QRI’s FAQ

These are the answers to the most Frequently Asked Questions about the Qualia Research Institute. (See also: the glossary).


(Organizational) Questions About the Qualia Research Institute

  • What type of organization is QRI?

    • QRI is a nonprofit research group studying consciousness based in San Francisco, California. We are a registered 501(c)(3) organization.

  • What is the relationship between QRI, Qualia Computing, and Opentheory?

    • Qualia Computing and Opentheory are the personal blogs of QRI co-founders Andrés Gómez Emilsson and Michael Johnson, respectively. While QRI was in its early stages, all original QRI research was initially published on these two platforms. However, from August 2020 onward, this is shifting to a unified pipeline centered on QRI’s website.

  • Is QRI affiliated with an academic institution or university?

    • Although QRI does collaborate regularly with university researchers and laboratories, we are an independent research organization. Put simply, QRI is independent because we didn’t believe we could build the organization we wanted and needed to build within the very real constraints of academia. These constraints include institutional pressure to work on conventional projects, to optimize for publication metrics, and to clear various byzantine bureaucratic hurdles. It also includes professional and social pressure to maintain continuity with old research paradigms, to do research within an academic silo, and to pretend to be personally ignorant of altered states of consciousness. It’s not that good research cannot happen under these conditions, but we believe good consciousness research happens despite the conditions in academia, not because of them, and the best use of resources is to build something better outside of them.

  • How does QRI align with the values of EA?

    • Effective Altruism (EA) is a movement that uses evidence and reason to figure out how to do the most good. QRI believes this aesthetic is necessary and important for creating a good future. We also believe that if we want to do the most good, foundational research on the nature of the good is of critical importance. Two frames we offer are Qualia Formalism and Sentientism. Qualia Formalism is the claim that experience has a precise mathematical description, that a formal account of experience should be the goal of consciousness research. Sentientism is the claim that value and disvalue are entirely expressed in the nature and quality of conscious experiences. We believe EA is enriched by both Qualia Formalism and Sentientism.

  • What would QRI do with $10 billion?

    • Currently, QRI is a geographically distributed organization with access to commercial-grade neuroimaging equipment. The first thing we’d do with $10 billion is set up a physical headquarters for QRI and buy professional-grade neuroimaging devices (fMRI, MEG, PET, etc.) and neurostimulation equipment. We’d also hire teams of full-time physicists, mathematicians, electrical engineers, computer scientists, neuroscientists, chemists, philosophers, and artists. We’ve accomplished a great deal on a shoestring budget, but it would be hard to overestimate how significant being able to build deep technical teams and related infrastructure around core research threads would be for us (and, we believe, for the growing field of consciousness research). Scaling is always a process and we estimate our ‘room for funding’ over the next year is roughly ~$10 million. However, if we had sufficiently deep long-term commitments, we believe we could successfully scale both our organization and research paradigm into a first-principles approach for decisively diagnosing and curing most forms of mental illness. We would continue to run studies and experiments, collect interesting data about exotic and altered states of consciousness, pioneer new technologies that help eliminate involuntary suffering, and develop novel ways to enable conscious beings to safely explore the state-space of consciousness.

Questions About Our Research Approach

  • What differentiates QRI from other research groups studying consciousness?

    • The first major difference is that QRI breaks down “solving consciousness” into discrete subtasks; we’re clear about what we’re trying to do, which ontologies are relevant for this task, and what a proper solution will look like. This may sound like a small thing, but an enormous amount of energy is wasted in philosophy by not being clear about these things. This lets us “actually get to work.”

    • Second, our focus on valence is rare in the field of consciousness studies. A core bottleneck in understanding consciousness is determining what its ‘natural kinds’ are: terms which carve reality at the joints. We believe emotional valence (the pleasantness/unpleasantness of an experience) is one such natural kind, and this gives us a huge amount of information about phenomenology. It also offers a clean bridge for interfacing with (and improving upon) the best neuroscience.

    • Third, QRI takes exotic states of consciousness extremely seriously whereas most research groups do not. An analogy we make here is that ignoring exotic states of consciousness is similar to people before the scientific enlightenment thinking that they can understand the nature of energy, matter, and the physical world just by studying it at room temperature while completely ignoring extreme states such as what’s happening in the sun, black holes, plasma, or superfluid helium. QRI considers exotic states of consciousness as extremely important datapoints for reverse-engineering the underlying formalism for consciousness.

    • Lastly, we have a focus on precise, empirically testable predictions, which is rare in philosophy of mind. Any good theory of consciousness should also contribute to advancements in neuroscience. Likewise, any good theory of neuroscience should contribute to novel, bold, falsifiable predictions, and blueprints for useful things, such as new forms of therapy. Having such a full-stack approach to consciousness which does each of those two things is thus an important marker that “something interesting is going on here” and is simply very useful for testing and improving theory.

  • What methodologies are you using? How do you actually do research? 

    • QRI has three core areas of research: philosophy, neuroscience, and neurotechnology 

      • Philosophy: Our philosophy research is grounded in the eight problems of consciousness. This divide-and-conquer approach lets us explore each subproblem independently, while being confident that when all piecemeal solutions are added back together, they will constitute a full solution to consciousness.

      • Neuroscience: We’ve done original synthesis work on combining several cutting-edge theories of neuroscience (the free energy principle, the entropic brain, and connectome-specific harmonic waves) into a unified theory of Bayesian emotional updating; we’ve also built the world’s first first-principles method for quantifying emotional valence from fMRI. More generally, we focus on collecting high valence neuroimaging datasets and developing algorithms to analyze, quantify, and visualize them. We also do extensive psychophysics research, focusing on both the fine-grained cognitive-emotional effects of altered states, and how different types of sounds, pictures, body vibrations, and forms of stimulation correspond with low and high valence states of consciousness.

      • Neurotechnology: We engage in both experimentation-driven exploration, tracking the phenomenological effects of various interventions, as well as theory-driven development. In particular, we’re prototyping a line of neurofeedback tools to help treat mental health disorders.

  • What does QRI hope to do over the next 5 years? Next 20 years?

    • Over the next five years, we intend to further our neurotechnology to the point that we can treat PTSD (post-traumatic stress disorder), especially treatment-resistant PTSD. We intend to empirically verify or falsify the symmetry theory of valence. If it is falsified, we will search for a new theory that ties together all of the empirical evidence we have discovered. We aim to create an Effective Altruist cause area regarding the reduction of intense suffering as well as the study of very high valence states of consciousness.

    • Over the next 20 years, we intend to become a world-class research center where we can put the discipline of “paradise engineering” (as described by philosopher David Pearce) on firm academic grounds.

Questions About Our Mission

  • How can understanding the science of consciousness make the world a better place?

    • Understanding consciousness would improve the world in a tremendous number of ways. One obvious outcome would be the ability to better predict what types of beings are conscious—from locked-in patients to animals to pre-linguistic humans—and what their experiences might be like.

    • We also think it’s useful to break down the benefits of understanding consciousness in three ways: reducing the amount of extreme suffering in the world, increasing the baseline well-being of conscious beings, and achieving new heights for what conscious states are possible to experience.

    • Without a good theory of valence, many neurological disorders will remain completely intractable. Disorders such as fibromyalgia, complex regional pain syndrome (CRPS), migraines, and cluster headaches are all currently medical puzzles and yet have incredibly negative effects on people’s livelihoods. We think that a mathematical theory of valence will explain why these things feel so bad and what the shortest path for getting rid of them looks like. Besides valence-related disorders, nearly all mental health disorders, from clinical depression and PTSD to schizophrenia and anxiety disorders, will become better understood as we discover the structure of conscious experience.

    • We also believe that many (though not all) of the zero-sum games people play are the products of inner states of dissatisfaction and suffering. Broadly speaking, people who have a surplus of cognitive and emotional energy tend to play more positive sum games, are more interested in cooperation, and are very motivated to do so. We think that studying states such as those induced by MDMA that combine both high valence and a prosocial behavior mindset can radically alter the game theoretical landscape of the world for the better.

  • What is the end goal of QRI? What does QRI’s perfect world look like?

    • In QRI’s perfect future:

      • There is no involuntary suffering and all sentient beings are animated by gradients of bliss,

      • Research on qualia and consciousness is done at a very large scale for the purpose of mapping out the state-space of consciousness and understanding its computational and intrinsic properties (we think that we’ve barely scratched the surface of knowledge about consciousness),

      • We have figured out the game-theoretical subtleties in order to make that world dynamic yet stable: radically positive, without just making it fully homogeneous and stuck in a local maxima.

Questions About Getting Involved

  • How can I follow QRI’s work?

    • You can start by signing up for our newsletter! This is by far our most important communication channel. We also have a Facebook page, Twitter account, and Linkedin page. Lastly, we share some exclusive tidbits of ideas and thoughts with our supporters on Patreon.

  • How can I get involved with QRI?

    • The best ways to help QRI are to:

      • Donate to help support our work.

      • Read and engage with our research. We love critical responses to our ideas and encourage you to reach out if you have an interesting thought!

      • Spread the word to friends, potential donors, and people that you think would make great collaborators with QRI.

      • Check out our volunteer page to find more detailed ways that you can contribute to our mission, from independent research projects to QRI content creation.

Questions About Consciousness

  • What assumptions about consciousness does QRI have? What theory of consciousness does QRI support?

    • The most important assumption that QRI is committed to is Qualia Formalism, the hypothesis that the internal structure of our subjective experience can be represented precisely by mathematics. We are also Valence Realists: we believe valence (how good or bad an experience feels) is a real and well-defined property of conscious states. Besides these positions, we are fairly agnostic and everything else is an educated guess useful for pragmatic purposes.

  • What does QRI think of functionalism?

    • QRI thinks that functionalism takes many high-quality insights about how systems work and combines them in such a way that both creates confusion and denies the possibility of progress. In its raw, unvarnished form, functionalism is simply skepticism about the possibility of Qualia Formalism. It is simply a statement that “there is nothing here to be formalized; consciousness is like élan vital, confusion to be explained away.” It’s not actually a theory of consciousness; it’s an anti-theory. This is problematic in at least two ways:

      • 1. By assuming consciousness has formal structure, we’re able to make novel predictions that functionalism cannot (see e.g. QRI’s Symmetry Theory of Valence, and Quantifying Bliss). A few hundred years ago, there were many people who doubted that electromagnetism had a unified, elegant, formal structure, and this was a reasonable position at the time. However, in the age of the iPhone, skepticism that electricity is a “real thing” that can be formalized is no longer reasonable. Likewise, everything interesting and useful QRI builds using the foundation of Qualia Formalism stretches functionalism’s credibility thinner and thinner.

      • 2. Insofar as functionalism is skeptical about the formal existence of consciousness, it’s skeptical about the formal existence of suffering and all sentience-based morality. In other words, functionalism is a deeply amoral theory, which if taken seriously dissolves all sentience-based ethical claims. This is due to there being an infinite number of functional interpretations of a system: there’s no ground-truth fact of the matter about what algorithm a physical system is performing, about what information-processing it’s doing. And if there’s no ground-truth about which computations or functions are present, but consciousness arises from these computations or functions, then there’s no ground-truth about consciousness, or things associated with consciousness, like suffering. This is a strange and subtle point, but it’s very important. This point alone is not sufficient to reject functionalism: if the universe is amoral, we shouldn’t hold a false theory of consciousness in order to try to force reality into some ethical framework. But in debates about consciousness, functionalists should be up-front that functionalism and radical moral anti-realism is a package deal, that inherent in functionalism is the counter-intuitive claim that just as we can reinterpret which functions a physical system is instantiating, we can reinterpret what qualia it’s experiencing and whether it’s suffering.

    • For an extended argument, see Against Functionalism.

  • What does QRI think of panpsychism?

    • At QRI, we hold a position that is close to dual-aspect monism or neutral monism, which states that the universe is composed of one kind of thing that is neutral, and that both the mental and physical are two features of this same substance. One of the motivating factors for holding this view is that if there is deep structure in the physical, then there should be a corresponding deep structure to phenomenal experience. And we can tie this together with physicalism in the sense that the laws of physics ultimately describe fields of qualia. While there are some minor disagreements between dual-aspect monism and panpsychism, we believe that our position mostly fits well with a panpsychist view—that phenomenal properties are a fundamental feature of the world and aren’t spontaneously created only when a certain computation is being performed.

