Discussion with Michael Johnson 1
Thoughtforms Life Podcast / Michael Levin
April 17, 2025
Source: https://thoughtforms-life.aipodcast.ing/discussion-with-michael-johnson-1/
Canonical transcript: https://opentheory.net/transcripts/thoughtforms-life-discussion-1/
Machine-generated transcript · not human-reviewed

[00:00:00.000] uh yeah so um i was thinking uh just kind of uh could have a conversation about um uh sort of intersections between your work and my work yeah um i think that uh like i'm i've been thinking so much about distributed stress minimization uh i think it's just an absolutely beautiful frame um and sort of the the various um fields of of the body we can say um and so i'm coming from
[00:00:30.000] the perspective of uh sort of uh this research into sort of formal research into consciousness we can say and then uh this frame of vasocomputation this idea that um tension in the vascular system essentially uh sort of holds bayesian priors about the the appropriate range of the neural system uh so yeah i guess uh to begin like i i
[00:01:00.000] i'd love to hear uh distributed stress minimization from uh directly from you uh what's what's going on there yeah yeah um so you know as you know we're interested in uh mechanism various cognitive mechanisms and really unconventional substrates and uh i've been thinking a lot about uh most most recently about um how to measure stress in unconventional systems
[00:01:30.340] and in particular the use of for for us uh the use of stress as a uh stress reduction as a driving variable for um anatomical homeostasis so the idea of cells and tissues having to adjust and uh both both anatomically but also transcriptionally and uh you know physiologically to instantiate specific goal states that that they have that then pulled you know and so this process pulls them along so um our latest our latest frame of
[00:02:01.000] still unpublished, but what's coming out now is that we can use various stress markers
[00:02:07.680] to identify, to basically show how these loops, in embryogenesis, you sort of go stage by
[00:02:17.180] stage and we can actually see the stress.
[00:02:19.920] So at the beginning, there's a particular target morphology and we study the one that's
[00:02:25.200] set by bioelectrics.
[00:02:26.100] There may be others, but we study the bioelectric one and the target morphology doesn't match
[00:02:31.280] the current anatomy at all.
[00:02:32.480] And so the stress goes up and it works like hell to get there.
[00:02:35.160] And when it does, it goes down.
[00:02:36.240] But by then the bioelectric state has moved on again.
[00:02:38.520] So it happens again and again and again, and it keeps pulling it along through all the
[00:02:42.480] stages and then eventually it sort of equalizes.
[00:02:45.060] And then we can talk about aging and things like that separately.
[00:02:47.440] But this is the kind of thing we study.
[00:02:50.720] And in particular, what I wanted to also discuss.
[00:02:55.460] First of all, I want to talk about the bioelectric.
[00:02:56.100] I've told everything that you've seen in the vascular system I want to hear about, but
[00:02:58.540] also how do we measure it in really weird systems that are not biologicals at all?
[00:03:04.180] So what does what the stress look like in the gene regulatory network?
[00:03:07.260] What does it look like in the physiological circuit?
[00:03:09.260] What does it look like in in other kinds of processes?
[00:03:11.880] Some sort of generic metric that we can apply across substrates?
[00:03:15.720] Right.
[00:03:16.680] Yeah, fascinating.
[00:03:17.660] So I was thinking yesterday about how sort of this there's this sort of distributed stress
[00:03:23.220] minimization they talk about.
[00:03:26.100] if a cell isn't in the right place, then there'll be stress and like cells kind of
[00:03:29.660] emergently coordinate to sort of get the cell into the right place. And kind of the bound of
[00:03:36.080] a system is the empathy of the cells in some sense. And I was thinking about how different
[00:03:42.980] systems may have sort of different currencies of stress. So first of all, I'm like, how does that
[00:03:49.920] work? And sort of what is stress in endothelial cells and what is stress in neurons and what is
[00:03:57.260] stress in glial cells and so on? And how do they communicate? And I guess I had the loose hypothesis
[00:04:04.780] that there's sort of an inter-system sort of stress exchange. And it would be interesting
[00:04:13.900] to see if one of the kind of core levers the body
[00:04:19.920] has to sort of adjust, we can say mood or strategy or Bayesian priors or whatnot, is sort of adjusting
[00:04:27.980] the exchange rate of stress between systems. Like maybe in a rest and digest mode, stress in the
[00:04:37.120] stomach gets weighted 5x normal. So it's like stress gets distributed, but the stress that the
[00:04:44.020] stomach is holding sort of gets pushed into other tissues, where maybe in a fight or flight mode,
[00:04:50.500] like that's reversed or something like that. So just a side note there. I can say that...
[00:05:00.000] think that you're touching on something that feels extremely important when you talk about the cognitive glue and things that bind the system together, we can say.
[00:05:17.900] And I would also say that, like, to sort of bring in vasocomputation, one kind of core hypothesis that I'm sort of trying to poke at is that it feels like it's kind of a very efficient compression to think of the vasculature as an agent.
[00:05:44.800] And, you know, we can say that vasculature is a very efficient compression.
[00:05:46.880] And I think that the body is kind of this amalgamation of agents, all kind of emergently cooperating.
[00:05:53.400] And I guess, like, I think of the vasculature as it's like, to what degree can you sort of ascribe different personalities to different agents of the body?
[00:06:05.720] I think it's a very interesting question.
[00:06:08.440] And so I guess I...
[00:06:11.000] So the sort of vasocomputation baseline is that...
[00:06:17.580] Vascular tension stabilizes local neural patterns.
[00:06:21.640] Or like whatever the sort of local substrate of compute is, we can say.
[00:06:25.720] That's the slightly more general frame.
[00:06:29.220] Vascular tension basically freezes the patterns there.
[00:06:32.720] And so it's sort of a neat way to sort of hold certain things as constants where there's tension.
