My impression was a bit different. It seems to me his thought process was:
I deduce consciousness is not a computable process.
Consciousness is a real physical process.
Quantum Mechanics may include processes that are not computable.
It is possible to have scientifically understandable processes that are not computable.
My deduction is not disproved by our current understanding of physics.
I think your reaction is reasonable. It seems to be a (well justified) conditioned response to "God of the gaps" arguments. I don't understand A therefore I need God to explain A. It is common now to see people replace God with Quantum Mechanics. I don't think that is the case with Penrose.
Yes, going back and reading the relevant articles on Wikipedia [1][2], my original post is a bit of a simplistic summary of his argument.
I agree that there could be quantum effects that are involved in the operation of the brain somehow. (I gather that some enzymes have been observed using quantum tunneling).
But it seems to go on to state that conciousness arises as a result of collapsing waveforms and some sort of special space-time that has the ability to perform/enable "super" computational feats.
So really I'm just not comfortable that we've tried to explain away conciousness by delegating it away to another ... plane.
Who can say. Certainly at the moment, opinion seems to be against this theory as the quantum effects suggested can't sustain long enough for the time period the theory requires.
Not quite. As I understand Penrose he only claims quantum mechanics is a reasonable place to look for an explanation of consciousness. He doesn't propose that QM is the explanation. In other words, he believes consciousness is not computable and uses QM to fight against the argument that "the universe is computable, consciousness is a part of the universe, therefore consciousness is computable".
he thinks consciousness is not computable, but is quantum computable?
umm, all quantum computers do is run code in many universes, then the output can draw from on the total results. it doesn't really change what kinds of things are computable, besides letting some new things run fast enough to be practical.
Summary: Psychedelic drugs increase the processing of the quantum annealing box because some biological plants like green tea have been shown to produce quantum effects. Further evidence can be found by analyzing lots of reports from people tripping and dismissing other explanations. Also this stuff must be in some way responsible for creativity and consciousness.
The video really ain't that bad and I even like that some people study if this might lead up to something. But lets just say once you start researching quantum effects in the brain you should maybe try to find arguments which are not based on yet another topic which sounds like typical 70's esoterica.
The basic building blocks for quantum computing could be there... yeah, sure, but unless he can present some stronger evidence this ain't worth the hour needed to watch the video. Though I liked the first half hour which was a nice overview on the topic.
Qubits decohere at anything like room temperature.
From my background in Neuroscience:
100 trillion connections is sufficient for all cognition.
Unfortunately, understanding why this claim is absurd takes a fairly strong background in both sciences. Rather than ask you to take my word for it however, I suggest you pick up a copy of they Feynman lectures and Principles of Neural Science. Read them both and we can then all have a good laugh at quantum-as-thought.
Most research into brain function and AI employs models that are at the lowest level atomic (i.e., basic chemistry). In ignoring quantum mechanical effects, we are like physicists who approximate a function with a Taylor's series and then truncate the series' higher-order terms. They shouldn't be significant but there's that nagging thought that sometimes they are.
In the book "Tending Adam's Garden : Evolving the Cognitive Immune Self" Irun Cohen discusses (among other things) how evolutionary processes may set up shop in any available ecological niche, regardless of how inefficient or convoluted the process may be. More nagging thoughts.
Degrees of freedom involving quantum mechanical operations are available; there's no reason they wouldn't be used. And as others have stated here, evidence is accumulating that they are used in biological systems. Hard evidence that the brain uses such processes would be revolutionary.
Many models of atomic and electromagnetic theory lie strewn by the roadside as science moves forward: ">Maxwell's molecular-vortex model of electromagnetism ( http://victorian.lang.nagoya-u.ac.jp/victorianweb/science/ma... ) and William Thomson's (Lord Kelvin) vortex model of the atom come to mind. I hope we're not as far off the correct path with current theories of intelligence and the mind.
The problem is that the basic assumption of neurons working as binary units is likely wrong. The next best guess is qubits, but that will take another generation, at least, to even begin to understand the implications let alone test. With two possible states for every neuron the exponential capacity to explain is ridiculous. The problem is that the basic assumption of neurons working as binary units is likely wrong. The next best guess is qubits, but that will take another generation, at least, to even begin to test let alone understand. With two states the exponential capacity to explain is ridiculous. The range of possibilities to decode gets absurd, if not infinite, with an additional supposition of the two. Godspeed to those trying!
I just came away with the impression that the train of thought was:
And with no explanation of how QP would actually solve the problems that regular computation couldn't.