Consciousness and The Interface Theory of Perception

S

Sciborg_S_Patel

Consciousness and The Interface Theory of Perception, Donald Hoffman

Despite substantial efforts by many researchers, we still have no scientific theory of how brain activity can create, or be, conscious experience. This is troubling, since we have a large body of correlations between brain activity and consciousness, correlations normally assumed to entail that brain activity creates conscious experience. Here I explore a solution to the mind-body problem that starts with the converse assumption: these correlations arise because consciousness creates brain activity, and indeed creates all objects and properties of the physical world. To this end, I develop two theses. The interface theory of perception states that perceptual experiences do not match or approximate properties of the objective world, but instead provide a simplified, species-specific, user interface to that world. Conscious realism states that the objective world consists of conscious agents and their experiences; these can be mathematically modeled and empirically explored in the normal scientific manner.

In support of the interface theory of perception, I present Monte Carlo simulations of evolutionary games in which perceptual strategies that see the truth compete with perceptual strategies that do not see the truth but are instead tuned to fitness. The result is that natural selection drives true perceptions to swift extinction. Our perceptions have evolved to guide adaptive behaviors, not to report the truth.

In support of conscious realism, I present a dynamical theory of consciousness in which the observer and the observed have precisely the same mathematical structure, i.e., in which there is a mathematically precise nondualism. I then derive the quantum wave function of the free particle from the asymptotic behavior of the conscious dynamics. This is a step toward solving the mind-body problem from the assumption that consciousness, not physics, is fundamental.
 
Hoffman's interface theory has come up before. I have not see that video; thanks for the link. Here is his paper:

www.cogsci.uci.edu/~ddhoff/interface.pdf‎

So why do we discover a certain level of illusion when we study, say, vision, but not an arbitrarily deep illusory experience?

~~ Paul

According to Hoffman we've edited out the part of reality that would show us we're in an arbitrarily deep illusion.

I'm going through the paper, Conscious Realism, now. I'll see where he goes with this. This is interesting:

The theories so far proposed by scientists are, at best, hints about where to look for a genuine scientific theory. None of them remotely approaches the minimal explanatory power, quantitative precision, and novel predictive capacity expected from a genuine scientific theory. We would expect, for instance, that such a theory could explain, in principle, the difference in experience between, e.g., the smell of a rose and the taste of garlic. How, precisely, is the smell of a rose generated by a 40 Hz oscillation, a reentrant thalamocortical circuit, information integration, a global-workspace entry, the quantum state of microtubules, or the collapse of evolving templates ? What precise changes in these would transform experience from the smell of a rose to the taste of garlic ? What quantitative principles account for such transformations ? We are not asking about advanced features of consciousness, such as self-consciousness, that are perhaps available to few species. We are asking about an elementary feature available, presumably, to a rat. But no current theory has tools to answer these questions and none gives guidance to build such tools. None begins to dispel the mystery of conscious experience. As Pinker (1997, p. 564) points out, “. . . how a red-sensitive neuron gives rise to the subjective feel of redness is not a whit less mysterious than how the whole brain gives rise to the entire stream of consciousness.”

In short, the scientific study of consciousness is in the embarrassing position of having no scientific theory of consciousness. This remarkable situation provokes several responses...
 
According to Hoffman we've edited out the part of reality that would show us we're in an arbitrarily deep illusion.
So he's taken a big step toward making his own hypothesis unfalsifiable? That's convenient.

What has he done other than copied various equations and replaced the low-level elements with conscious agenty-sounding new terms? Heck, for all we know, he worked backward. Does it make any interesting predictions?

Also, I'm not sure he has a grip on falsifiability. Didn't he say that the Church-Turing Thesis is falsifiable? I don't think so.

Also, I wish he'd stop throwing the word true around so cavalierly. The desktop interface better give us a fundamentally true picture of what's going on, or we're going to make lots of mistakes. It may be significantly abstracted, but that is not the same as "not true."

Time to look around for some critique on the interface theory.