[00:06:41.420] And then other things can be left as variables where there's not tension.
[00:06:46.880] But the other thing is...
[00:06:47.200] Vascular muscle, of course, is a form of smooth muscle.
[00:06:50.500] And smooth muscle has this latch bridge mechanism.
[00:06:53.480] Where it can basically glue itself shut.
[00:06:56.680] Like the action of the myosin basically stops sliding and gets chemically glued.
[00:07:02.420] So then...
[00:07:03.580] This can persist.
[00:07:06.200] And like...
[00:07:07.720] I guess from...
[00:07:08.720] My expectation is that these sort of latches...
[00:07:12.780] When...
[00:07:13.680] When vascular muscle sort of engages this latch.
[00:07:16.880] bridge mechanism and these maybe it's sort of actively defended or actively repressed or
[00:07:22.360] there's some sort of like metadata from the body like we need that then I expect that this can last
[00:07:31.340] for you know anywhere between minutes to hours to potentially decades so kind of a very long-term
[00:07:39.380] prior and then of course it's like vascular tension is sort of gatekeeping the body's
[00:07:49.160] central currency blood flow and so if neurons don't get blood flow they just don't have the
[00:07:56.800] the like resources to adapt to rewire etc and so I've been you know thinking about that as sort of
[00:08:08.240] um
[00:08:09.400] kind of a backup or kind of a uh almost like a an RLHF uh system for the neurons that um if if some
[00:08:19.660] like big prediction error happens or whatnot uh the vasculature can kind of leap in and and say
[00:08:26.020] okay like that didn't work we're not gonna let that happen again if you get bitten by a dog or
[00:08:30.400] something and um it can kind of latch uh latch patterns into kind of a known safe mode or
[00:08:37.660] um
[00:08:39.380] um maybe bring in this active inference frame like tension as a prediction
[00:08:44.060] uh so you can say um you you latch a prediction that you will be safe in a certain situation
[00:08:51.880] but of course that reduces overall system dynamism uh you basically block out certain
[00:08:59.460] um parts of your dynamic range which which has downsides
[00:09:06.020] um so I guess like
[00:09:09.380] um the the story that I want to tell is that you know we're sort of this um amalgam amalgamation of
[00:09:17.140] systems and these systems are sort of amalgamations of smart parts as as you've described and then
[00:09:24.920] what I want to say is the vascular system or like people think of intelligence as uh sort
[00:09:31.120] of embedded in the neural system and I think your your critiques have been absolutely spot on that
[00:09:36.840] and like even neuroscience there shouldn't be any kind of
[00:09:39.380] maybe about neurons uh only um so I want to say that maybe but like neurons are sort of specialized
[00:09:48.200] for long-range communication um and I guess I would propose sort of looking at uh smooth
[00:09:57.220] muscle cells in like with fresh eyes
[00:10:00.000] are these cells? What are they doing? And what is their functional role? I guess I see them as playing neurotic protector to the neural system. If the neural system can't handle something, the basal muscular system jumps in and tries to make it that.
[00:10:28.000] Make it manageable.
[00:10:30.600] So I want to pause there.
[00:10:32.660] Does that make sense?
[00:10:33.840] Yeah, absolutely.
[00:10:35.620] Absolutely fascinating stuff.
[00:10:37.220] Really important, I think.
[00:10:38.880] Do you think...
[00:10:40.220] So I was thinking about something a little related recently in terms of gene expression.
[00:10:47.200] So you have 20,000 different genes or whatever, and you don't have the metabolic resources to transcribe them all.
[00:10:54.320] And so one way to think about it is that the cell is making decisions about what to transcribe, fine.
[00:10:59.960] But another way to think about it is that could the genes actually be in competition for the attention of the transcription machinery?
[00:11:07.080] Right?
[00:11:07.560] So if there is this obvious overseer that is going to decide who actually gets transcribed, you would think that there would be some forces to start to hack it, right?
[00:11:19.100] Where the genes are trying to get the attention of the system.
[00:11:22.720] So my question, and you can answer it.
[00:11:24.320] And this is your case, is do you think that neurons, in addition to doing whatever it is that they're doing, are actually trying to get the goodies from the vascular system?
[00:11:36.280] Is that a thing?
[00:11:36.720] Are they trying to hack the vascular cells at all?
[00:11:39.420] Yeah.
[00:11:40.020] I mean, they'd have to be, right?
[00:11:42.460] And I think that the powerful thing here is the vascular system is holding the purse strings.
[00:11:52.340] Yeah.