    • However, even with this view, there still are very important questions, such as: what makes a unified conscious experience? Where does one experience end and another begin? Without considering these problems in the light of Qualia Formalism, it is easy to tie animism into panpsychism and believe that inanimate objects like rocks, sculptures, and pieces of wood have spirits or complex subjective experiences. At QRI, we disagree with this and think that these types of objects might have extremely small pockets of unified conscious experience, but will mostly be masses of micro-qualia that are not phenomenally bound into some larger experience.

  • What does QRI think of IIT (Integrated Information Theory)?

    • QRI is very grateful for IIT because it is the first mainstream theory of consciousness that satisfies a Qualia Formalist account of experience. IIT says (and introduced the idea!) that for every conscious experience, there is a corresponding mathematical object such that the mathematical features of that object are isomorphic to the properties of the experience. QRI believes that without this idea, we cannot solve consciousness in a meaningful way, and we consider the work of Giulio Tononi to be one of our core research lineages. That said, we are not in complete agreement with the specific mathematical and ontological choices of IIT, and we think it may be trying to ‘have its cake and eat it too’ with regard to functionalism vs physicalism. For more, see Sections III-V of Principia Qualia.

    • We make no claim that some future version of IIT, particularly something more directly compatible with physics, couldn’t cleanly address our objections, and see a lot of plausible directions and promise in this space.

  • What does QRI think of the free energy principle and predictive coding?

    • On our research lineages page, we list the work of Karl Friston as one of QRI’s core research lineages. We consider the free energy principle (FEP), as well as related research such as predictive coding, active inference, the Bayesian brain, and cybernetic regulation, as an incredibly elegant and predictive story of how brains work. Friston’s idea also forms a key part of the foundation for QRI’s theory of brain self-organization and emotional updating, Neural Annealing.

    • However, we don’t think that the free energy principle is itself a theory of consciousness, as it suffers from many of the shortcomings of functionalism: we can tell the story about how the brain minimizes free energy, but we don’t have a way of pointing at the brain and saying *there* is the free energy! The FEP is an amazing logical model, but it’s not directly connected to any physical mechanism. It is a story that “this sort of abstract thing is going on in the brain” without a clear method of mapping this abstract story to reality.

    • Friston has supported this functionalist interpretation of his work, noting that he sees consciousness as a process of inference, not a thing. That said, we are very interested in his work on calculating the information geometry of Markov blankets, as this could provide a tacit foundation for a formalist account of qualia under the FEP. Regardless of this, though, we believe Friston’s work will play a significant role in a future science of mind.

  • What does QRI think of global workspace theory?

    • The global workspace theory (GWT) is a cluster of empirical observations that seem to be very important for understanding what systems in the brain contribute to a reportable experience at a given point in time. The global workspace theory is a very important clue for answering questions of what philosophers call Access Consciousness, or the aspects of our experience on which we can report.

    • However, QRI does not consider the global workspace theory to be a full theory of consciousness. Parts of the brain that are not immediately contributing to the global workspace may be composed of micro qualia, or tiny clusters of experience. They’re obviously impossible to report on, but they are still relevant to the study of consciousness. In other words, just because a part of your brain wasn’t included in the instantaneous global workspace, doesn’t mean that it can’t suffer or it can’t experience happiness. We value global workspace research because questions of Access Consciousness are still very critical for a full theory of consciousness.

  • What does QRI think of higher-order theories of consciousness?

    • QRI is generally opposed to theories of consciousness that equate consciousness with higher order reflective thought and cognition. Some of the most intense conscious experiences are pre-reflective or unreflective such as blind panic, religious ecstasy, experiences of 5-MeO-DMT, and cluster headaches. In these examples, there is not much reflectivity nor cognition going on, yet they are intensely conscious. Therefore, we largely reject any attempt to define consciousness with a higher-order theory.

  • What is the relationship between evolution and consciousness?

    • The relationship between evolution and consciousness is very intricate and subtle. An eliminativist approach arrives at the simple idea that information processing of a certain type is evolutionarily advantageous, and perhaps we can call this consciousness. However, with a Qualia Formalist approach, it seems instead that the very properties of the mathematical object isomorphic to consciousness can play key roles (either causal or in terms of information processing) that make it advantageous for organisms to recruit consciousness.

    • If you don’t realize that consciousness maps onto a mathematical object with properties, you may think that you understand why consciousness was recruited by natural selection, but your understanding of the topic would be incomplete. In other words, to have a full understanding of why evolution recruited consciousness, you need to understand what advantages the mathematical object has. One very important feature of consciousness is its capacity for binding. For example, the unitary nature of experience—the fact that we can experience a lot of qualia simultaneously—may be a key feature of consciousness that accelerates the process of finding solutions to constraint satisfaction problems. In turn, evolution would hence have a reason to recruit states of consciousness for computation. So rather than thinking of consciousness as identical with the computation that is going on in the brain, we can think of it as a resource with unique computational benefits that are powerful and dynamic enough to make organisms that use it more adaptable to their environments.

  • Does QRI think that animals are conscious?

    • QRI thinks there is a very high probability that every animal with a nervous system is conscious. We are agnostic about unified consciousness in insects, but we consider it very likely. We believe research on animal consciousness has relevance when it comes to treating animals ethically. Additionally, we do think that the ethical importance of consciousness has more to do with the pleasure-pain axis (valence), rather than cognitive ability. In that sense, the suffering of non-human animals may be just as morally relevant, if not more relevant than humans. The cortex seems to play a largely inhibitory role for emotions, such that the larger the cortex is, the better we’re able to manage and suppress our emotions. Consequently, animals whose cortices are less developed than ours may experience pleasure and pain in a more intense and uncontrollable way, like a pre-linguistic toddler.

  • Does QRI think that plants are conscious?

    • We think it’s very unlikely that plants are conscious. The main reason is that they lack an evolutionary reason to recruit consciousness. Large-scale phenomenally bound experience may be very energetically expensive, and plants don’t have much energy to spare. Additionally, plants have thick cellulose walls that separate individual cells, making it very unlikely that plants can solve the binding problem and therefore create unified moments of experience.

  • Why do some people seek out pain?

    • This is a very multifaceted question. As a whole, we postulate that in the vast majority of cases, when somebody may be nominally pursuing pain or suffering, they’re actually trying to reduce internal dissonance in pursuit of consonance or they’re failing to predict how pain will actually feel. For example, when a person hears very harsh music, or enjoys extremely spicy food, this can be explained in terms of either masking other unpleasant sensations or raising the energy parameter of experience, the latter of which can lead to neural annealing: a very pleasant experience that manifests as consonance in the moment.

  • I sometimes like being sad. Is QRI trying to take that away from me?

    • Before we try to ‘fix’ something, it’s important to understand what it’s trying to do for us. Sometimes suffering leads to growth; sometimes creating valuable things involves suffering. Sometimes, ‘being sad’ feels strangely good. Insofar as suffering is doing good things for us, or for the world, QRI advocates a light touch (see Chesterton’s fence). However, we also suggest two things:

      • 1. Most kinds of melancholic or mixed states of sadness usually are pursued for reasons that cash out as some sort of pleasure. Bittersweet experiences are far more preferable than intense agony or deep depression. If you enjoy sadness, it’s probably because there’s an aspect of your experience that is enjoyable. If it were possible to remove the sad part of your experience while maintaining the enjoyable part of it, you might be surprised to find that you prefer this modified experience more than the original one.

      • 2. There are kinds of sadness and suffering that are just bad, that degrade us as humans, and would be better to never feel. QRI doesn’t believe in forcibly taking away voluntary suffering, or pushing bliss on people. But we would like to live in a world where people can choose to avoid such negative states, and on the margin, we believe it would be better for humanity for more people to be joyful, filled with a deep sense of well-being.

  • If dissonance is so negative, why is dissonance so important in music?

    • When you listen to very consonant music or consonant tones, you will quickly adapt to these sounds and get bored of them. This has nothing to do with consonance itself being unpleasant and everything to do with learning in the brain. Whenever you experience the same stimuli repeatedly, most brains will trigger a boredom mechanism and add dissonance of its own in order to make you enjoy the stimuli less or simply inhibit it, not allowing you to experience it at all. Semantic satiation is a classic example of this where repeating the same word over and over will make it lose its meaning. For this reason, to trigger many high valence states of consciousness consecutively, you need contrast. In particular, music works with gradients of consonance and dissonance, and in most cases, moving towards consonance is what feels good rather than the absolute value of consonance. Music tends to feel the best when you mix a high absolute value of consonance together with a very strong sense of moving towards an even higher absolute value of consonance. Playing some levels of dissonance during a song will later enhance the enjoyment of the more consonant parts such as the chorus of songs, which are reported to be the most euphoric parts of song and typically are extremely consonant.

  • What is QRI’s perspective on AI and AI safety research?

    • QRI thinks that consciousness research is critical for addressing AI safety. Without a precise way of quantifying an action’s impact on conscious experiences, we won’t be able to guarantee that an AI system has been programmed to act benevolently. Also, certain types of physical systems that perform computational tasks may be experiencing negative valence without any outside observer being aware of it. We need a theory of what produces unpleasant experiences to avoid inadvertently creating superintelligences that suffer intensely in the process of solving important problems or accidentally inflict large-scale suffering.

    • Additionally, we think that a very large percentage of what will make powerful AI dangerous is that the humans programming these machines and using these machines may be reasoning from states of loneliness, resentment, envy, or anger. By discovering ways to help humans transition away from these states, we can reduce the risks of AI by creating humans that are more ethical and aligned with consciousness more broadly. In short: an antidote for nihilism could lead to a substantial reduction in existential risk.

    • One way to think about QRI and AI safety is that the world is building AI, but doesn’t really have a clear, positive vision of what to do with AI. Lacking this, the default objective becomes “take over the world.” We think a good theory of consciousness could and will offer new visions of what kind of futures are worth building—new Schelling points that humanity (and AI researchers) could self-organize around.

  • Can digital computers implementing AI algorithms be conscious?

    • QRI is agnostic about this question. We have reasons to believe that digital computers in their current form cannot solve the phenomenal binding problem. Most of the activity in digital computers can be explained in a stepwise fashion in terms of localized processing of bits of information. Because of this, we believe that current digital computers could be creating fragments of qualia, but are unlikely to be creating strongly globally bound experiences. So, we consider the consciousness of digital computers unlikely, although given our current uncertainty over the Binding Problem (or alternatively framed, the Boundary Problem), this assumption is lightly held. In the previous question, when we write that “certain types of physical systems that perform computational tasks may be experiencing negative valence”, we assume that these hypothetical computers have some type of unified conscious experience as a result of having solved the phenomenal binding problem. For more on this topic, see: “What’s Out There?

  • How much mainstream recognition has QRI’s work received, either for this line of research or others? Has it published in peer-reviewed journals, received any grants, or garnered positive reviews from other academics?

    • We are collaborating with researchers from Johns Hopkins University and Stanford University on several studies involving the analysis of neuroimaging data of high-valence states of consciousness. Additionally, we are currently preparing two publications for peer-reviewed journals on topics from our core research areas. Michael Johnson will be presenting at this year’s MCS seminar series, along with Karl Friston, Anil Seth, Selen Atasoy, Nao Tsuchiya, and others; Michael Johnson, Andrés Gómez Emilsson, and Quintin Frerichs have also given invited talks at various east-coast colleges (Harvard, MIT, Princeton, and Dartmouth).

    • Some well-known researchers and intellectuals that are familiar and think positively about our work include: Robin Carhart-Harris, Scott Alexander, David Pearce, Steven Lehar, Daniel Ingram, and more. Scott Alexander acknowledged that QRI put together the paradigms that contributed to Friston’s integrative model of how psychedelics work before his research was published. Our track record so far has been to foreshadow (by several years in advance) key discoveries later proposed and accepted in mainstream academia. Given our current research findings, we expect this trend to continue in the years to come.

Miscellaneous

  • How does QRI know what is best for other people/animals? What about cultural relativism?

