The Evolutionary Argument Against Reality

[Yerda]

The gist is this: Evolutionary game theory can be used to argue that our perceptions are selected to hide reality from us. Or reality is what we see, and it aint the world.

Hoffmann and his PhD students have been running simulations of competition between organisms where one group is geared to percieve reality as is and the other is only able to percieve objects based on evoltionary fitness. The result is that fitness drives reality to extinction every time.

Among the implications is the possibility that what we think is reality is infact something like a species specific interface. Going deeper he is proposing a mathematically rigorous scientific model that "predicts" that no physical system has causal power, that physical objects are akin to icons on your interface desktop, that space and time are not fundamental constituents of the world but part of how we model it, and ultimately all of the world can be modelled as conscious observers.

This probably sounds like grade A fruitbattery but I'm pretty sure it isn't. Anyway, if you're interested have a look and leave a comment.

Further info:

Aye, he has a Ted talk like everyone else:

Here he is in chat with renowned philosophers Dan Dennett and David Chalmers:

His website at UCI with links to his papers (click on the link title Vita for loads more): Donald D. Hoffman | University of California, Irvine

He's also a bloody good photographer, hae a look at his flickr account: Donald Hoffman

 

[A Vestigial Mote]

Hoffmann and his PhD students have been running simulations of competition between organisms where one group is geared to percieve reality as is and the other is only able to percieve objects based on evoltionary fitness. The result is that fitness drives reality to extinction every time.

Among the implications is the possibility that what we think is reality is infact something like a species specific interface. Going deeper he is proposing a mathematically rigorous scientific model that "predicts" that no physical system has causal power, that physical objects are akin to icons on your interface desktop, that space and time are not fundamental constituents of the world but part of how we model it, and ultimately all of the world can be modelled as conscious observers.

You basically wrote word-for-word what his dialogue was in the video - I was sort of expecting him to delve into how the "simulations" worked, in what ways they manipulate the surroundings of the organisms involved to make sure that one group is seeing "reality like it is" and that another group is seeing things according to "fitness"... but there was no explanation forthcoming. I have no idea how the man expects anyone to just take his words at face value and accept that he was successful in managing those distinctions to some qualified degree. It makes the whole thing highly suspect, and - this may sound petty of me - especially because he had the digital "waterfall" of green symbols (à la "The Matrix") falling behind and in front of him throughout the video. Having spent a bit of time on that effect, I am sure, he could have, instead, focused on trying to get his message across more succinctly, with a bit more backing than "Hey, we did this, and we did it because I say we did it - did you see the cool effect behind me in the video? Neat right?"

That said, there is some support for what he's saying, but I don't think we are as abstracted from reality as he is proposing. Obviously, at its most fundamental, all matter is composed of atomic particles, and then sub-atomic beyond that, etc. And, naturally, our bodies are built to ignore those levels of "reality." As an example - more important is the ability to sense heat and light, than is the ability to sense interactions on the sub-atomic level. As another, we can't sense infrared or ultraviolet light, but there are other creatures who can. That is a segment of "reality" we don't even experience - because we have never had the need enough for it to become part of our evolutionary outcome. Does this make our interaction with the universe any less relevant than it needs to be? No. Because the only being to which our interaction with the universe matters is ourselves. If there are other facets of reality that would be game changers for us, it almost doesn't matter. You'd still just be living a life to be lived and then die in the same general "place" (aka - The Universe) as everything else.

 

[Nakosis]

The gist is this: Evolutionary game theory can be used to argue that our perceptions are selected to hide reality from us. Or reality is what we see, and it aint the world.

I've been arguing along the same lines for a while now. Like a bat uses sound, birds use magnetic fields, Dogs use smell. I suspect their brains interprets these senses to an interface as detail vivid as sight does for humans. A bat doesn't evaluate the sound pluses, it's brain paints an interface the bat uses to navigate.

Unfortunately since we are stuck with the senses we have, it's impossible to get beyond the interface imposed by our senses.

Best we can do is use a mechanical process to detect reality and convert this to something usable by our interface. It can trigger a sound or print a light image across a screen.

We're just getting better at interpreting reality into a format usable by our interface.

 

[Corvus]

What is reality anyway? An atom is 99.999% empty space. Perhaps we are just a computer simulation, we wouldn't know. We are subjectively within the box along with what we observe, thus we can only know what the limitations on our objectivity permit.

