thresholds, gradients and fuzzy logic. To escape this probabilistic haze into a clean world of ones and zeroes -- I read HN.
The links that you provide make o lot of bold, unsubstantiated claims (outright dismissing protein shedding as 'trivial' borders on funny) so I am not sure whether you are agreeing with me or disagreeing. They are also from the 90's and the understanding of the physiological complexity of the brain has changed a bit since then.
Ahh, the swampman! I always thought of this problem when watching characters in StarTrek using teleportation, didn't know it was an existing, famous problem.
IMHO the teleported person ceases to exist at some point and a _new_ , identical person is being "created" by the transporter device. It's more of a gut feeling, hard to explain in words why I feel this way.
It's kinda intuitive, but I'm pretty sure it's wrong.
Imagine you fall in coma and sleep long enough that all your body cells have been changed. Do you wake up a different person? I don't think so.
What you perceive as yourself is the experience of being aware of everything you have learned in the world. An identical brain, whether created by teleport or running in simulation, is you just as much.
So yes, both simulations will be sure they are you (and in a sense, both will be you—if we allow the simulations, we must abandon the idea of there being only one “you” at the moment). The funny thing, these simulations will immediately diverge by obtaining different experiences, and thus can be thought of “forks” of you.
Imagine you fall in coma and sleep long enough that all your body cells have been changed. Do you wake up a different person?
I believe you do. In fact I believe you are a different person moment-to-moment. The concept of 'you' is simply a mental convenience. This is exactly why Buddhists deny the 'self.' It's not that you don't exist- it's just that there is no single thing that remains unchanging, and the idea of a 'you' that exists through time is illusory.
I occasionally find myself wondering whether it is the same "me" that wakes up in the morning as falls asleep the night before. So is it really the same instance of my consciousness that boots up in the morning as was shut down the night before?
Not the most productive thing to think about as you fall asleep :-)
Yep. It's a consequence of information entropy. Both start receiving different stimuli or have other subtle differences that make each different like twins. The ethics seem to be that each should have rights equal to exist (run) as egalitarian as any other, regardless of origin. Stopping is deleting is killing.
Let's say we are actually able to do that by 2100. Somebody is scanning in patterns and current state of my neural activity. He copies that into a simulation - more than once. Which of the simulations will be me? All of them?
Also, it reminds me a bit of Swampman problem : http://en.wikipedia.org/wiki/Swampman
If we keep writing down common sense datums until 2100, we can make computers as smart as people.
Would it make them smart or just filled with hard coded knowledge? I would expect that in this case the goal would be to create a system that is able to acquire common sense knowledge on its own from a given environment and reason from it. What we might consider as common sense in our daily lives might not be fully applicable in different situations.