[00:11:54.320] like if the neurons don't get blood uh there's just a hard cap on sort of what they can do and
[00:12:00.200] i expect that they can i i expect that if we poke into neuron physiology we'll find that
[00:12:06.560] uh there will be like uh i'm not sure if it'll be continuous or discrete but they'll sort of drop
[00:12:12.460] into safe mode in a very clean elegant way like they're they're sort of designed to sort of uh
[00:12:17.900] go through metabolic winter uh we can say um that there was this um uh this sequence of papers um
[00:12:26.560] sort of starting with the the hemoneural hypothesis by by moran cow um and then moving into
[00:12:33.120] i think cognition is entangled with metabolism jacob at all and talking about how um
[00:12:40.120] like a couple of kind of crazy uh unexpected findings uh one being that um uh blood flow
[00:12:47.900] in in healthy tissue can vary by uh over a factor of 20. um so not 10 but 20x plus and so
[00:12:58.380] um we should expect sort of cells to behave like have different sort of modes of operation
[00:13:04.580] depending on that um and then the the second finding was that um actually changes in blood
[00:13:11.100] flow precede changes in neural activity um that it's not the neurons you know fire and use a bunch
[00:13:17.720] of uh you know uh uh uh uh uh uh uh uh uh uh uh uh uh uh uh uh uh uh uh uh uh uh uh uh uh uh uh uh
[00:13:18.900] it's that um the blood changes like the blood comes and then neurons change um yeah so so you
[00:13:30.640] know that sounds like um you know it would be it would be really interesting and maybe it's
[00:13:35.340] already been done but it would be really interesting to do some kind of a a multi-scale
[00:13:39.940] neural network model where the cells are age where the the neurons are agents that also you know
[00:13:48.900] there's some there's some pressure to compute whatever it is but there's also right if you want
[00:13:52.740] to do the computation that you're supposed to do you also have to navigate your local environment
[00:13:57.600] and get get the purse strings uh so to speak to uh to feed you right yeah yeah um interesting and
[00:14:06.300] i guess i'm sort of pushing on that uh
[00:14:10.620] i do think that like a very interesting sort of near proxy for attention maybe blood flow uh sort
[00:14:18.900] of turning up uh more cow describe it as sort of turning up the sensitivity or the gain on
[00:14:23.640] networks um and it'd be interesting to see okay like how do neurons sort of compete for blood flow
[00:14:31.540] and how do the algorithms of the body sort of punish defectors on that um yeah fascinating um
[00:14:41.820] i think uh yeah i'm i i guess like the the like
[00:14:48.900] cooperation deflection dynamics uh get get pretty deep pretty quickly
[00:14:54.540] are there um by the way uh something else you said about the the different personalities
[00:15:00.160] of the body and everything.
[00:15:02.520] I just, just an hour ago, I had a conversation with Frank Putnam and Alexey Tolchinsky,
[00:15:09.340] and they were talking about this business of dissociative disorders and the different
[00:15:15.540] personalities that can live within one body and how they navigate their relationships.
[00:15:19.960] And, and one, one of the, one of the most interesting, the thing, interesting things
[00:15:24.480] that he said was that they have shifted clinically from, instead of trying to integrate, what
[00:15:31.760] they now do is provide, they, they, they literally have a, they, they make a bulletin board where
[00:15:37.360] the different personalities can leave each other messages.
[00:15:39.880] Wow.
[00:15:40.340] And so this is actually, it, it's, you know, you achieve this, it's, I mean, it's
[00:15:44.460] a stigmatic medium where you get to like leave these messages for each other and communicate
[00:15:47.920] that way.
[00:15:49.960] And so the collective becomes more, more functional and integrated towards goals, but
[00:15:55.200] it's not the thing where you hope the pieces disappear in favor of the, of the whole, right?
[00:16:00.540] Right.
[00:16:01.120] So I wonder, do you, do you see in these vascular networks, do you see, how many agents do you
[00:16:08.140] see?
[00:16:08.360] Do you see, you know, multiple agents at the, at the same level?
[00:16:11.140] I mean, of course there's multiple levels, but, but at the same level, are there regions?
[00:16:14.900] Are there, is the whole thing kind of tightly integrated?
[00:16:18.220] Has anybody done things like that?
[00:16:19.960] Like, like you know, any kind of causal emergence metrics on the data, that kind of stuff.
[00:16:24.680] Right.
[00:16:26.240] I haven't I haven't seen that.
[00:16:28.220] That'd be fascinating.
[00:16:29.380] My expectation is that a lot of this logic is local that and, and maybe may sort of have
[00:16:38.140] sort of local organs as kind of the big attractor or like kind of the, the sever semi-sovereign
[00:16:45.100] domain.
[00:16:45.560] I guess I would say.
[00:16:47.100] And then sort of the, the tissue.
[00:16:51.320] that domain. So the stomach being a domain and the heart being a domain, brain being a domain,
[00:16:58.600] and so on. So yeah, it'd be fascinating to dig into that. But the sort of mechanisms of how the
[00:17:09.620] muscle cells are organized are like the vasculature is not one big muscle. It's like a
[00:17:15.300] bunch of tiny muscles. So yeah. And I guess part of this thesis is that a lot of the challenges of
[00:17:31.620] being a human living in modernity is that sometimes these vascular clenches and latches,
[00:17:42.080] these sort of crystallized priors,
[00:17:46.520] they're too sticky. An absolute central challenge for humans is how do we release this tension?
[00:17:58.800] It's like not all of it gets properly garbage collected, we can say. And we can say, okay,
[00:18:05.160] no, I got to pick up my daughter from school. I'm going to remember that. And that's sort of
[00:18:09.960] instantiated with a prediction slash tension.
[00:18:15.300] certain part of my, my nervous system. Um, and, and, you know, the, the vasculature reaches
[00:18:19.900] everywhere, you know, anywhere there's blood flow there, anywhere there's neuron neurons,
[00:18:24.520] there's blood flow and anywhere there's blood flow, there's this, uh, this muscle, but, um,
[00:18:31.020] but, uh, you know, I, I suspect that as we go through, uh, our day making predictions,
[00:18:37.200] making active inference predictions, um, we sort of naturally like, uh, you know, uh, clench and
[00:18:45.900] it's kind of seamless, uh, but sometimes sort of cruft builds up that if we, if we look at this
[00:18:53.420] as kind of, um, a form of side channel memory, um, that, uh, of the, the proper dynamic range
[00:19:00.560] or like, you know, don't use a big part of the range or like, uh, remember this pattern,
[00:19:06.080] or whatnot, then, um, you know, over time you'll just naturally get, uh, situations where
[00:19:14.240] you're
[00:19:15.300] uh, the resources aren't released to the system. They're just, uh, you know, you, maybe,
[00:19:22.640] maybe you were interrupted in doing a task and like your, your body was sort of holding something
[00:19:28.720] and it never got the like task completed trigger such that it could release. Um, so my expectation
[00:19:37.700] is that, uh, this is a pretty clean description of, uh, first of all, um,
[00:19:45.300] what we can call trauma, uh, the sort of, you know, uh, shards of information sort of, uh,
[00:19:52.160] held by tension in your nervous system. And like, uh, I have a friend, uh, Warren winter
[00:19:57.420] who, um, described, like,
[00:20:00.000] explain what Wilson's affordance was.