 

[Evangelicalhumanist]

You basically wrote word-for-word what his dialogue was in the video - I was sort of expecting him to delve into how the "simulations" worked, in what ways they manipulate the surroundings of the organisms involved to make sure that one group is seeing "reality like it is" and that another group is seeing things according to "fitness"... but there was no explanation forthcoming. I have no idea how the man expects anyone to just take his words at face value and accept that he was successful in managing those distinctions to some qualified degree. It makes the whole thing highly suspect, and - this may sound petty of me - especially because he had the digital "waterfall" of green symbols (à la "The Matrix") falling behind and in front of him throughout the video. Having spent a bit of time on that effect, I am sure, he could have, instead, focused on trying to get his message across more succinctly, with a bit more backing than "Hey, we did this, and we did it because I say we did it - did you see the cool effect behind me in the video? Neat right?"

That said, there is some support for what he's saying, but I don't think we are as abstracted from reality as he is proposing. Obviously, at its most fundamental, all matter is composed of atomic particles, and then sub-atomic beyond that, etc. And, naturally, our bodies are built to ignore those levels of "reality." As an example - more important is the ability to sense heat and light, than is the ability to sense interactions on the sub-atomic level. As another, we can't sense infrared or ultraviolet light, but there are other creatures who can. That is a segment of "reality" we don't even experience - because we have never had the need enough for it to become part of our evolutionary outcome. Does this make our interaction with the universe any less relevant than it needs to be? No. Because the only being to which our interaction with the universe matters is ourselves. If there are other facets of reality that would be game changers for us, it almost doesn't matter. You'd still just be living a life to be lived and then die in the same general "place" (aka - The Universe) as everything else.

I must say, I find this to be a very good response.

 

[Polymath257]

The gist is this: Evolutionary game theory can be used to argue that our perceptions are selected to hide reality from us. Or reality is what we see, and it aint the world.

And I agree to a large extent. Our sensory apparatus was evolved to promote survival in some particular environments. That is one reason we tend to see faces where no faces exist (clouds, toast, etc), why there are so many different optical illusions, and why logic is so hard for most people (I teach math--believe me, logical thinking is hard for most people).

This is also probably why it took so long for the human race to discover science: to probe deeper than the surface means we have to be able to put aside, through practice, the biases that come along with our senses. We are subject to superstitious thoughts, leaps of logic, and heard mentality. These are difficult to overcome and almost impossible at the individual level.

But, that said, our senses *do* pick up on some aspects of reality, even if only dimly. We may not be able to see more than a tiny part of the spectrum of light, hear only a tiny part of the spectrum of sound, taste and smell only a few chemicals, but those are enough to get some grasp. After that, rigorous testing of ideas and determined skepticism of all deductions is important for deeper learning.

It isn't easy, but it is something that can be done and it does lead to a better understanding of reality.

Now, whether that is healthy in an evolutionary sense is another question.

 

[Polymath257]

I've been arguing along the same lines for a while now. Like a bat uses sound, birds use magnetic fields, Dogs use smell. I suspect their brains interprets these senses to an interface as detail vivid as sight does for humans. A bat doesn't evaluate the sound pluses, it's brain paints an interface the bat uses to navigate.

Unfortunately since we are stuck with the senses we have, it's impossible to get beyond the interface imposed by our senses.

Best we can do is use a mechanical process to detect reality and convert this to something usable by our interface. It can trigger a sound or print a light image across a screen.

We're just getting better at interpreting reality into a format usable by our interface.

Exactly. And over time, we manage to get more and more of that reality into a format that we can use. We cannot see ultra-violet, but we can detect it in other ways. And we can learn that even if our senses don't pick it up, we can still be affected by it. The same is true for any number of other aspects of reality that we cannot directly perceive.

 

[Yerda]

You basically wrote word-for-word what his dialogue was in the video - I was sort of expecting him to delve into how the "simulations" worked, in what ways they manipulate the surroundings of the organisms involved to make sure that one group is seeing "reality like it is" and that another group is seeing things according to "fitness"... but there was no explanation forthcoming. I have no idea how the man expects anyone to just take his words at face value and accept that he was successful in managing those distinctions to some qualified degree. It makes the whole thing highly suspect, and - this may sound petty of me - especially because he had the digital "waterfall" of green symbols (à la "The Matrix") falling behind and in front of him throughout the video. Having spent a bit of time on that effect, I am sure, he could have, instead, focused on trying to get his message across more succinctly, with a bit more backing than "Hey, we did this, and we did it because I say we did it - did you see the cool effect behind me in the video? Neat right?"

