I find Deep Mind AI output to be pretty damn close.
Although I do find synesthesia hard to portray digitally.
Its not just the 5-ish senses that get mixed together. There are things like non-visual ideas that can get encapsulated into balls and other geometric shapes and infinitely replicated. Although perceivable in the mind at the time, this is not the same medium.
People talk about the visuals a lot because that's what you can see, but the visuals are actually one of the less interesting things about the experience.
For instance, why do I meet entities (nominally my self conscience) that appear smarter and wiser than myself? And why do they explain systems that then match up with other people's experiences? Why do different people's subconsciences come across such similar high level concepts during the experience?
If you know you know and if you don't you don't. It's an experience few will ever have and most people are not ready for. So they seek to reduce it to something pedestrian.
One thing I like about legalization pushes is that the weird gatekeeping and sometimes invalidation from the psychedelic community will get diluted in favor of a broader range of information, a broader range of possible experiences, what inputs lead to what outputs, etc.
>I find Deep Mind AI output to be pretty damn close.
Hmm, I don't know that I'd agree. I think a deepdream AI output approximates a breakthrough DMT experience probably less than the prisoners of the Allegory of the Cave watching shadows dance on the wall experienced the true outside world.
The combination of 5 senses and non-visual ideas into visible objects makes me wonder: If/When a direct-brain-input interface is created, will what the person perceives be like a DMT trip?
This is why I enjoy listening to music and watching a visualization at the same time. Marrying senses together but in an abstract way is strangely appealing.
Indeed! (author of the Hyperbolic Geometry of DMT Experiences article here). I think you might enjoy my video on the Free Energy Principle and Psychedelics. It IMO explains in a novel, meaningful, and non-trivial way the effect you are describing.
Friston's Free Energy Principle (FEP) is one of those ideas that seem to offer new perspectives on almost anything you point it at.
It seems to synthesize already very high-level ideas into an incredibly general and flexible conceptual framework. It brings together thermodynamics, probabilistic graphical models, information theory, evolution, and psychology. We could say that trying to apply the FEP to literally everything is not a bad idea: it may not explain it all, but we are bound to learn a lot from seeing when it fails.
So what is the FEP? In the words of Friston: "In short, the long-term (distal) imperative — of maintaining states within physiological bounds — translates into a short-term (proximal) avoidance of surprise. Surprise here relates not just to the current state, which cannot be changed, but also to movement from one state to another, which can change. This motion can be complicated and itinerant (wandering) provided that it revisits a small set of states, called a global random attractor, that are compatible with survival (for example, driving a car within a small margin of error). It is this motion that the free-energy principle optimizes."
Organisms that survive over time must minimize entropy injections from their environment, which means they need to minimize surprise, which unfortunately is computationally intractable, but the information theoretic construct of variational free-energy provides an upper bound on this ground truth surprise, meaning that minimizing it will indirectly minimize surprise. This cashes out in the need to maximize "accuracy - complexity" which prevents both overfitting and underfitting. In the video we go over some of the classical ideas surrounding the FEP: the dark room, active inference, explicit vs. implicit representations, and whether real dynamic systems can be decomposed into Markov blankets. Finally, we cover how the FEP naturally gives rise to predictive coding via hierarchical Bayesian models.
We then talk about Reduced BEliefs Under pSychedelics (REBUS) and explain how Carhart-Harris and Friston interpret psychedelics and the Anarchic Brain in light of the FEP. We then discuss Safron's countermodel of Strengthened BEliefs Under pSychedelics (SEBUS) and the work coming out of Seth's lab.
So, that's how the FEP shows up in the literature today. But what about explaining not only belief changes and perceptual effects, but perhaps also getting into the actual weeds of the ultra bizarre things that happen on psychedelics?
I provide three novel ideas for how the FEP can explain features of exotic experiences:
(1) Dissonance-minimizing resonance networks would naturally balance model complexity due to an inherent "complexity cost" that shows up as dissonance and prediction error minimization when prediction errors give rise to out-of-phase interactions between the layers.
(2) Bayesian Energy Sinks: What you can recognize lowers the (physical) energy of one's world-sheet. I then blend this with an analysis of symmetrical psychedelic thought-forms as energy-minimizing configurations. On net, we thus experience hybrid "semantic + symmetric" hallucinations.
