 Alright, hello everyone. Welcome to the Active Inference Livestream. It is Active Inference Livestream 13.2 on January 12, 2021. Today is going to be a really awesome discussion. So thanks everyone who has come out to participate. Welcome to the Active Inference Lab. We are an experiment in online team communication, learning and practice related to Active Inference. You can find us at our website, Twitter, Gmail, YouTube, or our Keybase public team and public username. And this is a recorded and an archived Livestream. So please provide us with feedback so that we can improve on our work. All backgrounds and perspectives are welcome here. And as far as video etiquette for Livestream goes, mute if there's noise in the background, raise your hand so we can hear from everybody who wants to speak and will use respectful speech behavior. All the active streams in 2021, at least the regularly scheduled ones are on Tuesdays from 7 to 9 Pacific Time. So check out this rb.gy link to see the schedule. Today we are in 13.2 on January 12th with Adam Saffron and Colin DeYoung talking about this cool paper about cybernetics and personality. And next week and beyond, we're going to be heading into different sets of papers. So check out this spreadsheet and see if you're interested in participating. So today, I'm tired from information technology. All right, I just had to mute Douglas. So yeah, please don't play a stream while you're on a live stream. This is a Cognito 2021. And this is a conference in 2021 that is being organized by several of our colleagues by Mao and by NS who has appeared on the discussion before. Welcome, Blue. And so this looks like a really cool conference that they are all organizing. So check out this conference if you're curious and hopefully we'll be able to have updates on this conference later on. So in the intro in warm-up session, we introduce ourselves. We can just give a short introduction or warm-up and then pass it to somebody who hasn't spoken yet. So hi, everyone. I'm Daniel. I'm in California and I'll pass it first to Blue. Hi, everyone. I am Blue Knight. I'm an independent research consultant based out of New Mexico. Oh, and I'll pass to Steven. Hello, I'm Steven. I'm based in Toronto. I'm doing a practice-based PhD at the Salomon School Institute of Applied Psychology. And I will pass it over to Sasha. Hi, I'm Sasha. I'm based out of Davis, California. I'm a neuroscience graduate student. And I will pass it to Colin. Hi, there. I'm Colin DeYoung. I'm a professor in the psychology department at the University of Minnesota. And I will pass it on to Adam. Hi, I'm Adam Saffron. I'm a post-doctoral researcher out of Bloomington, Indiana. Good to be with you. Cool. Go to Alex. Yeah. Hi, everyone. I'm Alex. I'm in Moscow, Russia. And I'm a researcher in systems management school. And I pass it to Scott. Thank you. Hello, everyone. My name is Scott David. I'm at the University of Washington Applied Physics Lab where I'm the director of the Information Risk Research Initiative. And I'll pass it to Marco. All right. Thank you. My name is Marco, Marco Lin. I'm from Holland. I graduated in cognitive neuroscience and I'm just very interested in that entrance. I'm not affiliated with the language, but I have to be here. Is there anyone left? Okay. I had to kick out somebody who's doing feedback. So please have it figured out. And yes, cool. All right. Maybe not. All right. Hello. Dave, are you there? Okay. He just got disconnected. Dave. I am here. Hello. Okay. Well, just go, just proceed. I'll figure it out. Fun. Great times. Thanks for being on. So for the warm up questions, people can just raise their hand if they want to say something that's related or unrelated. The same questions from last week will put up and take a little bit of a different go at them. So what is something you are excited about today? And what is something that you liked or remembered about the paper? Anyone can go for it. I'd be curious to either of the authors just looking back on it. It's not so long ago, but we are in a new year. So with the fresh lens of the numerology of the year and just maybe these discussions, how are you looking back at what you had done and seeing it in a new or a different way? Or any other participants who want to add something that they thought of? Go ahead. I might as well go ahead then. I was just going to say that I don't know that for me, maybe even more so than for Adam, this particular chapter was sort of right at the boundary of the things that I've been thinking about and working with Adam has really pushed me into making contact with this whole topic of active inference and the kind of Fristonian world. And then it's also made me think more about modeling because I've generally been fairly skeptical about the value of trying to model whole agents that are analogous to actual human beings for learning things about psychology just because the brain and people are so complex. If you have a model, it has to have a sufficient degree of verisimilitude in resembling the thing that you're modeling before you're likely to actually learn new things just from studying the model rather than learning about the limitations of the model in one way or another and your skill at modeling. But I guess one of the things that writing this chapter with Adam made me think about was Maher's different levels of analysis and the way that if you get a good enough resemblance at the level of function or computation, you might be able to learn things even if you don't have verisimilitude at the level of implementation or algorithm. We don't necessarily have to say that the brain is doing the exact kind of mathematical computations that FEP demands in the same way that we would do them to model it, but as long as we're capturing something that truly resembles the functional computation that a cybernetic system like a human being needs to carry out, we might actually be able to learn new things about the way that people work. Cool. Steven? And then anyone else who wants to raise their hand, but just maybe we could hear from everybody in the warm-up. So, Steven and then Adam, and then anyone else. I'm really excited by the fact that this is, you know, tapped into a very widely used tool in psychology. So, and then links this back to these new potentials of active inference, which, you know, we've talked a lot about maybe in more theoretical, philosophical, philosophical ways, but this really is showing how, you know, the tools can start to move into each other. So I'm curious today to see more ways in which, you know, there's questions about, you know, the potential for that going forward, you know, what possibilities in the field of even organizational psychology, which uses these traits as well, these traits are kind of used beyond kind of work with sort of therapeutic environment. So, and, you know, even with the Myers-Briggs and other types of things, how can a more dynamical kind of embodied, you know, what's the future basically for applied psychology? Cool. Adam, then Scott. So, yeah, this, actually working with Colin on this, it really stretched me also is at the, like, limits of what I was used to thinking about. And I think I never, like, really seriously engaged with personality science before this. Like, I was always interested, but then I think it was also, like, skeptical that they had any chance of carving nature at its joints. And I was just sort of like, eh, maybe. And then as we started exploring the correspondences between, like, the cybernetic view of personality and the other cybernetic perspective of active inference, I started to become more compelled. And then, I'd say, like, things are always kind of moving around for me in terms of, like, perspectives. But I think recently some of the things I've been thinking about is, like, how far can we stretch or apply these ideas of personality over what scales and what ranges of systems? Can we think of the personalities of societies, the personalities of any system? And so that's, I think, most recently been on my mind. But I'm excited to be here because I love these live streams and the last week was awesome. Cool, agreed. Scott, and then Sasha, that anyone else? So a couple of things. The active inference is totally new to me and has really been revelatory, as other people have said. A couple of things. I used to work in identity a lot, still do. Identity kind of system federated identity, kind of technical stuff, not on the technical side. What I realized is the attributes are carried in these Markov blankets, right? So the access to attributes is through this statistical views of the surface of other systems. And so that's been really useful to me. And then the other part I realized is, I realized that a lot of folks, Friston and others, are coming at this from the biology perspective of seeing, gee, how do these things emerge by themselves? What I realized is as a lawyer, all I've ever done is construct synthetic Markov blankets. Because contracts are revealing a surface of a system. In fact, the system doesn't exist until the contract exists. Forget corporations. If I want to put a supply chain in place, I write a contract and then the organism of the supply chain exists as a result. And so it's funny because in that idea of, if it's too rarefied for viewing human mind, I get it. And it's interesting because in the de-risking context where a lot of folks are getting together, they try to make it simpler. Worry about bankruptcy. Worry about notice. Worry about these specific things. Boom, boom, boom. That's all they care about. So it's kind of like a synthetic organism constructed by the contract. The Markov blanket defines the organism. It's like the opposite of the blind men and the elephant where they're trying to find the elephant. Here, there is no elephant until there are their parameters. So the attributes define the system, and it's a very bounded system, very few attributes to it, but it simplifies it. And that's what a lot of regulatory market issues are about. So that's been huge for me, really realizing that we can synthetically construct these systems and base them on shared story, shared narrative, shared meaning. Contracts is a formal way of doing that, but Folkways is an informal way of doing that. And I think that's absolutely a very rich avenue for my work and looking forward to working with others on it. Thanks. Cool. Thank you, Sasha. And then anyone else on the warm up? Yeah, Sasha, then Adam. Okay. Yeah. I'm just really kind of following up on what Adam mentioned, thinking about other applications for the big five theory and what other groups and organisms they might apply to. And I think that's very relevant for this group and the kind of projects that we're working on is personalities of a team and thinking about different behaviors. I'm sure this isn't a novel concept, but I'd love to hear more about it from the authors of other possible applications of these metrics for teams of varying sizes. And the other thing I've been thinking about since last week is this, if this paper aims to integrate these two concepts of active inference and the big five theory, are there parts that are incongruent? Is there something that doesn't, that can't work together from both of these systems? And is there something that active inference can glean or kind of take away from personality research in a way that hasn't really been tried before? Cool questions. Thanks. Adam, then Blue. Someone flipping what I said before related to what Sasha and Scott just said. And also not just applying personality to other systems where you might not normally think of them in terms of personality. I'm also interested in reversing that flow. So it's like how might we think of the personality of teams, but the things we know about teams, how might that influence how we think about persons and their personalities? Or contracts, like what are analogies or models that can be derived by thinking of what you described of this like an active contract, like creating the markup like out of the contract in the doing. To what extent can we have this as part of person formation and person evolution, personality evolution? I don't know if that makes sense, but that's another place I've been thinking is like this bi-directional flow of like taking things from persons, putting them in terms of things you might not normally think of as persons, taking things you might not normally think of as persons, and then putting it back to persons and their personalities. Yeah, it's kind of like finding the human in the non-human Adam, really interesting point. So personifying and enlivening things that are perceived as inanimate and then finding the mechanism inside of living systems too. So blue, then Colin. So going back to what Adam was saying, not just now, but earlier, and also just something that I liked about the paper. So for me, psychology has kind of always been like this soft science. You can't really do science on psychology, but really like what that means is that it's a wicked problem, right? It just falls into the complexity realm. Like people are super, super intricate, complicated. It's a convoluted like thing. So this paper for me really was awesome in the way that it kind of concretized or hardened up for me, like a quantitative way to really look at personality. Interesting. Colin, then anyone else? And I'll put up a third question as well. Well, first, I'm really glad to hear that this might have convinced somebody that psychology wasn't just soft science. It's definitely hard science in the sense that it's difficult. And that's largely about issues. Well, it's two things. I think it's issues with measurement, the kinds of measurements that we want to do and can do. They don't necessarily line up very well in terms of precision. So everyone has physics envy because you've got precision down to 10 decimal places on measurements, but we're lucky if we have precision down to one decimal place or if it's even clearer that our measurements are anything other than essentially heuristic. If you have a scale and you're measuring a personality trait, essentially