Hi. I summarised it because I'm confident that people wouldn't watch the video but might respond if I put the gist in the OP. The video was made by Quanta Magazine (they interviewed him after running an article originally published in The Atlantic) so the special effects are all on them. If you feel like a tough and mathsy read here's a link to a paper on the matter: http://cogsci.uci.edu/~ddhoff/PerceptualEvolution.pdf

 

[Yerda]

Further, I forgot to mention, if this is true physicalism is dead, idealism is nonsense and everything we thought we knew about the world outside of our experience is wrong. Oh the controversy.

 

[Polymath257]

Further, I forgot to mention, if this is true physicalism is dead, idealism is nonsense and everything we thought we knew about the world outside of our experience is wrong. Oh the controversy.

I don't see how that follows. Notice that the video starts by noting the distinction between 'seeing reality as it is' and 'seeing what is important to survival'. Evolution encourages the second, not the first.

But that is OK. It is a starting place. We will have some information, however incomplete, about some aspects of reality. We may not see everything about that tiger that is hunting us, but we definitely see that there is a tiger-like object that looks hungry and is approaching us. We may not see everything about that apple on the tree, but we see enough to know when it is ripe and when it is not.

Evolution selects for those things that keep us alive and breeding and will eliminate other things if they take too much energy; energy better spent elsewhere for survival.

But, humans have evolved big brains for a reason. Partly it is to be able to process social cues. Partly it is to develop language, that helps communicate threats and benefits. And partly it is to be able to plan ahead for rough times.

One of the side effects is that we can generalize concepts. This often leads us astray: superstitions, optical illusions, etc.

But, even though we don't see 'reality as it is', we see enough to be able to test our ideas. We can still learn the reality that our senses hide. We can learn how to test beyond what we can immediately see: to build microscopes and telescopes, and interferometers, and how to detect neutrinos. So, even though the interface is all we have, we can still learn what is behind that interface and where the interface gives us unreliable information.

 

[Yerda]

I don't see how that follows. Notice that the video starts by noting the distinction between 'seeing reality as it is' and 'seeing what is important to survival'. Evolution encourages the second, not the first.

But that is OK. It is a starting place. We will have some information, however incomplete, about some aspects of reality.

If we use the word reality to denote whatever it is we percieve then yes, we do. But Hoffmann seems to argue elsewhere that our perceptions are about as likely to give us a glimpse of objective reality (which he argues for) as the symbol "Dog" telling us anything about an actual dog. Physical objects are akin to symbols and they bear some relationship to the world but can't tell us anything about the world.

You follow me?

 

[Polymath257]

If we use the word reality to denote whatever it is we percieve then yes, we do. But Hoffmann seems to argue elsewhere that our perceptions are about as likely to give us a glimpse of objective reality (which he argues for) as the symbol "Dog" telling us anything about an actual dog. Physical objects are akin to symbols and they bear some relationship to the world but can't tell us anything about the world.

You follow me?

Yes, of course. There is no red apple in our brain. The redness we see is a representation for light of a certain wavelength. But the representation is sparked by something in the environment, and that something is light of a particular wavelength. Just like the word 'dog' is not a dog, but the word does still refer to something outside of us. I understand the analogy of physical objects being like icons. But I disagree with that terminology. The physical object, in that analogy, would be either the file or even the particular representation of the file on the disk. Our consciousness, however, only sees the icon.

In the third video above, I am much closer to Daniel Dennet's viewpoint than either David Chamber's or Donald Hoffman's. Consciousness is something that happens in our brains and is NOT a fundamental aspect of reality. It is how our brains process information. That information, as accumulated by our senses, is imperfect and our attention tends to ignore many aspects (for example, we have a persistent illusion that we see clearly in our whole visual field. In reality, we see clearly only in a very narrow region of our visual field).

 

[Yerda]

Yes, of course. There is no red apple in our brain. The redness we see is a representation for light of a certain wavelength. But the representation is sparked by something in the environment, and that something is light of a particular wavelength.

So what he's saying is that the entire environment is a part of the interface. Space and time are part of the interface and everything in space and time also. The representation is sparked by the world but nothing in the world is in space and time and is therefore fundamentally unlike anything that is represented.

I'm not arguing this is true. I think Hoffmann is though. It seems similar to the little bit of Kant's trancendental idealism that I've encountered.

In the third video above, I am much closer to Daniel Dennet's viewpoint than either David Chamber's or Donald Hoffman's. Consciousness is something that happens in our brains and is NOT a fundamental aspect of reality. It is how our brains process information. That information, as accumulated by our senses, is imperfect and our attention tends to ignore many aspects (for example, we have a persistent illusion that we see clearly in our whole visual field. In reality, we see clearly only in a very narrow region of our visual field).

