Integrated Information Theory is peak Camp #2 stuff
As a Camp #2 person, I just want to remark that from my personal viewpoint, Integrated Information Theory is sharing the key defect with Global Workspace Theory, and hence is no better.
Namely, I think that the Hard Problem of Consciousness has the Hart Part: the Hard Problem of Qualia. As soon as the Hard Problem of Qualia is solved, the rest of the Hard Problem of Consciousness is much less mysterious (perhaps, the rest can be treated in the spirit of the “Easy Problems of Consciousness”, e.g. the question why I am me and not another person might be treatable as a symmetry violation, a standard mechanism in physics, and the question why human qualia seem to normally cluster into belonging to a particular subject (my qualia vs. all other qualia) might not be excessively mysterious either).
So the theory purporting to actually solve the Hard Problem of Consciousness needs to shed some light onto the nature and the structure of the space of qualia, in order to be a viable contender from my personal viewpoint.
Unfortunately, I am not aware of any such viable contenders, i.e. of any theories shedding much light onto the nature and the structure of the space of qualia. Essentially, I suspect we are still mostly at square zero in terms of our progress towards solving the Hard Problem of Qualia (I should be able to talk about this particular shade of red, and this particular smell of coffee, and this particular psychedelic sound and what they are or what they are made of, so that the ways they are perceived do come through, and if I can’t, a candidate theory has not even started to address what matters personally to me).
So, as a temporary measure, I am just adjoining qualia as primitives to my overall world view and building on top of that. For example, as a temporary measure, I would model a subjective reality as a neural machine processing vectors representing linear combinations of qualia among other things, so I am talking about “a vector space generated by qualia and, perhaps, also by other base elements”. This is not a pathological assumption; if we eventually understand the nature of qualia, it is still a mathematically legitimate construction to consider a vector space generated by them.
As long as one does take a leap of faith and adjoins qualia in this fashion, classical neuroscience theories (e.g. the old 40hz Crick-Koch theory, or Global Workspace Theory) might just work for a Camp #2 person.
:-) But yes, I certainly do agree with the main punchline of this post :-)
I think a lot of Camp #2 people would agree with you that IIT doesn’t make meaningful progress on the hard problem. As far as I remember, it doesn’t even really try to; it just states that consciousness is the same thing as integrated information and then argues why this is plausible based on intuition/simplicity/how it applies to the brain and so on.
I think IIT “is Camp #2 stuff” in the sense that being in Camp #2 is necessary to appreciate IIT—it’s definitely not sufficient. But it does seem necessary because, for Camp #1, the entire approach of trying to find a precise formula for “amount of consciousness” is just fundamentally doomed, especially since the math doesn’t require any capacity for reporting on your conscious states, or really any of the functional capabilities of human consciousness. In fact, Scott Aaronson claims (haven’t read the construction myself) here that
the system that simply applies the matrix W to an input vector x—has an enormous amount of integrated information Φ
So yeah, Camp #2 is necessary but not sufficient. I had a line in an older version of this post where I suggested that the Camp #2 memeplex is so large that, even if you’re firmly in Camp #2, you’ll probably find some things in there that are just as absurd to you as the Camp #1 axiom.
(Some years ago I tried to search for “qualia” in IIT texts, and I think I got literally zero results; I was super disappointed to discover that indeed “it doesn’t even really try to make meaningful progress on the hard problem”.
So the theory purporting to actually solve the Hard Problem of Consciousness needs to shed some light onto the nature and the structure of the space of qualia, in order to be a viable contender from my personal viewpoint.
Unfortunately, I am not aware of any such viable contenders, i.e. of any theories shedding much light onto the nature and the structure of the space of qualia. Essentially, I suspect we are still mostly at square zero in terms of our progress towards solving the Hard Problem of Qualia (I should be able to talk about this particular shade of red, and this particular smell of coffee, and this particular psychedelic sound and what they are or what they are made of, so that the ways they are perceived do come through, and if I can’t, a candidate theory has not even started to address what matters personally to me).