what you've got is just a measure of people's degree of a particular kind of behavior or experience relative to other people. So the measurement is relative and it's not clear that the one point on the scale actually corresponds to some true interval rather than just a kind of heuristic approximation. So yeah, there are a lot of really difficult challenges in psychology. But nonetheless, in order to do science, you just have to have some kind of standardized measurement and then be able to replicate, to see if other people can get it. I used to think that you had to have hypothesis testing too, but then I realized you can really just do science if you do exploratory work and then replicate it because once you've found something that then de facto becomes the hypothesis for the replication attempt. So all you're really looking for are reliable patterns in reality fundamentally. And I think we can do that to some extent in psychology. I wanted to say something also about Sasha's question about whether there are incompatibilities between FEP and active inference and general approaches to personality and my theoretical approach specifically. And I think one of the things that's really nice about active inference is that it's very low level in some sense. It's this formalism that is extremely widely applicable. And the way that I see it is that it's essentially a formalism that captures the essence of all cybernetic systems. And so I haven't found anything that makes me say, oh, this isn't going to work. It's like there's a lot of detail about what people are that isn't captured in active inference, but only because it's so low level. But there's nothing about active inference that isn't crucial to what people are, right? Because the process of pursuing goals via feedback, that is just necessarily the core of what an agent is. And so I guess my concern isn't that there are incompatibilities. It is how to draw out the AI framework in such a way that it makes contact with a lot of the specifics that we know about human's function rather than the generalities about how all cybernetic systems function. Really interesting answer. We'll go to Dave and then Mark up. Hi, my professor of cybernetics, Gordon Pask, in the 1950s and 60s developed some very fine-grained measures of learning strategies, which is how individuals approach a specific concrete task as they're learning it and learning styles, which is a more systematic way that folks develop a fixed repertoire of approaches to a task. The extremes on that scale are serial learning and holistic learning. I believe the areas of his data, where he uses entropy as an important measure for certainty and for tolerance for ambiguity, should tie in with the active, what? Not the active insurance, the other major area and what we're talking about, free energy. And I suspect that the learning styles should correlate with personality, and that is an area that I believe he, Pask, did work on. So that's something I need to dig into the literature on and compare it. Cool. Marko? Thanks. Also, thank you, Scott. That was very well put. So in general, I just really like a lot of things that have been said about active inference as a framework. I also echo the notion that the beauty is that it's so generally applicable that you can kind of elaborate it, and that becomes a kind of game of if you take a bit further with a few assumptions, when does it touch or overlap or align with other existing theories or empirical data? But what has been mentioned and something I find personally intriguing possibility is not necessarily active inference, but if something like active inference would be true and it would indeed be the most approximate description of our neural dynamics, then naturally follow that it also can become quite intuitive. So just like Colin said, that you can elaborate until it kind of touches something else in terms of theory. You could also maybe consider the idea that you can elaborate your understanding of active inference relative to personality or other psychological notions until it starts to touch your kind of folk psychological notions of people, of yourself, of others, of society. And I think that's what I think someone in the beginning said. I think it was Steven about the future of applied psychology. And I think this is really, really crucial, not just from a scientific perspective, but also a social and a humanistic perspective, because I think most people agree that folk psychology isn't enough. And I think also most people agree that most scientific psychology, or clinical side, isn't very easily implementable in terms of actual understanding yourself on a first person. So I think that's definitely the most exciting aspect. I think this particular work is a very nice milestone in regards to that. I mean, I don't think anyone really understands this or that much better with some ratings on the big five. But knowing the mechanisms, I think it's a very, very beautiful lens to understand oneself. So cheers. Thank you for sharing, Steven, and thank you for coming off. Yeah, sort of following on from what Marco was saying there is like the fact that this paper also takes in the affordances and active influence, like you say, does this dynamics. You gives this generative potential to the big five, which I know is kind of sort of seen to be there. But you know, the challenges that they can end up being whether you like it or not prescriptive and kind of diagnostic to some extent and having a way to relate them opens up that phenomenological sense. Because I think what I'm really interested in is if people structure their metaphorical models of how they cognize in the world, like how they, whether they relate to time as time coming towards them or they're going towards time or whether they're trying to build a more firm career or they're trying to get on in the world and travel and go somewhere. So those kind of phenomenological ways of structuring thought, they seem to be more ways that you can then connect them as opposed to it being like completely disconnected or just too big a gap to bridge. So this mixture of the affordances, the dynamics, the blanket and the attention and the way that phenomenological sort of structuring could happen even at the individual level starts to become more plausible, which is kind of cool. Awesome. Yeah, a lot of cool ideas coming up. One theme to return to is that by looking at the patterns and thinking about the frameworks that span different systems of different levels of analysis, we can think about like the traits of organizations, teams, individuals, organs, cells. So that's the multi-scale kind of complex systems approach because we set out to look for patterns across systems. And then that led to all these interesting questions like would there be a big five for teams or how are the generative models of individual personality related to the generative models or the generative processes underlying group personalities. And so pretty interesting things that we're coming to. Any other points in this intro before we just return to some of the later pages? Colin, go ahead and then anyone else. I'll just make one point in relation to this issue of, you know, team dynamics and personalities at other scales. That's something that there are different ways to approach it, right? So people have certainly, psychologists have certainly studied what happens to teams that have different characteristics of the individual members. And so you can think of how do the personalities of individual members of a team influence the success of the team. So for example, people having people high in openness on the team tends to make it more effective in various ways. And it seems to be because of the, that's especially true for leaderless teams, right? So if you have a, it's not necessarily true if there's a clearly established kind of leadership structure in the team, but if you have leaderless teams, then having people who are more sort of cognitively flexible enables them to kind of organize themselves better. So, but I think that what we're thinking about here is something a little bit different. Like, is there a personality of a team as a whole that might be more than the sum of its parts? And, you know, they're probably going to be, well, you need to do a sort of parallel kind of research in terms of trying to come up with a comprehensive set of descriptors of the properties of teams and then seeing if they varied systematically in the same way that people did with individuals and leading to the big five. But I guess one of my intuitions is that the highest level of the personality hierarchy stability and plasticity, which are also the most replicable across different languages when people have done the kind of big five research. So there's this very clear way in which what I think of as the two fundamental properties of all cybernetic systems, you know, at least ones that have to be able to function in an unpredictable and complex environment, that is probably applicable at virtually any scale, even for something that might not be, well, I think it's an interesting question whether something like a team or a society is actually a cybernetic system itself or whether it is an emergent dynamic out of a set of cybernetic systems interacting with each other. But regardless, I think you can certainly describe the stability and plasticity of teams or even societies, right? You can describe them in terms of how much they focus on and value the lack of disruption of their functioning versus how much they value expansion and exploration and sort of dynamic growth. Cool. That reminds me so much of the ant research and do colonies have phenotypes or is it just the phenotypes of nestmates to selection shape colony traits or is it just shaping the nestmates and they're duking it out? So all these multi-level systems, we kind of can see these similarities. So Adam and then Scott and then anyone else? So in relation to what Colin just said, correct me if I'm wrong, but I believe in terms of like cross-cultural, like the robustness, like, is it the stability and plasticity? Like this, like that level of granularity actually is what is the most broadly applicable in terms of like showing the least variation, like if you look at it in those two? Yes. So in terms of the lexical research, like starting from a specific language and extracting descriptors and then doing the same kind of, you know, ratings and factor analytic work, the only truly universal solution that's been found is the two-factor solution. And those factors very closely resemble stability and plasticity. Okay, so yeah, so that would support the idea that like if we're trying to do like a multi-scale personality story or a quasi-anthropomorphizing of quasi-agents, or maybe we should just call them agents, depending on our view, that yes, we'll be most likely to find stability and plasticity. But, you know, because like you basically grounded the Big Five as part of cybernetic functioning, and if any, like from an active inference perspective, if any like sustaining system is doing some kind of either explicit or implicit goal pursuit, we actually might find like even the Big Five itself like being surprisingly applicable, although like as you're saying, like it's an empirical question. And so like, so I guess the idea is we actually do the research program of this sort of derivation of factors and then we should expect to find things that might be shockingly like the Big Five in places where you might not expect them. Nice. All right, I didn't raise my hand, but I just wanted to jump in and say like that. I think the best analogy for that that's going on in science right now is over the last let's say 20 years people have really started to take seriously the idea of personality and other species. Now, I mean whether personality is the right word there or you want some other one, it doesn't really matter, right? You have individual differences in psychological functioning of other species. And so people have been doing this attempting to come up with extensive sets of descriptors of other species and ratings of them or coding based on video or observation or whatever and then looking to see whether they're these similar patterns. And most of the Big Five are identifiable or fairly close analogs in other species. And the one that is the least commonly found is actually maybe sort of, you know, proof of concept in a way because conscientiousness is the dimension that least appears in other species, something like conscientiousness. And in fact, the only two species where something close to that has appeared as its own separate dimension is in bonobos and chimpanzees, right? And so you think like what really makes humans special while it's our ability to plan for the distant future and to prioritize goals that aren't immediate or near immediate and to really have these elaborate goal hierarchies. And so it might not be surprising that it's only our nearest evolutionary neighbors that have these identifiable patterns that are related to those abilities and that are separate from the things just like, you know, how emotionally reactive you are. Very interesting. Scott and then I think Adam and then Marco. So this is great. One of the things that I thought was interesting is like what gets measured gets done and what gets measured gets seen. And so it's kind of interesting because we're coming, it's like a bio theory of everything kind of idea here and we're like, wow, this is really explanatory. My recollection is that quantum mechanics was just a calculation convention originally but it kind of works and we kind of use it and even though we don't understand all of it, it's kind of helpful that way. So it's not so much is it true but is it useful and this is broadly useful. And so it's interesting in living systems or cybernetic systems. We start to get at these questions we ask and when we ask questions from the, I don't know, I'd like someone to tell me what the big five is after I make this comment. But if they're the big five of consciousness, then I always think of them as the