(3) Indra's Net: Each "competing cluster of coherence" needs to model its environment in order to synch up with it in a reinforcing way. This leads to attractor states where "everything reflects everything else".
Yet, some people try. There is "The Spirit Molecule" by Rick Strassmann, precisely about DMT, along with other books on "experimental spirituality".
“To most people who are even moderately experienced with entheogens, concepts such as awe, sacredness, eternity, grace, agape, transcendence, transfiguration, dark night of the soul, born-again, heaven and hell are more than theological ideas; they are experiences.” - Thomas Roberts
This phrase is quoted in "Sacred Knowledge: Psychedelics and Religious Experiences" by William A. Richards (https://www.goodreads.com/book/show/28121728-sacred-knowledg...), yet I find it the most suitable summary of this overview of scientific research on psychedelics and religion.
We hear about mystical visions from LSD ("acid"), psilocybin ("shrooms"), and DMT from many "spiritual but not religious" people and self-proclaimed shamans. But how does it relate to vision by ordinary people (ones who never tired, and wouldn't try if it weren't for legal, scientific research)? And how does it relate to prayer, mediation, and mystical visions by Christians, Jews, Buddhists, and Hinduists? How do monks and priests compare their psychedelic experience with their regular practice? Do they all turn to Zen Buddhism, or entrench in their religious background?
Regardless if you are deeply religious, or a non-spiritual atheist, I believe you will reconsider a few things after reading this book.
It's not a problem of the "amount" of analysis capabilities. The experience is orthogonal to rational. It's the same Chinese Room - no matter how many analytical brains you would throw on understanding what "red" qualia means, you won't be able to explain it and won't be able to experience it until you did.
We can analyze many “irrational” brain states from the outside. I’d suggest looking at modern fMRI results.
I’d also suggest forgetting about the Chinese Room. It might have been an interesting thing to discuss 40 years ago. It’s just sad to see people taking it seriously today.
Why not? If you get good at it, you can learn to quickly tell which wallpaper symmetry group a given tiling has. On most psychedelics the hallucinations are easily identifiable as belonging to one of the 17 wallpaper Euclidean symmetry groups. It turns out that on DMT you see symmetry groups that are none of those... and then you realize... wait, I'm seeing a flat surface tessellated by heptagons! This means... the space is hyperbolic! What's the epistemological issue here? :-)
I can't tell whether you're seriously asking, but just in case:
Geometry requires the ability to precisely measure angles and distances. Subjective hallucinatory experiences permit the witness to feel like they're seeing just about anything -- a sphere partitioned into five congruent squares, an ant that is an even number of ants, the square root of irony. But they don't admit any kind of measuring stick.
Although I guess if impossible-outside-of-hyperbolic-space tesselations were universal experiences of the users of DMT, that would do it. But I know a few, and they never mentioned repeating patterns. One talked about Gumby people a lot. Which, to be fair, probably involved some cool geometry, but I doubt it violated the parallel postulate.
One of the most fascinating things about DMT is that you get to experience the same level of visual resolution that you experience in your fovea but across your visual field. I'd say that for a trained observer/phenomenologist, we can know with the same level of trust that yes, they saw the *442 wallpaper symmetry group on DMT as we can trust that someone is seeing a square under normal conditions. And indeed, symmetries are already widely reported (see: https://psychonautwiki.org/wiki/Symmetrical_texture_repetiti...). It's a core effect. Next time you try DMT just pay attention... you'll see it all over the place.
Why are not more people reporting the hyperbolic features I've identified (e.g. hyperbolic folding of the worldsheet at the Magic Eye level, as described in the article)? And why instead do we mostly hear about reports of entities? The reason is simple: it's not what people are trying to bring back. We don't yet have a rational culture of inquiry for the structural analysis of psychedelic phenomenology.
As explained in the following article, currently people focus on the semantic content (the narrative) rather than the phenomenal character (the texture). But as we gather more rational, intelligent, and dedicated psychonauts, consistency of reports and consilience will increase: https://www.qualiaresearchinstitute.org/blog/rigorous-report...
If you’ve ever “broken through” you know it’s beyond anything an HN post could ever come close to.