Qualia Computing and The Hyperbolic Geometry of DMT Experiences

[remyman612]

Hello everyone,

I had a DMT experience a few days ago and was, once again, amazed by the strange geometric visuals I was experiencing. Out of a whim I searched "DMT geometry" and got to the following essay:

The Hyperbolic Geometry of DMT Experiences: Symmetries, Sheets, and Saddled Scenes

I wish I could specify what it's trying to say, but would fail to put it in proper terms (you'll see why). I'll give a quote instead:

This is an essay on the phenomenology of DMT. The analysis here presented predominantly uses algorithmic, geometric and information-theoretic frameworks, which distinguishes it from purely phenomenological, symbolic, neuroscientific or spiritual accounts. We do not claim to know what ultimately implements the effects here described (i.e. in light of the substrate problem of consciousness), but the analysis does not need to go there in order to have explanatory power. We posit that one can account for a wide array of (apparently diverse) phenomena present on DMT-induced states of consciousness by describing the overall changes in the geometry of one’s spatiotemporal representations (what we will call “world-sheets” i.e. 3D + time surfaces; 3D1T for short). The concrete hypothesis is that the network of subjective measurements of distances we experience on DMT (coming from the relationships between the phenomenal objects one experiences in that state) has an overall geometry that can accurately be described as hyperbolic (or hyperbolic-like). In other words, our inner 3D1T world grows larger than is possible to fit in an experiential field with 3D Euclidean phenomenal space (i.e. an experience of dimension R2.5 representing an R3 scene). This results in phenomenal spaces, surfaces, and objects acquiring a mean negative curvature. Of note is that even though DMT produces this effect in the most consistent and intense way, the effect is also present in states of consciousness induced by tryptamines and to a lesser extent in those induced by all other psychedelics.

Now, after reading this I was very much amused and confused. It sounded like a whole bunch of nonsense to me (admittely, I was under the waning effects of 400ugs of LSD at the time), noted it in the DMT nexus chat, and ignored it. After a few hours I came back to the article, read some more and fell into this 'Qualia Computing' rabbithole. These guys are... well, at least very ambitious and interesting. I'll quote again:

The Qualia Research Institute (QRI) is a non-profit based in the Bay Area close to San Francisco which seeks to discover the computational properties of experience. QRI has a "full-stack approach" to the science of consciousness which incorporates philosophy of mind, neuroscience, and neurotechnology. It's founder, Andrés Gómez Emilsson, studied Symbolic Systems at Stanford (and has a masters in Computational Psychology, also from Stanford). He has professional experience in data science engineering, machine learning, and affective science. His research at the Qualia Research Institute ranges from algorithm design, to psychedelic theory, to neurotechnology development, to mapping and studying the computational properties of consciousness. Andrés blogs at qualiacomputing.com.

I personally think that it's quite exciting to see a group of young, smart people approach the psychedelic experience and the exploration of consciousness in such a different way. Anyhow, on to the subject matter:

The article I quoted above, also has a video. It's a lecture on this framework of hyperbolic geometry given at the Harvard Science of Psychedelics Club (because apparently that's a thing):

[YOUTUBE]

He does a good point of explaining his framework even though the sound is quite low at times. I must admit I don't know anything about geometry so it's quite hard for me to touch on any specifics and there's a lot of I don't get (hopefully you people can explain it!).

The article and video basically deal with how the hyperbolic geometry (the article explains what that is) of DMT increases as the dosage increases and he describes by which geometric 'rules' this happens. It deals a lot with curves and triangles and all that good stuff (kinda makes me want to know more about geometry).
Another interesting claim he makes is that adding moar DMT is basically adding energy to your consciousness, and this increase in energy bring moar complexity. If you take too high a dose, the geometry becomes so complex you basically overload and forget what happened (amnesia). Now I've never really heard about people getting blackouts on nn,DMT but I suppose it sounds sensible. He also notes that the trajectory one takes through the DMT landscape effects how much you bring back from the experience (I suppose we already knew this due to changa).

The end of the article also deals with entities on the various levels and a lot more. It would be foolish of me to talk about 'Bayesian Energy Sinks', 'The Hamiltonian of a World-sheet' and 'Energy Invariants', since again, I know too little of these things.

