The Case Against Reality | Prof. Donald Hoffman on Conscious Agent Theory

#1

We have no clue how consciousness emerges from 3 pounds of wet goo. Cognitive scientist Don Hoffman takes us deep into his research suggesting we're attacking the problem backwards. The implications may challenge all you know about our place in reality. Take the Red Pill and see how deep the rabbit hole goes...

In this extended interview we dive into the Interface Theory of Perception (how evolution hides the truth about reality in favor of a dumbed down user interface that only shows us "fitness" payoffs that help us survive), the "Hard Problem of Consciousness" and how we simply have no idea how consciousness could emerge from physical matter, the theory of Conscious Realism (how reality is really a social network of conscious "agents" and our perceptions are simply the interface by which we exchange experience with other agents), the math behind conscious agent theory, implications for artificial intelligence (AI), psychedelics, spirituality, and much much more.
Don't be put off by mention of survival of the fittest or mathematical modeling. This is a fascinating interview, this is the meeting of science and spirituality.

Enjoy.
 
#2
Absolutely great interview - actually more of a dialogue.
Sometimes I had to pinch myself to prove I am really listening to a top scientist discussing consciousness as a fundamental long before humanity appeared.
NDEs were mentioned and Hoffman dared to say 'God' and equate it with scientific theorems and he says his theories fit survival beyond death
A few quotes....
Question So you're saying the moon doesn't exist when I don't look.

Hoffman That's exactly right. Space and time themselves do not exist, independent of us.

Space time is something that you create in this moment. So we are the authors of space and time, Space and time are constructs of our interface.

Space time has come to the end of its usefulness.

I was so impressed by this video that I bought the book and transcribed the whole interview.
Thanks so much for posting LoneShaman
 
#3
Thanks, LS - I don't know if you realised, but Alex has just finished an interview with this guy. There was a pre-discussion, and I have suggested that the discussion continues right here.

David
 
Last edited:
#7
I am not sure if he will be able to use mathematics on the non-material realm, but it is amazing he has come as far as he has.

I want to link this with Behe's "Darwin evolved" because he comes to a fascinating conclusion by applying random mutation plus natural selection (RM+NS).

http://www.skeptiko-forum.com/threads/behes-argument-in-darwin-devolved.4317/

Basically he argues against RM+NS by pointing out that a random mutation can very rarely be helpful, and when it is, it almost invariably because it damages or breaks something.

This leads to the idea that a species can only evolve so much before it destroys too much of its heritage of genetic machinery. This is a startling, but seemingly inevitable conclusion!

I feel the two ideas are related, but it is getting late here - I'll be back.

David
 
#8
I am not sure if he will be able to use mathematics on the non-material realm, but it is amazing he has come as far as he has.
I have no idea how, although mathematics itself is abstraction and non material. During a DMT experience when breaking through as they say, the world as it perceived (by user interface) shatters and dissolves into geometric patterns (math). The realm you find yourself in is populated with seemingly conscious agents, many extremely exotic or alien. A enormous variety of them. It makes me wonder if the user interface is actually removed or lessened as to allow one to see beyond the veil into true reality. He did mention how true reality is far too much for us to deal with in accord to our function of day to day life. That seems to be accurate in the case of the DMT realm.

I want to link this with Behe's "Darwin evolved" because he comes to a fascinating conclusion by applying random mutation plus natural selection (RM+NS).

http://www.skeptiko-forum.com/threads/behes-argument-in-darwin-devolved.4317/

Basically he argues against RM+NS by pointing out that a random mutation can very rarely be helpful, and when it is, it almost invariably because it damages or breaks something.

This leads to the idea that a species can only evolve so much before it destroys too much of its heritage of genetic machinery. This is a startling, but seemingly inevitable conclusion!
It stands to say quite a bit about evolution and the origin of life I would think. As I have mentioned ad nauseam, codes and in particular the representations of symbols can only be connected by minds, or in fact conscious agents. And is a single cell a conscious agent ? Or perhaps even a smaller scale, Is a molecular machine a conscious agent?
 