    • We think that, to a large extent, people and animals work under the illusion that they are pursuing intentional objects, states of the external environment, or relationships that they may have with the external environment. However, when you examine these situations closely, you realize that what we actually pursue are states of high valence triggered by external circumstances. There may be evolutionary and cultural selection pressures that push us toward self-deception as to how we actually function. And we consider it negative to have these selection pressures makes us less self-aware because it often focuses our energy on unpleasant, destructive, or fruitless strategies. QRI hopes to support people in fostering more self-awareness, which can come through experiments with one’s own consciousness, like meditation, as well as through the deeper theoretical understanding of what it is that we actually want.

  • How central is David Pearce’s work to the work of the QRI?

    • We consider David Pearce to be one of our core lineages. We particularly value his contribution to valence realism, the insistence that states of consciousness come with an overall valence, and that this is very morally relevant. We also consider David Pearce to be very influential in philosophy of mind; Pearce, for instance, coined the phrase ‘tyranny of the intentional object’, the title of a core QRI piece of the same name. We have been inspired by Pearce’s descriptions for what any scientific theory of consciousness should be able to explain, as well as his particular emphasis on the binding problem. David’s vision of a world animated by ‘gradients of bliss’ has also been very generative as a normative thought experiment which integrates human and non-human well-being. We do not necessarily agree with all of David Pearce’s work, but we respect him as an insightful and vivid thinker who has been brave enough to actually take a swing at describing utopia and who we believe is far ahead of his time.

  • What does QRI think of negative utilitarianism?

    • There’s general agreement within QRI that intense suffering is an extreme moral priority, and we’ve done substantial work on finding simple ways of getting rid of extreme suffering (with our research inspiring at least one unaffiliated startup to date). However, we find it premature to strongly endorse any pre-packaged ethical theory, especially because none of them are based on any formalism, but rather an ungrounded concept of ‘utility’. The value of information here seems enormous, and we hope that we can get to a point where the ‘correct’ ethical theory may simply ‘pop out of the equations’ of reality. It’s also important to highlight the fact that common versions and academic formulations of utilitarianism seem to be blind to many subtleties concerning valence. For example, they do not distinguish between mixed states of consciousness where you have extreme pleasure combined with extreme suffering in such a way that you judge the experience to be neither entirely suffering nor entirely happiness and states of complete neutrality, such as extreme white noise. Because most formulations of utilitarianism do not distinguish between them, we are generally suspicious of the idea that philosophers of ethics have considered all of the relevant attributes of consciousness in order to make accurate judgments about morality.

  • What does QRI think of philosophy of mind departments?

    • We believe that the problems that philosophy of mind departments address tend to be very disconnected from what truly matters from an ethical, moral, and philosophical point of view. For example, there is little appreciation of the value of bringing mathematical formalisms into discussions about the mind, or what that might look like in practice. Likewise there is close to no interest in preventing extreme suffering nor understanding its nature. Additionally, there is usually a disregard for extreme states of positive valence, and strange or exotic experiences in general. It may be the case that there are worthwhile things happening in departments and classes creating and studying this literature, but we find them characterized by processes which are unlikely to produce progress on their nominal purpose, creating a science of mind.

    • In particular, in academic philosophy of mind, we’ve seen very little regard for producing empirically testable predictions. There are millions of pages written about philosophy of mind, but the number of pages that provide precise, empirically testable predictions is quite thin.

  • What therapies does QRI recommend for depression, anxiety, and chronic pain?

    • At QRI, we do not make specific recommendations to individuals, but rather point to areas of research that we consider to be extremely important, tractable, and neglected, such as anti-tolerance drugs, neural annealing techniques, frequency specific microcurrent for kidney stone pain, and N,N-DMT and other tryptamines for cluster headaches and migraines.

  • Why does QRI think it’s so important to focus on ending extreme suffering? 

    • QRI thinks ending extreme suffering is important, tractable, and neglected. It’s important because of the logarithmic scales of pleasure and pain—the fact that extreme suffering is far worse by orders of magnitude than what people intuitively believe. It’s tractable because there are many types of extreme suffering that have existing solutions that are fairly trivial or at least have a viable path for being solved with moderately funded research programs. And it’s neglected mostly because people are unaware of the existence of these states, though not necessarily because of their rarity. For example, 10% of the population experiences kidney stones at some point in their life, but for reasons having to do with trauma, PTSD, and the state-dependence of memory, even people who have suffered from kidney stones do not typically end up dedicating their time or resources toward eradicating them.

    • It’s also likely that if we can meaningfully improve the absolute worst experiences, much of the knowledge we’ll gain in that process will translate into other contexts. In particular, we should expect to figure out how to make moderately depressed people happier, fix more mild forms of pain, improve the human hedonic baseline, and safely reach extremely great peak states. Mood research is not a zero-sum game. It’s a web of synergies.



Many thanks to Andrew Zuckerman, Mackenzie Dion, and Mike Johnson for their collaboration in putting this together. Featured image is QRI’s logo – animated by Hunter Meyer.

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

TL;DR

Some substances seem to be much better at treating psychological trauma than others, even when they are seemingly similar in nature. We have reason to believe that 5-MeO-DMT is significantly better suited for this task than N,N-DMT (“DMT” from now on). In order to gain insight into why this difference exists, we investigate the phenomenological differences and similarities between the experiences produced by these two tryptamine psychedelics. In particular, we develop 9 lenses that show promise for understanding how 5-MeO-DMT and DMT differ:

  1. Space vs. Form: 5-MeO is more space-like than DMT.
  2. Crystals vs. Quasi-Crystals: 5-MeO generates more perfectly repeating rhythms and hallucinations than DMT.
  3. Non-Attachment vs. Attachment: 5-MeO seems to enable detachment from the craving of both existence and non-existence, whereas DMT enhances the craving.
  4. Underfitting vs. Overfitting: 5-MeO reduces one’s model complexity whereas DMT radically increases it.
  5. Fixed Points and Limit Cycles vs. Chaotic Attractors: 5-MeO’s effect on feedback leads to stable and predictable attractors while DMT’s attractors are inherently chaotic.
  6. Modulation of Lateral Inhibition: 5-MeO may reduce lateral inhibition while DMT may enhance it.
  7. Diffuse Attention vs. Focused Attention: 5-MeO diffuses attention uniformly over large regions of one’s experiential field, while DMT seems to focus it.
  8. Big Chunks and Tiny Chunks vs. A Power Law of Chunks: 5-MeO creates a few huge phases of experience (as in phases of matter) with a few remaining specks, while DMT produces a more organic power law distribution of chunk sizes.
  9. Integration vs. Fragmentation: 5-MeO seems to give rise to “neural integration” involving the entrainment of any two arbitrary subnetworks (even when they usually do not talk to each other), while DMT fragments communication between most networks but massively enhances it between some specific kinds of networks.

All of this together suggests that 5-MeO-DMT is better at helping you “reconnect with yourself” than DMT. And this may be key to treating trauma effectively.


What is Trauma?

I will start out by briefly mentioning an interesting property of psychological trauma. You see, trauma has a lot of somatic manifestations. Feeling disconnected from yourself,  like you are full of blockages, that you have numb regions in your body despite no physical damage, and so on, are all quintessential ways in which trauma shows up in a person’s everyday life. Given these manifestations, do these suggest any new way of treating this? How about using something that facilitates the communication between parts of your nervous system that are not on “speaking terms” with each other? Would giving our nervous system a kind of vibration that simultaneously entrains any two of its regions to make them act as a unit be of any help?


Psychotropic Treatment of Trauma

Based on tens of interviews, hundreds of trip reports, and a literature review, I have arrived at a tentative short list of drugs that have the highest potential to heal trauma (in decreasing order):

  1. 5-MeO-DMT
  2. MDMA
  3. Ketamine

They are all synergistic combined with music, vibration, strobes, and olfaction. And when wisely used, they all have the ability to help you move on past pain: stop ruminating, stop feeling like your behavior is inhibited, and stop having panic attacks associated with your past experiences.

At some point in the future I will provide direct empirical evidence for the claim that these three substances are uniquely good for treating trauma. Arguably psilocybin, ayahuasca, and LSD can be helpful in processing traumatic experiences too. But my claim is that the options I listed are uniquely good at deeply resolving the issues at an emotional level and bringing to you the opportunity to feel a profound and lasting sense of inner peace.

DMT won’t help as much as 5-MeO-DMT.

MDA is not as good as MDMA.

And DXM, ok, perhaps it can also be quite useful for trauma… but ketamine has something “extra” that really helps.

What is this?


harmonic_society


The Koan: 5-MeO-DMT, MDMA, Ketamine?

Perhaps we could gain a new perspective by framing this as a Koan: what do MDMA, 5-MeO-DMT, and ketamine have in common? You HAVE to the figure this out in the next 24 hours, or your guru will literally cut your cat in half. Motivated enough?

What do you do? Well, you start out by borrowing psychopharmacology books from the library. But does that help? When it comes to trauma, in traditional textbook neuroscience MDMA is at best just a footnote. Ketamine is not even mentioned for the most part, unless the book is hip, but even then it will be mentioned in the chapter about anaesthetics and painkillers; its psychoactive effects will be glossed over as “emergent phenomena”. No! What are you doing? Wasn’t John C. Lilly already talking about the far-out, extraordinarily bizarre, perhaps even literally inter-dimensional properties of “vitamin K” way back in the 70s? Then why is my 2007 “Drugs and the Brain” textbook so totally lacking in any kind of genuine phenomenological insight about this stuff?

And what about 5-MeO-DMT? You are lucky if the term appears even once in your $800 textbook. And if it does indeed appear, you can bet it will also be in a footnote, this time concerning matters such as “psychoactive animals”, “other tryptamines”, and “mesoamerican entheogens”. You will neither see 5-MeO-DMT mentioned in a personal identity philosophy textbook, nor in a neuroscience treatise on “neural synchrony”, nor in the part of academia focused on “innovation in the treatment of mental illness”.

It is sad to admit, but the official main-lined level of interest in the three most promising therapeutic tools for trauma listed above is a matter of sorting and assembling footnotes.

I am exaggerating a bit, of course.

MDMA’s therapeutic potential is gaining traction thanks to the tireless work of MAPS. S-Ketamine is now approved as an anti-depressant. And while 5-MeO-DMT is gaining popularity at a glacial pace, it is at this point by no means a secret. An increasing number of vocal members of the psychedelic community have been talking about 5-MeO-DMT for some years. People who have publicly emphasized how different five is from other psychedelics include Hamilton Morris, James Oroc, Martin Ball, Leo Gura, and Rak Razam. But what we still lack is rigorous scientific backing for these claims. After all, everyone is likely to want to sell their aesthetic preferences as universal truths about beauty and bliss, right? Thankfully, there are some early scientific indications already:

The above graph comes from a 2018 study that investigated the therapeutic effects of 5-MeO-DMT-containing toad venom relative to psilocybin. The dose used (the amount of buffo venom vaporized) had an estimated content of 5-7mg of 5-MeO-DMT, and the researchers classified 75% of the resulting experiences as meeting the criteria for a “complete mystical experience”. It measured people’s level of response with the Mystical Experiences Questionnaire (MEQ30), and as you can see from the graph above, in every category 5-MeO-DMT seems to be more powerful than psilocybin. The level of effectiveness was indeed found to be higher than all but the highest dose of psilocybin, and chances are that the study couldn’t show it was more effective than that because it was underpowered to detect it, and not because there is no difference (in other words, the sample size was not large enough for the difference between high-dose psilocybin and 5-MeO-DMT to reach statistical significance). Also bear in mind the key difference that the trip lasts under 20 minutes in total, meaning that even if the trip fails to produce full effects, you can still afford to try it again ten more times in the same time interval that it would have taken you to experience a full psilocybin trip. More so, it is important to point out that the dose of 5-MeO-DMT taken by the participants of this study is considered to be at the edge between “light” and “common” in PsychonautWiki’s entry on the drug*. Indeed, for many people the “breakthrough” tends to happen around 10mg, and I’ve heard of people using up to 30mg of it at a time. (Beware: if you ever try this – please don’t jump straight to a high dose, as this can cause serious trauma as a result.) Therefore, I think it is reasonable to expect that future studies will confirm what anecdotal data is currently screaming: that 5-MeO-DMT is more “powerful” and “mystical” in its effects than psilocybin, LSD, DMT, 2C-B, and all the rest.