I'm in a similar position. I still think the brains cause consciousness perspective makes the most sense I just find this approach of Hoffmann's to be pleasant. There a lovely kind of wholeness to it or something.

 

[Polymath257]

So what he's saying is that the entire environment is a part of the interface. Space and time are part of the interface and everything in space and time also. The representation is sparked by the world but nothing in the world is in space and time and is therefore fundamentally unlike anything that is represented.

Well, as Dennet points out, the format for the information in the brain is spikes of neural activity. I think we have to made a distinction between the color red as we see it (the representation) and the wavelengths of light in reality. I disagree that the interface has NOTHING to do with reality. If it didn't there would not be *any* fitness for certain perceptions.

And that is what makes me wonder about his simulation. How does he distinguish between seeing 'some reality' and seeing 'what is fit'? Don't we at least need to see some reality to see those aspects that are fit?

I'm not arguing this is true. I think Hoffmann is though. It seems similar to the little bit of Kant's trancendental idealism that I've encountered.

I think you are right. And I think I disagree with him. An interesting position, though.

But then, I don't believe philosophical zombies are a coherent concept. Again, I follow Dennet more than Chambers or Hoffmann

I'm in a similar position. I still think the brains cause consciousness perspective makes the most sense I just find this approach of Hoffmann's to be pleasant. There a lovely kind of wholeness to it or something.

if only there was actual evidence of this. I notice that Hoffmann accepted that science is still possible, but I wonder what he sees science as doing. He also pointed out that he only focused on perception, not on things like rationality and planning.

So, we perceive this table as solid. We know full well that it is mostly empty space. But we also have a theory of solids made of atoms that are mostly empty space. So, even though we don't *perceive* the reality, we can still know the reality and understand it enough to use it to create new methods to probe it.

 

[LukeS]

Transcendental surrealism?

 

[Yerda]

Well, as Dennet points out, the format for the information in the brain is spikes of neural activity. I think we have to made a distinction between the color red as we see it (the representation) and the wavelengths of light in reality. I disagree that the interface has NOTHING to do with reality. If it didn't there would not be *any* fitness for certain perceptions.

And that is what makes me wonder about his simulation. How does he distinguish between seeing 'some reality' and seeing 'what is fit'? Don't we at least need to see some reality to see those aspects that are fit?

There's a link to a paper a few posts up. It's quite technical but even a quick scan will give you some idea of what he's been upto.

In English the symbol 'apple' denotes the fruit but tells us zero about it. The analogy seems to hold for his view of the empirical world in that we should only expect the symbols that we call reality to tell us what symbols we use. I apologise if I'm being repetitive here.

You're well within your rights to disagree here but I think if his assumptions are correct he's correct about this too. All this fits quite well with ideas about the unity of the world (i.e. the world is no thing in particular) that I find quite attractive so maybe that's why I like it so much.

if only there was actual evidence of this.

It's quite hard to imagine how we could even test it. Hoffmann seems to think quantum mechanics can be derived from his models which would be a good start but it's still going to be a kinda mathematical interpretation rather than an empirically validated theory.

I notice that Hoffmann accepted that science is still possible, but I wonder what he sees science as doing. He also pointed out that he only focused on perception, not on things like rationality and planning.

The same argument he uses to support his idea that our perceptions are not of reality also lead to reliable maths and logic. I guess science remains what it always was just with the proviso that we are aware we aren't modelling the world, only the interface.

 

[Polymath257]

There's a link to a paper a few posts up. It's quite technical but even a quick scan will give you some idea of what he's been upto.

In English the symbol 'apple' denotes the fruit but tells us zero about it. The analogy seems to hold for his view of the empirical world in that we should only expect the symbols that we call reality to tell us what symbols we use. I apologise if I'm being repetitive here.

You're well within your rights to disagree here but I think if his assumptions are correct he's correct about this too. All this fits quite well with ideas about the unity of the world (i.e. the world is no thing in particular) that I find quite attractive so maybe that's why I like it so much.

OK, I agree that our symbol, the word 'apple' gives us no information about the fruit. But the fact that we can and do have certain types of interaction with that apple *does* tell us something about it. So, even if it is false to think that a file represented by a rectangle on my screen is *actually* a rectangle, it is still true that there is *something* that holds the information for that file and that can interact with my GUI. Now, the rectangle is part of the GUI, but the information is part of the file.

