Koch vs Penrose
In the 30 March issue of Nature, Christof Koch and Klaus Hepp brought out an article ("Quantum Mechanics in the Brain") which puts a fairly strong case for why quantum mechanics is unlikely to play a significant role in mediating conscious experience.
(See: http://www.nature.com/nature/journal/v440/n7084/index.html; page 611)
They offer an interesting thought experiment:
Imagine you have a cat. You put it in a box a-la Schroedinger, that notorious animal lover (i.e. you accompany it with some quantum mechanical system, typically an atom undergoing radioactive decay, coupled to a device which will kill the cat if a particular quantum transition occurs: Hey presto, you have an alive-dead cat.)
You now observe the cat through a window in the box with one eye, and, say, a picture of a zeppelin with the other.
This cunning idea allows the separation of `observation' (i.e. photons striking the retina) from consciousness, thanks to a fascinating visual phenomenon called binocular rivalry.
Essentially, binocular rivalry occurs whenever two dissimilar images are presented to the eyes simultaneously. Instead of seeing a mish-mash of both images, one often observes the images to alternate between complete suppression and complete salience.
The reason this is interesting is because, for a long while, physicists have speculated that quantum superpositions (i.e. the alive-dead cat) only collapse into definite classical states (i.e. alive cat or dead cat) when information about the superposition reaches consciousness. Koch and Hepp's idea is that if `observation' occurs without concommitant conscious perception, it should be argued that the cat will remain in an indefinite superposition state until the image of the cat is consciously perceived.
They actually go on to claim that this should be empirically testable.
Now; while I certainly give them full marks for imagination, and I am not overly fond of the `quantum consciousness' idea, I have to say that I don't think this quite hits the mark. The issue is that this doesn't actually provide any additional leverage to the problem (though I'm not sure about that... please leave a comment if you disagree!).
Von Neumann made an illuminating analysis of the measurement problem; he showed that, strictly speaking, measurement involves the coupling of successively more `macroscopic' systems to the quantum system undergoing measurement. These systems are coupled in such a way that one state of the quantum system (say, a decay event) will give one result (say, a dead cat) while another state of the quantum system (no decay) will give another result (a live cat). Technically, the quantum system and the measurement apparatus are said to be `entangled'.
The problem is `how do we go from a superposition of entangled states ( i.e. dead-cat+decayed-atom entangled with live-cat+nondecayed-atom ) to a single unentangled state (i.e. dead-cat+decayed-atom)?' Unfortunately, the best we can do is to say that the superposition collapses when we become conscious of the fact that the cat is dead.
We could try to get around this by having a computer which recorded the `life' or `death' of the cat for us, so that we then just look in its memory some time in the future... the problem then is that the computer itself is in a superposition with the cat and the quantum system, and this superposition only breaks down when we actually look at the computer's memory!
What Koch and Hepp are suggesting is that by dissecting the initial `detection' event on the retina from the subsequent `consciousness' event, one could potentially check whether the cat died at the moment you became conscious of its state, or at the moment photons from the cat reached your eye.... but of course, that's impossible to do! How could you make that measurement? You'd only know the results of it when you became conscious of them, and that leaves us back where we started... before we become conscious of the results, technically, we're still in a superposition!
I have two comments to make on this:
1) I think Koch and Hepp's idea is still interesting because it points out that at least part of the brain can be in superposition with the outside world without wave-function collapse, which in turn reduces the `wriggling room' for where consciousness resides (if that question still makes sense...) or how it is generated...
2) Perhaps one way to test their idea would be to have a quantum system for which the probability of obtaining a particular outcome varied periodically with time, with a suitable period (about the same as the typical binocular rivalry switch time...). i.e. the quantum state is something like: cos(wt)|0> + sin(wt)|1>. Then by performing multiple trials (perhaps not using cats... the ethics would be very difficult to obtain) we could look at the statistics of the collapse process; if collapse occurred when photons first hit the retina, we'd expect to see a mixture of 1s and 0s with a constant probability distribution (i.e. no dependence on when the human observer reported the result) however, if collapse occurred only when the human observer became conscious of the result, we'd expect to see a periodic dependence on the result depending on when the observer made the report.