[00:20:02.420] Like, as we perceive something, we don't see it as it is.
[00:20:08.140] We see the object in terms of what I can do to the object and what that object can do
[00:20:16.900] to me.
[00:20:18.380] So if, you know, if you see a dog, maybe you see, oh, like, I could pet that dog.
[00:20:24.640] But also, oh, that dog could bite me.
[00:20:27.960] And of course, if something bad happens, then kind of the negative sort of Wilson affordance
[00:20:33.900] might get latched into the system.
[00:20:37.240] But yeah, I guess, like, I see it as kind of a neat explanation of trauma, which would
[00:20:45.160] be pretty exciting if, you know, this is kind of the home system where that lives.
[00:20:50.500] But also, just in terms of, you know, I've been kind of, you know, I've been kind of
[00:20:55.420] trying to dig into the Buddhist frame for a while.
[00:20:59.700] And it feels like, you know, what they what is talked about in terms of sort of, you know,
[00:21:07.180] there's sort of this self, this sort of immutable, you know, set of constants that sort of prevent
[00:21:15.340] maybe a wider aperture on reality, you could say, that this might be a particularly good
[00:21:24.260] way to explain that as well.
[00:21:25.820] And that it's, it's not that, you know, if we had a magic wand, and we could open all
[00:21:30.340] your vascular attention, like good things would automatically happen, like, probably
[00:21:34.040] some of it is very load bearing.
[00:21:37.240] But I do expect that, you know, if, if we sort of, like, if we were to track the sort
[00:21:47.860] of fine vascular tension of meditators, as they kind of go through the path, we'd see
[00:21:53.240] some sort of
[00:21:55.000] opening where maybe some networks that were sort of latched when you were two,
[00:22:01.820] when you were three, et cetera, may reopen. And sort of you get this much wider sense of
[00:22:10.680] possibility. So those are the two applications that I would be very excited about.
[00:22:17.960] Yeah. Yeah. You know, Eric Hull in our center has been developing some more recent new metrics
[00:22:26.640] for things like uncovering new levels of causality and so on. I wonder,
[00:22:35.860] presumably there are lots of data that this could be applied to, yeah,
[00:22:39.580] to see what is actually going on in that system as far as integration and so on.
[00:22:46.540] Yeah. I've been...
[00:22:47.960] I've been a huge fan of Eric. I think his EC 2.0 looks really good. So yeah,
[00:22:55.400] I'll have to think about that.
[00:23:01.600] And, you know, another interesting thing to look at maybe, and I don't know how hard or easy it is
[00:23:08.260] to do these in humans, but... And also there's not that many of these patients, but there are
[00:23:13.600] some really interesting exceptional human cases where people have
[00:23:17.200] very, very high levels of causality. And so I think that's a really interesting thing to look at.
[00:23:17.960] Very diminished brain, brain volume, and yet normal, normal cognition. And I would be
[00:23:24.940] interested to know what's going on, right, in the vasculature and the muscle in those, in that,
[00:23:29.840] in that case, is it like taking over some of the processing? Does it matter?
[00:23:35.920] Right. Right. Yeah. I mean, one question that seems like it's at the intersection of what I'm
[00:23:43.620] looking at and your work is this idea of...
[00:23:48.240] distributed stress minimization between types of contractile tissue in the body and also neurons.
[00:23:56.860] So, I mean, I do have the expectation, first of all, that like this system is trying to minimize
[00:24:04.600] net expected stress. I mean, if you see a tiger and like your stomach clenches up and, you know,
[00:24:12.660] it's like it is a high stress state, but it's lower stress than being eaten.
[00:24:19.220] And so I think like it's kind of trying to multiply stress across, you know, we can say
[00:24:26.120] Bayesian futures or whatever and trying to sort of minimize total stress. But I guess I have.
[00:24:36.000] Hmm.
[00:24:40.540] Oh, I lost my train of thought.
[00:24:46.140] Yeah, I mean, you know, Richard, Richard, Richard Watson and and Chris Buckley and those folks have
[00:24:54.640] done some models. I don't know how much of it is published, but they've done some models on
[00:24:59.620] the
[00:25:00.040] computation in networks of springs and the physical stress dynamics and how they allow the network to compute.
[00:25:12.080] So I think that's very interesting.
[00:25:15.460] And I think this intersection of stress as it's understood in cognitive science, stress
[00:25:23.840] as it's understood in biology and physics, and also in mathematics, right?
[00:25:30.260] So geometric frustration and so on.
[00:25:33.300] I think it would be very, very interesting and helpful to actually connect all of those
[00:25:39.980] because I guess the null hypothesis is that those are all different things that we just
[00:25:43.840] call it stress.
[00:25:45.460] But my suspicion is that it actually is all under...
[00:25:47.960] I mean, I think geometric frustration is real frustration.
[00:25:50.840] So on the very end of the cognitive spectrum, like I think it's actually, right?
[00:25:54.960] That's alignment, misalignment of your parts is probably a fundamental aspect of stress
[00:26:01.200] and frustration in composite cognitive systems.
[00:26:03.940] Yeah, absolutely.
[00:26:05.720] And so I remembered what I was going to say, but just a quick note.
[00:26:13.760] So in...
[00:26:15.460] In 2016, I wrote just a short book, Principle Equality, that laid out what I call the symmetry
[00:26:24.760] theory of balance, and this idea that if we have kind of a formalism for an experience,
[00:26:31.300] and I mean, this is a hard thing to construct, but if we sort of had a perfect mathematical
[00:26:36.820] representation of what it feels like to you or what it feels like to me, then the symmetry
[00:26:43.780] of this mathematical representation, it's really important.