~~ Paul
 
So he's taken a big step toward making his own hypothesis unfalsifiable? That's convenient.

That's a weird leap, as it directly contradicts what he says. In the presentation he presents a way to falsify his theory.

Also, I wish he'd stop throwing the word true around so cavalierly. The desktop interface better give us a fundamentally true picture of what's going on, or we're going to make lots of mistakes. It may be significantly abstracted, but that is not the same as "not true."

He already discusses this distinction. Did you read the Conscious Realism paper?

Time to look around for some critique on the interface theory.

I'd make sure you aren't presenting a caricature of his position before trying to cast doubt on it.
 
That's a weird leap, as it directly contradicts what he says. In the presentation he presents a way to falsify his theory.
Listening again. At what timestamp does he discuss this? Perhaps you are referring to his statement that he has to be able to derive QM.

He already discusses this distinction. Did you read the Conscious Realism paper?
It's been awhile. What does he say?

I'd make sure you aren't presenting a caricature of his position before trying to cast doubt on it.
Yes, that is a good idea.

~~ Paul
 
Last edited:
Can you try explaining what you mean by '...but not an arbitrarily deep illusory experience...' I'm not sure what that is?
Why is it that we can discover illusions down to a certain level, but apparently no further?

He says that he has to be able to derive all of quantum mechanics from his theory. But if all he does is come up with a different model for quantum mechanics that produces exactly the same predictions, what has he accomplished? In particular, would he agree that our perceptions are true down to the level of QM?

Note that he says he's presenting his model in the spirit of the Church-Turing Thesis, which is not a theorem but a thesis. In fact, he calls it a "framework."

~~ Paul
 
Last edited:
Regarding the idea that an icon doesn't have to represent true aspects of the thing being represented, Hoffman says:

"This theory addresses the natural question, "If our perceptions are not
accurate, then what good are they?" The answer becomes obvious for user
interfaces. The colour, for instance, of an icon on a computer screen does
not estimate, or reconstruct, the true colour of the file that it represents
in the computer. If an icon is, say, green, it would be ludicrous to
conclude that this green must be an accurate reconstruction of the true
colour of the file it represents. It would be equally ludicrous to conclude
that, if the colour of the icon doesn't accurately reconstruct the true
colour of the file, then the icon's colour is useless, or a blatant
deception. This is simply a naive misunderstanding of the point of a user
interface. The conventionalist theory that our perceptions are
reconstructions is, in precisely the same manner, equally naive."

What he says is certainly true, but it misleads us into thinking that nothing about the icon represents true aspects of the file. Of course, we can list a dozen things about the icon/file relationship that are true, which the icon represents just fine. And we could easily add a few more truths about the file if we enhanced the icon (e.g., which disk it's on, how big it is).

So this leaves me not understanding how he divides the world of perceptions into true perceptions and not-true perceptions.

~~ Paul
 
Last edited:
Interesting. But given skeptics tendency to glide over uncomfortable truths - see Churchland trying to brush past Alex mentioning Koch's panpsychism - I'll have to send this to Hoffman and see if I can get a response.
Yes, see if you can. Particularly since Hoffman is more of a skeptic about current physics than is Kaznatcheev.

Consider a fundamental problem here. Hoffman's mathematical model of reality is, according to him, an interface to the underlying reality. It therefore has no more evolutionary reason to represent the truth than does any other model. And, as usual, the whole conscious agent thing is just an interpretation of his math.

Again, though, as I explained in post #11, I don't really understand what he means by truth.

~~ Paul
 
Last edited:
Why is it that we can discover illusions down to a certain level, but apparently no further?

He says that he has to be able to derive all of quantum mechanics from his theory. But if all he does is come up with a different model for quantum mechanics that produces exactly the same predictions, what has he accomplished? In particular, would he agree that our perceptions are true down to the level of QM?

Note that he says he's presenting his model in the spirit of the Church-Turing Thesis, which is not a theorem but a thesis. In fact, he calls it a "framework."