But what this “power” and “mysticism” exactly amounts to still lacks clear and useful definitions. More so, is there any concrete reason why 5-MeO-DMT may be also superior at healing trauma relative to, eg. LSD or psilocybin? Technically, one could currently argue that since the presentation of “complete mystical states” is a mediating factor in whether psilocybin has long-lasting psychological benefits, that 5-MeO-DMT is more effective simply because it has a higher probability of causing this effect. But I would argue that the texture of 5-MeO-DMT peak experiences is different and not only just more intense, and that the way in which it is different matters for its therapeutic value.

To investigate this particular difference, we now move on to examining the phenomenological difference between 5-MeO-DMT and DMT.


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

My experience is that a reasonable ~20% of people I talk to who have a long-standing interest in psychedelics have heard about 5-MeO-DMT’s special properties. However, only a much smaller percentage of people have actually tried it. At Qualia Computing we have talked about its exceptional phenomenological properties a number of times. Yet it remains that most readers who reach out have not themselves experienced it. Hence I have not really had access to quality trip reports in order to say anything meaningful about the way in which it is different from DMT.

Thankfully, I’ve recently interviewed someone who has a decent level of experience with 5-MeO-DMT (20+ trips), along with a significant level of experience with vaporized DMT (100+ trips), and is also acquainted with the combination (10+ trips with both substances at once).

Given the incredibly intense psychoactive effects of 5-MeO-DMT (both for good and bad), most people struggle to put into words anything meaningful about the state. That said, as it has been the case with a number of other states of consciousness (e.g. LSD, DMT, and MDMA) I feel compelled to try to offer a sane, rational, agnostic, and pragmatic description of its phenomenology. In particular, I think that 5-MeO-DMT’s unique trauma-healing potential really deserves a close look. I believe that it sheds light on a wide range of topics of interest such as neural annealing, the Symmetry Theory of Valence, and the pseudo-time arrow (video). With this in mind, I inquired with my interviewee about the differences between N,N-DMT and 5-MeO-DMT. Together, after a lengthy open-ended discussion, we found the following ways to compare them:

1. Space vs. Form

QRI - Art and Consciousness copy 33QRI - Art and Consciousness copy 29

One of the first things that stands out when you take DMT at even small doses is the way in which colors get intensified. This is a special case of a very generalizable effect: in fact, every perceptual feature you can point to is enhanced while on DMT, from the sharpness of edges and pointy things to the feeling of movement and rotation. The interviewee indeed said that a brain on DMT “becomes a powerful qualia machine“.

After trying 5-MeO-DMT for the first time, most people already familiar with DMT mention something along the lines of “I was surprised that I didn’t see many colors.” The visual component of five is rarely very colorful; if color appears, it is in the form of a golden or sometimes light faint magenta or green hue. For the most part, the visual component of the experience is black and/or white. At times, one can see rainbow halos, but like rare subatomic particles, on 5-MeO-DMT rainbow colors come in and out of the vacuum, as if somehow equivalent to it. The bulk of the visuals manifest in a dazzling sense of spaciousness, as if there was a cosmic paint called “transparent/translucent”. The space often feels immeasurable due to a lack of a reference frame from which to make a judgement in terms of known comparisons. But what inevitably stands out is that the space seems large, uniform, harmonious, smooth, and luminous. Somatic feelings blend with this space, and the uniformity and symmetry of it allows for energy to seamlessly move throughout it. It really is a remarkable effect, one which can easily give rise to the felt-sense of Open Individualism. Yet, despite the engrossingly engaging character of these feelings, there is very little narrative complexity in sight.

Who knew that empty space could be so much fun? That you could fit so much love and bliss in an (experiential) vacuum? More so, the more you are able to relax into it, the more you embrace the waves of equanimity, the more you allow the space to become perfectly smooth and seamless… the more blissful it all gets!**

2. Crystals and Quasi-Crystals

Here is an interesting thing – ultrasound has been used in order to bias the way in which water crystals form, and thus creating much more “cubical” water than is otherwise possible. More generally, the phenomenon of vibration affecting crystallization processes is worth considering as an explanatory framework. DMT comes with a particular vibe that some have identified as having a characteristic frequency somewhere between 20 and 30Hz, whereas 5-MeO-DMT’s vibe seems to be a notch higher, perhaps in the range of 30 to 40Hz. On these drugs, your attention is jittered back and forth at a certain frequency, and this affects your ability to focus on any given part of your experience. The specific jittering itself makes it harder and easier to construct and manipulate certain thought-forms over others.

Speculatively, this model says that the jittering of attention caused by 5-MeO-DMT and DMT give rise to crystal and quasi-crystal building blocks, respectively, for phenomenal objects in one’s experience.

Phenomenologically, it seems that the vibratory signature of DMT effects doesn’t wrap around your experience an integer number of times. Thus, what we will call, for lack of a better term, the qualia crystals that form while on DMT seem to be inherently unstable and alien to your normal way of cogitating. The fact that the vibrations don’t fit perfectly in one’s experiential field forces it to bend out of shape to accommodate such vibrations. The result is constant chaos – fluid instabilities as the core effect.

Spatial_KH_2D_vorticity

Kelvin–Helmholtz instability (source)

On the other hand, with 5-MeO-DMT, it’s as if the vibration activates parts of the field of experience in exactly the right way for them to blend, unite, and resonate with one another. The vibrations fit perfectly inside one’s experiential field, and allows it to relax into its own natural shape. And this allows for perfect crystals of awareness to peacefully grow, multiply, and synchronize.

It is of course surprising that a tiny difference in the frequency of the vibe could have such large effects in the way phenomenology crystallizes. But this is true for other systems. When one talks about the complexity of shapes in resonant systems, for instance, Lissajou curves can provide a helpful intuition pump: in Lissajou curves, merely changing one of the frequencies by a small relative amount can result in a huge difference between the pictures drawn. From a simple circle to a complex mesh.

1_o75epumclddyHiiJfCYNIg

According to the interviewee, 5-MeO-DMT would neatly correspond to one of the simplest Lissajou curves above, while the DMT sensations would be better evoked by one of the meshy ones.

With that said, it is worth mentioning that 5-MeO-DMT is probably not hitting the right frequencies merely by chance. It’s probably more that it is activating a system whose attractor is self-correcting and results in the kind of symmetrical crystallization that gives rise to deep feelings of bliss.

3. Attachment to Existence and Non-Existence

5-MeO seems to point at a super general sense of “relaxation”. A meditation teacher, Ajahn Brahm, talking about jhanas said that his mantra to achieve states of deep concentration and peace was: “Relax… to the maaaaaaaxxxx…”.

And according to the interviewee, the thrust of 5-MeO is that it feels like your nervous system is paradoxically injected with a lot of energy and yet the vibe of this energy is that of total and complete- ultimate?- relaxation. But our body and mind are not used to relaxing deeply. The contrast between this energy and one’s usual neurotic state can itself produce a lot of dissonance and resistance. In most cases, this is transient and contained to the first couple of minutes of the experience, though in higher doses and with bad luck, it can also spiral out of control. On small and moderate doses the energetic vibe of “relax to the max” takes over one’s experiential center of mass and teaches the rest of your nervous system how to relax.

It vibrates your nervous system in just the right way that all of your tensions, and hidden knots, and internalized stresses, bubble up to the surface, and you have the chance to try to “unravel” all of that tension.

After taking it a couple of times, you get the very nice feeling of being OK with whatever happens. 5-MeO-DMT might be described as a drug that allows you to reduce “thirst, craving, desire” in a very generalized kind of way. In Buddhist terms, this would be to reduce “Taṇhā“, which comes in three different kinds: “kāma-taṇhā (craving for sensual pleasures), bhava-taṇhā (craving for existence), and vibhava-taṇhā (craving for non-existence).” While we are used to hearing religious figures and moralists talk about the importance of not experiencing cravings for sensual pleasure, popular culture still lacks legible myths about the craving for existence and non-existence. So it comes as a shock for someone who has never developed meditative introspective insight just how much of our suffering has the flavor of either craving for existence or non-existence. Dosing on 5-MeO-DMT gives you a glimpse for what a mind devoid of these more subtle forms of craving feels like.

DMT comes with a vibratory frenzy that directly causes a lot of knots and tiny stress points throughout your entire experiential field. And to a large extent, how the experience unfolds is the result of you trying to manage all of those knots and stress points to avoid having them accumulate and concentrate in painful ways. The effect of this is that DMT acutely increases attachment to existence and non-existence. You wonder “what am I in this world?” and then cling to it, along with an intense fear of losing yourself in the world of vibration. The fear is quite involuntary and primal. And because you are clinging to who or what you are, there is something that you can lose, and that contributes to the feeling that the stakes are very high.acacia-trees-grown-on-the-african-savannah-stock-video-footage-png-tree-savannah-1920_1080

On 5-MeO you look at a landscape- say a tree in a Savannah- and you can see it in the most Zen way you can possibly imagine (and I don’t mean cutting cats in half). You think: “there is neither someone nor nobody in there” and “it’s all just arising and passing of ephemera”.

On DMT you look at the same landscape and you feel: “Oh gosh, what AM I in here? Am I that rock over there? That rock seems threatened by erosion! Am I the tree? But what if something comes and eats the tree?” and so on.

On DMT you feel like you are one of many little beings in a vast ecosystem. On 5-MeO-DMT you step back and you “realize” that you are the entire ecosystem.

In turn, this may explain to some extent the fact that the content of DMT hallucinations is often filled with exotic beings. And almost always, these beings have hyper-specific ways of life, tastes, intentions, and beliefs. The realms you experience on DMT are all saturated with attachment to existence and non-existence, and the beings you interact with are no exception. In fact, they may be a manifestation of those intensified cravings! From the interviewee:

“I’ve taken DMT about 100 times and have encountered many different vibes and kinds of intelligences in those realms. I’d love to map out the possible narratives – there are so many! Loosely speaking, I’ve had many different encounters with intelligent beings: from blissfully angelic and benevolent to outright demonic. Most of the beings I’ve encountered are somewhere in-between, and for the most part, tend to have pretty dualistic mindsets.

There is this whole class of beings I’d identify as harlequins/jesters that just love to play tricks with perception (I’m sure that’s what Terrence McKenna was pointing at).

Then there are “artists” which have a particular style that they explore and can range from emotionally self-sufficient to aggressively in-your-face about their work. The “look at THIS and look at THIS and look at THIS!!” kind of stance, where you are not given enough time to process what they’ve already shown you before they thrust even more stuff into you, and then attach ‘cookies’ into your etheric body to track you in future trips so you “like” and “subscribe” to their “interdimensional channel” or something like that.

There’s also a lot of beings that seem to want to tell you that they are confident that God does not exist and that “everything is allowed”. And I’ve felt that they are really indifferent to morality, but still have powerful abilities and unique qualia of a more scientific bent.

I once also encountered what felt like a true sadistic demon that played some really nasty tricks on my perception, and filled me with “etheric bugs” and had hundreds of little minions to attack me in many unpleasant ways. That said, I blame this on the fact that I was sleep deprived when I took DMT that one time, and it’s never happened before or since. But that experience gave me a lot of respect for the drug.

I’ve also encountered realms where they actually do consciousness research as such, and are benevolent and into engineering paradises. I even asked one of them if they knew my favorite philosopher, and they said “yes, what a nice fellow – we hope he will be more widely known in the future. We just wished that he wasn’t so sad a lot of the time.”

I have found that my pre-existing mood is the single most important variable that determines the kind of intelligences I encounter. So I’d really like to someday try MDMA and DMT combined. I suspect those would be very angelic beings most of the time.

Interestingly, I feel that while DMT feels profoundly spiritual, to a large extent it is less “nondual” than most other psychedelics. A lot of beings I’ve encountered simply don’t seem to care about oneness at all. But on LSD, mushrooms, and of course 5-MeO-DMT, the Golden Rule seems to play a very central role in the experience. Those experiences are much more of a “teaching” than the wacky stuff one encounters in the DMT realms.