So, to continue with the analogy, when I see an apple, the image I have in my head is not the 'real' apple. But that real fruit has the ability to interact in a way that produces, say, the sensation of red. The problem is to determine how much and what aspects of my sensation are from the GUI, how much from the OS and how much from the apple itself.

It's quite hard to imagine how we could even test it. Hoffmann seems to think quantum mechanics can be derived from his models which would be a good start but it's still going to be a kinda mathematical interpretation rather than an empirically validated theory.

The same argument he uses to support his idea that our perceptions are not of reality also lead to reliable maths and logic. I guess science remains what it always was just with the proviso that we are aware we aren't modelling the world, only the interface.

At that point, I'm not sure the difference makes a difference. If, for example, we are only brains in a vat, the interface is all there is for us. So we *define* reality by what happens in that interface. And until that interface breaks down, that definition is a perfectly usable one. Once we get to use science to form hypotheses and test those hypotheses, the interface becomes the reality.

 

[Yerda]

OK, I agree that our symbol, the word 'apple' gives us no information about the fruit. But the fact that we can and do have certain types of interaction with that apple *does* tell us something about it.

Even if there is no apple?

If we take Hoffmann's word then we don't interact with an apple when we see one. We don't interact with any physical object as there are no physical objects to interact with. That includes what we think of as "We".

So, even if it is false to think that a file represented by a rectangle on my screen is *actually* a rectangle, it is still true that there is *something* that holds the information for that file and that can interact with my GUI. Now, the rectangle is part of the GUI, but the information is part of the file.

I agree. But does the nature of the icon reveal anything about the nature of the file?

In the analogy the file is the world. None of the properties of the icons are present in the file just as he is arguing none of the properties of physical objects are present in the world. That implies every possible percept is "not real" or non-veridical as he describes it.

So, to continue with the analogy, when I see an apple, the image I have in my head is not the 'real' apple. But that real fruit has the ability to interact in a way that produces, say, the sensation of red. The problem is to determine how much and what aspects of my sensation are from the GUI, how much from the OS and how much from the apple itself.

This is how I generally think about this kind of thing which is why I found these ideas so startling. Maybe I'm just weird but I'd like others to be able to enjoy these ideas as much as I do...even if they're wrong...which I'm confident they must be. The most remarkable thing is that he doesn't appear to be a crank.

I see you mentioned in another thread that you teach maths. You're better positioned than I am to make sense of the models he is using. It's all Bayesian frameworks and Markovian Kernels which means little to me.

You could try these if you have time:

The Interface Theory of Perception

Probing the interface theory of perception: Reply to commentaries

http://cogsci.uci.edu/~ddhoff/PerceptualEvolution.pdf

http://cogsci.uci.edu/~ddhoff/interface.pdf

http://cogsci.uci.edu/~ddhoff/ConsciousRealism2.pdf

 

[Yerda]

Transcendental surrealism?

Conscious realism, apparently.

 

[gnomon]

The gist is this: Evolutionary game theory can be used to argue that our perceptions are selected to hide reality from us. Or reality is what we see, and it aint the world.

Hoffmann and his PhD students have been running simulations of competition between organisms where one group is geared to percieve reality as is and the other is only able to percieve objects based on evoltionary fitness. The result is that fitness drives reality to extinction every time.

Among the implications is the possibility that what we think is reality is infact something like a species specific interface. Going deeper he is proposing a mathematically rigorous scientific model that "predicts" that no physical system has causal power, that physical objects are akin to icons on your interface desktop, that space and time are not fundamental constituents of the world but part of how we model it, and ultimately all of the world can be modelled as conscious observers.

This probably sounds like grade A fruitbattery but I'm pretty sure it isn't. Anyway, if you're interested have a look and leave a comment.

Further info:

Aye, he has a Ted talk like everyone else:

Here he is in chat with renowned philosophers Dan Dennett and David Chalmers:

His website at UCI with links to his papers (click on the link title Vita for loads more): Donald D. Hoffman | University of California, Irvine

He's also a bloody good photographer, hae a look at his flickr account: Donald Hoffman

Personally I didn't care about his inaccurate statements about the history of science. They were absolutely useless as well as wrong.

The idea of a flat earth.......which is a recent phenomenon and it was Eratosthenes who laid out a nearly accurate circumference of the Earth......

Never mind. He is obviously inadequate in understanding the history of science. That first video was utterly boring and quite frankly offensive to anyone who took a high school science class.

Philosophically speaking.......whatever rocks your boat. There was no hard evidence of anything and on the metaphysical level he is far preceded by other cultures.

It was boring.