7 ms·

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 tog

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.

oarabbus_ 5y ago
>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.
dave_sullivan 5y ago
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.
- scyzoryk_xyz 5y ago
  This. If you’re focusing on the visuals you are missing out on the most important therapeutic value of the experience.
- vmception 5y ago
  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.
bloopernova 5y ago
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.
- vmception 5y ago
  I'm curious to know more about what neuron networks are communicating that usually aren't
andresgemilsson 5y ago
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. Here it is: https://youtu.be/45tG1oVigVo https://youtu.be/45tG1oVigVo And video description: 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". ~Qualia of the Day: Indra's Net~

oarabbus_ 5y ago

>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.

dave_sullivan 5y ago

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.

scyzoryk_xyz 5y ago
This. If you’re focusing on the visuals you are missing out on the most important therapeutic value of the experience.
vmception 5y ago
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.

bloopernova 5y ago

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.

vmception 5y ago
I'm curious to know more about what neuron networks are communicating that usually aren't

andresgemilsson 5y ago

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. Here it is: https://youtu.be/45tG1oVigVo https://youtu.be/45tG1oVigVo And video description: 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". ~Qualia of the Day: Indra's Net~