#9
It stands to say quite a bit about evolution and the origin of life I would think. As I have mentioned ad nauseam, codes and in particular the representations of symbols can only be connected by minds, or in fact conscious agents. And is a single cell a conscious agent ? Or perhaps even a smaller scale, Is a molecular machine a conscious agent?
Yes, but science manages to dodge most arguments about this by invoking vast numbers of organisms over vast time spans. However Behe's argument seems very finite. Every evolutionary change (almost every) is accompanied by a loss of useful genetic information - so an evolving species gets stuck rather quickly. It is a rather neat argument that I describe in the Behe thread.

In a way I suspect Hoffman's evolutionary argument could be extended in various ways too. For example, are we as 'fit' as a bacterium, which can reproduce so fast and with so little excess baggage? Our brain power is a luxury, rather like seeing reality roughly as it is might be a luxury, and his argument would imply that neither luxury is possible.

I guess you have done DMT! Curiously I have a chemist friend who, many years ago, synthesised DMT without realising its reputation. He still didn't know about that property of the molecule until he described this work to me, and I asked him if he realised what DMT was.

He did say he felt rather peculiar after finishing that reaction!!!

David
 
#10
Yes, but science manages to dodge most arguments about this by invoking vast numbers of organisms over vast time spans. However Behe's argument seems very finite. Every evolutionary change (almost every) is accompanied by a loss of useful genetic information - so an evolving species gets stuck rather quickly. It is a rather neat argument that I describe in the Behe thread.
I do remember an article he wrote on this before he wrote the book. Sometimes the degenerative mutation can be positive, usually by breaking a gene to allow for a shorter pathway for a particular function. It's the easy fix but still a loss of information. I haven't read the book so you'd now all about that I expect. Still evolution has such a broad meaning. At times it refers to different things. What I mean is can adaption be really classified as evolution? Well yes and now depending on the context. I assume this theory deals only with RM & NS? That would make sense. I mean you can't actually create anything that way, only adapt what has already been created. These days we have a bunch of other tools in the kit that Darwinism has ignored. such as symbiogenesis, horizontal DNA transfer, action of mobile DNA and epigenetic modifications. Yet still no answer to biological novelty. Inherent creativity?

In a way I suspect Hoffman's evolutionary argument could be extended in various ways too. For example, are we as 'fit' as a bacterium, which can reproduce so fast and with so little excess baggage? Our brain power is a luxury, rather like seeing reality roughly as it is might be a luxury, and his argument would imply that neither luxury is possible.
Well if you have a network of conscious agents it's an entirely new playing field. I expect there would be memory which then things may start to resemble Sheldrakes morphogenic fields in some aspects. Mutation, similar to the immune system could be directed with memory and direction Protein sequence and folding could be directed or orchestrated via agency. Who knows, the thing is you would have the extra layer of intelligence beyond physical law in play and beyond the biological information. You'd just have to be careful not to use it where ever an immediate solution was not apparent. As Hoffman said we don't see consciousness only it's icon, so I'm not sure how these things could be recognized as agency.
 
#11
I assume this theory deals only with RM & NS? That would make sense. I mean you can't actually create anything that way, only adapt what has already been created. These days we have a bunch of other tools in the kit that Darwinism has ignored. such as symbiogenesis, horizontal DNA transfer, action of mobile DNA and epigenetic modifications. Yet still no answer to biological novelty. Inherent creativity?
Yes, he confines himself to RM+NS. However I would conjecture that evolution by X+NS, where X is something that does not involve intelligence, would hit similar constraints, but they might be messier to derive - not least because all the other mechanisms seem to be rather up in the air right now. Epigenetics is maybe the most interesting one, but am I right that nobody has found a mechanism to translate those tags on DNA/histone proteins back into DNA - so epigenetic changes last a few generations and then fade?