The blog has a lot of various interesting other essays, one I particularly liked was this one:


5-MeO-DMT vs. N,N-DMT: The 9 Lenses


It's not up to me to say if he's right or wrong, but I thought it was very interesting nonetheless. Admittedly, I do have a bit of a 'why are the deconstructionists ruining the magic of DMT' mixed feelings, but in the end more research (initiatives) are a good thing imho.

 

[Hank Scorpio]

Wow.

I had watched that lecture before, and found it really interesting. There is no doubt that some peoples visions on DMT show some interesting geometries/symmetries which we don't encounter in the normal world.

But that article they wrote is .... verifiably bonkers. (Source: I studied theoretical physics).

The talk appears to be a lot more restrained than the article. It's one thing to say that peoples hallucinations appear hyperbolic, which they do in the talk, and another thing entirely to suggest that a DMT trip somehow alters your relationship with spacetime itself, as they do in the article.

Sorry if my comment is a bit negative. I would definitely agree that more research is good. As far as I know, nobody exactly knows the mechanism by which DMT visuals get their distinctive geometry and I (reductionist ) would love to find out.

 

[remyman612]

It's one thing to say that peoples hallucinations appear hyperbolic, which they do in the talk, and another thing entirely to suggest that a DMT trip somehow alters your relationship with spacetime itself, as they do in the article.

Interesting! Isn't that claim sort of right though? The effects of DMT obviously effects how one experiences (phenomenological) space and time. It obviously doesn't change the world 'as it is' (not trying to get into deep philosophical waters; I'll drown). The article is still full of mumbo-jumbo to my eyes, so I could totally be wrong; perhaps he does claim that somehow. In the essay Algorithmic Reduction of Psychedelic States he goes more in-depth about that subject. I'll quote:

For the purpose of this article I will assume that direct realism, in all of its guises, is wrong. That is, I will assume that any mind-independent object can only be experienced indirectly. What we experience is not the object (or beings) themselves, but a qualia-furnished representation entirely contained within one’s mind (this is often called the simulationist account of perception). Furthermore, I will also assume that the behavior of the universe can be fully described with the Standard Model of physics (or a future version of it).

What are your opinions on the second part of the geometric essay though? ('2) Dynamic System Account: Energy Sources, Sinks and Invariants')

He asks the reader to 'assume' that visuals (not just on DMT) require a certain amount of energy:

For instance, the brightness of a point of colored light in one’s visual field is energy-dependent. Likewise, the information content in a texture, the number of represented symmetrical relationships, the speed by which an object moves (plus its acceleration), and even the curvature of one’s geometry.

How bold of a claim is that? It sounds quite reasonable but I'm surprised one would even have to make an assumption about a rather dull statement. I guess might just not really know enough of the brain to conclude this using hard science.

Then he claims that DMT is, other than 'regular' states which modulate mental energy all the time, is a constant high-energy state. DMT also impairs those 'mental energy sinks' while increasing the throughput of its energy sources. That is, he says, what eventually accounts for the mental/visual effects of DMT. Which sounds interesting, but doesn't really seem based on anything.

Energy sinks are still present and they struggle to capture as much of the energy as possible. In particular, one energy sink is “recognition” of objects on the world-sheet.

This model postulates that attention functions as an energy source, whereas pattern recognition functions as an energy sink.

Now this gets more interesting because there certainly is something going on with 'focusing' on the certain detail while in DMT space. He gets into an example of how the curvature of the 'world sheet' we experience in everyday life is 'arousal-dependent'. So de-focusing your vision for instance diminishes the energy it takes to sustain a curved world-sheet with a lot of information. Which is important for the next part, which gets good:

One essential property of our minds is that our level of mental arousal decreases when we interpret our experience as “expected”. People who can enjoy their own minds do so, in part, by finding unexpected ways of understanding expected things. In the presence of new information that one cannot easily integrate, however, one’s level of energy is adjusted upwards so that we try out a variety of different models quickly and try to sort out a model that does make the new information expected (though perhaps integrating new assumptions or adding content in other ways). When we cannot manage to generate a mental model that works out a likely model of what we are experiencing we tend to remain in an over-active state.

