Posts
Comments
Aristotle scooped you, this is an og fallacy: https://en.m.wikipedia.org/wiki/Accident_(fallacy)
(But this is an excellent essay and I love it)
Well, there's the fact that people have lots of seizures, which as far as we can tell are very chaotic patterns of electrical activity that scramble all information contained in ongoing oscillatory patterns. (Note the failure of spike sorting algorithms upon recruitment of neurons into seizure activity. http://m.brain.oxfordjournals.org/content/early/2015/07/17/brain.awv208.abstract) Not only that, but TMS (https://en.m.wikipedia.org/wiki/Transcranial_magnetic_stimulation -- effectively introducing large random currents in large chunks of brain tissue) doesn't seem to produce any long term effects as long as you don't start actually causing tissue damage through hydrolysis.
On the molecular side, we know that our core personality is resilient to temporary flooding of the brain with a large array of different transmitter analogs, antagonists, and other chemicals. (All of the drugs that people do) Many of these chemicals are synthetic ones that we didn't co-evolve with.
I think it's very reasonable to suspect that most of the important information that composes the individual is stored in genetic regulatory networks, and in the connectome. Chemical gradients aren't very information dense, and while we might a priori expect there to be a lot of information in ephemeral electrical activity, I think seizures and TMS are both good demonstrations that this information can at least be restarted given the structure of the network.
Final thing to consider: there's much more individual variation at the level of anatomy than there is at the level of electrophysiological properties. There are a relatively small number of morphological categories of neurons (100s), that are fairly stereotyped across humans. But brain anatomy varies enormously from subject to subject. (Take into account that as a Cognitive neuroscientist, I'm probably biased in this regard)
There's still some missing pieces, like working memory CAN'T be stored in the connectome because plasticity mechanisms and genetic mechanisms aren't fast enough.
At the very least though, I think there's a lot of hope. After all, the connectome and genetic information can be well preserved even with plasticization and slicing. My money's on those being the critical pieces of information.
Hm, I have a lot problems with Searle's argument. But even if you skip over all of the little issues, such as "The Turing Test is not a reasonable test of conscious experience", I think his biggest flaw is this assumption:
The intuition that the Chinese room follows a purely syntactic (symbol-manipulating) process rather than a semantic (understanding) one is a correct philosophical judgement.
If you begin with the theory that consciousness arises from information theoretical properties of a computation(such as Koch and Tononi's Integrated Information Theory), then while you may reach some unintuitive conclusions, you certainly don't reach any contradiction, meaning that Searle's argument is not at all a sufficient disproof of AI's conscious experience. Instead, you simply hit the conclusion that for some implementations of rulesets, the human-ruleset system IS conscious, and DOES understand Chinese, in the same sense that a native speaker does. I think we can undo the intuition scrambling by stating that the ruleset is analogous to a human brain, and the human carrying out the mindless computation is analogous to the laws of physics themselves. Do we demand that "the laws of physics" understand Chinese in order to say that a human does? Of course not. So why does it make sense to demand that the human (who, in the chinese room is really playing the same role as physics) understand Chinese in order to believe that the room-human system does?