As another Camp #2 person, I mostly agree—IIT is at best barking up a different wrong tree from the functionalist accounts—but Russellian [1]monism makes it at least part of the way to square 1. The elevator pitch goes like this:
On the one hand, we know an enormous amount about what physical entities do, and nothing whatsoever about what they are. The electromagnetic field is the field that couples to charges in such and such a way; charge is the property in virtue of which particles couple to the electromagnetic field. To be at some point X in space is to interact with things in the neighborhood of X; to be in the neighborhood of X is (among other things) to interact with things at X. For all we know there might not be such things as things at all: nothing (except perhaps good taste) compels us to believe that “electrons” are anything more than a tool for making predictions about future observations.
On the other hand, we’re directly acquainted with the intrinsic nature of at least some qualia, but know next to nothing about their causal structure. I know what red is like, I know what blue is like, I know what high pitches are like, and I know what low pitches are like, but nothing about those experiences seems sufficient to explain why we experience purple but not highlow.
So we have lawful relations of opaque relata and directly accessible relata with inexplicable relations: maybe there’s just the one sort of stuff, which simultaneously grounds physics and constitutes experience.
Is it right? No clue. I doubt we’ll ever know. But it’s at least the right sort of theory.
If your intuitions about the properties of qualia are the same as mine, you might appreciate this schizo theory pattern-matching them to known physics.
Neutral monism does sound like a good direction to probe further.
I doubt we’ll ever know.
If we survive long enough, we might live to see a convincing solution for the “Hard Problem”.
If we don’t solve this ourselves, then I expect that advanced AIs will get very curious about what is this thing (“subjective experience”, “qualia”) those humans are talking about and they will get very curious about finding ways to experience those things themselves. And being very smart, they might have better chances to solve this.
But groups of humans might also try to organize to solve this themselves (I think not nearly enough is done at present, both theoretically and empirically; for example, people often tend to assume that Neuralink-style interfaces are absolutely necessary to explore hybrid consciousness between biological entities and electronics, but I strongly suspect that a lot can be done with non-invasive interfaces (which is much cheaper/easier/quicker to accomplish and also somewhat safer (although still not quite safe) for participating biological entities))...
That’s for experiments. For theory, we just need to demand what we usually demand of novel physics: non-trivial novel experimental predictions of subjectively observable effects. Some highly non-standard ways to obtain strange qualia or to synchronize two minds, something like that. Something we don’t expect, and which a new candidate theory predicts, and which turns out to be correct… That’s how we’ll know that a particular candidate theory in question is more than just a philosophical take...
So the theory purporting to actually solve the Hard Problem of Consciousness needs to shed some light onto the nature and the structure of the space of qualia, in order to be a viable contender from my personal viewpoint
I get “the nature” part, but why the structure is a part of the Hard Problem? It sure would be nice to have more advanced neuroscience, but we already have working theories about structure and can make you see blue instead of red. So it’s not a square zero situation.
Because qualia are related to each other. We want to understand that relation at least to some extent(otherwise it is unlikely that we’ll understand what they are reasonably well). Our subjective reality is made out of them, but their interrelations probably do matter a lot.
can make you see blue instead of red
But this example is about relationship between physical stimuli and qualia (in this particular instance, “red” is a not a quale, only “blue” is a quale (and “red” is a physical stimulus which would result in a red quale under different conditions).
But yes, we do understand quite a bit about color qualia (e.g. we understand that we can structure them in a three-dimensional space if we want to do so based on how mixed colors are perceived in various psychophysical experiments (so that’s a parametrization of them by a subset of physical stimuli), or we can consider them independent and consider an infinite-dimensional space generated by them as atomic primitives, and it’s not all that clear which of these ways is more adequate for the purpose of describing the structure of subjective experience (which seems to be likely to be much older historically than humanity experience with deliberately mixing the colors)).
Quoting my old write-up:
what is the dimension of the space of subjective
colors? This depends quite a bit on whether a particular person wants to state that brown is
a kind of dark orange (then one is probably heading towards three-dimensional space of
subjective colors), or if, contrary to that, one wants to state that brown is a very particular sensation,
quite independent from orange (then one is probably heading towards a high-dimensional or,
perhaps, even infinitely-dimensional space of subjective colors).
However, if one fancies to choose the infinitely-dimensional space, some colors are still similar to each other, they do change gradually, there is still a non-trivial metric between them, this space is not well-understood...