This general principle applies to the world-sheet. One of the predominant ways in which a world-sheet reduces its energy (locally) is by morphing into something you can recognize or interpret. Thus the world-sheet in some way keeps on producing objects, at first familiar, but in higher energies the whole process can seem desperate or hopeless: one can only recognize things with a stretch of the imagination. Since humans in general lack much experience with hyperbolic geometry, we usually don’t manage to imagine objects that are symmetric on their own native geometry. But when we do, and we fill them up with resonant light-mind-energy, then BAM! New harmonics of consciousness! New varieties of bliss! Music of the angels! OMG! Laughter till infinity and more- shared across the galaxy- in a hyperbolic transpersonal delight! It’s like LSD and N2O! Wow!

As one does not know any object that the world-sheet can reasonably be able to generate in high doses, and the world-sheet has so much energy on its own, energy can seem to spiral out of control. This explains in part the non-linear relationship between experienced intensity and DMT dose.

That would account for the extremely large range of experiences one can have on DMT and also the 'speed' at which one experiences the visuals. I've been wondering why sometimes my visuals are almost 'photo realistic' (or at least the memory of em do) while often it's very obviously a buncha self-rotating, colorful geometry.

On DMT, anything that attention focuses on will begin branching, copying itself and multiplying, a process that quickly saturates the scene to the point of filling more spatial relationships than would fit in Euclidean 3D. The rate at which this happens is dose-dependent. The higher the dose, the less inhibiting control there is and the more intense the “folding” property of attention will be.

Again; I realize this is not a formal method or the end-of-all framework to describe these things. But its certainly food for (my) thoughts. I wouldn't mind living in a world where "the value of hyperbolic phenomenal spaces will be proportional to the level of wellbeing and bliss that can be felt in them."

Sorry if my comment is a bit negative. I would definitely agree that more research is good. As far as I know, nobody exactly knows the mechanism by which DMT visuals get their distinctive geometry and I (reductionist ) would love to find out.

Don't be sorry! I applaud any and all civil discourse, this is a place of learning after all. Feel free to point out some more obvious flaws in this framework/theory. The people at the Qualia Institute are, as far as I've gathered from their blog, very interested in any (constructive) criticism aimed their way.

 

[bismillah]

Hyperbolic space... that's such a perfect description! Brings back my visions of rooms inside rooms, hallways which turn at a right angle more than 4 times without looping, and other such strange spaces. The rest of it all seems very speculative, though... they claim to be approaching it scientifically rather than "spiritually", but they don't really provide the hard analysis to back that up.

Still interesting.

 

[Hank Scorpio]

Interesting! Isn't that claim sort of right though?

Hmmm. I actually read the article a bit too quickly. I missed the bit at the start where they say that they call a persons spatiotemporal representation a "world sheet". That's a term borrowed from string theory, and I thought they were literally talking about world sheets . Even then, it still doesn't make a lot of sense. They say things like:

Even when one manages to push most of the curvature out of a given modality (e.g. vision) it is likely to quickly return in another modality (e.g. kinesthetic or auditory landscape)

It is sensible to say that a particular pattern looks like a representation of some hyperbolic geometry, but how the hell can an auditory landscape be curved in any well/defined sense? It's sort of nonsense by definition. They say stuff like this a lot in the article.

They try and talk about the energy and acceleration of the world-sheet using language that sounds like physics, but they're trying to apply physics to something they defined as being not physically real; the world sheet is just one particular human's subjective perception of reality. As bismallah said, they need some hard analysis there to back it up, because as it stands, there's just too many completely unfounded assertions.

My opinion of Dynamic System Account: Energy Sources, Sinks and Invariants is that there is some grains of truth mixed in with a lot of mumbo jumbo. For example, this part:

One essential property of our minds is that our level of mental arousal decreases when we interpret our experience as “expected”.

Is, as far as I know, correct. I'm not a neuroscientist, but I've read some articles that say that the human brain is constantly trying to minimise the amount of suprise it experiences by constantly trying to predict what comes next. Being suprised uses more energy because the brain needs to work harder to figure out what's going on.

It also probably makes sense to say that an excited/suprised/aroused brain is more "energetic", because there is more neurons firing. So, it wouldn't be a stretch to say that a brain on DMT is very suprised and so probably has a lot more neurons firing than normal, and so is more energetic, but I don't think we can say more than that.

It's getting late, tomorrow I might try and dig up some of those articles I mentioned, they were really interesting!