[00:26:46.460] would uh would sort of correspond to like exactly correspond to the pleasantness of the experience
[00:26:52.800] so basically it's kind of a formal mathematical uh way of saying harmony in the mind is the thing
[00:27:00.680] that feels good um and so yeah i what you say about uh geometric frustration uh definitely
[00:27:12.040] so so what do you think about uh
[00:27:19.180] systems you know simple systems in which everything is is aligned and harmonious so
[00:27:24.280] you know a magnet or something where everything is nicely nicely aligned um what's the you know
[00:27:32.120] is that is that for that minimal system is that can we say that it's somehow maximally that it
[00:27:37.540] has surveillance at that point right uh well
[00:27:40.760] i would say that um there's going to be a couple requirements um so uh one being the system has to
[00:27:48.840] be conscious uh so this like the symmetry theory of valence is not a theory of consciousness
[00:27:53.920] um it's a theory of valence um so if we if we can point to a conscious system then sort of
[00:28:01.800] um and then kind of construct a formalism for for what it feels like to be that system then I do
[00:28:09.220] think that symmetry and
[00:28:10.760] formalism corresponds like there's like an identity relation with uh with valence
[00:28:16.060] um but um yeah uh just kind of a simple magnet may not uh sort of may or may not uh lead to
[00:28:26.300] uh consciousness um I would also say that uh so I mean uh Eric uh studied with tony and like one of
[00:28:35.380] tony's frames is that um you need integrated information for consciousness
[00:28:40.760] um and sort of I do expect that there to be an interesting trade-off between
[00:28:46.990] um sort of enough complexity such that you have some integrated information
[00:28:54.270] and then enough simplicity such that everything is like nice and symmetrical and harmonious so
[00:29:00.940] um yeah that's that's maybe worth noting
[00:29:08.160] yeah um
[00:29:10.760] um in back in on the the sort of distributed stress minimization uh frame um I am uh I'm
[00:29:19.460] reminded of sort of the different types of muscles and uh I I'm a big fan of like um Joe Bullock's
[00:29:27.180] work on sort of maybe like muscles are a little bit different than we when we think they are and
[00:29:32.300] so on um and uh like I I do suspect that there's going to be this interesting sort of
[00:29:40.760] uh we could say leakage of stress from one muscle type to another muscle type
[00:29:46.620] um and uh so like if if you have like a lot of skeletal muscle stress maybe some of your smooth
[00:29:54.340] muscle um also sort of picks up some of that tension and and so on and that this
[00:30:00.020] may be like computationally interesting in a lot of ways, especially in so far as
[00:30:06.920] Michael Johnson, contractile tissue that's not sort of finely regulating the neural system bleeds into this vascular muscle that I think is finally regulating the neural system.
[00:30:21.920] Title, Zuzalu, and there's also fascia, which it's kind of a wildcard topic. A lot of people describe very interesting properties to fascia. I think it's very interesting that it's so electrically active. It's very conductive.
[00:30:39.880] Title, Zuzalu, and I guess one thing that I've been sort of wondering, and I'm really curious what your instincts say.
[00:30:48.020] Title, Zuzalu, So fascia can't exactly latch, but it can be a very conductive property.
[00:30:53.960] It's much sort of slower than VSMCs.
[00:30:57.280] VSMCs can operate on the scale of hundreds of milliseconds, whereas fascia maybe a minute or so to contract.
[00:31:07.540] But once contracted, they can sort of rewire, and that's the new default.
[00:31:12.620] And I guess I'm wondering, I suspect that VSMCs might have this fine regulatory effect on neurons, but what do you think fascia regulate?
[00:31:27.740] And what do you think contractions in fascia might sort of, quote unquote, latch?
[00:31:34.020] Yeah, boy, that's an interesting question.
[00:31:36.340] I don't know.
[00:31:41.620] Yeah.
[00:31:42.620] I don't know.
[00:31:44.100] It's really, I think, you know, one area that might be relevant to this is acupuncture.
[00:31:52.580] So I've seen some good, Elaine Alange-Van in Vermont has these really interesting experiments.
[00:31:59.060] She's got this full thickness skin tissue model where she puts the needle in and, you know,
[00:32:04.320] you get this lateral view of all the layers of the fibroblasts and everything, and you put the needle in and you sort of twiddle it.
[00:32:08.780] And what she shows is that the fibroblasts grab onto the needle.
[00:32:12.620] And then by twisting it, you're pulling, you're basically making these tensile forces that spread very long range.
[00:32:20.340] So I'm not an expert on any of that connective tissue, but it's almost certainly there's going to be some kind of mechanical computation there.
[00:32:27.700] Right.
[00:32:27.880] I don't know what the endpoint is, but there's got to be.
[00:32:29.860] I would assume nature is using it.
[00:32:32.680] I mean, one thing that Josh Mongard and I have been developing recently is this idea of poly computation, where the body is a collection of observable tissue.
[00:32:42.600] that is interpreting each other and all the physical events in every which way, right?
[00:32:46.900] So I would be shocked if a dynamic like that had no observers paying attention to it.
[00:32:53.320] Right. Interesting.
[00:32:55.020] Yeah. But I don't know what exactly, I mean, I don't know which processes exactly
[00:32:58.000] would are tuned into it, but I'm sure something's watching it.
[00:33:03.460] Yeah. Yeah. Nice. That's great. Oh, so one thing that comes to mind also,
[00:33:10.220] I think you're one of the experts, if not the expert on sort of the electric fields of the body.