accounting firms but that's just because I was a lawyer for so many years. But it's like we're using the Turing test in a way. It's too much hubris about our form of intelligence, right? And so if we point our measurement instrument at a certain thing, that's what we're going to see. If we point our measurement instrument at human cognition, human consciousness, our sense of human consciousness, that's what we're going to see. And so this is a great discussion. To me it's not a truth finding thing but a usefulness of the paradigm kind of issue. And if someone can tell me what the big five are, just list them, that would be good because I'm not familiar with the term as in parlance. Sure. It's the big five dimensions that are empirically observed when you ask people personality trait surveys and their openness, consciousness, extraversion, agreeableness, neuroticism, ocean. And then one metaphor on the big five and it's kind of like the big five continents. That's what we have in the world as we have it. And so that is specifics. That's related to history and that changes. But the principle is water and land. And so maybe in a different system, a different sphere with a different type of water level, you still would have water and land, but the actual number of big dimensions might be empirically seen to be different. So that's kind of an interesting thing. I think Adam, then Marco, then Colin. I don't know if this would be useful but perhaps in a bit actually might be good to have Colin talk about the specific mappings of the big five onto cybernetic function that he describes in his original paper. That could be useful for people. But it's interesting about conscientious like being less likely to generalize potentially across species or systems. Because it seems like that makes some sense to me that in terms of preserving goal hierarchy integrity as you're confronted with all these different circumstances, but you want to maintain this overall like broadly applicable goal hierarchy that's benefiting you overall. In order for that to be the case, you have to have a system capable of generative modeling with or modeling with a certain degree of temporal depth and the policies is pursuing in counterfactual richness such that it could actually get itself into trouble and get confused and go down like wrong paths. And so you wouldn't expect this for all systems. You would expect this in a certain degree of sophistication, even maybe sophisticated effective inference in order for you to really actually need something like conscientiousness. So that one actually might not generalize as much. And related to what Scott just said, there seems to be this interesting combination. It's like realist versus instrumentalist. Or it's like you can think of like this personality is almost like, like the big five is like an effective field theory over persons where it's a good approximate description, a useful course screening. But now the question I would have those, is it like just completely... So there's a sense in which it's model relative. It's based on our interests. But there's like another sense in which it's like, well, like what worked in evolution? And like, and it's like, you don't have to take the perspective of evolution, but it's a, we can all kind of converge on it. It's like quasi-objective. Like it's one of many perspectives, but it's a pretty darn good perspective or an important one to keep in mind. And so that's since the big five actually might be carving nature at its joints in a really powerful way. And so I guess one more thing along those lines is, I think we're good for probably calling to talk about this since you discovered them, which is the aspects. And so one thing I'm wondering is, so it seems like in a lot of studies, when you do things at the level of the, so correct me if I'm wrong, but it's kind of, so under, so the big five has this hierarchical structure over them. You'll have these two penetrates, which we've been discussing of stability as shared variance of conscientiousness, agreeableness and inverse neuroticism and plasticity as shared variance of extroversion and openness. But then underneath the big five, confidently that you'll get, if you like force it to have the water level go down or up, I'm messing up Daniel's analogy, but you'll get two and only two aspects underneath each of them. And what's interesting is they oftentimes kind of read as being at cross purposes with each other. It's like these are very different things. They're very different. It's very different to be like industrious versus orderly. Like these are very different ways of being in the world of neurotic. It's like very different to being like withdrawing or being like more volatile and like reactive. And so I guess one thing I'm wondering is like both in doing active inferential modeling of persons and trying to suss out personality, would it be better to actually in some ways focus on the aspects in terms of finding joints there, but also maybe across other cybernetic systems, like could there be more purpose purchase from the aspects? Who and I just flipped to the slide. So just for anyone can just say go to this figure or go to this other thing. So Colin and then Marco, sorry you're muted. Thank you for the reminder. Yeah, so while there's a lot to think about in what's been said in the last couple of comments now. So in terms of this idea of continents and water and land, I do think that there's a way in which stability and plasticity are like the equivalent of water and land so that pretty much any organism you're going to be able to identify, if there's any kind of individual variability, which of course there must be in order for evolution to proceed. Like so probably if you take amoebas for example, they're not going to have a lot of different personality dimensions, but nonetheless you might find some that are a bit more of the same species that are a bit more exploratory and others that are a bit more cautious, let's say. So even at this very simple level, you may get this trade off between do you do things that keep your functioning running without disruption or do you do things that lead to greater expanse and discovery of new resources? So those probably do trade off. I would hope that when we're within a species though that the big five are not like just the continents. In other words, they're not just a historical accident. I think that they actually reflect variation in parameters that have evolved and that are present in every intact organism of that species. The issue though in terms of carving nature at its joints is just that nature in this case is so complicated. So if you look at, this is a good figure actually that Daniel put up to make this point, which is that when we draw these kinds of hierarchical taxonomies of personality traits, they are necessarily over simplifications especially at the levels below the big five because we know that personality does not have simple structure. And what that means from a factor analytic perspective is that if you took the taxonomy exactly literally that's depicted there, you would say, well, for example, no trait that is underneath extroversion that's a subcomponent of extroversion can be related to any trait that is a subcomponent of agreeableness, right? Because there are no lines connecting them in that figure. But in fact, the pattern of association and co-variation among more specific personality traits below the big five is considerably more complicated than that would suggest. And so for example, assertiveness is negatively correlated with politeness, right? So assertiveness is part of extroversion, politeness is part of agreeableness. Those are correlated with each other. However, it happens that the enthusiastic or sociable component of extroversion is positively correlated with the other aspect of agreeableness which is compassion and caring about other people. And so you have two different components of extroversion correlating in opposite directions with two different components of agreeableness and those end up canceling each other out. And so at the first, at the other level you see no correlation between agreeableness and extroversion essentially. So this whole thing, you know, the structure it looks simple in the picture, but you know it's a lot more complicated when you're down you go. And so I would have, you know, it seems to me that this aspect level distinctions are important. They are supported by some behavioral genetic research as well as some of my own research just at the phenotypic level. I think that they're probably the most important distinction within each of the big five in terms of discriminant validity. In other words, you often get things that are predicted differently at that level. For example, politeness and compassion are the two aspects of agreeableness. And well, it turns out that if you look at people's political orientation, people who identify as more left-wing or liberal score higher in compassion, people who identify as higher in conservatism, more right-wing score higher in politeness, right? So conservatives are more polite, liberals are more compassionate. When you go up to agreeableness, you see no correlation with political attitudes, which people found surprising for a long time because issues around altruism and cooperation seem very relevant to political issues and policy. Well, so it turns out that that's because something lower than the big five actually is associated with political attitudes in opposite directions. So I definitely think that these lower levels of the hierarchy can be useful. You know, there's less, the evidence for the big five is remarkably extensive, right? Whereas the evidence for that next level down that I've worked on is much less extensive. Nonetheless, you know, again, I think there isn't just one way to carve this system up because the patterns of interrelatedness are so complicated. And I do think that these are useful, useful subdimensions at any rate, even if you could focus on others. So I'll leave that there. Cool. Thanks, Marco. Then Scott. Thanks. The last few comments were very, very thought-provoking. So let me know if I ramble on for too long. So I love metaphors. So if you don't mind, I'll try to provide a complementary one. So instead of maybe landscapes, I have this thing with imagining the brain is one big dynamical system. And so all those dynamical flows of patterns, they need to, of course, be a pattern to order structures in the adaptive sense. So this kind of free structure could maybe be seen as a kind of grip, a kind of handle, a kind of heuristic well of gravity to kind of heuristically bias particular dynamics. And at least for me, that would make sense because of these big five factors that are quite segregate in terms of why I think they would be relevant. And that leads me back to the matter of personality and kind of anthropomorphization. Because I agree that there's a big problem of our theory-ladenness in the manner in which we approach psychology. And that also should exhibit itself in, as mentioned, the observations when it comes to animals. But for example, when it comes to conscientiousness, as Adam already pointed out, there's a big problem in the world that and I think he did allude to it, but another thing to emphasize is also the kind of field or a range of possible alternatives to care about. To someone who would be low on conscientiousness or care-free would perhaps not simply be lacking goal-orientedness, but be more aware of in touch or have a bigger affinity with the notion of play of simply engaging in behavior that is reducing expected free energy. Yet, not necessarily defined in terms of a particular goal. So not all animals would have to deal with that. Because the evolutionary simply are entrained or evolved to have awareness or sensitivity to a multitude of possibilities in what to do in environments. So the very emergence of a particular big five factor might be also related to the complexity of the imperatives that they encounter in their life, in their niches. So one maybe interesting example is like dogs or other social animals, although in general I don't think it would be appropriate to project or impose our anthropomorphic notion of personality. I do think at least dog owners can agree that dogs tend to have a personality that seems to develop or evolve and even seemingly be similar to their owners. Possibly because the imperatives of their environment is expanded by being in relation with human agents. And so by the expansion of possible niches, possible situations, the relevance or the variation specific to where conscientiousness is a relevant factor because more parents previous than it might not be the case. Does that make sense? Yeah, I'll leave that that because I forgot everything else wanted to say. No, it's really interesting that Big Five is not the essence of an individual. It's actually in a cybernetic framework making the individual's phenotype in relationship to the goals and the context of their niche. So it actually moves away from some of the essentialist types of critiques that people might naively bring up against Big Five. So it's a historical pattern with a generative underlying cybernetic reality. So Scott, then Adam, then anyone else. So fantastic conversation. A couple of observations. One, personality is for a person and so it's kind of interesting when we take personality and pose it on other systems. I'm just playing linguistics there a little bit. But in a way what is the delta between personality and living form ality? I wonder. So that might be because are we assuming the human elements? Again, it's that subjectivity question. One of the things also it occurred to me that Friston kind of allows for a notion of situated personality beyond situated cognition. And so maybe those are overlapping. Maybe they're not. I'm not sure if all personality elements are seen as cognitive or if there's some things that are, you know, cognitive and hormonal and whatever other causation chains there are. So it's interesting that the behavioral elements we ascribe to personality may or may