When you take 5-MeO-DMT and DMT at the same time, you can really feel the contrast between the dualistic “us vs. them” vibe that underlies DMT and the unitive sense that underlies 5-MeO-DMT. I’ve experimented with the combo and found it to be super informative. And usually, I realize that while DMT turns your brain into a high-octane “qualia machine”, 5-MeO is in fact much more peaceful and happy in a deeper sense. I’d like to understand both, but my preferred “home” would be for sure the 5-MeO realms.”

4. Underfitting vs. Overfitting

One interesting lens with which to make sense of the difference between people who are open to experience and people who are not is that of model complexity, which casts this difference in terms of the statistical concepts of underfitting and overfitting.

Having narrow views, simple explanations, and enduring preferences is very good when the world itself is either very simple or impossible to understand. But having complex views, multi-layered explanations, and flexible preferences is more adaptive than the alternative in a world that is both complex and can be understood with some effort.

Indeed, some speculation about the nature of sleep from the predictive coding paradigm of cognition is that dreaming is a process of model complexity reduction. The information that we accumulate over the span of a day is incorporated in an ad-hoc fashion while awake, and only properly integrated (and pruned) after a good night’s sleep. This at least provides the theoretical precedent for describing a specific state of consciousness in terms of its effects on model complexity. And here we would propose that as a very general effect profile, 5-MeO-DMT reduces model complexity while N,N-DMT increases it.

As a consequence, we would posit that people who take five a lot will converge towards compelling yet over-simplified models of realty, whereas people who use DMT a lot will converge to overly complex and unnecessarily detailed explanation for even the simplest of phenomena.1__7OPgojau8hkiPUiHoGK_w

Perhaps in the future people could be diagnosed as chronic overfitters and underfitters. In turn, these two drugs could be given by prescription, for the maladies of improper model selection practices:

N,N-DMT would be given to the sufferers of too much worldview simplicity. People like this believe that the world is dominated by the struggle between capitalism and communism. They think that there is a 50% chance that God almighty exists. They assign zero probability to unlikely events, such as lizard people power conspiracies. In people like this, DMT is a powerful mind enhancer capable of challenging cartoonishly simple background assumptions and introducing a healthy dose of skepticism in mainstream narratives.

5-MeO-DMT would instead be given to people who are overwhelmingly embroiled in complex interpretations of the nature of reality. Whether it is in the realm of conspiracy theories, religious cults, the biochemistry of aging, or any such hopelessly convoluted field of research, a little five will unscramble the mind of the compulsive overfitter. Thanks to the drug, the Bayesian puncture, the Occam’s cut, and the pragmatic so what coalesce into a decimating hit to the load-bearing hub-nodes that feed unfalsifiable belief systems. The model complexity reduction effect dissolves entire subfields, assimilates clusters, and seamlessly mends discontinuities in the reality mappings of the patient. At moderate dosages and treatment regimes, the sufferer recovers fully. The sufferer often ends up healed of their traumas, and occasionally healed of many more things than expected. At levels much above those of the therapeutic standard of care, there is a risk that the treatment may result in the healing of the fundamental traumas of conscious experience. The drug may offer the patient a chance to relinquish phenomenal reality in exchange for an extemporal “ultimate relief”. To extinguish the flame of existence, as they say.41592_2016_Article_BFnmeth3968_Fig1_HTML

Importantly, after 5-MeO-DMT therapy, the patient is, let’s say (for the sake of speculating), 20 times as likely as members of the general public to say yes to questions like “Are we all one consciousness?”, “Is the world a process created for the refinement of our souls?” and “Is the universe made of infinite consciousness?”.

So where does that leave us? The good news is that this may have game-theoretical benefits for the side of consciousness in the eternal battle between consciousness and pure replicators. The bad news is that it can overwrite important information obtained from the senses, one’s education, and logical reasoning:

Becoming the God of “I-AM-Now-ness” and filling your entire experiential field with that flavor of awakening is a recipe for ecstasy, not for good epistemology.

Indeed, the patient may become a bit hooked to the simplification of their model complexity; to make reality as they know be replaced by a simpler, yet more intense, version of perceived reality is very tempting. It can be seductive to embrace a view like “you are God and you have created everything for your own amusement” or “you are the dream of God”. Rather than compassion, why not indifference? Being the “way God entertains itself” is both poetically satisfying and super trippy. A lot of people would find that such belief adds some spice to their lives. Overfitting-and-underfitting-effect-on-error

But the price of truth is everything. In turn, it would be ideal to complement any model complexity reduction that goes too far with a healthy amount of prediction errors.

5. Fixed-Points and Limit Cycles vs. Chaotic Attractors

The brain contains many self-correcting feedback systems. Psychedelics in general can be modeled as drugs that mess with the inhibitory component of excitation-inhibition feedback systems. They accomplish this, quite possibly, by disrupting the inhibitory serotonergic connections coming from the cortex that gate the excitatory input coming from the thalamus. This may account for why tracers look the way they do – the failure to inhibit the thalamus results in looped replays of recent states. This may go a long way in explaining why people find “video feedback” so trippy and fascinating. Namely, because a lot of psychedelic effects can be understood as feedback getting out of control, literally.

oscillation_1_5_5_75_75_1_10_0-05_signal_trailing

More concretely, at the Harvard talk on the Hyperbolic Geometry of DMT experiences I introduced this notion that each DMT experience as a whole can be thought of as a trajectory in the energy vs. complexity landscape. Here, the vertical axis indicates the degree of energy of the experience (roughly corresponding to the intensity, brightness, and amount of qualia), while the horizontal axis represents how much information is encoded in the experience. One interesting operationalization of information is through the concept of symmetry breaking***, in which case the horizontal axis approximately tracks the “distance from perfectly symmetrical spaces in terms of number of symmetry breaking operations”.

I then postulated that we could generate an ontology based on feedback + noise to explain how two DMT trips of roughly the same level of intensity can nonetheless contain very different amounts of information.

Hyperbolic Geometry of DMT Experiences copy 50

What I did not mention at the time was that while DMT does have the ability to send you to any place in the energy-complexity landscape, most of the experiences are in the region of middle-complexity. In fact, especially strong DMT trips tend to become fully chaotic, so there is even a mild correlation between dosage and complexity. On the other hand, 5-MeO-DMT tends to propel you towards the low-complexity region of the space.

Using a feedback system ontology we could thus posit that 5-MeO-DMT modifies the feedback parameters of your brain in such a way that the states it gives rise to are ether fixed points or short limit cycles:

The video above depicts a fixed point (as with the other animations below, I took this from an old 1984 video about video feedback you can see here). A fixed point is a configuration of the system that is stable upon feedback iterations. In the video above we see a fixed point consisting of a cross (presumably the result of the camera having a 90 degree tilt) that is then perturbed and eventually collapses into just a single circular dot at the center.

The above are limit cycles. The first (left) is a comparatively simple limit cycle in that every stage along its reproduction cycle is very similar to each other. The second (right) one is a bit more complex, yet despite a long winded path, it really does repeat more or less perfectly over and over. 5-MeO-DMT limit cycles are more akin to the one on the left, but on occasion may be a bit more complex and rhythmic over the span of seconds. Either way, there is often a strong pull towards a simple resonant pattern with remarkable stability.

Contrast that with chaotic attractors:

tumblr_p0bmdl9R0x1shq9dbo1_500

Chaotic pseudo-time arrow

When we have actual chaos, the states continue to change indefinitely. We could then posit that DMT’s characteristic jittering seems to prevent the formation of stable fixed points or short limit cycles. Despite both DMT and 5-MeO-DMT disrupting feedback in the brain, the subtle differences in the way parameters of this feedback are modified can make all the difference between perfectly simple and smooth results and the endless generation of chaotic structures.

6. Modulating Lateral Inhibition

Another exciting lens with which to look at the difference between these psychedelics is by allusion to lateral inhibition: according to a couple of recent trip reports I received from another anonymous source, there is a remarkable difference between the tracers of 5-MeO-DMT and those of DMT. In particular, the anonymous tripper points out that DMT tracers are chains of concatenated positive and negative after-images of the stimulation source, whereas 5-MeO-DMT only produces positive after-images.

afterimage (1)

Stare at the cross. The green ball you see as a result is the “negative after-image” of the missing purple dot. 

In other words, when you see a blue ball moving on a screen, on DMT you will see tracers of that ball that change in color from blue to yellow to blue again and so on, all following after the original blue ball. But on 5-MeO-DMT, one will only see a long blue tracer. This is a remarkable difference, and if true, it would seem to be an important hint.****

13408_2015_22_Fig1_HTML

Orientation selectivity map in the visual cortex

Lateral inhibition in the cortex prevents the overlapping of incompatible features in one’s own experience. For example, the primary visual cortex shows a map of orientation selectivity as shown above. Each hypercolumn is selective to only a specific orientation, and the surrounding hypercolumns are selective to different orientations. More so, via lateral inhibition, when a hypercolumn is activated, it inhibits the surrounding ones.

In this model, DMT would perhaps leave this process untouched, so that when the brain is massively energized, it still maintains this flip-flopping between each feature and its complement. Indeed, some people have described DMT as turning half of your neurons fully on, the other half of your neurons fully off, then inverting the activation so that the off neurons are turned on and the on ones are turned off, and flip-flopping between these two steps many times per second. Could this be the phenomenal expression of having energized checkerboard patterns of excitation and inhibition saturating your cortex? That is, taking a system with in-built lateral inhibition into over-drive?

5-MeO-DMT, on the contrary, seems to allow for “all colors to blend together into pure white light”, and “the past and the future to collapse into the present”, and the “self-other distinction to be dissolved”, and so on. Intuitively, if the drug is biochemically disabling lateral inhibition, that could be reflected as a profound sense of unity and interconnectedness at the phenomenological level, “transcending every last barrier”.

7. Diffuse Attention vs. Focused Attention

As we mentioned earlier, DMT tends to come with a feeling of “becoming tense” while 5-MeO-DMT has a primary vibe of “becoming relaxed”. One way in which we can model this is in the way the drugs influence whether your attention is focused or diffuse. Now, I should say that this model will be necessarily incomplete because at the peak of a 5-MeO experience one often does in fact feel super focused in some way. But I would posit that this sense of focus is much more holistic than the way our common-sense use of the term would suggest. The focus on DMT, on the other hand, does feel very much akin to the “conventional” sense of focus, where you are able to precisely position figure and ground in such a way that they have as much contrast as possible.dmt_from_scalable_locally_injective_maps

In a way, the beings one encounters on DMT could be thought of as “attentional attractors”. As you create a lot of little focal points during the experience, these begin to build up and define the contents of your mind. Each focal point makes it easier for you to create another one nearby. This snowballs into an effect where there are clusters of focal points that become the “centers of mass” of the narrative. These could very well be what underlies the “beings” on DMT. Using Buddhist terms again, DMT beings might perhaps be thought of as exotic “nimittas“: attention hubs.

Also, because the experience is high-dimensional and changes your sense of what “understanding” even means, it seems that the feeling of super-intelligence on DMT might be a projection of one’s own super-intelligence (of a certain kind) in the state.

In contrast, 5-MeO-DMT makes it easy to de-focus on anything. To let go, and experience it in a diffuse way.

8. Big Chunks and Tiny Chunks vs. A Power Law of Chunks

Geology uses the word matrix in way that is very different from either math or science fiction aficionados. In geology, a matrix is the entire mass of materials on a rock within which crystals, grains, and clasts are found. As Anders and Maggie discussed recently, the way in which minerals form depends to a large extent on the presence of water in the process of crystallization. The huge diversity of minerals we see on Earth’s surface is partly a result of the availability of water in the mantle. Perhaps a lens with which to see DMT is as playing the same role in the brain as water does in fractional crystallization. It lubricates the matrix of your mind, which enables the crystallization of countless qualia exotica. 5-MeO-DMT instead homogenizes the kinds of crystals that can form.

In brief, DMT is to 5-MeO-DMT as a matrix of diverse minerals is to a mono-phasic large enclave. DMT is like complex music (cf. music as an ordered phase of sound) while 5-MeO-DMT is like a single mantra repeated over and over. Is this metaphor useful? It seemed to resonate with the interviewee.