But when I say “we are at square zero”, “color qualia” are not just “abstract colors”, but “subjectively perceived colors”, and we just don’t understand what this means… Like, at all… We do understand quite a bit about the “physical colors ⇒ neural processing” chain, but that chain is as “subjectively colorless” as any theoretical model, as “subjectively colorless” as words “red” and “blue” in this text (in my perception at the moment)...
I do hope we’ll start making progress in this direction sooner rather than later...
As a Camp #2 person, I just want to remark that from my personal viewpoint, Integrated Information Theory is sharing the key defect with Global Workspace Theory, and hence is no better.
Namely, I think that the Hard Problem of Consciousness has the Hart Part: the Hard Problem of Qualia. As soon as the Hard Problem of Qualia is solved, the rest of the Hard Problem of Consciousness is much less mysterious (perhaps, the rest can be treated in the spirit of the “Easy Problems of Consciousness”, e.g. the question why I am me and not another person might be treatable as a symmetry violation, a standard mechanism in physics, and the question why human qualia seem to normally cluster into belonging to a particular subject (my qualia vs. all other qualia) might not be excessively mysterious either).
So the theory purporting to actually solve the Hard Problem of Consciousness needs to shed some light onto the nature and the structure of the space of qualia, in order to be a viable contender from my personal viewpoint.
Unfortunately, I am not aware of any such viable contenders, i.e. of any theories shedding much light onto the nature and the structure of the space of qualia. Essentially, I suspect we are still mostly at square zero in terms of our progress towards solving the Hard Problem of Qualia (I should be able to talk about this particular shade of red, and this particular smell of coffee, and this particular psychedelic sound and what they are or what they are made of, so that the ways they are perceived do come through, and if I can’t, a candidate theory has not even started to address what matters personally to me).
So, as a temporary measure, I am just adjoining qualia as primitives to my overall world view and building on top of that. For example, as a temporary measure, I would model a subjective reality as a neural machine processing vectors representing linear combinations of qualia among other things, so I am talking about “a vector space generated by qualia and, perhaps, also by other base elements”. This is not a pathological assumption; if we eventually understand the nature of qualia, it is still a mathematically legitimate construction to consider a vector space generated by them.
As long as one does take a leap of faith and adjoins qualia in this fashion, classical neuroscience theories (e.g. the old 40hz Crick-Koch theory, or Global Workspace Theory) might just work for a Camp #2 person.
:-) But yes, I certainly do agree with the main punchline of this post :-)
I think a lot of Camp #2 people would agree with you that IIT doesn’t make meaningful progress on the hard problem. As far as I remember, it doesn’t even really try to; it just states that consciousness is the same thing as integrated information and then argues why this is plausible based on intuition/simplicity/how it applies to the brain and so on.
I think IIT “is Camp #2 stuff” in the sense that being in Camp #2 is necessary to appreciate IIT—it’s definitely not sufficient. But it does seem necessary because, for Camp #1, the entire approach of trying to find a precise formula for “amount of consciousness” is just fundamentally doomed, especially since the math doesn’t require any capacity for reporting on your conscious states, or really any of the functional capabilities of human consciousness. In fact, Scott Aaronson claims (haven’t read the construction myself) here that
So yeah, Camp #2 is necessary but not sufficient. I had a line in an older version of this post where I suggested that the Camp #2 memeplex is so large that, even if you’re firmly in Camp #2, you’ll probably find some things in there that are just as absurd to you as the Camp #1 axiom.
Yes, I agree with all this.
(Some years ago I tried to search for “qualia” in IIT texts, and I think I got literally zero results; I was super disappointed to discover that indeed “it doesn’t even really try to make meaningful progress on the hard problem”.