not may not drive through all living systems. I tend to define living systems and it's not anything reasoned. It's just what I've done for the last 10 years and it's been effective for me is I define living systems as auto-catalytic and entropy secreting. So they can be in a variety of different forms and so I think of markets and corporations as living systems. And so to me one of the questions that with the definitions in the Friston notion we have kind of a fractal notion of things that show scale independence I guess is one of the elements of fractal systems and so to what extent can the definition of using these personality traits at different levels help us understand the scale independent possibilities there? That's one issue. Then the last point these are kind of random but I think they're clustered around the idea of the model ascribing. The last one is, you know, when we talk if I'm talking to Daniel there's at least four identities involved there my sense of me, his sense of me, his sense of him and my sense of him. So one of the challenges when you're talking to corporations, markets maybe not dogs because they may have quality, is that the other the other entity doesn't have qualia. If you're talking to a corporation it does not have qualia. The people in the corporation have qualia but the corporation itself does not. And so the that's obviously we can talk about that but my point is if they don't have qualia can they have empathy and if they don't have empathy maybe there are differences in kind of the Friston stack of fractal stack of living things and so maybe that is one way to start to distinguish between markets and corporations and again gets back to the animals issue. Do animals have qualia is the quality of their qualia slightly different than ours such that maybe there's a subset of personality that they show things like that anyway just a couple of random thoughts I think but they again that definition of living beings I think is bound up in here and again not being to us not being too stubborn in our perspective that the personality is something that would describe all the attributes that work up and down the biological stack. Thanks. We got Adam then Steven Marco Collin. So go ahead Adam. Hi. Okay there's a lot to try to rein it in for me. I like so what Marco is saying with this basically this attractor network formation perspective supposed metaphor or really good model the he mentioned like niche construction and he mentioned I think this perspective is really important in terms of thinking of the dynamic character of the processes that might give rise to these attractors over the course of development and situations and with development potentially being understandable and an open-ended sense that never stopped. And so I guess along the thinking of that causal process model that Collin has in the original paper where it's going from genes and environment to you're having these cybernetic function and then traits and characteristic adaptations but then there's this feedback as and one way of understanding this feedback is the kind of niche construction relationship where you're modifying your environment you're specializing yourself for these environments as you're trying to get along in the world as a goal-seeking system one more thing with relation to what Scott was saying I think that's actually I don't want to also while I'm interested in seeing how far we can stretch personality I wouldn't want to overstate it either because there were major transitions in evolution and one of them would be basically the ability to have certain types of generative modeling where there might be a qualia and qualia of different varieties and then we might go beyond that to not just qualia but things like access consciousness and if you like that distinction I do, like being able to know your experience and report on it and things like being autonoetic to reflect on even your knowing and so this would be very different cybernetic systems once you introduce this so like you can have like and if you have something like the big 5 showing up or the big 10 or the medium 10 I don't know what you want to say but they might show up differently so like you can think of like generalized agreeableness as like tendencies towards like non-zero interaction and like not defecting in generalized game theory something like that like in a multi-agent context you're more of a cooperator and you coordinate better but very different thing if you're actually having something like the feeling of compassion or empathy these are different processes getting into the mix and you want a different handling and some things might not actually you want a very different handling yep interesting like would rather somebody act compassionately and be perceived as high percentile compassion or perceived overwhelming experience of compassion Steven, Marco, Colin, Blitt yeah this brings in the idea of how people relate in a task based environment or a more task orientated environment I suppose where it might structure things a bit more clearly between the things in the world where the person is and whatever mechanism or structure is interacting between the two am I picking up at all am I there's some sort of delineation and one thing that sort of comes to mind when we talk about so you've got the generative model you've got these hidden states of the world and you've got this Markov blanket and the thing with the Markov blanket is unlike it's not a system boundary you never can really define it you've got this and what I thought it was quite interesting really when by having the stability and plasticity am seeing sort of the plasticity with this extraversion which you could say is like influencing the hidden states in a way you're perturbing out there and the the intellect is making models is the sort of the other side but the the stability in some ways is really stabilizing with the Markov blanket like the nature of the regime because you've got neuroticism is it like a very hilly landscape of actually like inferring so maybe we because I'm really interested in how we might have mental space psychology approaches which there might be some sort of intervening spatial mediator to how we process and that spatial mediator of how big and scary things are around us and in our metaphors it could be sort of mediating so the neuroticism the agreeableness of conscientiousness seem to have more structuring about how I might take data in through the Markov blankets the other side and the other two seem to have a bit more about what's out there and what's in here in a really crude way that's what I'm just saying that came up for me I thought it was really interesting Thank you, Marco, Colin, Liv Thanks Yeah, I also want to build up on that so I fully agree so I think the prospects that are most attractive with this kind of approach to personality is as Stephen alluded to that you can associate these aspect factors into particular challenges as it were and I think it's nice also to connect that with what was mentioned in the beginning about Mars' levels of analysis so Mars' levels of analysis is a bit of a hot topic always so for getting a bit biased but personally I don't believe that active inference should be seen as quintessentially computational because I think it would discredit the fact that it implies a lot about the algorithmic level so just like Stephen kind of already said, you could have these intermediate spaces and that's basically asking you how these computational imperatives or functions are going to be realized and I really want to emphasize the notion of mediation because I want to connect that to what Scott was talking about so it should be emphasized that from what I can tell this model of personality is kind of neutral about consciousness we should be very very mindful about not conflating the first person signs or perceptual models of how empathy arises within us with the process empathy as such because if you see the process of empathy within the brain then there's two ways to perceive it at the very least, third person view from the measures whether physiological or questionnaires and first person of course it stands to reason that it's not necessary to have the first person encounter with these processes for these processes to happen but having said that, it would be very interesting to imagine that consciousness would be like intermediate space as well such that the components factors and the respective challenges of the big fine can actually equilibrate in the sense of it's not a trivial challenge to balance the respective aspects of these factors and the factors between each other for the autonomous being of concern and so it would stand to reason that indeed there might be another intermediate space intermediate between these five factors which would emphasize the cybernetic function of something like consciousness as a space which I think it might be interesting for you Scott because I think then you can generalize it for non-living beings or composites or organizations because if you then associate personality as heuristics for certain domains of challenges or domains of imperatives and then having to bind them together through a certain integrative space then you could have I think a very interesting approach organizational science if that makes sense Thanks, that reminds me a lot of the projective consciousness model paper that we read in a previous active stream of cybernetic organisms or cybernetic systems other than organisms if they have a projective geometry do they also have that awareness? So Colin, Blue, Scott What I wanted to say actually ties in nicely to the point that Marco was just making I was going to start by saying that I think it's important that the existence of personality is not dependent on the existence of qualia so this is I think a mistake that people often make in the big five and similar personality systems they think well these are just about how people perceive themselves or how people perceive other people and maybe they're useful categories in terms of what you need to know about a person but actually the claim is that assuming that these descriptions of people's behavior have validity they describe patterns of behavior that exist regardless of whether people are accurately aware of them or not and so you could have differences in persistent differences in behavior from organism to organism in any species that would be directly analogous to what we think of as personality in human beings now of course it is true that we have this additional layer in which one kind of characteristic adaptation that we have is that we tell a story about ourselves consciously and we can describe ourselves and we actually make use of that when we try to measure people's personalities because we often just ask them to describe themselves but we're not interested just in their description of themselves as such, we're interested in what we hope and we know from a fair amount of evidence that those descriptions are reasonably accurate so it's not just that I think that I talk a lot I actually do talk a lot if you take some kind of objective measurement of that, that's part of what being extroverted is about and so consciousness is obviously something that is mysterious in various ways but I think that if we have that division between the easy and hard problems of consciousness I think we actually have some pretty important things to say about the so-called easy problems of consciousness like how does it function, what does it do and I think Marco is right that one of the things that consciousness does is allow people to deal with conflicting information, conflicting goals and to create a sufficiently broad model of the world and our attention of it that we can then control our attention effectively. I really like Jeffrey Gray's perspective on consciousness, his last book was just called consciousness creeping up on the hard problem, that's really excellent and then I also like what is his name, Michael Graziano's theory of consciousness it ties in very nicely to this idea that a cybernetic system to be a good cybernetic system you have to have some kind of physical representation of the variables that are being controlled that's the good regulator of this theorem. And so I think that one useful way to think about consciousness is as a model of our attentional field that allows us to direct our attention effectively to things that are important for our goals or things that are potentially conflicting between goals and so in that sense yes it would certainly be relevant to conflicts that arise because of different needs or goals that are related to different personality traits but it's not something external to the traits specifically because the traits are descriptions of patterns of variation in a population and so when we ascribe them to the person the person has a level of the trait and so those traits actually should be reasonably comprehensive descriptors of functioning in all of the important categories of human function and that includes consciousness and in fact if you look at the contents of the Big Five openness and intellect is the trait that is really related to the qualities of consciousness people who are high in openness have our conscious of more things their consciousness is broader it's deeper it's more complex and you know so there's the openness side of that trait in the intellect side and so you've got this sort of pattern detection and then pattern analysis and all of that is basically about in large part the qualities of our consciousness. Cool very interesting Blue then Scott then Adam So I am going to totally bypass consciousness because it's just a it's a hard problem right but I do want to just consider for a minute what would happen if we maybe not asking what is a person but what is an individual could we then like scale over like non-human systems like you know organizations or even to other animals like Adam brought up like transitions in evolution and how do these like personality traits maybe influence evolution that would be interesting to think about like can