The Wiki article has plenty of links to papers and news articles, including the Integrated Information Theory website. Scott Aaronson's blog article, 'Why I am not an Integrated Information Theorist' is an interesting critique.
I think the difficulty with consciousness studies is the irreconcilable subjective-objective divide. Subjective experience (Nagel's 'what it is like', and the accompanying qualia) is inaccessible to all but the experiencer and can only be communicated indirectly through metaphor and appeal to shared objective experience (in the hope that another human having the same objective experience will have a roughly similar subjective experience). We can never know how similar different individual's subjective experiences are, but we do know that given the same objective experience they can be very different, e.g. the same roller-coaster experience will give different people very different subjective experiences.
The inaccessibility of subjective experience means that a scientific approach can only deal with correlates of consciousness, so may be prone to the error I think IIT makes, of suggesting that, having detected some correlate (e.g. integrated information), any system that shows that feature must, ipso facto, be conscious. This is like finding that excellent fitness and good reflexes are highly correlated with being a good footballer, and suggesting that therefore anyone with excellent fitness and good reflexes must be a good footballer. "To a man with a hammer, everything looks like a nail"
I think the difficulty with consciousness studies is the irreconcilable subjective-objective divide. Subjective experience (Nagel's 'what it is like', and the accompanying qualia) is inaccessible to all but the experiencer and can only be communicated indirectly through metaphor and appeal to shared objective experience (in the hope that another human having the same objective experience will have a roughly similar subjective experience). We can never know how similar different individual's subjective experiences are, but we do know that given the same objective experience they can be very different, e.g. the same roller-coaster experience will give different people very different subjective experiences.
The inaccessibility of subjective experience means that a scientific approach can only deal with correlates of consciousness, so may be prone to the error I think IIT makes, of suggesting that, having detected some correlate (e.g. integrated information), any system that shows that feature must, ipso facto, be conscious. This is like finding that excellent fitness and good reflexes are highly correlated with being a good footballer, and suggesting that therefore anyone with excellent fitness and good reflexes must be a good footballer. "To a man with a hammer, everything looks like a nail"
Last edited:
Upvote
0