Thanks Paul, I've never heard of this idea before, so it's new to me. I listened to his theory and I like bits of it, but I can't see how he translates these ideas right up to my everyday experiences, and how, or what, consciousness is. Again, he also seems to just start with consciousness as existing, and no explanation of how it comes into being. He moves from one, to two, to three conscious agents etc. but they are apparently already conscious? I'm OK with feedback loops, and that we interact with each other, but how...?

It's a lot of maths which goes way over my head, and as you say, he seems to feel the similarity he has found with quantum physics as significant, but as we don't understand quantum physics, and indeed we have fundamental problems with it (measurement problem etc.) I don't really understand what he's showing me. Other than consciousness has something to do with our perception of reality (i.e. quantum physics).

I definitely accept that consciousness is all that science has been observing. Perception is indirect in my view, so I can't see any way around that. But I don't get his claim that there are no public physical objects... that appears to defy my experience of the world... but perhaps he's just not explaining this claim very well.

I have many problems with this, yet I do like parts of it...
 
Yes, see if you can. Particularly since Hoffman is more of a skeptic about current physics than is Kaznatcheev.

Consider a fundamental problem here. Hoffman's mathematical model of reality is, according to him, an interface to the underlying reality. It therefore has no more evolutionary reason to represent the truth than does any other model. And, as usual, the whole conscious agent thing is just an interpretation of his math.

Again, though, as I explained in post #11, I don't really understand what he means by truth.

~~ Paul

I absolutely agree with that... somebody in the audience seized on the same issue as to how he was defining truth... and he just dismissed it by saying he could model truth, and was therefore able to play god? With that I just found myself back at square one.
 
Thanks Paul, I've never heard of this idea before, so it's new to me. I listened to his theory and I like bits of it, but I can't see how he translates these ideas right up to my everyday experiences, and how, or what, consciousness is. Again, he also seems to just start with consciousness as existing, and no explanation of how it comes into being. He moves from one, to two, to three conscious agents etc. but they are apparently already conscious? I'm OK with feedback loops, and that we interact with each other, but how...?
It appears that he starts with a fundamental agent whose consciousness is a given. Notice that he does not make any attempt to calculate the various mappings between the parts of the agents: W (world), X (experiences), G (actions).

It's a lot of maths which goes way over my head, and as you say, he seems to feel the similarity he has found with quantum physics as significant, but as we don't understand quantum physics, and indeed we have fundamental problems with it (measurement problem etc.) I don't really understand what he's showing me. Other than consciousness has something to do with our perception of reality (i.e. quantum physics).
He's trying to show that there is no reason why our perceptions have to be closely related to the underlying reality.

I definitely accept that consciousness is all that science has been observing. Perception is indirect in my view, so I can't see any way around that. But I don't get his claim that there are no public physical objects... that appears to defy my experience of the world... but perhaps he's just not explaining this claim very well.
I have many problems with this, yet I do like parts of it...
There are no public physical objects separate from the interface. There are public aspects of the underlying reality.

~~ Paul
 
I absolutely agree with that... somebody in the audience seized on the same issue as to how he was defining truth... and he just dismissed it by saying he could model truth, and was therefore able to play god? With that I just found myself back at square one.
Was that in his description of the simulations they ran? Those were an attempt to show that evolution favors fitness rather than truth.

~~ Paul
 
I absolutely agree with that... somebody in the audience seized on the same issue as to how he was defining truth... and he just dismissed it by saying he could model truth, and was therefore able to play god? With that I just found myself back at square one.

When does this happen?
 
It appears that he starts with a fundamental agent whose consciousness is a given. Notice that he does not make any attempt to calculate the various mappings between the parts of the agents: W (world), X (experiences), G (actions).


He's trying to show that there is no reason why our perceptions have to be closely related to the underlying reality.


There are no public physical objects separate from the interface. There are public aspects of the underlying reality.

I don't know what any of that means... sorry...
 
Back
Top