Anyway... Not the best written post, but nonetheless, an article I found very interesting...
(See: http://www.nature.com/nature/journal/v440/n7084/index.html; page 611)
They offer an interesting thought experiment:
Imagine you have a cat. You put it in a box a-la Schroedinger, that notorious animal lover (i.e. you accompany it with some quantum mechanical system, typically an atom undergoing radioactive decay, coupled to a device which will kill the cat if a particular quantum transition occurs: Hey presto, you have an alive-dead cat.)
You now observe the cat through a window in the box with one eye, and, say, a picture of a zeppelin with the other.
This cunning idea allows the separation of `observation' (i.e. photons striking the retina) from consciousness, thanks to a fascinating visual phenomenon called binocular rivalry.
Essentially, binocular rivalry occurs whenever two dissimilar images are presented to the eyes simultaneously. Instead of seeing a mish-mash of both images, one often observes the images to alternate between complete suppression and complete salience.
The reason this is interesting is because, for a long while, physicists have speculated that quantum superpositions (i.e. the alive-dead cat) only collapse into definite classical states (i.e. alive cat or dead cat) when information about the superposition reaches consciousness. Koch and Hepp's idea is that if `observation' occurs without concommitant conscious perception, it should be argued that the cat will remain in an indefinite superposition state until the image of the cat is consciously perceived.
They actually go on to claim that this should be empirically testable.
Now; while I certainly give them full marks for imagination, and I am not overly fond of the `quantum consciousness' idea, I have to say that I don't think this quite hits the mark. The issue is that this doesn't actually provide any additional leverage to the problem (though I'm not sure about that... please leave a comment if you disagree!).
Von Neumann made an illuminating analysis of the measurement problem; he showed that, strictly speaking, measurement involves the coupling of successively more `macroscopic' systems to the quantum system undergoing measurement. These systems are coupled in such a way that one state of the quantum system (say, a decay event) will give one result (say, a dead cat) while another state of the quantum system (no decay) will give another result (a live cat). Technically, the quantum system and the measurement apparatus are said to be `entangled'.
The problem is `how do we go from a superposition of entangled states ( i.e. dead-cat+decayed-atom entangled with live-cat+nondecayed-atom ) to a single unentangled state (i.e. dead-cat+decayed-atom)?' Unfortunately, the best we can do is to say that the superposition collapses when we become conscious of the fact that the cat is dead.
We could try to get around this by having a computer which recorded the `life' or `death' of the cat for us, so that we then just look in its memory some time in the future... the problem then is that the computer itself is in a superposition with the cat and the quantum system, and this superposition only breaks down when we actually look at the computer's memory!
What Koch and Hepp are suggesting is that by dissecting the initial `detection' event on the retina from the subsequent `consciousness' event, one could potentially check whether the cat died at the moment you became conscious of its state, or at the moment photons from the cat reached your eye.... but of course, that's impossible to do! How could you make that measurement? You'd only know the results of it when you became conscious of them, and that leaves us back where we started... before we become conscious of the results, technically, we're still in a superposition!
I have two comments to make on this:
1) I think Koch and Hepp's idea is still interesting because it points out that at least part of the brain can be in superposition with the outside world without wave-function collapse, which in turn reduces the `wriggling room' for where consciousness resides (if that question still makes sense...) or how it is generated...
2) Perhaps one way to test their idea would be to have a quantum system for which the probability of obtaining a particular outcome varied periodically with time, with a suitable period (about the same as the typical binocular rivalry switch time...). i.e. the quantum state is something like: cos(wt)|0> + sin(wt)|1>. Then by performing multiple trials (perhaps not using cats... the ethics would be very difficult to obtain) we could look at the statistics of the collapse process; if collapse occurred when photons first hit the retina, we'd expect to see a mixture of 1s and 0s with a constant probability distribution (i.e. no dependence on when the human observer reported the result) however, if collapse occurred only when the human observer became conscious of the result, we'd expect to see a periodic dependence on the result depending on when the observer made the report.
Anyway... Not the best written post, but nonetheless, an article I found very interesting...