Could you copy the brain to a computer?

[gmpd2000]

The question is simple, Could you copy the brain to a computer? like remplacing neurons with electronic copies.

If possible, Would that Copy have the same memories or opinions?

and most important, It would have thoughs or not?

[NGTOne]

Well, depends on what you mean. I don't see why you couldn't copy the connection network (While it's too complex for today's computers, it's certainly possible), but you'd have to copy the current state of the brain as well, at a given time, or you'd end up with what was basically a dead network. If you had a sufficiently powerful computer, I don't see why it couldn't be fully sentient.

[K^2]

Regular home computer? No. Not enough storage. A server or cluster with huge storage capacity? Yes, in principle.

As far as running it, that's a more complicated question. On one hand, there is a lot that we still don't know about the exact way messages are sent between the neurons. There is just so much going on. But on the other hand, human brain keeps mostly functioning across a range of moods, illnesses, chemical imbalances, and under influences of poisons and drugs. We can certainly simulate neuron-to-neuron communication with enough precision to fall somewhere within that range.

The other factor is that there is no computer in the world that can simulate neural activity in human brain at anywhere remotely close to real time. Human brain has hundreds of billions of neurons, each connected to hundreds or even thousands of its neighbors. Each neuron refreshes at something like 100Hz, but if you want a good simulation, you'll need a much shorter time step, on the order of 10kHz or better. This dwarfs any supercomputer by itself. If you take into consideration what sort of a royal mess this does to memory mapping, it could easily take days of real time to simulate 1s of brain activity on largest clusters.

But given that you are ok with slowing the time down for the simulated brain, and you don't care if simulated brain turns out to be depressed or high as a kite, yes, in principle, we can simulate a human brain with all of the thoughts in it.

Of course, this assumes you have a full map of a brain of a grown human brain, which we have no chance to obtain. And we can't really start from scratch and teach a "blank" brain either. Not only would it take way too long, because we can't simulate it at real time, but learning involves changing the way the neurons are connected, and we don't understand that part of the process enough to build something that grows and develops the way human brain does.

So in practice, we just don't have the data to work with to make such a simulation worthwhile.

[jwenting]

nope. There's simply no computer that has the required storage and/or processing power, and within the bounds of reality there's unlikely to ever be one.

There's simply not enough matter in the universe to create a computer that can replicate the complexity of the human brain without replicating the human brain, in which case we'd be building human brains, not computers.

[Kryten]

n

There's simply not enough matter in the universe to create a computer that can replicate the complexity of the human brain without replicating the human brain, in which case we'd be building human brains, not computers.

True, if you assume that the position of every single atom in the brain needs to be simulated to have a working model of the brain. Given how much change the brain can undergo (chemically and physically) and still remain generally functional, this isn't a very credible assumption.

[Dodgey]

The better question is weather we can reverse the procedure afterwards.

[SargeRho]

Short: Yes.

Long: No. You can copy the precise structure of a human brain onto a computer, but I doubt you could actually simulate it using that. Brains and Computers are fundamentally different constructs. You'd have to translate the structure of the brain into something a computer can use.

[Nibb31]

You can simulate pretty much every natural or physical phenomenon at various degrees of fidelity. The processing power relates to the complexity of the model which relates to the

For example, we have weather simulations that are quite capable, although they are simplified models of reality. There is no reason we couldn't simulate the atmosphere down to the movement of each molecule of air and model its reactions with neighboring molecules. They physics are well understood, so there is no theoretical limit to what we can simulate. The only limit is raw computing power, which is why we use simplified models.