I was particularly disappointed because it came from Christof Koch, and their “40hz paper” from 1990 has been a revelation and a remarkable conceptual breakthrough Francis Crick and Christof Koch, “Towards a neurobiological theory of consciousness”, so I had all those hopes and expectations for IIT because it was from Koch :-) :-( )
As another Camp #2 person, I mostly agree—IIT is at best barking up a different wrong tree from the functionalist accounts—but Russellian [1]monism makes it at least part of the way to square 1. The elevator pitch goes like this:
On the one hand, we know an enormous amount about what physical entities do, and nothing whatsoever about what they are. The electromagnetic field is the field that couples to charges in such and such a way; charge is the property in virtue of which particles couple to the electromagnetic field. To be at some point X in space is to interact with things in the neighborhood of X; to be in the neighborhood of X is (among other things) to interact with things at X. For all we know there might not be such things as things at all: nothing (except perhaps good taste) compels us to believe that “electrons” are anything more than a tool for making predictions about future observations.
On the other hand, we’re directly acquainted with the intrinsic nature of at least some qualia, but know next to nothing about their causal structure. I know what red is like, I know what blue is like, I know what high pitches are like, and I know what low pitches are like, but nothing about those experiences seems sufficient to explain why we experience purple but not highlow.
So we have lawful relations of opaque relata and directly accessible relata with inexplicable relations: maybe there’s just the one sort of stuff, which simultaneously grounds physics and constitutes experience.
Is it right? No clue. I doubt we’ll ever know. But it’s at least the right sort of theory.
As in Bertrand Russell
If your intuitions about the properties of qualia are the same as mine, you might appreciate this schizo theory pattern-matching them to known physics.
Yes.
Neutral monism does sound like a good direction to probe further.
If we survive long enough, we might live to see a convincing solution for the “Hard Problem”.
If we don’t solve this ourselves, then I expect that advanced AIs will get very curious about what is this thing (“subjective experience”, “qualia”) those humans are talking about and they will get very curious about finding ways to experience those things themselves. And being very smart, they might have better chances to solve this.
But groups of humans might also try to organize to solve this themselves (I think not nearly enough is done at present, both theoretically and empirically; for example, people often tend to assume that Neuralink-style interfaces are absolutely necessary to explore hybrid consciousness between biological entities and electronics, but I strongly suspect that a lot can be done with non-invasive interfaces (which is much cheaper/easier/quicker to accomplish and also somewhat safer (although still not quite safe) for participating biological entities))...
That’s for experiments. For theory, we just need to demand what we usually demand of novel physics: non-trivial novel experimental predictions of subjectively observable effects. Some highly non-standard ways to obtain strange qualia or to synchronize two minds, something like that. Something we don’t expect, and which a new candidate theory predicts, and which turns out to be correct… That’s how we’ll know that a particular candidate theory in question is more than just a philosophical take...
I get “the nature” part, but why the structure is a part of the Hard Problem? It sure would be nice to have more advanced neuroscience, but we already have working theories about structure and can make you see blue instead of red. So it’s not a square zero situation.
Because qualia are related to each other. We want to understand that relation at least to some extent(otherwise it is unlikely that we’ll understand what they are reasonably well). Our subjective reality is made out of them, but their interrelations probably do matter a lot.
But this example is about relationship between physical stimuli and qualia (in this particular instance, “red” is a not a quale, only “blue” is a quale (and “red” is a physical stimulus which would result in a red quale under different conditions).
But yes, we do understand quite a bit about color qualia (e.g. we understand that we can structure them in a three-dimensional space if we want to do so based on how mixed colors are perceived in various psychophysical experiments (so that’s a parametrization of them by a subset of physical stimuli), or we can consider them independent and consider an infinite-dimensional space generated by them as atomic primitives, and it’s not all that clear which of these ways is more adequate for the purpose of describing the structure of subjective experience (which seems to be likely to be much older historically than humanity experience with deliberately mixing the colors)).
Quoting my old write-up:
However, if one fancies to choose the infinitely-dimensional space, some colors are still similar to each other, they do change gradually, there is still a non-trivial metric between them, this space is not well-understood...
But when I say “we are at square zero”, “color qualia” are not just “abstract colors”, but “subjectively perceived colors”, and we just don’t understand what this means… Like, at all… We do understand quite a bit about the “physical colors ⇒ neural processing” chain, but that chain is as “subjectively colorless” as any theoretical model, as “subjectively colorless” as words “red” and “blue” in this text (in my perception at the moment)...
I do hope we’ll start making progress in this direction sooner rather than later...