[00:33:19.620] And I appreciated your somewhat recent tweet about, it's really hard to measure fields and it's often,
[00:33:29.580] you can measure VVAM, the membrane potential and so on. But so caveats about,
[00:33:39.670] it's,
[00:33:40.220] sort of hard epistemologically to approach this topic. But I'm really wondering sort of
[00:33:48.340] what's sort of the basic electrical layout of the body. So, you know, as I understand it,
[00:33:55.520] you know, you have cells have a strong membrane where there's, I guess like the
[00:34:03.700] charge density of the membrane is equivalent to like a lightning. So it's like, it's like,
[00:34:10.220] very small, but very potent for its size. And then mitochondria have even stronger cell
[00:34:15.720] membranes or mitochondrial membranes. And then you also have things like
[00:34:23.740] fascia, which conduct electricity. I understand that bone is essentially calcified fascia. So
[00:34:30.780] that's also very electrically conductive. You have things like, you know, wound healing seems,
[00:34:40.220] or like wounds generate a stronger electrical field, such that, you know, maybe that helps
[00:34:47.940] things kind of reorganize and heal. But I guess I'm wondering, like, from the perspective of
[00:34:58.940] consciousness research,
[00:35:00.000] where I sort of like electromagnetism is a sort of particularly
[00:35:06.980] interesting sort of way to sort of describe the organism and like whether or not consciousness
[00:35:17.500] sort of lives in the EM field you know that's that's a rabbit hole but I'm looking at this
[00:35:23.880] in terms of Wolfram's branchial space where sort of an object's true shape lives in this branching
[00:35:33.700] space of possible ways to decohere and so if like I guess I have the idea that a mind is kind of
[00:35:44.240] a shape in branchial space and I'm very curious what the body's shape in branchial space could be
[00:35:54.660] and I guess it's my expectation that the EM fields of the body
[00:36:02.100] would say a lot about what our sort of true shape is our our shape and branchientease
[00:36:08.660] so you know I don't exactly know the questions to ask you but yeah that's kind of the setup
[00:36:16.760] yeah so so I haven't I haven't yet figured out the relationship between my model and
[00:36:23.880] and Wolfram's model of that space.
[00:36:27.120] I think that what I currently think is similarly
[00:36:30.360] that there is a basically,
[00:36:34.440] what used to be called a platonic space of forms, right?
[00:36:38.540] And that what we are building when we make cells, embryos,
[00:36:45.480] biobots, AI's, whatever, is we're making interfaces.
[00:36:48.680] Yeah, we're making interfaces to specific patterns
[00:36:51.280] in that space.
[00:36:53.880] And so, in the cellular,
[00:36:55.260] it looks like the bioelectric circuits
[00:36:58.510] are very sort of versatile in that way.
[00:37:02.580] And they can pull down lots of different patterns.
[00:37:04.140] So in the body, you have a huge number of these things.
[00:37:07.420] So you have static forces among the molecules.
[00:37:10.820] So you have different distributions of static charges,
[00:37:13.000] and then you have voltage gradients
[00:37:15.300] across intracellular membranes.
[00:37:17.240] So Golgi, ER, nuclear, envelope, of course, mitochondria,
[00:37:20.640] all of those things have a voltage gradient.
[00:37:22.480] Then the cell itself,
[00:37:23.880] has a voltage gradient across the membrane,
[00:37:25.880] but even that isn't one gradient.
[00:37:27.660] A typical cell has many different voltage domains
[00:37:30.020] across its surface.
[00:37:31.060] It's like a soccer ball of different domains.
[00:37:34.180] And then the cells come together into tissues and epithelia.
[00:37:38.960] We will have an epithelial,
[00:37:41.000] a trans-epithelial potential across them on top of that.
[00:37:45.660] Then some people, I mean, Becker and other people,
[00:37:48.180] measured this weird longitudinal electrical potential,
[00:37:51.940] which is, you know,
[00:37:53.880] basically body scale voltage differences, right?
[00:37:56.560] You got a very, very long, very long range.
[00:37:59.940] And those are all the static electric things.
[00:38:03.260] And then on top of that,
[00:38:04.620] you've got the electromagnetic components,
[00:38:07.760] which some of the voltages change very slowly.
[00:38:10.400] So the induced magnetic field is very low,
[00:38:13.080] but there are other events that produce natural EMFs
[00:38:16.420] coming off of living tissue
[00:38:18.950] and going into the ultra weak photon range, right?
[00:38:22.320] So UV photons.
[00:38:23.900] So it's sort of like just an enormous amount
[00:38:27.680] of these kinds of things going on at different scales,
[00:38:30.280] at different frequencies, all the way from DC,
[00:38:33.140] which is what we studied to light basically.
[00:38:36.800] And all of those are kind of interpenetrating
[00:38:40.160] at the same time.
[00:38:41.680] And the cells and other systems
[00:38:44.080] are trying to make sense of all of it.
[00:38:45.540] You know, they're kind of swimming in the soup of signals.
[00:38:49.220] Yeah, fascinating.
[00:38:50.540] And yeah, I guess like,
[00:38:53.880] I'm cognizant that I'm sitting with the expert here.
[00:39:00.520] And like what to you has been the most surprising
[00:39:05.060] in kind of studying this system?
[00:39:09.800] Interesting question.
[00:39:12.620] Well, I guess the most surprising thing to me so far
[00:39:19.820] has been first of all,
[00:39:21.680] just how plastic it all is.