we think of maybe clams as like ultra neurotic features and ants as like high in consciousness could we take like these personality traits and map them on to organisms that have evolved over time and what would that look like in terms of evolution and what is essential openness to exploration is obviously key for you know some of the major transitions to happen that reminds me of the principal components the big fives they're based upon the patterns of empirical variation in the actual population as Colin brought up and then the animal spirits are without making a metaphysical claim they're like archetypes like of bravery or of extraversion or social butterfly or you know a couch potato it's still a living creature it's a potato it's in the ground and so how we model different types of metaphors it's like that's on our Procrustian bed but then what does it look like to have it on their Procrustian bed for their personalities or individualities so Scott then Marco then Colin Lou I love the notion of neurotic clams that's my favorite yeah this is this is great and just along the same lines a couple of things that leaped into my head I was thinking of Irving Goffman stuff you want to be perceived so that's kind of modifying your environment in terms of the persona perception that was one thing and also a lot of self other things in here the identity kind of notion in that and that idea presentation of self one of the things I was wondering about is the DSM Diagnostic Statistical Manual DSM 5 just came out and really how it plays into the what I guess they would define as anomalous persona traits and usually DSM seems to be the ones that aren't the ones that make you not productive in the capitalist society kind of things because there's a lot of things in there called cultural in there but it's interesting that those are the outer bounds of what's expected expectations of persona and I think a lot of the survival in this questions have to do with expectations and the rising of these variables in the corollaries in animals and in people have to do with survival and increasingly in social survival but also just that survival of the organism and so that idea of presentation of self might be even related to camouflage and issues of that you know how do you present yourself or not present yourself in society so anyway just that host of thoughts how that naturally evolved is kind of interesting also and this is not a study thing I'm not an expert on this but just functionally again I think of the mind and consciousness well I think of the consciousness side just like blue said but I think of the mind as not in the brain I think of the mind as in language and material culture and the brain as an antenna tuned into the local mind so if you're my sister adopted a child at the age of 5 months and now she's a was raised in central Pennsylvania and she was adopted from China she doesn't speak Chinese she speaks Spanish English she's a U.S. person but she looks like a Chinese person so that their mind was developed in the in Pennsylvania and so if there's no feral consciousness which take that as an assumption for a second or no feral mind and if it's a learned phenomenon it's interesting that we have this framing and then we keep re-upping the framing re-affirming and valorizing the framing it's interesting to wonder if there's a lag in terms of survivability what were the persona characteristics that were selected for and I don't just mean in biological evolution but socially are they identical to the ones that will be selected for in a much more dense interaction environment for instance so there may be some interesting questions if and to the extent that persona is not on board in the brain inherently but is a learned social phenomenon what is the how malleable is it and I mean I guess that's what education is about and why colonialists try to get indigenous people in their schools their children in schools early to render the externality innocuous but it'd be interesting from a perspective of current situations that we have in current policy challenges and public health challenges etc. information challenges how might we handle those recognizing a more plastic persona just a few thoughts cool interesting Marco Colin Adam thanks wait I forgot what I was going to we'll come back okay Colin Adam Marco okay well I think that there are certainly true that there are aspects of the mind and personality that are learned but then there are others that are evolved essentially and sufficiently genetically programmed that they would be present even in you know feral children but and that connection back to the idea of the evolution of personality when I when you know blue was talking about neurotic clams I was thinking about some of the work that's actually been done on the evolution of personality in animals some of it is very interesting and it's a paper from almost 15 years ago now by Daniel nettle that talks about the evolution of personality in humans and other animals and so one of the things that he talks about are these experiments where you take a bunch of fish and if you just naturalistically observe them you will see that there are differences in personality some of the fish are relatively bold some of them are relatively timid and cautious but the difference is that if you systematically manipulate their environment you can actually change evolution of that population over time in interaction with the personalities of the individual and so the way that I think about this is that personalities drive evolution but evolution also drives changes in personalities at the population level so what happens is that if you have those fish in an environment that has a lot of predators then the bold ones get eaten and the frequency of the more cautious timid ones increases in the population but if you put them in an environment that has very few predators then what happens is that the bold ones tend to monopolize a lot more resources and the frequency of the bold personalities increases in the population yes so always the ecological selection on the development of personality not even just personality something that can be reached out and grasped but actually how personality becomes developed in relation so Adam then Marco then Steven trying to tie together a few threads before I lose them so like earlier I was saying you know evolution is one of multiple perspectives but a really special perspective and it is a really special perspective like I it almost makes me move towards like while on some level I'm an instrumentalist but with respect to like the things like what is the ontological status of the Marco blanket I think there is also like a realism a quasi realism and maybe a realism so strong we can sometimes throw away with the word quasi although we should always kind of keep in our back pockets and we should keep in mind that there are multiple ways of carving up the world actually infinite ways it's just somewhere to be more or less good for being such as us and one really good carving would be the one that basically relates to functional cycles that achieve closure on time scales where they can be subject to cumulative evolution broadly construed whether being the development of the organism or phylogeny and so what that might take out some particularly universally agreed upon while still subjective joints but to kind of bring it together more among those and with relation to your personality characteristics there's an interesting thing that can happen for species capable of modeling each other themselves with a certain kind of richness having certain types of consciousness in the sense of knowledge where without it potentially that they can come aware of these reliable core screenings of themselves and others and then this forms a basis for your planning and then this forms a basis for basically not just the niches that you're trying to construct but the way your niche is constructing you and so for all this like active inference with its like embrace and embracing of circular causality and this neutrality between like system and environment in terms of each being its own kind of active inferential system and exploring this interplay I think that's really essential for understanding personality in both the specific and the general case before I let you all move on like one final to the threads would be like when talking about something like consciousness so like I recently kind of took on the hard problem it is debated how much success I had but in trying to make way on this it seems useful to not and probably with respect to discussions of personality all of our words are just words they're core screenings that we agree upon and they don't exist in the world and so consciousness part of the issue is like we mean very different things when we use the word sometimes and so so you might say Scott was suggesting like you know is we might want to distinguish between mind and consciousness or intelligence and consciousness and so you could say like your mind is distributed through the world and that your brain is an antenna I think that's a good perspective but if you're thinking of like consciousness as like a type of model where there is something that it's like from the inside to be that kind of modeling process that might not be extended that might be like a subsystem and a sub model and so similar kind of building up this onion of complexity with respect to consciousness and people are kind of kind of engaging at different levels but not always being clear on the level that they're engaging in and so you could so recently I was engaging at the level of like basic phenomenality itself and Rudolf in like the project of consciousness model very similar spirit it's you know specifically what are the ways we can describe your subjectivity as a particular reference frame and point of view in the world as you as a cybernetic system how can we characterize that and so then we go like another level still and we're getting to what Colin's talking about with things like attention schemas and like models of your awareness for modeling yourself and others and share and shared attention schemas and so I guess the and we can go further still into things like self-consciousness and each of these will be kinds of major transitions in evolution and development which I understand is a kind of evolution and each of these will now dock with personality in different ways now the one final thing with it's interesting is the way in which personality traits if you're thinking them as kind of core screenings it's like which is like core screen descriptions and so when can you core screen you can core screen when you have something like a renormalization process when you have some sort of like emergent attractor structure such that the lower level or the smaller fluctuations smaller dynamics can be kind of like averaged out because now you're pointing at this bigger thing and the bigger slower attracting manifold didn't care about the particulars so rather that's not relevant for this course or phenomena that you're trying to observe and so there's a sense in which personality is a kind of core screenings like what are the nature of these like normalization processes and so one of them it would seem like so what is inducing this attracting structure it seems like a good part of it would be so you can model yourself across situations so you know what kind of agent am I that I am likely to be doing what and what circumstances and this will be pretty essential for pursuing any kind of complex plan but similarly for coordination other or working around other agents or cooperation or competition what kind of agents is this I'm dealing with but for the sake of cooperation you might you know ask someone to be responsible they should be able to respond and report in a reliable and consistent way we actually demand of each other in a way a certain non-erratic character we demand dimensionality reduction of each other just because otherwise we get surprised and as the Fringe principle in active interest we don't like being surprised because then we can't plan and then we can't outsource our cybernetic entropy and it's bad so there's a sense in which to what degree like there's the same way that like personality traits are influencing they're emerging from evolution then influencing the course of evolution with development your personality traits are emerging from your interaction with your environment as you're carving out this niche but also they can also become like attracting targets in of themselves on different levels like if you were if you have this expectation of yourself as being this type of agent it can be a self-fulfilling prophecy I think I'm an erotic agent I think I'm an extroverted agent now other people think I'm an erotic agent other people think I'm an extroverted agent and now they're treating me a certain way and I'm acting a certain way and this keeps going yes no a lot of awesome stuff that reminds me of the scripts it reminds me of the phrase the common saying the absolute power to act absolutely brings responsibility or something like that like power and responsibility but now we're talking about the power to act and the responsibility to stimuli and that means that we respond so Marco then Steven thanks Adam for that fantastic model so first a bit of a meta comments or question so since you said we can't use the chat in stream I've been putting some links in the youtube chat but I'm wondering if you had a recommended way to do that because a lot of people mentioned a lot of interesting research I think people are modest to the point that they won't advertise their own research or link it so what would you prefer Daniel to respond to the meta to respond to the meta and just to clarify so in the jitsie for the live stream it will make a if you use the chat so it's very disruptive to use the chat during the jitsie so the single source of truth for the video is the youtube page or on keybase so definitely everybody should make a keybase username and join us at actinflab.public and check out actin actinference.org to find out