The same goes for the brain. We are already capable of simulating individual neurons and synapses to pretty high degree of fidelity. The chemical and physical ways in which they react to each other are well known nowadays. There is nothing mysterious or magical about it. We are already capable of assembling millions of simulated virtual neurons and observing how they react:

http://singularityhub.com/2012/12/10/scientists-create-artificial-brain-with-2-3-million-simulated-neurons/

It's just a matter of time until we have the processing power to simulate an entire brain. The first models will probably work at a much slower scale than a real brain and only for short durations. For example, they might run a computer simulation for 2 months to get 10 seconds of brain activity. But as available processing power increases, dedicated hardware is designed, and the simulation model is optimized, it's pretty inevitable that we will reach the Singularity event during the next couple of decades.

Edited February 20, 2014 by Nibb31

[Nuke]

this is where i usually go into a tirade about the difference between a neural net and a cpu. but i dont feel like doing that today.

[Nibb31]

A CPU can simulate a neural net.

There is a difference between any object and a CPU. However, we use computers to simulate all sorts of non-computer things every day.

You can theoretically break down any object into a model that can be simulated by an algorithm: weather models, astronomical models, nuclear reaction models, cars, planes, bridges... Brain models are no different. It's just a matter of understanding how the physical object interacts with its environment and having the computing power to simulate that interaction.

Edited February 20, 2014 by Nibb31

[xrayfishx]

Well, the tech for that isn't far, I guess, but it would not be a good idea (I'm looking at you, GLaDOS).

[Seret]

The only possible answer to this question is: we don't know. We know too little about the brain, and it's not clear whether we'll be able to overcome the technical hurdles to create a machine with equivalent complexity.

[K^2]

The only possible answer to this question is: we don't know. We know too little about the brain, and it's not clear whether we'll be able to overcome the technical hurdles to create a machine with equivalent complexity.

Nonsense; nonsense.

There's simply not enough matter in the universe to create a computer that can replicate the complexity of the human brain without replicating the human brain

More nonsense!

It's a finite state machine with quite a manageable state vector. As I've pointed out, we're already well capable of storing all required information to describe the brain's state. Simulation is a bigger hurdle, but nothing like what you're saying.

[MartGonzo]

There is the bluebrain project which is attempting to simulate the neural connections of a human brain inside a computer as a long term goal but I haven't looked into it recently so I'm not sure how far along they are.

http://bluebrain.epfl.ch/

Edited February 20, 2014 by MartGonzo

[lodestar]

There is the bluebrain project which is attempting to simulate the neural connections of a human brain inside a computer but I haven't looked into it recently so I'm not sure how far along they are.

http://bluebrain.epfl.ch/

Could you copy all the currently quantifiable properties of a living brain into a sufficiently large computer? Sure, why not? It only depends on how much information you can measure. But copying all the currently quantifiable properties of a living brain isn't copying the brain itself, nor whatever phenomena arise from it. How could you tell if it would have thoughts or not if we can't even tell for sure anyone else besides ourselves is thinking?

Curiously, the prevailing model of brain function always follow the dominant technology of the time. Future scientists will be as amused by our expectations to reduce the brain functions to an electronic computer as much as we are amused by 18th century scientists who had similar expectations with hydraulic systems.

Edited February 20, 2014 by lodestar

[MartGonzo]

Could you copy all the currently quantifiable properties of a living brain into a sufficiently large computer? Sure, why not? It only depends on how much information you can measure. But copying all the currently quantifiable properties of a living brain isn't copying the brain itself, nor whatever phenomena arise from it. How could you tell if it would have thoughts or not if we can't even tell for sure anyone else besides ourselves is thinking?

Curiously, the prevailing model of brain function always follow the dominant technology of the time. Future scientists will be as amused by our expectations to reduce the brain functions to an electronic computer as much as we are amused by 18th century scientists who had similar expectations with hydraulic systems.

I agree completely, I just mention it because it's the closest real world project to what the OP was asking that I know of.

We have such an incomplete understanding of the brain that I suspect if artificial brains comparable to human brain power are ever created they will be of a different type of intelligence in no way similar to the human brain.

[Seret]

Nonsense; nonsense.