Well if you have a network of conscious agents it's an entirely new playing field. I expect there would be memory which then things may start to resemble Sheldrakes morphogenic fields in some aspects.
Mutation, similar to the immune system could be directed with memory and direction Protein sequence and folding could be directed or orchestrated via agency. Who knows, the thing is you would have the extra layer of intelligence beyond physical law in play and beyond the biological information. You'd just have to be careful not to use it where ever an immediate solution was not apparent. As Hoffman said we don't see consciousness only it's icon, so I'm not sure how these things could be recognized as agency.
We probably need to distinguish 'real intelligence' from 'intelligence which is isomorphic with a computer program.

I mean the problem as I see it, is not in anything that a cell does - however intricate - it is how can such a construction evolve. You seem to need 'real intelligence' of a very high order to know how to make changes in such a mechanism.

I think Sheldrake's morphic fields are supposed to contain 'real intelligence'.

More generally, it is fascinating that Hofmann has started to really stick his neck out and discuss the idea that reality might be made of consciousness!

David
 
#12
Yes, he confines himself to RM+NS. However I would conjecture that evolution by X+NS, where X is something that does not involve intelligence, would hit similar constraints, but they might be messier to derive - not least because all the other mechanisms seem to be rather up in the air right now. Epigenetics is maybe the most interesting one, but am I right that nobody has found a mechanism to translate those tags on DNA/histone proteins back into DNA - so epigenetic changes last a few generations and then fade?
I agree, none of those other mechanism account for biological novelty. We have things that can shuffle what already exist in order to adapt but it doesn't say anything about evolution in the sense of macro evolution. This was sort of what I was getting at with the use of the word evolution. They would be two entirely different things. You certainly can't continue to develop something new if you just keep breaking things and degrading information

We probably need to distinguish 'real intelligence' from 'intelligence which is isomorphic with a computer program.

I mean the problem as I see it, is not in anything that a cell does - however intricate - it is how can such a construction evolve. You seem to need 'real intelligence' of a very high order to know how to make changes in such a mechanism.
Yes, I get the sense that perhaps both types of intelligence may be involved. Biology is clearly teleological, implying that there is a predefined goal projected by some active agency. I think biology is definitely one place more than any other we can get sense of this network of conscious agency. Of course I'm not talking about brains and nervous systems in relation to the consciousness of the organism, rather the biological cellular functions that our outside the scope of individual consciousness. Although depending on how you see the structure of consciousness it may well be a gradient that may even extend upward to a unifying whole of cosmic consciousness.
 
#13
I am reading Hoffman's book - which is fairly dense - and I'm starting to waver a bit. Unfortunately he makes a lot of use of the Holographic Principle, which is an idea from string theory. However Peter Woit (Columbia) repeatedly points out on his blog that after about 3 decades of study there is no evidence that string theory is correct!

http://www.math.columbia.edu/~woit/wordpress/

I think this mathematical proof that RM+NS doesn't result in veridical perception, should be thought of as yet another proof that Neo Darwinism is wrong! Our eyes, ears, touch sensors, and olfactory sensors were designed by something!

His theory more or less tears up the whole of science, just to leave Neo Darwinism intact - that seems a bit daft to me!

David
 
#14
I am reading Hoffman's book - which is fairly dense - and I'm starting to waver a bit. Unfortunately he makes a lot of use of the Holographic Principle, which is an idea from string theory. However Peter Woit (Columbia) repeatedly points out on his blog that after about 3 decades of study there is no evidence that string theory is correct!
Actually most serious thinkers think String Theory is all knotty and completely unravelled.
 
#15
His theory more or less tears up the whole of science, just to leave Neo Darwinism intact - that seems a bit daft to me!
I think, suspect that what he is doing is playing both sides of philosophical approach in order to stay within particular academic circles. He talks a lot about how evolution does not care about truth only survival and propagation of the species. But clearly in the case of human beings there is much about us that is not about survival. For example art, music, philosophy etc.. and more importantly big questions that has shaped science itself, like quantum physics and the nature of reality. His own work has nothing to do with propagating the species but is reflective of the philosophy of mind and reality that the species seems to have always pondered.