more but the keybase at this point is our actual organizational hub so that's where we're doing the open participatory active inference lab like this is a project of the communications branch of active inference lab continue yeah I didn't know what keybase is but thank you glad I asked building on Saffron's great comments so I fully agree that personality is definitely something scaffolded so personally I do see the relevance of course grading but I think it's also interesting to use another example or metaphor of bootstrapping so one interesting thing is so in AI we keep dealing with interactability but fairly we are still living with not defying laws of computation so there needs to be this kind of leak of faith of how to start off acting around in the world so one way to see it instead of course grading in terms of temporal scale maybe it's simply let me put it this way if going to behave over time, over time, over time across different contexts there should be overlap in the kind of parameters or hyper parameters that govern your behavior to the world and those that are more consistent or more covariance or synergetic they will consolidate into this kind of heuristic core to further elaborate and specify and on top of that building on to that I was wondering because Adam earlier asked Colin if he could maybe explicitly map the the big five to mechanisms or inferential terms but I was wondering if it's appreciated or useful to also put the question for everyone since everyone is relatively familiar with active inference or Fristone terminology how they would characterize the big five factors because mostly it's in very simplistic adjectives or purely behavioral but I think would be interesting to maybe at least attempt faithfully to understand them from the inside out what kind of functions or processes or inferential problems would be associated with a few people already did that but I was wondering if people would like to do that more especially together great question and also a lot of it is enclosed in the paper which is why this is so fun to have as a journal club because we can always superfluously add on top but then the paper also references a lot of these things so we'll go to Steven Colin Scott a lot of really cool stuff I think well as well as the modeling idea I think going back to this the idea of a timid fish and a bold fish and the affordances of the environment one thing that's really been I've been really striking me has been trying to integrate drama based projects and sort of experiential workshops with psychology is that we've got this kind of reflective perspective of the big five or aspect to psychology like you can take a perspective and reflect and you get a sense of it the big five will use that in the reporting yet it's happening to some sub personal dynamics and other dynamics which may not be available at the conscious reflective level and I think that that then starts to get really interesting with the affordances again with the timid and bold of how this all sort of scales because it's not necessarily a timid generative model that creates the timid if I take too many caps of coffee okay now my I don't know if it's my brain that's changed or whether it's the sensitivity to my of my perceptual and active action reflexes which are kind of now or maybe it's my stomach or so you've got this there's an interesting way with this that it can bring in the kind of the dynamics and the affordances and sort of one thing that's quite interesting in theater you've got an acting method which is basically very like you hear a very text orientated and what did my character want to do but another way is you should actually get into the body of an animal so you'd be a spider and they actually have in the lecock school they do like set a whole series of weeks in the zoo and everyone has to be animals and inhabit animals bodies and but you know there's something different about if you're on four legs and you're moving around smelling things okay you're going to be balancing your inputs and your options and affordances for action are quite different right so it might well change the way when I'm not going to speculate but just just basically coming back to the point of the timid and the bold and this idea that the way the active influence or the inactive sort of dynamics of the body with the environment can be a lot of that could be sitting at a sub personal level like you're not really sure how you know that bright light has affected my retinal or my tiredness has affected or my alcohol consumption you know some of these types of things so I thought sort of say that that kind of could connect in with this the more consciousness reflective piece and the sort of phenomenological cybernetic subconscious piece cool Colin Scott Adam I just wanted to say that the reason that I didn't take up Adam's suggestion to go through each of the big five and talk about the cybernetic processes that we think are probably related to them is precisely that you know that is all discussed in the paper I didn't want to take up a bunch of extra time with things that many people have already read in terms of this kind of idea of the you know the richness of the psychological processes that are underlying each of the big five that is something that you know it's true that a lot of people have this very shallow perspective on them partly due to the way that they were derived in the first place and the stereotypes that people have about them as you know scientific entities but I could spend the rest of the live stream talking about these processes in great detail but I think it would probably be better just to refer people to the original paper that I wrote in 2015 on cybernetic big five theory and then also maybe to a theory that we wrote on the cybernetic perspective on psychopathology because the big five turn out to be equivalent also to the major dimensions of risk for different forms of psychopathology and we talk a lot about the specifics that might be important there as well and all my papers or most of my papers are available on my website cool great and that's also why again the journal club format it's not the only format we're going to experiment with in active inference lab but this will actually allow us to have conversations that are rooted in a lot of specifics and various papers will be amenable to that in different ways so Scott and then Adam so and I wanted to apologize for not reading not having an opportunity to read the paper before this one so it would have been better informed had I done that one I wanted to just occurred to me something that might be going on here and it's a little bit stretch so let me give me a second here so this is going to be where I'm going with this is whether living forms are entropy engines okay that's where I'm going to go with this so there was entropy I'm talking about now using Shannon and von Neumann and entropy notions generally not by analogy but rather saying you talk about general disorder and so I'll be a little fast and loose with entropy for a second here so Penrose wrote a paper book recently we talked about the what we get from the sun is not energy but rather neg entropy because the yellow light of the sun comes in infrared light goes out the difference is the frequency the information carrying capacity of yellow light is higher this is higher frequency and so it's a net energy neutral except for global warming or climate change but it's but what we get is the structure of plants reflects that higher neg entropy opportunity I guess is one way to put it so let's so starting with that what if entropy was driving a lot more and so what if the remember recall my definition of living forms as auto catalytic entropy secreting entities so entropy universally is increasing living forms are localized neg entropy lower entropy locally and try to preserve it what if the drive both the physical symbiogenesis where free existing bacteria migrate on board to eukaryotes and then become mitochondria in human in animals and chloroplasts in plants so they come together because the better deal basically a better way to be auto catalytic and entropy secreting at the larger scales what if that extends to where we are now with the with the what if those states stages of personality have those functions we're saying they do and then it has to do with both entropy accounting so when you're meeting someone who shares personality states or has predictable and X personality states you can predict there's a lower differential so from a second law of thermodynamics perspective you have like for Carnot's equation you need a hot and cold differential to perform work what if the information differentials that you have between entities that are meeting each other help to cultivate various responses and those responses are we are cataloging in those five big persona traits and so that's ways of being or ways of encountering so it helps you navigate the social landscape and that the what what if what we have really is what I call an entropy engine which is that the personality goes out and harvests disorder through these mechanisms dissipating it obviously but also growing from it and what I mean by that is all institutions all of them are dis de-risk some subset of interactions whether it's grocery stores or hospitals or banks or whatever that's what institutions do at larger scales so their neg entropy expressions at larger scales but we have those those feelers that go out and maybe the personality types and our physical attributes are the feelers that go out to figure out where there's opportunities for proliferation into a community of interest around a shared risk thanks wow very interesting with the Jordan Peterson the darkness into light ordering the universe and you know order disordered order as a thermodynamic inference process so hopefully that will be something we can like continue thinking about in our last 20 minutes so we'll do Adam then Marco and anybody who's watching live can leave a comment anyone else who's on here can prepare any like last thoughts but this has been a super awesome discussion so Adam then Marco with relation to entropy you might want to check out the work of in terms of life Jeremy England he has a really interesting handling of that's free energy principle compatible of entropy maximization and you might also enjoy Terence Deacon's work Colin hooked me up with him it's in complete nature he gets into discussions like cough auto kind of discussions of auto catalysis like Kaufman talks about and work cycles as ways of like describing systems in their reality because you know if you're talking about realities you know differences that make a difference being able to contribute to a work cycle is a pretty good criteria and if we're able to identify system on some sense it's engaging probably in some kind of work cycle and you can do an entrap like a handling of it that's going to involve entropy also in terms of entropy there is maybe two papers you might like one is I posted them in the YouTube feed which is psychological entropy from Hirsch Mar and Peterson 2012 and then a more recent attitudinal entropy framework by Dalleg, Boersboom, Harvill and Wander Maussam forgive my pronunciation but to now look back around to some of the things that Stefan was saying that still on some level of abstraction will involve entropy one of the places I'm actually really thinking now into and trying to understand is the conditions under which personality might be more or less plastic the conditions under which you can get trade change, altered traits and not just altered states and along these lines of notions of like acting like technical acting versus method acting like to what extent is this sort of self-fulfilling like the self perpetuating attractor dynamic that makes up persons and their personalities to what extent is it changeable even if there are these endurance parameters to what extent can you take those arrows and Collins calls the diagram and really work them so Collins basically talked about if you want to have his well-being work and his psychopathology work about character adaptations being where you have the most leverage like the things you tend to do the design of your life and the situations you show up and how you show up and this is where you like clinicians and yourself you might have the greatest purchase in an acting change but and so this would work on multiple levels one would be like it would change the nature of what you experience and we're learners we update it would also change your niche construction it would change the way that the context in which you're in that would be leverage there but I guess one more thought would be bring it back to consciousness like to what extent it's not just change to change your characteristic adaptations would be to change the things you tend to do but what you're describing this sort of like method acting view or you bringing that up but like acting to what extent can personality be changed by really paying close attention to this what it is it like to be me and actually what could it be like to be a different kind of me and then imagining yourself in that way I'm going to call it faking it until you make it and to what degree is this a substantial part of how personality forms to begin with like children at play copying their parents pretending to be that way but now the question I would have is in the bootstrapping can we still leverage this as adults and one thing along those lines would be that makes me think you might be able to I believe in the Tibetan tradition they'll I think it's part of jokes I don't know how to pronounce it but basically they'll do these kinds of like deity practices where like let's say you have some kind of like affliction that's like not working for you you might pretend to be a kind of complimentary deity or spirit but you don't pull this off on your own like the social context and so you create this kind of subpersona that you live into I believe they call it tulpa Marco I think actually pointed me to this but but you don't do it on your own you have a holding and