[00:39:23.880] is. And, and the idea that evolution apparently has really
[00:39:31.680] spent most of its effort on creating a system that is able
[00:39:36.820] to creatively interpret the memories that it has, whether
[00:39:41.760] those are genetic, genetic memories, or behavioral
[00:39:44.080] memories. You know, I've been I've been playing with us with
[00:39:46.860] this bow tie architecture thing where basically, at any given
[00:39:49.860] poem moment, you don't have access to the past to but but
[00:39:53.880] you're aimed forward in prediction and, and you have to
[00:39:56.320] take the prompts that you've been given, whether those are
[00:39:58.580] your genes or your n grams
[00:40:00.140] from previous experiences or whatever they are and you have to construct the story right and so just just the ability of every day we see these amazing things that have no evolutionary precedent as such but what you're seeing is
[00:40:12.820] the evidence that that uh basically living living things are these amazing um sense making uh
[00:40:19.960] systems at multiple scales and they're using all of these computational uh affordances all the
[00:40:25.400] different layers of the body to to to tell coherent stories that may or may not bear any
[00:40:30.380] relationship to the previous story they were they were given right fascinating wow uh this kind of
[00:40:36.760] reminds me of uh sort of looking at the the vessel muscular system and sort of this idea that maybe
[00:40:44.000] in sort of some ideal idealized setup um uh the the nervous system is kind of this uh
[00:40:52.700] i think of it like a set of wind chimes and like as sensations kind of come in and hit it
[00:40:59.460] then sort of by the the presence or absence of certain sort of homes you can build a model of
[00:41:06.820] your environment your environment um and then you have these uh these sort of latches these sort of
[00:41:14.920] uh tiny areas of chronic tension and like the the resolution on these uh it's like between you know
[00:41:22.140] 100
[00:41:22.700] to 400 micrometers so um that would say just in the brain uh you could have somewhere between
[00:41:29.680] 21 million to 1.3 billion um i'm calling them vascular addressable units yeah um of
[00:41:36.900] differential tension uh and so like these can kind of save save patterns and over time we sort
[00:41:44.620] of accumulate uh these sort of little points of tension uh which sort of become a predictive story
[00:41:52.700] um they sort of not only encode uh what we expect from our environment descriptively but what we
[00:42:02.540] expect from our environment prescriptively uh that we they're sort of very sort of active inferency
[00:42:09.260] uh like i will i will make my environment into this uh not just i i'll like expect this uh to
[00:42:17.520] happen um and so like over time it does seem like it's a very very very very very very very very
[00:42:23.700] sort of begin to live in this story. Yeah. And sometimes, like, I guess, part of the Buddhist
[00:42:32.700] critique is that, actually, this is bad in some ways. And that it's actually hard to not live in
[00:42:42.520] a story. Yeah. And this, I think it's interesting, hearing you, you say this, that, you know, you
[00:42:53.700] are doing something that feels very similar. And that there are stories upon stories in a lot of
[00:43:00.720] systems. And this is right. And this relates to, I'm working on something I jokingly have
[00:43:06.940] provisionally titled Femto Buddhism, where, you know, you sort of asked the question, and I've had
[00:43:13.100] this discussion with, you know, the Buddhist scholars and say, well, so all living beings,
[00:43:17.980] right, under delusion, then must be liberated and all that. Yes. So, cells, right? So,
[00:43:23.700] right. Okay, yeah, I guess so. So, molecular networks inside of cells, like, what does it mean?
[00:43:29.700] Right, you know, chemistry, presumably doesn't make mistakes, developmental biology definitely
[00:43:34.000] makes mistakes. And so, is there a point in which you say, and I mean, I think people's typical
[00:43:41.520] intuitions is that, no, no, no, that stuff doesn't, it doesn't accrue karma, it doesn't get liberated,
[00:43:45.280] it just kind of does what it does. But then you have this amazing living stuff, and it has these
[00:43:49.420] issues of, right. But actually, can you actually take this all the way back to the beginning?
[00:43:53.700] Right? So, I'm interested in taking some of these concepts of what it means to be in delusion about
[00:44:01.040] your environment, and what does it mean to be in a flow state, and, you know, and have sort of more
[00:44:07.840] or less direct access to this, you know, to this information, as opposed to sort of painstakingly
[00:44:14.000] and mistakenly, often we're trying to work it out. And how does that relate to kind of least action
[00:44:19.660] laws, you know, when you have a photon that doesn't have to worry about calculating all the different
[00:44:23.700] things, right? And then you have this, you know, this, this kind of always goes in the right in the
[00:44:25.940] least action path, and what's happening, right? So, so from there, we sort of dip into this, as from
[00:44:31.120] from from the particles, we dip into this place where we as living beings have to work really hard
[00:44:36.800] to, you know, we fight for, for, for, for information, and for trying to figure out what to do.
[00:44:41.200] And then, you know, some of us, some of the geniuses, or the, the exceptional people, then
[00:44:45.980] they get into the flow state, and they're like, like the photon again. So what, what, you know,
[00:44:49.760] what does that, what does that curve look like? I think, I think, I think there's,
[00:44:53.700] there's a lot to be said about what, what, what chemistry is doing from, from, from that,
[00:44:57.580] you know, from, from that perspective. And
[00:45:00.040] the other thing that um uh i think now now i'm thinking having talked to you about this i think we uh we have something coming out with um where we took xenobots which are these like novel frog based constructs we introduced a nervous system and we wanted to see what does a nervous system look like in a being that has never had evolution shape the structure of its nervous system. Right. But we don't have a
[00:45:22.980] InScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScript
[00:45:24.380] And so now I'm thinking maybe the next thing we gotta do
[00:45:27.680] is we gotta introduce some vasculature
[00:45:29.990] and see what that looks like.
[00:45:32.200] Amazing.
[00:45:33.720] There's something else I wanted to show you
[00:45:35.220] before we break.
[00:45:37.120] Yeah, please.
[00:45:37.840] You wanna see a really weird vascular hydraulic computer?
[00:45:43.100] Absolutely.
[00:45:43.900] I don't know if you've ever seen this.
[00:45:45.160] This is something Naorosh and Murugan worked in my lab
[00:45:49.220] on this.
[00:45:49.760] This is physarum, the slime mold.
[00:45:51.580] And you know, physarum, it does mazes and things like that.
[00:45:55.260] I wanna show you one particular thing
[00:45:58.400] that you might appreciate.
[00:46:00.300] So here's a branch of the slime mold.