Another point is that what Darwinism uses in its theory are by his classifications not elements of true reality but the icons of perception and not the actual truth.

The more I have pondered what he has to say, the more I see that it is also reflective of other lines of thought, even the very roots of all religions such as animism. What does resonate with me is this network of conscious agents. Like I mentioned in my last post concerning cellular function at one level of consciousness to consciousness at the level of the organism, and extending upward to a unifying cosmic consciousness. I see this network of agency as fractal.

You had asked me not too long ago about my acceptance of animism, I explained how it was about a connection I found with my own experience and that of the serpent mythology in many ancient traditions. In my experiences, all of them in fact I perceive these incandescent tendrils, the permeate me and everything, it is a network of living conscious energy. When my consciousness is shifted I can then see it with eyes open in 3d space. I think in his terms that my interface has loosened somewhat and I am seeing a part of reality that is hidden from the normal interface. This is the same as what Bernardo Kastrup would say when it comes to mind altering events. I just thought I would mention this because it does fit very well with my own experiences. I think what he is saying is not much different, it is just dressed in different clothes.
 
#16
Another point is that what Darwinism uses in its theory are by his classifications not elements of true reality but the icons of perception and not the actual truth.
Right - and for a bit I toyed with the idea that he was imagining a true reality in which fitness space would be fairly smooth, so that NS really could search for an optimum - but then he started discussing the Holographic principle with great enthusiasm.

I wonder if he realises the incredible combinatorial problems when RM+NS is applied to DNA and proteins - does he realise that there is a problem here for traditional neo Darwinism?
For example art, music, philosophy etc.. and more importantly big questions that has shaped science itself, like quantum physics and the nature of reality. His own work has nothing to do with propagating the species but is reflective of the philosophy of mind and reality that the species seems to have always pondered.
Exactly, and by almost ignoring this problem, he is following the majority of evolutionists. At one point in his book he does discuss whether the FBT theorem applies to other aspects of the intellect, but then dismisses this idea almost immediately.

The book gets tedious, because it never even considers alternative possibilities except ones that he can dismiss out of hand.

I think mathematics can achieve wonders, but it can also drag intelligent people down into a deep, deep rabbit hole! String theory and the FBT may be perfect examples of this phenomenon!

David
 
#17
After listening to the video LS posted, I have a question, one I hope Alex will ask Donald Hoffman: what, to him, is the exact meaning of a "conscious agent"? Is it an agent that is conscious? Or something whose agency somehow (and possibly quite accidentally) generates consciousness?

I get the impression that DH may be a supporter of panpsychism, an ontological principle that incurs a second hard problem of consciousness: the combination problem. How can consciousness, if it is a property of all matter, even at the subatomic level, combine to produce higher degrees of consciousness/intelligence. And, more intriguingly, why?

I suppose it has something vaguely going for it, and that is the fact that what we think of as matter can and does complexify, and as it does, ends up having ever more complex properties. NH3 (ammonia) has somewhat more interesting properties than Nitrogen or Hydrogen, its constituents, alone do. And that goes in spades for Carbon, hydrogen, Nitrogen and Oxygen, which together can produce (God alone knows why) almost single-handedly* all the varied proteins we find in animate beings (*but there may be, for example, some contribution from other elements such as Sulphur, Magnesium and Iron).

Is an amino acid somehow more conscious than any of its constituent parts? Why does every amino acid have a -COOH group at one end and an -NH2 group at the other, and how come the -COOH group of one amino acid can interact with the -NH2 of another to produce a peptide bond, accompanied by the production of water, and thereby another peptide in the string that will eventually produce a protein?

We don't really answer any "why" questions merely by being able to describe chemical reactions or coming up with explanatory models of how energy exchanges allow/enable them to occur. Regardless, the panpsychist seems somehow to be associating such reactions with a noticeable (if quite small) increase in consciousness. I'd ask myself whether such a putative increase is an accidental result of complexification of genuinely existent material objects, or whether pre-existing consciousness is incidentally exhibiting some of the processes occurring in its mind, which processes are perceived by us as if they were concrete objects?

