6 ms·

A Neuroscientist’s Radical Theory of How Networks Become Conscious

12 points by npalli 13y ago

protonfish 13y ago
This is another example of the poor quality work being done on higher intelligence research. He defines consciousness so that it has no testable characteristics. Saying that it is an emergent property of multiple connections in a system smacks of magic smoke. This hypothesis is not "radical" at all but a lazy throwback to pre-science philosophy. I hope that this guy is just a suit and it is not representative of the work being done at the Allen Institute. If it is, Paul should think about spending his money elsewhere.
- KingMob 13y ago
  There aren't really any good definitions of consciousness that are easily falsifiable. It's in the nature of the problem. All of our collected data about other people's/organism's consciousness is third-person, while consciousness is definitely a first-person phenomenon. I don't know for sure that anyone else is conscious; the world could be populated by zombies (in the philosophical sense) for all I know. Anyway, the root of the word "science" comes from "knowledge", not falsifiability. These are still very early days for brain science and consciousness is its hardest problem. Integrated information theory (IIT) in particular, does offer some falsifiable predictions, btw. IIT's phi metric requires both interconnectedness, but also heterogeneity. E.g., Koch alluded to the cerebellum. The cerebellum has half of the neurons in the brain, but they are arranged in a simple, almost crystal-like pattern. People who have had cerebellums removed due to cancer show no alterations in consciousness. If we were to find someone who had their cerebellum removed and it did impact their consciousness, IIT would be incorrect. Likewise, IIT fits with data about the lack of consciousness during slow wave sleep and generalized seizures (simple, rhythmic firing patterns), but again, if we were able to find counterexamples, it would be disproved.
colanderman 13y ago
That’s associative memory. What is the simplest explanation for it? That consciousness extends to all these creatures, that it’s an imminent property of highly organized pieces of matter, such as brains. Um, I'm pretty sure there are mathematical explanations for how associative memory works, that don't invoke the paranormal: http://en.wikipedia.org/wiki/Hopfield_network http://en.wikipedia.org/wiki/Hopfield_network

12 points by npalli 13y ago

This is another example of the poor quality work being done on higher intelligence research. He defines consciousness so that it has no testable characteristics. Saying that it is an emergent property of multiple connections in a system smacks of magic smoke. This hypothesis is not "radical" at all but a lazy throwback to pre-science philosophy. I hope that this guy is just a suit and it is not representative of the work being done at the Allen Institute. If it is, Paul should think about spending his money elsewhere.

KingMob 13y ago
There aren't really any good definitions of consciousness that are easily falsifiable. It's in the nature of the problem. All of our collected data about other people's/organism's consciousness is third-person, while consciousness is definitely a first-person phenomenon. I don't know for sure that anyone else is conscious; the world could be populated by zombies (in the philosophical sense) for all I know. Anyway, the root of the word "science" comes from "knowledge", not falsifiability. These are still very early days for brain science and consciousness is its hardest problem. Integrated information theory (IIT) in particular, does offer some falsifiable predictions, btw. IIT's phi metric requires both interconnectedness, but also heterogeneity. E.g., Koch alluded to the cerebellum. The cerebellum has half of the neurons in the brain, but they are arranged in a simple, almost crystal-like pattern. People who have had cerebellums removed due to cancer show no alterations in consciousness. If we were to find someone who had their cerebellum removed and it did impact their consciousness, IIT would be incorrect. Likewise, IIT fits with data about the lack of consciousness during slow wave sleep and generalized seizures (simple, rhythmic firing patterns), but again, if we were able to find counterexamples, it would be disproved.

colanderman 13y ago

That’s associative memory. What is the simplest explanation for it? That consciousness extends to all these creatures, that it’s an imminent property of highly organized pieces of matter, such as brains. Um, I'm pretty sure there are mathematical explanations for how associative memory works, that don't invoke the paranormal: http://en.wikipedia.org/wiki/Hopfield_network http://en.wikipedia.org/wiki/Hopfield_network