Friday, October 14, 2011

Why Your Brain is NOT a Quantum Computer

Ervin Laszlo has written a number of articles declaring that human brains are quantum computers. As a computer scientist who is also interested in psychology I found the topic to be very interesting. However, as I read the articles I became frustrated with the lack of detail and the way they seemed to put forth provocative ideas with no sound evidence or reasoning behind them. They were still useful to me, sometimes we can learn as much from a poorly thought out idea as a good one. Prior to reading the article I had no idea what a quantum computer was. Unfortunately, I didn’t gain much insight on that from the articles but I did do some research to better understand them. I would like to describe what I found here and describe why I strongly believe your brain is not a quantum computer.

The first important point is that everything I or anyone is about to say about quantum computers is all about the idea of a quantum computer. To date there are no quantum computers and there are many complex unsolved problems (indeed they may be unsolvable problems) that need to be addressed before such computers exist.

The most interesting thing I found on researching quantum computers is that in many ways they aren’t all that different from traditional computers. A traditional computer consists of memory and a Central Processing Unit (CPU) that manipulates the memory. The CPU does things like basic math and moving information from one place to another. The memory is represented as electrical impulses (bits) that can be either on or off, 0 or 1. The essential difference between a traditional computer and a quantum computer is that a quantum computer would take advantage of the very mysterious phenomenon of quantum entanglement. The significance of entanglement is that it offers a model where particles can be described as being in multiple states at the same time. An entangled bit (known as a qubit) can be in the state of on and off at the same time. Qubits allow highly parallel search of a problem space. Rather than investigating each solution one at a time, using a qubit you can test many different alternatives all at the same time.

While qubits and quantum computers offer exciting possibilities to solve certain very complex problems from a computer science standpoint they aren’t that revolutionary. In fact as I looked on the Internet I found some incorrect statements about how Qubits differ from traditional bits. JR Minkel at Scientific American stated in a video “even today’s fastest computers ultimately do things one at a time”. This is absolutely and patently false. Supercomputers in fact work precisely because they do many computations at the same time. In fact its even wrong when considering personal computers. If you have a desktop computer it most likely has at least two CPU’s. That’s one of the things that enables you to have multiple applications up and running at the same time. For supercomputers this is taken to the extreme, they have hundreds or more CPU’s that work in parallel. The difference is that those computers still work with traditional bits that can be either 0 OR 1. A quantum computer working with qubits that can represent both values at the same time would be able to be much more massively parallel than even our fastest supercomputers.

The thing about parallel computing is that it sounds wonderful in principle but its a bit more complicated in practice. Adding 100 CPUs does not at all guarantee that you will cut the time required to solve a problem by 100. There is effort to write a special program that can exploit the benefits of parallelism, essentially divide a big problem into a bunch of smaller problems, solve the smaller problems than combine those solutions into one big solution. Not all problems lend themselves to such a divide and conquer approach. In fact no computer scientist that I came across is talking about quantum computers as a significant step toward artificial intelligence or better understanding the human mind. The emphasis is on specific mathematical problems such as factoring integers which could be useful in code breaking.

So with that background on quantum computers we can address the question: “is there any evidence at all that our brain functions as a quantum computer?” The answer to that is an emphatic no. We do have a model for how a computer can mimic our brain. That model is neural networks, mathematical representations that simulate the biochemical neurons in our brains. That model has proven extremely powerful, to the point where we can use computer neural networks to replicate the pattern matching capabilities, the ability to recognize edges, shapes, even faces that our brains do. To date there is absolutely zero evidence that the neurons in our brain have any ability to communicate at the quantum level. In fact there is no evidence that any biological system has the capability to interact at the quantum level. This is not surprising, Entanglement is a phenomenon that is extremely difficult to measure in the laboratory. It requires setting up complex highly sensitive and sophisticated electronic devises such as beam splitters and the entangled states are very susceptible to being disentangled by such measurements. That is the whole point of quantum entanglement, that such states represent multiple possible probability functions before being measured but the probability waves collapse into one real value after they are measured. The idea that a biological capability to detect such entanglements could somehow evolve seems highly improbable.

I would like to close with a more general discussion of why I think its important to rigorously examine such claims as “your brain is a quantum computer”. As I and others with a scientific background commented on Dr. Laszlo’s articles we were often maligned as people with closed minds and no imagination. I think its important to distinguish between having an open mind and an uncritical mind. I believe that I’m very open to all sorts of possibilities. I do think that the potential philosophical implications of quantum physics and string theory for example may be mind blowing and are worth serious investigation. However, there is a difference between serious investigation and arbitrary speculation. Many authors on the Huffington Post these days love to attach the word quantum to whatever their pet theory is. Quantum consciousness, quantum alternative realities, quantum brains,... These authors take advantage of the fact that quantum theory is difficult to understand to give any pet theory the appearance of scientific validity. This is pseudo-science at its worst. It does nothing to educate people on the very interesting topics of quantum theory nor on their potentially very interesting philosophical implications.

No comments:

Post a Comment