Consciousness theory slammed as "pseudoscience"

[John Clark]

Consciousness theory slammed as "pseudoscience"

John K Clark    See what's on my new list at  Extropolis

jqq

[Dylan Distasio]

Having read that letter, I don't find it very becoming of the scientists writing it who should know better.   Regardless of what you think about IIT and its merits or lack thereof, it results in some predictions that can be tested, and while some of the work is certainly hard to carry out, it provides a framework for future experiments.

The letter sounds to me like it was written by a bunch of high schoolers who are butthurt that their pet theories of consciousness are not the leading one in the press.   To label it pseudoscience arbitrarily is a bridge too far and does not reflect well on the signees.  

--
You received this message because you are subscribed to the Google Groups "Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to everything-li...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAJPayv3KhiA9tJXvhos5RRSStwh6hSWm6SFVfz-vpRZDrYGM%2BA%40mail.gmail.com.

[Jason Resch]

By its own definitions IIT is not falsifiable, for it proclaims that a computer program that gave identical behavior in all situations to another conscious system, would not be conscious. But since it has identical behavior there is no objective way to prove this assertion of IIT (that one system is conscious while the other is not).

This also implies the possibility of philosophical zombies (which IIT proponents freely admit), which also implies consciousness is epiphenomenonal, with all the problems of philosophical zombies and epiphenomenonalism entail.

So is it pseudoscience? I don't know if I would call it that, but I think it is almost certainly wrong as it is currently framed.

I do find some strengths in some of the ideas that have come out of it, in particular how a system must be capable of affecting itself for it to be aware of its consciousness. I also think it is right to put the focus on information.

I think where it errs is in confusing a logical-informational state with a instantaneous physical state. This leads to the mistaken belief that a parallel computation is more conscious than a serial computation, even when they compute the exact same function (IIT proponents don't consider space-time symmetry).

I think that if IIT corrected these problems, it would be no more than functionalism. I think of IIT is a kind of "functionalism in denial", as it makes many similar claims to functionalists, placing emphasis on the causal organization of a system, but at the last moment, it insists that a computer implementing that same causal organization would not be conscious.

Jason 

[John Clark]

On Thu, Sep 21, 2023 at 2:43 PM Dylan Distasio <inte...@gmail.com> wrote:

> Having read that letter, I don't find it very becoming of the scientists writing it who should know better.   Regardless of what you think about IIT and its merits or lack thereof, it results in some predictions that can be tested

What about consciousness did the ITT theory, or any other consciousness theory,  predict?  

John K Clark    See what's on my new list at  Extropolis

wcp

[John Clark]

On Thu, Sep 21, 2023 at 3:00 PM Jason Resch <jason...@gmail.com> wrote:

> By its own definitions IIT is not falsifiable, for it proclaims that a computer program that gave identical behavior in all situations to another conscious system, would not be conscious. But since it has identical behavior there is no objective way to prove this assertion of IIT (that one system is conscious while the other is not).

Exactly, and that's why I can't prove that IIT is wrong but I can prove it's a silly theory. 

> So is it pseudoscience? I don't know if I would call it that 

I would.  

> This also implies the possibility of philosophical zombies (which IIT proponents freely admit), which also implies consciousness is epiphenomenonal,

If Darwin is right, and I think he is, then consciousness must be epiphenomenal, a side effect of intelligence, because random mutation and natural selection cannot detect consciousness and yet it managed to produce it at least once (in me) and probably many billions of times, and the only reason it was able to do that is because Evolution COULD detect intelligent behavior. It also implies that either philosophical zombies are impossible or it would be easier to make a conscious Artificial Intelligence than an Artificial Intelligence that was not conscious. So if you observe a computer behaving intelligently your default position should be that it's conscious, which is the exact same position we take when we observe one of our fellow human beings behaving intelligently.  

> I do find some strengths in some of the ideas that have come out of it, in particular how a system must be capable of affecting itself for it to be aware of its consciousness.

It's pretty obvious that being able to affect yourself is a necessary condition for consciousness, but is it sufficient? A thermostat can affect itself but .... 

> I also think it is right to put the focus on information.

That is also pretty obvious, I don't need to read  an ITT paper to know that.  

> I think where it errs is in confusing a logical-informational state with a instantaneous physical state. This leads to the mistaken belief that a parallel computation is more conscious than a serial computation, even when they compute the exact same function

I strongly agree with you about that.  

John K Clark    See what's on my new list at  Extropolis

gms

wen

[Brent Meeker]

Scott Aaronson is far from "a bunch of highschoolers".

Why I Am Not An Integrated Information Theorist (or, The Unconscious Expander)

Recently, lots of people have been asking me what I think about IIT—no, not the Indian Institutes of Technology, but Integrated Information Theory, a widely-discussed “mathematical theory of consciousness” developed over the past decade by the neuroscientist Giulio Tononi.  One of the askers was Max Tegmark, who’s enthusiastically adopted IIT as a plank in his radical mathematizing platform (see his paper “Consciousness as a State of Matter”).  When, in the comment thread about Max’s Mathematical Universe Hypothesis, I expressed doubts about IIT, Max challenged me to back up my doubts with a quantitative calculation.

So, this is the post that I promised to Max and all the others, about why I don’t believe IIT.  And yes, it will contain that quantitative calculation.

Read the rest at:
https://scottaaronson.blog/?p=1799

Brent

To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAJrqPH94En0oZRkN891QadG8NSX%3DBTQ-bUw4sBij1JfjPJShqg%40mail.gmail.com.

[Dylan Distasio]

I love Aaronson.   I also don't see him as a signatory on that letter.   I was commenting on the authors whining and winging about IIT in a publication versus just getting on with the science and publishing evidence for their theories of consciousness instead of crying about "scientific misinformation" and "protecting the public" because IIT got mentioned in Nature and the popular press.   Let the science speak for itself by publishing actual papers.

To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/92b5d59c-2e92-7c41-c34c-c6c4a4a4ea58%40gmail.com.

[spudb...@aol.com]

Well, Observer Moments go back, at least, to Schrödinger & Wigner, if not to Boltzmann. There may be, I speculate because I am a Quantum Woo kind of guy, something baked-in to the Universe, that triggers things. 

To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAJrqPH_p6Rt5gu8xLwLyZa40eM3akt_DmfOSfTiQv3nPjL-YVw%40mail.gmail.com.