9. Integration vs. Fragmentation

In Neural Annealing, Mike Johnson argues that what makes MDMA special for healing trauma is what at QRI we call integration:

On MDMA’s strangely powerful therapeutic effects, I’d suggest MDMA shares the ‘basic psychedelic package’ with substances like LSD and psilocybin (albeit a little weaker at common doses). Anything with this ‘baseline’ package significantly increases the energy parameter of the brain, which both allows escape from bad local minima and canalizes the brain’s core CSHWs, which both should be highly therapeutic. My intuition is MDMA may also have a particular effect on stochastic firing frequencies of neurons, and that this effect essentially acts as an emergent metronome – and this metronome will drive synchronicity between diverse brain regions. Given the presence of such a region-spanning ‘clean’ metronomic signal, brain regions that have partially ‘stopped talking to each other’ will re-establish integration, and some of this integration will persist while sober (or rather, some of the reasons for the lack of integration will have been negotiated away during the MDMA-driven integration). Plausibly this ‘emergent metronome’ effect may also underlie the particular phenomenological effects of 5-MeO-DMT as well, particularly in terms of sense of unity, high valence, and therapeutic potential. (HT to Steve Lehar for pointing at this ’nystagmus’ phenomenon as being somehow linked to MDMA’s mood-lifting effect, and to Andrés for calling my attention to Lehar’s work and suggesting 5-MeO-DMT may also share this mechanism.)

Like most other psychedelics, N,N-DMT also shares the same ‘basic psychedelic package’ and can have beneficial therapeutic effects. But it lacks this ‘special’ ability that allows arbitrary parts of your nervous system to rhythmically entrain with one another. This “emergent metronome” on MDMA and 5-MeO-DMT works as a kind of universal “vibratory currency” and results in a reduction of inner dissonance to a much greater extent than (relatively) simple “energizers” like DMT.

To Wrap Up

We hope that the above discussion has given you an idea about the difference between DMT and 5-MeO-DMT and why this matters for their therapeutic potential. The above is just the start of a deep inquiry into the topic that will certainly take many years, but we believe that it is a novel way of seeing the contrast between these two substances that may be generative for others. We also believe that it is very worth trying: nailing down this difference may be incredibly important to develop novel ways of treating mental illness. While DMT will undoubtedly continue to dazzle and amaze people curious about the state-space of consciousness, the superlative potential of 5-MeO-DMT to heal trauma puts it on a different level of importance altogether.

In the future we shall also explain why MDMA and ketamine have this potential. And ultimately, as we begin to understand what makes these substances so special, we hope to find ways of creating effective therapies from first principles. Stay tuned.



* The toad venom dose was 50mg, with an estimated 5-7mg of 5-MeO-DMT. This is admittedly likely to produce somewhat more potent effects than just 5 to 7mg of pure 5-MeO-DMT. But the extent of this enhancement is currently poorly understood, and you can find many people online saying that the difference is tiny and others who argue it is enormous. Given just how intense and qualitatively unique pure 5-MeO-DMT already is, I think that applying Occam’s razor would tell us that “it’s just the 5-MeO-DMT itself”. So while I am ready to accept the possibility of profound synergy between other tryptamines in toad venom and 5-MeO, I am not holding my breath for it. I, rather, expect that the difference between 5-MeO alone and the full-spectrum stuff will be akin to the difference between drinking 10 shots of vodka and drinking 10 shots of vodka and one chamomile tea. Namely, a real but largely inconsequential difference.

** This, of course, blends well with the Symmetry Theory of Valence, a subject to which we shall return in the near future.

*** This is where information-less states are those which are perfectly symmetrical, and the information content of a state is defined as the minimum number of symmetry breaking operations needed to transform an information-less state into it.

**** This is admittedly very weak evidence so far. If you have experience with both of these compounds and have explored the way in which they give rise to after-images, please let me know if you can confirm or deny the effect here mentioned.


Picture: Andrés & Claudia Silva Ruiz

Breaking Down the Problem of Consciousness

Below you will find three different breakdowns for what a scientific theory of consciousness must be able to account for, formulated in slightly different ways.

First, David Pearce posits these four fundamental questions (the simplicity of this breakdown comes with the advantage that it might be the easiest to remember):

  1. The existence of consciousness
  2. The causal and computational properties of experience (including why we can even talk about consciousness to begin with, why consciousness evolved in animals, etc.)
  3. The nature and interrelationship between all the qualia varieties and values (why does scent exist? and in exactly what way is it related to color qualia?)
  4. The binding problem (why are we not “mind dust” if we are made of atoms)
david_pearce_criteria_for_scientific_theory_of_consciousness

David Pearce’s Four Questions Any Scientific Theory of Consciousness Must Be Able to Answer


Second, we have Giulio Tononi‘s IIT:

  1. The existence of consciousness
  2. The composition of consciousness (colors, shapes, etc.)
  3. Its information content (the fact each experience is “distinct”)
  4. The unity of consciousness (why does seeing the color blue does not only change a part of your visual field, but in some sense it changes the experience as a whole?)
  5. The borders of experience (also called ‘exclusion principle’; that each experience excludes everything not in it; presence of x implies negation of ~x)
Axioms_and_postulates_of_integrated_information_theory

Giulio Tononi’s 5 Axioms of Consciousness


Finally, Michael Johnson breaks it down in terms of what he sees as a set of what ultimately are tractable problems. As a whole the problem of consciousness may be conceptually daunting and scientifically puzzling, but this framework seeks to paint a picture of what a solution should look like. These are:

  1. Reality mapping problem (what is the formal ontology that can map reality to consciousness?)
  2. Substrate problem (in such an ontology, which objects and processes contribute to consciousness?)
  3. Boundary problem (akin to the binding problem, but reformulated to be agnostic about an atomistic ontology of systems)
  4. Scale problem (how to connect the scale of our physical ontology with the spatio-temporal scale at which experiences happen?)
  5. Topology of information problem (how do we translate the physical information inside the boundary into the adequate mathematical object used in our formalism?)
  6. State-space problem (what mathematical features does each qualia variety, value, and binding architecture correspond to?)
  7. Translation problem (starting with the mathematical object corresponding to a specific experience within the correct formalism, how do you derive the phenomenal character of the experience?)
  8. Vocabulary problem (how can we improve language to talk directly about natural kinds?)
Eight-Problems2

Michael Johnson’s 8 Problems of Consciousness

Each of these different breakdowns have advantages and disadvantages. But I think that they are all very helpful and capable of improving the way we understand consciousness. While pondering about the “hard problem of consciousness” can lead to fascinating and strange psychological effects (much akin to asking the question “why is there something rather than nothing?”), addressing the problem space at a finer level of granularity almost always delivers better results. In other words, posing the “hard problem” is less useful than decomposing the question into actually addressable problems. The overall point being that by doing so one is in some sense actually trying to understand rather than merely restating one’s own confusion.

Do you know of any other such breakdown of the problem space?


27983216_842901315900851_2411647783839521476_o

Is the Orthogonality Thesis Defensible if We Assume Both Valence Realism and Open Individualism?

Ari Astra asks: Is the Orthogonality Thesis Defensible if We Assume Both “Valence Realism” and Open Individualism?


Ari’s own response: I suppose it’s contingent on whether or not digital zombies are capable of general intelligence, which is an open question. However, phenomenally bound subjective world simulations seem like an uncharacteristic extravagance on the part of evolution if non-sphexish p-zombie general intelligence is possible.

Of course, it may be possible, but just not reachable through Darwinian selection. But the fact that a search process as huge as evolution couldn’t find it and instead developed profoundly sophisticated phenomenally bound subjectivity is (possibly strong) evidence against the proposition that zombie AGI is possible (or likely to be stumbled on by accident).

If we do need phenomenally bound subjectivity for non-sphexish intelligence and minds ultimately care about qualia valence – and confusedly think that they care about other things only when they’re below a certain intelligence (or thoughtfulness) level – then it seems to follow that smarter than human AGIs will converge on valence optimization.

If OI is also true, then smarter than human AGIs will likely converge on this as well – since it’s within the reach of smart humans – and this will plausibly lead to AGIs adopting sentience in general as their target for valence optimization.

Friendliness may be built into the nature of all sufficiently smart and thoughtful general intelligence.

If we’re not drug-naive and we’ve conducted the phenomenological experiment of chemically blowing open the reducing valves that keep “mind at large” out and that filteratively shape hominid consciousness, we know by direct acquaintance that it’s possible to hack our way to more expansive awareness.

We shouldn’t discount the possibility that AGI will do the same simply because the idea is weirdly genre bending. Whatever narrow experience of “self” AGI starts with in the beginning, it may quickly expand out of.


Michael E. Johnson‘s response: The orthogonality thesis seems sound from ‘far mode’ but always breaks down in ‘near mode’. One way it breaks down is in implementation: the way you build an AGI system will definitely influence what it tends to ‘want’. Orthogonality is a leaky abstraction in this case.

Another way it breaks down is that the nature and structure of the universe instantiates various Schelling points. As you note, if Valence Realism is true, then there exists a pretty big Schelling point around optimizing that. Any arbitrary AGI would be much more likely to optimize for (and coordinate around) optimizing for positive qualia than, say, paperclips. I think this may be what your question gets at.

Coordination is also a huge question. You may have read this already, but worth pointing to: A new theory of Open Individualism.

To collect some threads- I’d suggest that much of the future will be determined by the coordination capacity and game-theoretical equilibriums between (1) different theories of identity, and (2) different metaphysics.

What does ‘metaphysics’ mean here? I use ‘metaphysics’ as shorthand for ‘the ontology people believe is ‘real’. What they believe we should look at when determining moral action.’

The cleanest typology for metaphysics I can offer is: some theories focus on computations as the thing that’s ‘real’, the thing that ethically matters – we should pay attention to what the *bits* are doing. Others focus on physical states – we should pay attention to what the *atoms* are doing. I’m on team atoms, as I note here: Against Functionalism.

My suggested takeaway: an open individualist who assumes computationalism is true (team bits) will have a hard time coordinating with an open individualist who assumes physicalism is true (team atoms) — they’re essentially running incompatible versions of OI and will compete for resources. As a first approximation, instead of three theories of personal identity – Closed Individualism, Empty Individualism, Open Individualism – we’d have six. CI-bits, CI-atoms, EI-bits, EI-atoms, OI-bits, OI-atoms. Whether the future is positive will be substantially determined by how widely and deeply we can build positive-sum moral trades between these six frames.

Maybe there’s further structure, if we add the dimension of ‘yes/no’ on Valence Realism. But maybe not– my intuition is that ‘team bits’ trends toward not being valence realists, whereas ‘team atoms’ tends toward it. So we’d still have these core six.

(I believe OI-atoms or EI-atoms is the ‘most true’ theory of personal identity, and that upon reflection and under consistency constraints agents will converge to these theories at the limit, but I expect all six theories to be well-represented by various agents and pseudo-agents in our current and foreseeable technological society.)

Quantifying Bliss: Talk Summary

Below I provide a summary of the Quantifying Bliss talk at Consciousness Hacking (video360 degree live feed record), which took place on June 7th 2017. I am currently working on a longer and more precise treatment of the topic, which I will be posting here as well. That said, since the talk already makes clear, empirically testable predictions, I decided to publish this summary as soon as possible. After all, there is only a small window of opportunity to publish one’s testable predictions online before the experiment is run and they turn into “retrodictions”. By writing this out and archiving it on time I’m enabling future-me to say “called it!” (if the results are positive) or “at least I tried” (if the experiment fails to show the predicted effects). Better do this quick, then, for science!

The Purpose of Life

We begin by asking the question “what is the purpose of life?”. In order to give a sense for where I am coming from, I explain that I think that the purpose of life is…

  1. To Understand the Universe
  2. To be Happy, and Make Others Happy

I admit that for the first half of my life I thought that the only purpose of life was to understand the universe. If anything, in light of this exclusive goal, happiness could be seen as a temporary distraction rather than something to pursue for its own sake. Thankfully, as a teenager I was exposed to philosophy of mind, was introduced to meditation, and experimented with psychedelics, all of which pointed me to the fact that (a) we don’t understand consciousness yet, and (b) happiness is really a lot more important than we usually think, even if one is only concerned with the most theoretical and abstract level of understanding possible.