hardness social context that supports your act and you're able to kind of collaboratively and active and that seems to be we can empirically establish this as a mechanism of personality change this seems to be part of what makes the magic happen so yeah that's the things I'm sort of thinking what are all of the levers we have to actually design our personalities going forward so what mental techniques psychedelics like meditation what's at our disposal so that we can have like you know we can be good meta regulators over ourselves and like realizing like our deep values thank you Adam I'm just going to throw up that closing slide and then we'll hear from everybody who wants to raise their hand with the final note this has really been an awesome act in stream 13.1 and 2 so in the calendar event for those who are participating live there's a feedback form a survey which would be appreciated and all of you are always welcome whether you're the author or not to participate in a future act in stream you can go to the spreadsheet and find out the calendar but all of you would be extremely welcome to be in a group act in stream like this again and then if anybody has an idea for an event that they want to arrange or like an interview or a follow-up discussion or a few minutes isn't enough during this conversation to talk about drama or consciousness or whatever it happens to be so let's just plan that as an event so that we can always make sure to hold this base for hearing about everyone's perspective because it's just such a great stew when we can have it all going so Marco and then Stephen and then anyone else who wants to raise their hand on a last note thank you I agree this has been very very enjoyable so thank you Daniel Scott thank you for the nice metaphorical pictures so I fully agree that there's this beautiful notion of the but I think it's also interesting to first distinguish reflexive feelers and intentional or expressive feelers right so how do you navigate to the world and in that challenge how do you navigate yourself and so I think personality or yourself model regarding you as a person are more of those reflexive feelers and then subsequently you are more able to grip or control or regulate the way you navigate the world with your perceptual feelers or specifically the recognition models by which we are able to take in the world because they are what are being compared with the sense data and you also said something very nice about I really like Carnot engines as a picture I fully loved that you mentioned that and you mentioned information differentials now information differentials is exactly what active inference is about right so the beauty is also there's a notion you mentioned entropy a lot but there's also a notion of relative entropy which is also known as the Kublack library which is exactly underpins the entire formalism because exactly like the Carnot engine you want to minimize the temperature differential but it brings temperatures constant right similarly the energy consumption doesn't vary with more than the order of magnitude and so the interesting question here is okay to what extent is it like a Carnot engine to what extent does the minimization of information differentials play a part in its functional behavior so one very nice notion that I like which I got from a friend is coordinated equilibration so both in the case of normal conventional thermodynamic equilibration such as in Carnot engines as well as in my opinion for active inference you're basically equilibrating the whole collection the complex covariance and co-conditional collection of information differentials and so no matter what by virtue of it being physically instantiated it needs to equilibrate physically which has some relation with informational equilibration so in other words the one thing I like to say is basically we're letting nature do the work we're already nature it has to equilibrate and it happens that that finding of equilibria is in itself homologous or analogous to particular algorithmic associations which if you're interested is also interesting in computer science because apparently there's some nice works that connect icing spin models with self-organized criticality how the brain operates via avalanches or cascades of activity apparently these are formally implementable to solve basically most famous algorithmic challenges so I think hopefully you might find that interesting and so I'm not sure what your exact phrasing was but I want you to know that I think it's also a nice picture to imagine that what we're doing effectively oh yeah you're talking about negentropy so the effect of what you're doing is basically you're partially deciding how you're going to be imprinted by the world and that's the whole power of the whole predictive picture because the patterns that are out there in the world are effectively the challenges themselves so when we internalize the patterns then we're more gripped or attuned or have a bigger affinity with the world so you're exactly right in that regard sorry I forgot the rest of what you said and in regards to Adam's beautiful points about the pasticity of the personality I also very much echo this kind of criticism question of the whole assumption that personality traits is static, rigid but I think what he said earlier about coarse graining and what I called bootstrapping kind of maybe is a path to understand better how it shouldn't be seen as so static because if it indeed is a bootstrap in a time where it lacks self models and basically the accessibility of changing that initial bootstrapping model is inhibited or it's obstructed effectively right because if you want to reflect if you act upon these models then you need to have a path towards these models so that you have to act upon whatever influence your personality through layer and layer and layer from superficial to deeper and that's a big challenge and I think indeed things like toll pass things like visualization exercises are one example but I also really like Steven's notes on drama and more situated experiments situated design where I think the notion of porosity is also very interesting to what extent do you allow the environment to influence you and finally I just wanted to note that's because I mentioned it psychedelics yeah absolutely because the whole phrase is psychedelics the revealing of the mind but that's only one step the second step is reveal such that it becomes accessible to systematic change so yeah so those are my thoughts and thank you for giving me these thoughts nice it's really great how you tied so many of those elements together so in the last couple of minutes just quickly from anyone who wants to speak we have Steven then Scott then anyone else who raised then Adam then anyone who raises their hand I mean this is what this paper is so interesting about is the way that it gives these dynamic mechanism which is good for active inference but it's also also quite an intuitive way to break down all these generative models and different things it gets hard when you start to layer different areas I suppose the two things that come up when I was thinking about the comments on the drama or acting is I think sometimes the acting or having a chance to go and do a performative event or action that's more like you're going on stage and that could be like that mounting car problem I don't know if you see the paper of exploring a landscape using your model and seeing different possibilities so it could extend out the possibilities and that mounting car problem was quite good for looking at how that issue of expanding out can be important and then there's also and there's different ways for instance method action is actually can be quite dangerous a lot of people have psychological issues if they go down that route so there's basically putting on an act is like you're still a performer and you know that you're in a sort of a field of practice presenting and being aware of where the audience is how close they are what size of you they can see however just by doing that you will have that experience so you will have expanded your range your entropy of possibilities or states of the world there's also participatory type drama where you could revisit a real life experience and go back into through psycho drama social drama and actual experience that you had and look at that and then that also could give different ways to maybe understand your actual lived life now whether that would change the big five I don't know it may not be about changing the big five for someone it may be something else within that you know so I think it's really I just thought I'd add that to the great combination thank you Scott and then Adam then Colin just in the last couple of minutes this is great very rich discussion I like the fact that we raised more things and this is really the beginning of a discussion then we resolved and that's beautiful the tulpa really struck me because it's something where there's an imagination in an internal which is re valorized reified by the external in a in a process way and it really reminded me of law in law if you say you have a right but there is no duty of anyone else to respect that right like if I say I have a right to free speech or right to sit anywhere I want on the bus and no one else respects it I don't really just I just have words on paper I really need the duties and the rights and that dyad I think is really interesting that reification that valorization and it started to wonder whether it's it feels like something like behavioral catalysts or consciousness catalysts where you have that external and internal element are necessary for that coordinated equilibrium equilibrium that as Marco was looking to some fascinated by the discussion here I feel like there's we're just beginning and looking forward to other ways for us to continue because there's a lot of pieces that we still need to explore it feels like totally agreed so we'll have Adam then Colin and if anyone has thoughts on where we could explore next or what would be fun to explore they can mention it too Adam then Colin and a special thanks to both of you authors for coming for both of these conversations it was just a really made the whole thing come together so Adam Colin blue I've been loving this thank you all of you I'm really glad Marco brought up criticality because it gives me an opportunity to hold forth for as what Colin knows is my religion the and I think this someone harkens back to last week's discussion of meditation and mindfulness so one of the ways I view mindfulness is a kind of like psychological manifestation of self-organized criticality so it's like there's this edge of chaos or this inter regime where you have enough variation enough disorder to give birth to a dancing star or rather to have variation for evolution to work with but then you have enough stability to get this cumulative evolution to have structure build upon structure and so on and so if it's the one regime where evolution is possible yeah this is a multi-scale phenomena this is probably the universality class like you'll find it everywhere at all scales and so at the psychological level whatever you're really doing in some sense I think it's going to be involved with respect to optimality and if one of the things you're doing is trying to like for instance try out something new and like regulate like the amount of psychological entropy you're experiencing this basically setting up for yourself this return to this metastable place where you can balance stability and plasticity where you can sort of ride along the edge of chaos and not fall in or go too straight too far away from it this is I think if you want to either maintain your personality or change your personality or work within the bounds of it this seems to be really essential and this final thing is I've been increasingly thinking like the way we think of mindfulness is like all wrong and that rather than thinking of it as like this is a really nice state to be in and you're watching your breath and everything is like super great that's good but like more like the face space description and more as process and rather than like a thing you get it's a return to like a kind of a necessary home and really an essential part of cybernetic functioning of balancing stability and plasticity that you're going to need for anything Thanks so much Adam Colin Ben Blip I'm glad this whole idea of the edge of chaos came up because Adam earlier you said at some point we're talking about active inference so we're talking about free energy and we don't like surprise and of course as soon as I hear that you know what I think well no we do like surprise we seek out surprise often we like surprises we like uncertainty we like risk people enjoy these things that's exploration and the reason that we have both one I think maybe the most important fact in psychology that we have an innate ambivalent response to uncertainty and increase in entropy which is simultaneously positive and negative right we obviously don't like surprise we fear entropy we need things to be predictable or everything disintegrates but at the same time we like surprise we like uncertainty we like exploration and if we don't do that then we don't adapt and eventually we get too rigid and everything disintegrates right so you have to seek out entropy in order to avoid eventual catastrophic increases in entropy right you have to seek it out and master it and so that's why you need to be at that line between order and chaos right between order and disorder and you know as Adam's pointing out that's what evolved systems evolved toward is toward the edge of chaos maximizes the possibility for computation maximizes the possibility for adaptation and there is to make a segue from the abstract to the more concrete because I feel like I sort of like what my collaboration with Adam it's like I'm time is to more concrete psycho psychology is time to more abstract theory there's a lot of work in the last 20 years in personality psychology on personality change and especially personality trait change one of the