[russdill]

That's the problem with Penrose's thinking though. He's absolutely convinced that consciousness cannot boil down to something computable. So he reaches for the quantum shelf, but not just the quantum shelf, the quantum processes we don't yet understand since otherwise it'd just be something computable, but with more steps.

[sarchertech]

I mean most AI researchers are utterly convinced that the human brain is a Turing machine. There’s not reason to presuppose that. And if everyone presupposes that and it isn’t true, we’ll be stuck spinning our wheels until someone questions it.

[russdill]

This isn't a viewpoint coming from AI, is been a view that's been around for quite some time. The standard model does an exquisite job of predicting the day to day physics we experience. To find search for new physics where we might find discrepancies, we have to build absolutely huge colliders, and even those don't really effect our "day to day" physics.

For consciousness to be based on some non computable function there would have to be some unknown physics occurring in the brain. Sure, that's hard to disprove, but it also strains credulity, hard.

Penrose picking "quantum" for being the element of physics that would have to change to allow this non computibility is just woo. Why not just say magic?

[pmontra]

I agree with you that picking quantum was probably the fad of the time and it was an easy pick as quantum is everywhere. He cold have said "electrons". However the argument against unknown physics is not very sound IMHO.

There has been unknown physics at play inside brains since forever and it still is and always will be, by definition of science.

The point is that we don't even know how to define consciousness and humanity doesn't have a shared agreement about which living beings are conscious or are not. We're still like engineers building things millennia or centuries ago with only a shadow of a theory of why their creations worked. And yet we still walk on bridges from 2000 years ago and we had electric batteries and power plants before knowing how an electro magnetic wave moved.

[russdill]

If it were known that there where physical interactions occurring within the brain that deviated from the standard model, we wouldn't bother building didn't colliders.

[sarchertech]

Even if the brain showed a real deviation from the Standard Model, we’d still need colliders. The brain is a messy, low energy environment, and observation isn’t identification.

Colliders provide clean, high energy tests to pin down masses, spins, and couplings, and we already need them for open problems. A bio anomaly would set targets, not replace colliders.

[sarchertech]

> This isn't a viewpoint coming from AI

It has always been related to AI. Shortly after Church and Turing formalized computability, people started squaring off into 2 camps. People who believed strong AI was possible and those who didn’t.

We know the standard model is incomplete. Penrose’s ideas come directly from his his explorations of the gaps he suspects exist.

[russdill]

There are people that come to it from the direction of AI because they think they know something about consciousness. That's certainly putting the conclusion first.

But there are also physicists who come at the problem from the direction of physics. We know the standard model is incomplete, but we also know it covers everyone we experience with exquisite precision. Unless there is a black hole or temperatures on par with the big bang going on within our brain, the standard model will tell you what you need to know

[sarchertech]

You are vastly overstating the certainty of your position. Many physicists, including Penrose (even if ignoring his consciousness arguments) are deeply unsatisfied with the standard model and believe that it is incomplete. Despite understanding how useful it can be.

It is very unlikely that it can tell you “everything you need to know” or even everything that could possibly be useful.

Even if that were the case we have no way of knowing that or even supporting that with our current understanding. That takes that statement thoroughly outside of the realm of science and into philosophy.

>Unless there is a black hole or temperatures on par with the big bang going on within our brain, the standard model will tell you what you need to know

Do you think that if it were that simple, that a world renowned physicist would capable of doing the back of the envelope math?

Also do you think that if it were that simple that this world renowned physicist could consistently convince other world renowned physicists to engage with him beyond simple 1 paragraph rebuttals? To the point where they will write entire chapters in books published by him?

You may not find his argument convincing, but his arguments aren’t dismissed as crank science outside of edgy internet posts.

[russdill]

The standard model is universally understood to be incomplete, but the problem is that it's perfectly able to predict all known interactions we can measure. We need to go to much higher energies to find violations. And the refinements we have been doing (higgs, etc) have no effect on our understanding of biology.

Penrose claims that violations of the standard model must exist within table top experiments. Not because of any specific objections to the standard model, but because of philosophical objections related to the nature of consciousness. And so it doesn't point to where those violations must be, but instead just a blind search.

As far as I know, professionals are too kind to refer to Penrose as a crank due to his extensive contributions to physics and mathematics. But his claims here are related to neither physics not mathematics but the philosophy of consciousness, an area where he hasn't made any discoveries. Professionals are happy to refer to his ideas as "highly implausible" or even as useful as "pixie-dust in the synapses".