I now regard “to understand the universe” and “to be happy and make others happy” on an equal footing. More so, these two life goals complement each other. On the one hand, understanding the universe will allow you to figure out how to make anyone happy. And on the other hand, being happy and making others happy can allow you to stay motivated in order to figure out the nature of reality. Hence one can think of these two life goals as synergistic rather than as being in opposing camps (of course, at the edges, one will be forced to choose one over the other, but we are nowhere near the point where this is a concern).

By taking these two “purposes of life” seriously we are then faced with a crucial question: What makes an experience valuable? In other words, for someone who is both trying to understand the universe and trying to make its inhabitants as happy as possible, the question “how do you measure the value of an experience?” becomes important.

At Qualia Computing we generally answer that question using the following criteria

  1. Does it feel good? (happy, loving, pleasant)
  2. Does it make you productive (in a good way)?
  3. Does it make you ethical?

That is to say, the value that we assign to an experience is guided by three criteria. In brief, a valuable experience is one that feels good (i.e. has positive hedonic tone), improves your productivity (in the sense of helping you pursue your own values effectively), and makes you more ethical – both towards yourself and others. That said, for the purpose of this talk, I make it explicit that I will only discuss how to measure (1). In other words, we will concern ourselves with what makes an experience feel good; ethics and productivity are discussed elsewhere.*

What is Bliss?

So what makes an experience feel good? The “feel good” quality of an experience is usually called valence in psychology and neuroscience (also described as the “pleasure-pain axis”). This quality is to be distinguished from arousal, which refers to the amount of energy expressed in an experience. Four examples: Excitement is a high-valence, high-arousal state. Serenity is a high-valence, low-arousal state. Anxiety is low-valence, high-arousal. And depression low-valence, low-arousal.

For some people valence and arousal are correlated (either negatively or positively as shown by Peter Kuppens). Likewise, one’s culture can have a large influence on the way one conceptualizes of valence (or ideal affect, as demonstrated in the extensive work of Jeanne Tsai). That said, valence is not a cultural phenomenon; even mice can experience negative and positive valence.

Even though valence and arousal do seem to explain a big chunk of the differences between emotions, we can nonetheless find many cases where the “texture” of two emotions feel very different even though their valence and their arousal are similar. Hence we ask ourselves: How do we explain and characterize the textural differences between such emotions?

And across all of the possible intensely blissful states on offer (encompassing all of the possible inner meanings present), what exactly is shared between them all at their very core?

Some interpret holistic feelings of wellbeing as a sort of spiritual signal. In this interpretation, feeling at a very deep level that the world is good, that things fall into place perfectly, that you don’t owe anything to anyone, etc. is a sign that you are on the right (spiritual) track. Undoubtedly many people use the (often extreme) positive shift in their valence upon religious conversion as evidence of the validity of their choice. Intense positive valence may not throw Bayesian purists off-balance, but for the rest of the world, blissful experiences are often found as cornerstones of worldviews.

Other people say that bliss is “just chemicals in your brain”. Some claim that it’s more a matter of the functional state of your pleasure centers (themselves affected by dopamine, opioids, etc.) rather than the chemicals themselves. Many others are focused on what usually triggers happiness (e.g. learning, relationships, beliefs, etc.) rather than on what, absolutely, needs to happens for bliss to take place in the simplest experiential terms possible. Most who study this closely become mystics.

Could it be that there’s something structural that makes the experiences feel good? Let’s say that there exists a good-fitting mathematical object that translates brain states to experiences. What mathematical property of that object would valence look like? Our proposal is very simple. In some sense it is the simplest possible theory for the important theory of consciousness. We propose the symmetry theory of valence.

(The important theory of consciousness is the question that asks why experience feels good and/or bad, vs. e.g. the hard problem of consciousness, why consciousness exists to begin with).

The Symmetry Theory of Valence

We are pretty confident that consciousness is a real and a measurable phenomenon. That’s why Consciousness Hacking is such a good venue for this kind of discussion. Because here we can talk freely about the properties of consciousness without getting caught up about whether it exists at all. Now, symmetry is a very general term, how is that precise?

Harmony feels good because it’s symmetry over time. In reality, our moments of experience contain a temporal direction. I call this a pseudo-time arrow, since its direction is likely encoded in the patterns of statistical independence between the qualia experienced. And by manipulating the symmetrical connectivity of the micro-structure of one’s consciousness, one can change the perception of time. It’s a change in the way one evaluates when one is and how fast one is going. 

In this model, the pleasure centers would work as “tuning knobs” of harmonic patterns. They are establishing the mood, the underlying tone to which the rest needs to adapt. And the emotional centers, including the amygdala, would be strategically positioned to add anti-symmetry instead. Hence, in this framework we would think of boredom is an “anti-symmetry” mechanism. It prevents us from getting stuck in shallow ponds, but it can be nasty if left unchecked. Cognitive activity may be in part explained by differences in boredom thresholds.

Connectome-Specific Harmonics

I was at the Psychedelic Science 2017 conference when I saw Selen Atasoy presenting about improvisation enhancement with LSD. She used a paradigm previously developed, whose methods and empirical tests were published in Nature in 2016 but now applied to psychedelic research. For a good introduction check out the partial transcript of her talk.

In her talk she shows how one can measure the various “pure harmonics” in a given brain. The core idea is that brain activity can be interpreted as a weighted sum of “natural resonant frequencies” for the entire connectome (white matter tracks together with the grey matter connections). They actually take the physical structure of a mapped brain and simulate the effect of applying the excitation-inhibition differential equations known for collective neural activity propagation. Then they infer the presence and prevalence of these “pure harmonics” in a brain at a given point in time using a probabilistic reconstruction.

Chladni plates here are a wonderful metaphor for these brain harmonics. This is because the way the excitation-inhibition wavefront propagates is very similar in both Chladni plates and human brains. In both cases the system drifts slowly within the attractor basin of natural frequencies, where the wavefront wraps around the medium an integer number of times. I was in awe to see her approach applied to psychedelic research. After all, Qualia Computing has indeed explored harmonic patterns in psychedelic experiences (ex. 1, ex. 2, ex. 3), and the connection was made explicit in Principia Qualia (via the concept of neuroacoustic modulation).

giphy-downsized-large

But how do these harmonics look like in the brain? Show me a brain!

 

Notice the traveling wave wrapping around the brain an integer number of times in each of these numerical solutions (source). The work by these labs is incredible, and they seem to show that the brain’s activity can be decomposed into each of these harmonics.

At the Psychedelic Science 2017 conference, Selen Atasoy explained that very low frequency harmonics were associated with Ego Dissolution in the trials that they studied. She also explained that emotional arousal, here defined as one’s overall level of energy in the emotional component (i.e. anxiety and ecstasy vs. depression and serenity), also correlated with low frequency harmonic states. On the other hand, high valence states were correlated with high frequency brain harmonics.

These empirical results are things that I claim we could have predicted with the symmetry theory of valence. I then thought to myself: let’s try to come up with other predictions. How should we consider the mixture of various harmonics, beyond merely their individual presence? How can we reconstruct valence from this novel data-structure for representing brain-states?

The Algorithm for Quantifying Bliss

Starting my reasoning from first principles (sourced from the Symmetry Theory of Valence), the natural way to take a data-structure that represents states of consciousness and recover its valence (in cases where samples occur across time in addition to space), is to try to isolate the noise, then proceed to quantify the dissonance, and what remains becomes what’s consonant. Basically, one will estimate the rough amount of symmetry (over time), as well as the degree of anti-symmetry, and the level of noise total.

In other words, I prophesize that we can get an “affective signature” of any brain state by applying an algorithm to fMRI brain recordings in order to estimate the degree of (1) consonance, (2) dissonance, and (3) noise within and across the brain’s natural harmonic states. This will result in what I call “Consonance-Dissonance-Noise Signatures” of brain states (“CDNS” for short) consisting of three histograms that describe the spectra of consonance, dissonance, and noise in a given moment of experience. The algorithm to arrive at a CDNS of a brain state is as follows:

Remove some of the noise in the brain state by applying the technique in Atasoy (2016) and recovering the distribution of the best approximation possible for the harmonics present (you may apply some further denoising on the harmonics when taken as a collective). Then estimate the total dissonance of the combination of harmonics by taking each pair of harmonics and quantifying their mutual dissonance. Finally, subtract the dissonance from “all of the interactions that could have existed” and what’s left ends up being the consonance. This way you obtain a Consonance, Dissonance, Noise Signature.

 

Each of these three components will have their associated spectral power distribution. The noise spectrum is obtained during the first denoising step (as whatever cannot be explained by the harmonic decomposition). Then the dissonance spectrum is a function of the minimum power of pairs of harmonics that exist within the critical band of each other (see slides 18; possibly upgraded by 20), as well as the frequencies of the beating patterns.

Quantifying Dissonance?

In order to quantify dissonance we use a method that may end up being simpler than what you need to calculate dissonance for sound! E.g. in Quantifying the Consonance of Complex Tones With Missing Fundamentals (Chon 2008) we learn that the human auditory system may at times detect dissonance even when there is no actual dissonance in the input. That is, there are auditory illusions pertaining to valence and dissonance. Based on the missing fundamental one can create ghost dissonance between tones that are not even present. That said, quantifying dissonance in a brain in terms of its harmonic decomposition may be easier than quantifying dissonance in auditory input, precisely because the auditory input (and any sensory input for that matter) contains many intermediary pre-processing steps. The auditory system is relatively “direct” when compared to, e.g. the visual system, but you will still see some basic signal processing done to the input before it influences brain harmonics. The sensory systems, being adapted to meet the criteria of both interfacing with a functioning valence system and representing the information adequately (in terms of the real-world distribution of inputs) serve the function of translating the inputs into usable signals. I.e. frequency-based descriptions, often log-transformed, in order to arrive at valence gradients. For this reason, the algorithm that describes how to extract valence out of a brain state may turn out to be simpler than what you need to predict the hedonic quality of patterns of sound (or sight, touch, etc).

In brief, we propose that we can compute the approximate amount of dissonance between these harmonics by seeing how close they are in terms of spatial and temporal frequencies. If they are within the critical window then they will be considered as dissonant. There is likely to be a peak dissonance window, and when any pair of harmonic states live within that window, then experiencing both at once may feel really awful (to quantify such dissonance more precisely we would use a dissonance function as shown in Chon 2008). If indeed symmetry is intimately connected to valence, then highly anti-symmetrical states such as what’s produced by overlapping brain harmonics within the critical band may feel terrible. Remember, harmony is symmetry over time. So dissonance is anti-symmetry over time. It’s worth recalling, though, that in the absence of dissonance and noise, by default, what remains is consonance.

Visualizing Emotions as CDNS’s of States of Consciousness

Above you can find two ways of visualizing a CDNS. Before we go on to the predictions, here we illustrate how we think that we will be able to see at a glance the valence of a brain with our method. The big circle shows the dissonance and consonance for each of the brain harmonics (the black dots surrounding the circle represent the weights for each state). If you want the overall dissonance in a given state, you add up the red-yellow arrows, whereas if you want the total consonance, you add the purple-light-blue arrows. The triangles on the right expand upon the valence diagram presented in Principia Qualia. Namely, we have a blue (positive valence/consonant), red (negative valence/dissonant), and grey (neutral valence/noise) component in a state of consciousness. Each of these components has a spectrum; the myriad textures of emotional states are the result of different spectral signatures for hedonically loaded patterns.

Testable Predictions

Quantifying Bliss (27)

We predict that intense emotions/experiences reported on psychedelics will result in states of consciousness whose harmonic decomposition will show a high amount of energy to be found in the pure harmonics (this was already found in 2017 as explained in the presentation, so let’s count that as a retrodiction). People who report being “very high” will have particularly high amounts of energy in their pure harmonics (as opposed to more noisy states).

The predicted valence for their experiences will be a function of the particular patterns (in terms of relative weights) of the various harmonics. Those which generate highly harmonic CDNS will be blessed with high valence experiences. And those who experience high dissonance, as empirically measured, will report negative feelings (e.g. fear, anxiety, nausea, weird and unpleasant body load, etc). In particular, we can explore the shape of highly harmonic states. In this framework, MDMA would be seen as likely to work by increasing the energy expressed by an exceptionally consonant set of harmonics in the brain.

