I wanted to repost something that was posted in the consciousness and science forum. I had an exchange with the author of a paper that I criticized, and I think that there is some good content for this thread. In particular, I address the issue of Integrated Information Theory and panpsychism.
Hi, I'm the author. I really appreciate all the great discussion I see here, and there is some great new information as well. However, there are also some false criticisms of the article. The examples I give are certainly correct. For any system that integrates information, IIT calculates a precise number that quantifies the amount of integrated information (phi). You can certainly calculate a phi number for a beetle, or a cat, or a dog, or a computer. This is the most fundamental principle in the entire theory. I suggest reading up on IIT if you don't believe so.
Thank you for taking the time to respond to my criticism.
Regarding your specific claim in your article, you said that "[...]the theory posits that any system that processes and integrates information, be it organic or inorganic, experiences the world subjectively to some degree. Plants, smartphones, the Internet--even protons--are all examples of such systems."
In the paper by Tononi and Koch (Consciousness: Here, There but Not Everywhere, 2014), they seemed to be saying that the internet would not have phi:
"A related question has to do with the Internet and whether it could be conscious (Koch 2014). One way to think about this is to assume that each computer connected to the internet is an element having real causal power at the macro-level. [...] In principle, this kind of organization could be arranged so that it gives rise to a complex of high [phi],
although this is certainly not the way the internet works right now." (pg 14, emphasis added)
Regarding computers, I did find a part that does support your claim that computers may have some level of phi:
"In that case, we suspect that the computer would likely
not form a large complex of high [phi max], but break down into many mini-complexes of low [phi max] (due to the small fan-in and fan-out of digital circuitry, fig. 5C), existing at the very fast temporal scale of the computer clock." (ibid, pg 8)
However, this is made a bit more clear in the abstract of the same paper:
"Also, in sharp contrast with widespread functionalist beliefs, IIT implies that digital computers, even if their behavior were to be functionally equivalent to ours, and even if they were to run faithful simulations of the human brain, would experience next to nothing."
So this still does not support the claim that computers would be conscious, rather that there may be a possibility of small complexes within a computer having a low phi.
Bobby Azarian said:
As far as Tononi goes, he and Koch are developing IIT together. Koch is very outspoken about IIT being a form of 'neo-panpsychism', if you will. Although Tononi keeps quiet about this, he is in agreement...
You are correct that Koch has said a lot about IIT being a form of panpsychism, and in his book, Consciousness: Confessions of a Romantic Reductionist, he does make statements such as the following:
"By postulating that consciousness is a fundamental feature of the universe, rather than emerging out of simpler elements, integrated information theory is an elaborate version of
panpsychism." (pg 132)
And,
"The entire cosmos is suffused with sentience. We are surrounded and immersed in consciousness; it is in the air we breathe, the soil we tread on, the bacteria that colonize our intestines, and the brain that enables us to think." (pg 132)
However, Tononi is quite different in opinion here:
"How close is this position to panpsychism, which holds that everything in the universe has some kind of consciousness? Certainly, the IIT implies that many entities, as long as they include some functional mechanisms that can make choices between alternatives, have some degree of consciousness. Unlike traditional panpsychism, however, the IIT does not attribute consciousness indiscriminately to all things." (Consciousness as Integrated Information: a Provisional Manifesto, 2008, pg 236)
He continues:
"Moreover, panpsychism hardly has a solid conceptual foundation. The attribution of consciousness to all kinds of things is based more on an attempt to avoid dualism than on a principled analysis of what consciousness is. Similarly, panpsychism offers hardly any guidance as to what would determine the amount of consciousness associated with different things (such as humans, animals, plants, or rocks), or with the same thing at different times (say wakefulness and sleep), not to mention that it says nothing about what would determine the quality of experience." (ibid, pg 236)
Even in the paper with Koch (Tononi and Koch, 2014), they state the following in the abstract:
"The theory vindicates some intuitions often associated with panpsychism - that consciousness is an intrinsic, fundamental property, and is graded, is common among biological organisms, and even some very simple systems may have some of it. However, unlike panpsychism, IIT implies that not everything is conscious, for example aggregates such as heaps of sand, a group of individuals or feed-forward networks."
Considering that panpsychism is defined as "the doctrine or belief that everything material, however small, has an element of individual consciousness" or, sometimes more literal to the definition, that
mind or psyche is fundamental to all things, IIT cannot be considered panpsychism under either of these definitions.
Bobby Azarian said:
Where there is integrated information, there is consciousness, plain and simple.
There is more required than integrated information, as states by Tononi:
"Second, (strong) integration is by no means the only requirement for consciousness, but must be complemented by information and exclusion." (pg 20, From the Phenomenology to the Mechanisms of Consciousness: Integrated Information Theory 3.0, 2014)
Bobby Azarian said:
The 'local maxima' concept is important when asking whether consciousness is everywhere, or only in certain systems. For example, the Internet will not be conscious if it, as a whole, does not integrate more information than any of its subcomponents, which includes individual computers. Therefore, it is likely that the internet is not conscious, but Koch ponders the idea since an exact calculation is intractable at this time.
But it is also stated in the paper (Tononi and Koch, 2014) that:
"On the other hand, if one considers the micro-elements inside each computer (say its transistors) as having real causal power, we are back to the situation in which they most likely would not form any large complex within each computer, let alone across connected computers." (pg 14)
Bobby Azarian said:
Why do I say a subatomic particle is conscious? It's a direct quote from Koch. As the article says, a triad of quarks constitutes an integrated system. Therefore, there is a modicum of subjective experience. So basically, consciousness IS everywhere, but it is to such a small degree that it is essentially negligible. That's why Koch and Tononi stress the whole local maxima thing so much. These are the only places where consciousness exists to a degree worth calling consciousness. It should also be known that Koch and Tononi have been very inconsistent in regard to IIT, and some very important features change every time they create a new iteration. I tend to side with Searle, and think that consciousness is unique to biological systems. Subjective experience is only supported by certain hardwares, which have unique molecular properties. I have an article coming out that expresses this view, and I will check back here when it is published. However, IIT is certainly plausible, and it is definitely an information-based form of panpsychism. Also, if Koch says, assume Tononi agrees, until Tononi says otherwise. They are also buddies with Chalmers, so you can see Chalmer's functionalism as being compatible with IIT.
Koch has made that claim about quark complexes, but I have not been able to find any such claim made by Tononi, or even in a paper in which Tononi is a co-author. Since this is a claim of Koch, that is why I said it is not a claim of IIT. I don't find it correct to assume that Tononi agrees just because Koch has made the comment. If none of the papers on IIT state this, then I cannot say that it is a statement of IIT.
The other problem with this comment by Koch is that even if a system of quarks in a proton constitute an integrated system, this system cannot possibly be conscious because this claim would violate one of the axioms of IIT: the Exclusion axiom. Subatomic particles exist in indeterminate or superposed states, and the exclusion axiom explicitly forbids this:
"[...] at any given time there is only one experience having its full content, rather than a superposition of multiple partial experiences [...]" (pg 3, From the Phenomenology to the Mechanisms of Consciousness: Integrated Information Theory 3.0, Tononi 2014).