[00:46:03.320] Here's another branch.
[00:46:04.100] This thing has been injected with these little,
[00:46:06.980] they're fluorescent beads, basically.
[00:46:09.000] Okay, and so we're just gonna track the fluorescent beads.
[00:46:11.460] So, okay.
[00:46:12.600] And so what you can see here is,
[00:46:15.820] right now it's just acting like a Y splitter in a hose.
[00:46:20.280] Right?
[00:46:21.060] But each one of these little things
[00:46:23.960] is independently addressable.
[00:46:25.720] Because you can see that it like shut this off.
[00:46:27.920] This thing's still going crazy.
[00:46:28.960] This one's completely shut off.
[00:46:30.240] So this amazing fractal thing that has just tons and tons
[00:46:34.660] of these branch points,
[00:46:37.260] if all of them are independently addressable,
[00:46:40.120] first of all, what network is controlling
[00:46:43.080] which things get turned on and off, right?
[00:46:45.420] And then, and you can even see like sometimes
[00:46:48.040] it goes backwards.
[00:46:48.900] And so I don't know if there's directionality,
[00:46:51.060] to the actual synapses here,
[00:46:53.260] but there's synapses basically,
[00:46:55.320] because you can easily imagine a mechanism where,
[00:46:57.700] based on prior experience,
[00:46:59.240] this thing gets turned on or off, right?
[00:47:01.300] Right.
[00:47:02.340] That's beautiful.
[00:47:03.260] So I thought that was kind of cool.
[00:47:04.840] And like, what the heck are they computing, first of all?
[00:47:07.220] And what, again, there should be these kind of two,
[00:47:10.480] at least two overlapping systems
[00:47:12.040] where there's the hydraulic system.
[00:47:13.480] But on top of that, something has to be guiding the,
[00:47:16.500] you know, the opening of the branch points
[00:47:18.520] and the kind of the,
[00:47:21.060] yeah, the decision-making there.
[00:47:22.500] Yeah, absolutely.
[00:47:24.080] Super relevant.
[00:47:25.100] And I mean, it does also kind of call into question,
[00:47:27.840] like how do you figure out the causality
[00:47:29.960] between the systems?
[00:47:31.020] I mean, everything is regulating everything else,
[00:47:33.740] but can we sort of say, you know,
[00:47:35.840] this is the dog and that's the tail or like,
[00:47:38.620] and you know, whether it's grandeur causality or,
[00:47:41.560] you know, there are different metrics.
[00:47:43.120] And I think Eric's work would be relevant.
[00:47:44.980] Yeah, for sure.
[00:47:46.220] For sure.
[00:47:47.040] Yeah, super interesting.
[00:47:49.320] Yeah, amazing.
[00:47:51.060] Thank you so much.
[00:47:51.520] I love your work.
[00:47:52.880] I think it really opens a new, you know,
[00:47:54.720] a whole new avenue of all this stuff.
[00:47:56.520] And I think, yeah, I think we really need to,
[00:47:59.000] A, start just looking more carefully
[00:48:01.520] at the vasculature and the muscle,
[00:48:05.300] but also this kind of broadens and helps us
[00:48:08.260] to try to define metrics of stress that,
[00:48:11.060] in systems that are not, you know,
[00:48:13.440] directly mappable onto each other.
[00:48:16.040] Yeah, likewise.
[00:48:17.480] I mean, I'm such a Mike Eleven fan.
[00:48:21.140] I think it's just, what you're doing is just amazing.
[00:48:23.540] So yeah, thanks.
[00:48:25.100] Yeah, thanks so much.
[00:48:26.160] I think, you know, I think at some point it would be cool.
[00:48:28.220] I'd love to, we're all full for this semester,
[00:48:31.160] but maybe in the fall,
[00:48:32.140] if you could give a talk to our center,
[00:48:33.980] I think that would be really awesome.
[00:48:35.400] That'd be amazing.
[00:48:36.320] You know, I think people would love to,
[00:48:38.500] I think they need to know about this stuff.
[00:48:41.060] So yeah, thank you.
[00:48:42.300] Let's keep chatting.
[00:48:43.820] If you have any thoughts on kind of general definitions
[00:48:46.680] of stress in diverse models, and things like that,
[00:48:51.060] in general, I would love to talk some more about it.
[00:48:53.840] Yeah, absolutely.
[00:48:54.820] And just to inject one more kind of observation that,
[00:48:59.900] I think that like this idea of boundary conditions
[00:49:04.360] in physics, in biology, it's sort of very important.
[00:49:08.040] And I think that you have kind of this very fresh perspective
[00:49:12.280] on sort of what defines the boundary
[00:49:14.700] and sort of what defines the boundary of cooperation
[00:49:17.840] or morphological boundary and so on.
[00:49:21.060] And I guess like one of the big core challenges
[00:49:27.440] in consciousness research is determining the boundary
[00:49:31.040] of a conscious system.
[00:49:32.480] Yes.
[00:49:33.100] And that I think that your work definitely seems relevant.
[00:49:38.160] And I would also say that, you know,
[00:49:40.060] from my bias perspective,
[00:49:42.420] that this sort of branchial space view,
[00:49:45.300] this sort of viewing objects as sort of shapes
[00:49:50.000] in branchial space, you know, is a very important part of it.
[00:49:52.060] like the true shape lives in branchial space perspective.
[00:49:55.760] And that I do expect that understanding
[00:50:00.580] your work in terms of what determines boundaries in organisms, biological systems, and then does that lead to a natural boundary condition in branchial space?
[00:50:15.980] This seems very… I don't know how to solve that, but maybe you do.
[00:50:23.100] Yeah, very interesting. Okay, well, it sounds like, yeah, let's… we should… we'll have another conversation about the boundary, the whole boundary thing. I think it's super, super important.