things we know from this research is that virtually everyone has at least one personality trait that they would like to change we also know that personality traits can and do change over time it might not always be easy to change traits they might not always change you know dramatically but we do see change in personality traits over time we know that things like therapeutic interventions change personality traits especially reducing neuroticism and personality psychologists have recently been trying to develop systematic programmatic interventions to help people to change different personality traits and of course a lot of it is through first just changing more specific habits more specific characteristic adaptations because that's what as Adam said you have more of an immediate grasp on is how to have to change your habits how to change your particular strategies but it is possible for those to generalize to these more general changes and I would think of that as you know you you learn specific things but eventually those things that you learn may actually tweak some of the underlying general parameters that actually enact those specific things over time thank you Colin for those who need to drop off thanks for coming just to respect the end of the official event but blue Dave and then we'll chill for a couple more minutes but thanks everyone so much for coming today so I want to thank the authors especially for for being here and live ending up this conversation but just as far as things that I think we could continue talking about or that would be interesting to pursue in the active inference framework are things like personality change or things that do change the personality right like trauma therapy psychedelics mindfulness meditation like all of these things can really have drastic effects on who a person is fundamentally I expect in like terms of big five as well and then also like maybe in a non active inference kind of way like the discussion on consciousness is is big and heavy and maybe not for this live stream but something that is definitely always worth talking about thank you blue Dave yeah Sean Carroll interviewed Carl Friston a few months ago and he brought up I think the same point that Professor DeYoung just mentioned is wait a minute we don't try to be certain about everything we want novelty we want we want to be confused and the professor Christian became even slower and more methodical and self critical and self qualifying than usual and really went I think was working through all these levels of mathematician that I'm sorry in my humble opinion he's kind of gotten trapped into that's something that Professor Edelman was terrified as this incredibly brilliant young kid and he's going to get his brain sucked by the algebra thank you Dave um you know it reminds me it's controlled novelty we can't just say more it's more is different is what we say in complexity and when it comes to cybernetic goal-driven edge chaos systems it's not so simple as just take a linear aggression and you know more ants is better for the colony no I mean selection has clearly in different niches resulted in colonies that have different shapes of everything so it's a local vector of selection but then the broader space is actually not quite so linear so I hear you on that Marco yeah thanks I'm glad to mention that podcast is actually in my opinion the best conversation with Carl because Sean being also a professor Carl also being a physicist he really was able to get Carl active and it's more playful and cheerful about talking about it but I'm not sure what you mean with the with the getting more quietness of critical I think it's more awareness that that's really the the main challenge in dealing with these kind of seeming tensions or paradoxes because I do think that's a very important attention to point out when it comes to formal modeling especially when it comes to formally modeling brains or humans as we are all all too familiar with which concerns as seen for example in the personality model and they're often these complementarities these trade-offs and often in folks ecology in normal communication we tend to pick a trade or pick a characteristic and just say that's good or bad it's always decontextualized it's rarely in relation to the trade-offs in relation to which they obtain an actual normative value normative value and probably something else oh yeah so I linked it on the YouTube chat but I really personally am very fond of the theme of pleasant surprises pleasant perturbations and this controls these control unknowns for example Stephen's gone out but he mentioned the idea of spaces right so we could imagine well first should be more formal so formally that's already resolved in 2017 a paper I was happy to be the co-author on because active influence doesn't concern itself with free energy per se it's expected free energy right and the more sophisticated your models the more sophisticated your temporal depth and your counterfactual imagination imagination is very formal but I like allowing us to talk about imagination as if it was formal right kind of liberating is for the folks of psychological connotations and way too many centuries of philosophy kind of dunking on this right because it's simply because we didn't have the grounding so it takes exactly in active influence that these tensions disappear if we allow ourselves to accept what the mathematicians say but there is this kind of trade-off but there is this kind of complementarity that indeed it is kind of the hippie notion of yin and yang and so it's no surprise that stability and toxicity is a method straight in my opinion and oh yes sorry and I linked that paper from Haileche who is at the free university of Brussels and a pretty prominent researcher and he has a very interesting notion which quite like about challenge propagation so that's continuing on the theme of pleasant surprises because we design challenges for each other social situations games, work all these cultural niches are designed challenges simply by virtue of the fact that they were selected by our social culture dynamics we allowed these to persist throughout time in culture knowing that in their persistence they enact positive effects on us as we try to adapt to them right and so in that sense I find it very very beautiful that these kind of controlled unknowns is controlled chaos is such a prominent theme in how we design each other and allow each other to cultivate and in that sense another theme we co-regulate each other and in the co-regulation of each other we develop these kind of super personal personalities which I think well hopefully people can agree that we sometimes really feel that viscerally the kind of personality or vibe as millenials us like to say and that's a certain vibe, that's an area and these vibes tend to resonate with their personalities because we already have these models of this kind of strategies goals and values we instantiate for particular situations and so it's not for a fact that the same models afford us also sense certain personality like themes in situations because that is literally the kind of brain system we use we use a brain system of persons to do this the brain system there was a letter from the area we have next to each other the brain area activated when we do self modeling and next to that's modeling other people and so that is also exhibited in how easily we attribute personality or intention to inanimate super simplified objects right and that speaks to in my opinion the validity of the core assumptions in this paper that personality can be decomposed and indeed that's an effect of values strategies and goals sorry I went on a ramble I was going to say but yeah our relational architecture the way that we talk about organic chemistry and this chlorine wants to do that and this galaxy wants to do that the intentional stance that relational thinking is so deep and the way that you put it with the super personality it's like the Superman is the collective leviathan it's not just the better individual nest mate it's that and in the context of something that's actually super or supra like above not just better Scott and then anyone else who wants to make a final comment but this has been really great Marco this is I mean that was so beautiful he's starting to think of the choreography of consciousness you know that's what it felt like you know it's such a beautiful notion I always used to tell my kids now they're in the 20s I used to say the only time you see reality is when you see paradox and any other time you're inside a model and the paradox and that goes right the paradox is the reality that's the reality of our interaction you know you're interacting then you know you're encountering something and so that notion of embracing the paradoxical and the dance and the negotiation of it it reminds me the last point I was going to make is it reminds me of the research they did on fish swimming they took a semi-rigid piece of plastic and they put it in a stream and they started increasing the speed of the stream and at a certain speed of the stream the plastic undulated with the periodicity that was akin to how fish swim and they realized that fish don't swim through water they pass through the water and because the water is causing them their swim similarly that choreography of consciousness is that right we're not we don't just situate our cognition we situate our persona and it's constructed externally and internally so the last thing I wondered about is is each system gadelion so each system is incomplete has questions within it that can't be answered from within it and maybe our novelty seeking is an intrinsic recognition that we're all gadelions that we need to go outside to find answers so anyway great discussion love it thank you guys thanks as always coming on Scott so let's just go from Dave, Colin, Marco and then we'll probably call it unless someone else has a burning desire the only time we need to be conscious is where we're in that zone of paradox we ought to understand it and we don't nice Colin well on the same theme and I really like that idea of paradox being reality too and that's the same as the primal materia of the alchemists right it's what is the thing behind everything that gives birth to those things and it's precisely the thing that we don't understand it's the thing that's unknown right and so that's the chaos that's on the you know the other side of the border where we're at the edge of chaos and yeah from especially from the perspective of consciousness what you need to be conscious of is precisely only those that you don't understand perfectly already Marco? yeah Scott thank you that was beautiful yeah I just wanted to riff on that so I laughed because I feel comfortable saying this here so you know me and my friends we were interested in psychedelics so I find it very very fascinating that there's a huge variety in what I call idiosyncratic encounters so a lot of my friends with the psychedelics have this one thing that they feel like oh wow that was so fundamental and one of my friends let's call her Lutin she had exactly that right she said I couldn't escape the idea that everything is paralyzed right and though I won't go into it but every other such quintessential counter idiosyncratic powder also in my opinion relates to something formalizable in terms of active inference and I really really like a notion of the narrative of dance I use the metaphor a lot too you might also be interested in Fritjof Kaplass' work who also wrote the Tao of Physics and he also had this moment of insight about physics simply being this dance right he used the metaphor of Gali deity in Hinduism but more formally and interestingly well my attempt to formally make it interesting is as we noted right cybernetics acts upon configurations of models simply because systems are composite compositional distributed and constantly reorganizing so the very least I can say that it's about configurations yet going back to an earlier thread if conscience indeed is intermediating cybernetics then they need to intermediates the balancing or the metaconfiguration of configuration which themselves are constantly changing so exactly in the example you took with the fish they are embedded or immersed in fluid dynamics right there are flows and I say that formally these are informational flows so they're immersed in all these flows and they need to find or the space needs to find the space needs to situate itself in between those flows and by virtue of doing that it is regulating them right and so I think that's a very beautiful idea and the last thing I think is also very beautiful so we all go deli and we're all incomplete which reminds me of conversations I've had with Simon De Deo at SFI Institute very very good researcher very interesting has some very very nice work on Bayesianism and the nature of knowledge and I think I said to him exactly what you said but how I phrase it is we're all each other's oracles so an oracle and if you know in computer science but they give you an answer that you can't actually compute but if you would have the answer then things would come easier so it is by virtue of the no way complete and that is allowed because other people's incomplete systems can act as oracle for us and so I think that's also something that speaks to our experience of how we are aware of certain things but it's in fact actually very incomplete but we trust that it's okay to be incomplete and if you're interested in that then I'm sure you know Hofstadter, Gilles Scherbach Terence Dieken Incomplete Nature as Adam mentioned and Deo as I mentioned so yeah thanks so much that was very stimulating I hope you met interesting I close it with William Blake quotation he wrote in his notebook do what you will this life's a fiction and is made up of contradiction very beautiful I applaud to that peace out everyone thanks for participating thanks for watching it hope to like have more participation join the stream if you're not on it already and we'll see you next time thank you so much for watching this bye thanks everyone