DH is sort of in the same ballpark as Bernardo Kastrup, but I can't help but wonder if he isn't to some extent still within in a no-man's land where one can have it both ways. If everything we perceive is simply an icon of reality, not reality itself, that is fairly consistent with Idealism. Both he and Bernardo think there are things-in-themselves, or noumena which the icons (phenomena) are representing. They'd probably both be able to describe themselves as objective idealists in that they both believe only noumena are real whilst phenomena are merely representational.

That said, DH does seem to rely heavily on mathematics, according it a possibly decisive role. I wonder if he fully realises that mathematics is itself phenomenal rather than noumenal? That any answers he gets that rely on maths are perforce going to be mere reflections of the actual truth rather than finally dispositive? Yes, he's gone a lot further, and is much more open, than the general run of scientists, but in the end he may be trying to prove his thesis based on representational, phenomenal principles, and his potential degree of success may be limited by this.

Maths is the least contaminated methodology for examining reality we've so far come up with, and possesses a certain cachet because so few of us are very capable at it. But in the end, it's based on accepting certain assumptions and elaborating the rules for manipulating its icons. Maths becomes the nearest we have to truth, but not actual truth itself; it remains phenomenal rather than noumenal. It's held in such incredible awe, as if it were much more than a bunch of internally consistent, manufactured icons. Because of this, I wonder whether DH will ever get at the truth rather than merely a better approximation of it.
 
#18
Just a few more thoughts on what both David and Michael have brought up.

Donald states many times that our interface (desktop) our perceptions is what evolution has built into us. However based solely on his theory shouldn't it be more accurate to say that it is what conscious agency has built into us? After all evolution, time, space and what we call physicality is "booted" up from conscious agency. In other words evolution is what is happening on the desktop it is happening in space and time. I think this is a definite contradiction.

That being said I don't think it is a problem that undermines the general line of reasoning. We just have to give different qualities to evolution, that being is that it too is derivative of conscious agency. Which according to his theory it must be?

I would love to hear what he would say about this.

What is a conscious agent? This is what I have gathered from his own words.

A conscious agent is something that has experiences, based on experience it can make decisions and can take actions, actions affect reality and in turn reality projects perceptions. Conscious agents are non physical existing outside space and time.

So from these interactions the formats can be extrapolated "booted up". Formats such as space, time and matter. This is a distinguishing feature from that of panpsychism. It would in my opinion be more in the philosophical realm of idealism.
 
Last edited:
#19
A conscious agent is something that has experiences, based on experience it can make decisions and can take actions, actions affect reality and in turn reality projects perceptions. Conscious agents are non physical existing outside space and time.
I could also put this into the same context of my own extrapolation of Pierces semiotics of information transfer.


The object is reality, not physical but conceptual. It is the representamen that is what we would perceive as physicality (icons). We are the interpretant of the symbols or icons.

This is also the heart of the irreducible mind, it is how the mind works. It is how biology works. I am also saying it is how reality works.

This is the most fundamental basics of information transfer . The only way to transfer information through space and time. Even in science these days, there is a growing acknowledgment that information is more fundamental than space and time. However information by itself is nothing without the interpretant. The information (concepts) itself flows from conscious agency outside of time and space.

This is consistent with some of the interpretations of quantum physics. The concept of local realism having being refuted experimentally. Demonstrating that perception cannot be removed from reality.
 
Last edited:
#20
A conscious agent is something that has experiences, based on experience it can make decisions and can take actions, actions affect reality and in turn reality projects perceptions. Conscious agents are non physical existing outside space and time.
I love all this stuff.
I used to think that rocks and plastic and such were the unconscious accoutrements that God made available on the stage that conscious objects play their part on. However, I like the idea that the Sun and Earth themselves are conscious, as do many others. So is a rock the equivalent of a chunk of dead wood? That once upon a time was conscious when existing as lava many years ago.
I think it’s important (for me) to think of everything, whether rocks or plants or human beings, as divine creation.
 
Last edited:
Top