Note that I wasn't saying that we wouldn't be able to achieve this in the future, I was saying we don't know enough now to know if we would or not. Yes, I was saying that simulation is an open question, you're just assuming that I wasn't. There are some interesting arguments that question whether simulating the brain is possible at all on our technology (eg: Roger Penrose in "The Emperor's New Mind"). If you're asserting that we do currently know enough about the brain to know how to build a mechanical one then I'm sorry but you're wrong. We've only just started to crack it open with new gizmos like fMRI, we've got a lot to learn yet.

As for your second "nonsense" then if you know how to shrink circuitry to the degree required to create a nanoelectronic device of equivalent complexity to a human brain then I assume you'll be picking your nobel prize up later in the week? Congrats, btw The brain is a highly sophisticated nanosystem, we don't have the proficiency in either nanotech or synthetic biology (or indeed know if either of those are the right approaches) to know what parts of the sci-fi vision will be practical.

[Nibb31]

Future scientists will be as amused by our expectations to reduce the brain functions to an electronic computer as much as we are amused by 18th century scientists who had similar expectations with hydraulic systems.

Nobody expects to "reduce brain functions to an electronic computer". What they are attempting to do is to create an abstract simulation model of brain cells. Whether the model is implemented as an algorithm on a silicon computer, a steam-powered turing machine, a holographic quantum computer, or a biological substrate is irrelevant.

The hard part is modelling how the brain works, and neurobiologists have most of it worked out by now.

[Nibb31]

As for your second "nonsense" then if you know how to shrink circuitry to the degree required to create a nanoelectronic device of equivalent complexity to a human brain then I assume you'll be picking your nobel prize up later in the week? Congrats, btw The brain is a highly sophisticated nanosystem, we don't have the proficiency in either nanotech or synthetic biology (or indeed know if either of those are the right approaches) to know what parts of the sci-fi vision will be practical.

Who says that an artificial brain has to be small or implemented with nanotech or synthetic biology or even if it has to run in real-time. Those are just implementation details.

[Seret]

Who says that an artificial brain has to be small or implemented with nanotech or synthetic biology or even if it has to run in real-time. Those are just implementation details.

Sure, but for any current or foreseeable future tech it would have to be small. Create a simulated brain on current microelectronic hardware just wouldn't work. There may be some wondertech in the future that allows us, but that's basically just hand waving the question away.

Sure, you might be able to run something not at real time, the OP wasn't specific, but I think the question implied a functionally equivalent simulation.

[lajoswinkler]

True, if you assume that the position of every single atom in the brain needs to be simulated to have a working model of the brain. Given how much change the brain can undergo (chemically and physically) and still remain generally functional, this isn't a very credible assumption.

That's because every change is done gradually and the network works through it. Synthesis de novo would be futile as it would introduce immediate and total differences.

Emulations are possible just like we do with weather prognosis, but when you apply that to brain, you can't expect to replicate an exact healthy person. I think the product would be horrific. Probably highly mentally disrupted individual, at best.

Neural tissue communicates not only by ionic impulses. There is chemical signalization, too, and the communication goes throughout the body. Brain is not a separate entity that just calculates stuff using neurons.

All that is subjected to chaos on a molecular level.

In short words - no. We can't make an exact copy.

[lodestar]

Nobody expects to "reduce brain functions to an electronic computer". What they are attempting to do is to create an abstract simulation model of brain cells.

Buddy... any simulation is a reduction.

The hard part is modelling how the brain works, and neurobiologists have most of it worked out by now.

Neurobiologists might have worked out pretty well how many known chemical interactions happen in the nervous system, but obviously, that's eons away from actually solving the many binding problems. There are many more fundamental philosophical issues to be solved before empirical knowledge of brain functions can actually answer anything.

[SargeRho]

What philosophical issues?

[lodestar]

What philosophical issues?

[SargeRho]

Conciousness is as physical as the programs running on a computer. Exept the brain operates mainly on chemical reactions, and computers purely on electricity.

There is nothing that is non-physical, but some things are more than the sum of their parts.