A point to make here is that predicting “pure harmonics” on psychedelics (evidently simple and ordered patterns), would seem to go counter to the recently accrued empirical data concerning entropy in the tripping brain.** But we also know that the psychedelic brain can produce ridiculously self-similar near-informationless yet highly intense moments of experience preceded by a symmetrification process. Indeed, there are several symmetric attractors for the interplay of awareness and attention at various levels of “consciousness energy” and quality of mood. These states, in turn, not only are hedonically charged, but also allow the exploration of high-energy qualia research (since the implicit symmetry provides an energy seal). Highly energetic states of consciousness can be encapsulated in a highly symmetrical network of local binding. More about this in a future article.

On the other hand, we predict that people on SSRIs will show an enhanced amount of noise in their CDNS. A couple of slides back, this was represented as a higher loading of activity in the grey component of the triangular visualization of a CDNS. Likewise, some drugs will have various effects on the CDNS, such as stimulants inducing more consonance in high frequencies, whereas opioids and hypnotics having signatures of inducing high consonance in the low frequencies.

Summary of Predictions About Drug Effects

  1. Psychedelic substances will increase the overall power of the brain’s pure harmonics, and thus result in a CDN Signature characterized by: (a) high consonance of all frequencies, (b) high dissonance of all frequencies, and (c) low noise of all frequencies. Criticality will be observed by way of the CDNS having high variance.
  2. MDMA will produce a very specific range of states that have on the one hand very pure harmonic states of high frequencies, and on the other, very small collective dissonance and noise. In other words: (a) high amounts of high-frequency consonance, (b) low amounts of dissonance of all frequencies, and (c) low noise of all frequencies.
  3. Any “affect blunting” agent such as SSRIs, ibuprofen, aspirin, acetaminophen, and agmatine, will produce CDNS characterized by: (a) reduced consonance of all frequencies, (b) reduced dissonance of all frequencies, and (c) increased noise in either some or all frequencies. We further hypothesize that different antidepressants (e.g. citalopram vs. fuoxetine) will look the same with respect to reducing the C and D components, but may have differences in the way they increase the N spectrum.
  4. Opioids in euphoric doses will be found to (a) increase low frequency consonance, (b) decrease dissonance for all frequencies but especially the high frequencies, and (c) slightly increase noise across the board.
  5. Stimulants will be found to (a) increase medium and high frequency consonance, (b) leave dissonance fairly unaltered, and (c) reduce noise for all frequencies but especially those in the upper end of the spectrum.

Predictions About Emotions

For now, here are the specific predictions concerning emotions that I am making:

  1. The energy of the consonant (C) component of a CDNS will be highly correlated with the amount of euphoria (pleasure, happiness, positive feelings, etc.) a person is experiencing.
  2. The energy of the dissonant (D) component will have a high correlation with the amount of dysphoria (pain, suffering, negative feelings, etc.) a person feels.
  3. The energy of the noise (N) component will be correlated with flattened affect and blunted valence (i.e. feeling neither good nor bad, like there is a fog that masks all feelings).
  4. If one creates a geometric representation of the relationships between various brain states using their respective CDNS similarities as a distance metric for emotional states using Multi-Dimensional Scaling (MDS) techniques, one will be able to recover a really good approximation of the empirically-derived dimensional models of emotions (cf. dimensional models of emotionWire-heading Done Right). In other words, if you ask your participants to tell you how they feel during the fMRI sessions and then associate those emotions to their instantaneous CDNS, and then you apply multidimensional scaling to the resulting CDNS, you will be able to recover a good dimensional picture of the state-space of emotions. I.e. “subjective similarity between emotions” will be closely tracked by the geometric distance between their corresponding CDNS:
    1. Applying MDS scaling to the C component of the CDNS will result in a better characterization of the differences between positive emotions.
    2. Applying MDS to the D component will result in a better characterization of the differences between negative emotions. And,
    3. Applying MDS to the N component will result in a better characterization of the differences between valence-neutral emotions.

The Future of Mental Health

Quantifying Bliss (28)

Sir, your 17th harmonic is really messing up the consonance of your 19th harmonic, and it interrupts the creative morning mood you recently enjoyed. I suggest taking 1mg of Coluracetam, listening to a selection of Diamond songs, and RD23 [stretching exercise]. Here’s your expected CDNS.

Penfield mood organs may not be as terrible as they seem. At least not if you’re given a good combination of personalized settings, and a manual to wire-head in the proper manner. In such a situation, among the options available, you will have the ability to choose an experientially attractive, healthy, and sustainable set of moods indefinitely.

The “clinical phenomenologist” of the year 2050 might look into your brain harmonics, and try to find the shortest paths to nearby state-spaces with less chronic dissonance, fishing for high-consonance attractors with large basins to shoot for. The qualia expert would go on to provide you various options that may improve all sorts of metrics, including valence, the most important of them all. If you ask, your phenomenologist can give you trials for fully reversible treatments. You sample them in your own time, of course, and test them for a day or two before deciding whether to use these moods for longer.

Personalized Harmonic Retuning

I assume that people will be given just about enough retuning to get back to their daily routines as they themselves prefer them, but without any sort of nagging dissonance. Most people will probably continue on with their preference architectures relatively unchanged. Indeed, that will be a valued quality for a personalized harmonic retuning product. Having adequate mood devices that don’t mess up your existing value system might eventually become a highly understood, precision-engineered aspect of mainstream mental health. At least compared to the current (pre-psychedelic re-adoption 2017) paradigms. Arguably, even psychedelic therapy is pretty blunt in a way. Not in the sense of blunting the hedonic quality of your experience (on the contrary). But in the sense of applying the harmonization process indiscriminately.

For the psychonauts (hopefully they are not too rare by then), who still want to investigate consciousness even though human life is already full of love (in the future), we will have a different arrangement. They are free to explore themselves while being part of a research institute. Indeed, pursuing the purpose of understanding the big picture (including consciousness) will require the experimental method. More so, exploring the state-space of consciousness will, for the foreseeable future, be a way to find new ways of making others happy. People will continue to explore alien state-spaces in the search of highly-priced high-valence states. At least for some scores of generations valence engineering is bound to continue to be economically profitable. As we discover new drugs, new treatments, new philosophical trances, new interpretations and expressions of love, and so on, the economy will adapt to these inventions. We already live in an informational economy of states of consciousness, and the future is likely to be like that as well. Except that consciousness technologies will be immensely more powerful.

Barring the unlikely emergence of anti-hedonist Spartan self-punishing transhumanist social movements enabled with genetic technology, I don’t anticipate major obstacles in the eventual widespread use of mood organs. In fact, the wide adoption of SSRIs in some pockets of society shows that the general public is willing and interested in minor self-adjustments to deal with chronic negativity. Hedonic technology is in its early days, but with a root understanding of the nature of valence, the sky is the limit.

Case studies – SSRIs & Psychedelics

Let’s take a closer look at SSRIs and psychedelics in light of the Symmetry Theory of Valence.

SSRIs have an overall effect of blunting one’s experience at pretty much every level imaginable. Usually just a little, enough to help people re-establish a new order between their harmonics, in a more noisy, less intense range of moods. Some people may benefit from this sort of intervention. Now, also it’s worth pointing out the possible side effects, which have the common theme of reducing the structural integrity of the micro-structure of consciousness. Thus, the highly ordered pleasant and unpleasant experiences get softened. Whether this generalized softening is beneficial depends on many factors. Psychonauts usually avoid them as much as possible in order to protect the psychoacoustical potential of their brain, were they to desire to use this potential sometime in the future.

Psychedelics, in this framework, would be interpreted as neuroacoustic enhancers. These agents trigger, via control interruption, a more “echo-ey acoustic environment for one’s consciousness”. Meaning, any qualia experienced under the influence lasts for longer (the decay of intensity of experience as a function of time since presentation of stimuli becomes a lot “slower” or “fatter”). On high doses, the intensity of each component of a cycle of an experience can feel just as intense, and thus one might find oneself unable to locate oneself in time. Sometimes intense feelings return cyclically, and ultimately at strong doses, experiential feedback dominates every aspect of one’s experience, and there isn’t anything other than standing waves of synesthetic psychedelic feelings.

Peak symmetry states with their associated valence would be predicted to be far more accessible on highly harmonic states of consciousness. So psychedelics and the like could be carefully used to explore the positive extreme of valence: Hyper-symmetrical states. That said, for responsible exploration, a euphoriant will be needed to prevent negative psychedelic experiences.

Final Thoughts

A Harmonic Society is a place where everyone recognizes what makes other sentient beings love life. It’s a place in which everyone deeply understands the valence landscapes of other beings. People in such a society would know that a zebra, an owl, and a salamander all share the pursuit of harmonic states of consciousness, albeit in their own, often different-looking, state-spaces of qualia. We would understand each other far more deeply if we saw each other’s valence landscapes as part of a big state-space of possible preference architectures. Ultimately, the pursuit of existential bliss and the ontological question (why being?) would incite us to explore each other through consciousness technologies. We will have an expanded state-space of available possible moods, both individual and collective, increasing our chances of finding a new revolutionary understanding of consciousness, identity, and what’s possible for post-hedonium societies.


*I will note that to define what’s ethical one ultimately relies on beliefs about personal identity; truly frame-independent systems of morality are exceptionally hard to construct.

**The Entropic Brain theory portrays psychedelia in terms of increased entropy, but also, and most importantly, focuses on criticality. Just thinking about entropy would not distinguish between adding white noise and adding interesting patterns. In other words, from the point of view of simple entropy without any spectral (or nonlinear) analysis, SSRIs and psychedelics are doing pretty much the same thing. So the sense of “entropy” that matters will have to be a lot more detailed, showing you in what way the information encoded in normal states of consciousness changes as a function of entropy added in various ways.

On psychedelics one does indeed find highly ordered crystal-like states of consciousness (which I’ve described elsewhere as peak symmetry states), and as far as we know those states are also some of the most positively hedonically charged. Hence, at least in terms of describing the quality of the psychedelic experience, leaving symmetry out would make us miss an important big-picture kind of quality for psychedelics in general and their connection to valence variance.

 

***→ see quote →

My hypothesis strongly implies that ‘hedonic’ brain regions influence mood by virtue of acting as ‘tuning knobs’ for symmetry/harmony in the brain’s consciousness centers. Likewise, nociceptors, and the brain regions which gate & interpret their signals, will be located at critical points in brain networks, able to cause large amounts of salience-inducing antisymmetry very efficiently. We should also expect rhythm to be a powerful tool for modeling brain dynamics involving valence- for instance, we should be able to extend (Safron 2016)’s model of rhythmic entrainment in orgasm to other sorts of pleasure.

– Michael Johnson in Principia Qualia, page 52

Thinking in Numbers

To this day, readers both of [my] first book [Born on a Blue Day] and of my second, Embracing the Wide Sky, continue to send me their messages. They wonder how it must be to perceive words and numbers in different colors, shapes, and textures. They try to picture solving a sum in their minds using these multi-dimensional colored shapes. They seek the same beauty and emotion that I find in both a poem and a prime number. What can I tell them?

Imagine.

Close your eyes and imagine a space without limits, or the infinitesimal events that can stir up a country’s revolution. Imagine how the perfect game of chess might start and end: a win for white, or black, or a draw? Imagine numbers so vast that they exceed every atom in the universe, counting with eleven or twelve fingers instead of ten, reading a single book in an infinite number of ways.

Such imagination belongs to everyone. It even possesses its own science: mathematics. Ricardo Nemirovsky and Francesca Ferrara, who specialize in the study of mathematical cognition, write that “like literary fiction, mathematical imagination entertains pure possibilities.” This is the distillation of what I take to be interesting and important about the way in which mathematics informs our imaginative life. Often we are barely aware of it, but the play between numerical concepts saturates the way we experience the world.

 

– Extract from the book Thinking in Numbers, by Daniel Tammet.

Daniel was diagnosed with high-functioning autistic savant syndrome. He also happens to have synesthesia. Unlike most people, he has complex multi-modal experiences when thinking about numbers. But unlike most synesthetes, Daniel’s synesthesia is computationally useful in arithmetic tasks. This makes his mind a proof of concept that one can develop innovative computational uses for otherwise bizarre qualia varieties connected in bizarre ways. I.e. his synesthesia shows the potential that lies dormant in the realm of qualia computing.