TSC 2025 - Barcelona - Plenary 2 - Brain Modulation

[Music] [Music] We are about 10 minutes behind. So we'll

we'll start about five minutes behind and try to make it up. Uh this is a very uh interesting session um on brain modulation and we have two speakers in

the technology uh area of modulating uh and stimulating and modulating the brain for uh benefit and uh

where's Julio? Okay, so our first

where's Julio? Okay, so our first speaker is from Barcelona. Uh he's

someone I've known for a long time actually. Uh his name is Julio Rafini

actually. Uh his name is Julio Rafini and he is from Star Lab in Barcelona and also Neuro Electrics in Barcelona and he will be talking about Kaggarov theory of

consciousness and brain stimulation.

Round of applause for Julio Rafini.

[Music] Thank you. Okay, thank you. Okay, good

Thank you. Okay, thank you. Okay, good

morning everybody. Um yeah so I'm going to be talking about comograph 3 and how it relates to neurom modulation brain stimulation and in particular in

computational uh approaches to it.

So first of all I as was mentioned by Stuart I I work at Starlab and Neuroelectrics in in which is where we develop more the clinically oriented stuff. Um our

mission is to deliver life-changing therapies for patients and we are uh currently addressing epilepsy, major depressive disorder and also Alzheimer's

and um our approach is computational. So

we try uh to build mechanistics models of how the brain works and how pathology impacts the brain and to start working from there to build neurom modulation solutions.

However, my background is physics and um I was attracted to science and to physics because of the big questions um and that's also what brings me here today. So the questions such as what is

today. So the questions such as what is time, space, what is reality, why is math so powerful in explaining things and who is this famous observer that

pops out every now and then including in u in quantum mechanics and then who am I? And I purposely write an AI in there

I? And I purposely write an AI in there just to uh keep step with the future. So

the background of my work has been the electrical brain. As a physicist, this

electrical brain. As a physicist, this is very natural. Looking at the brain as an electrical system that appears to compute electrically and this raises the issue of what do electric field patterns

have to do with mind uh if anything and can we harness them for communication or therapy?

In our company, we have developed systems now to read and write into the brain using transcranial electrical stimulation and EEG. And these are complex systems that can produce

different patterns of stimulation in the brain, which raises the issue of how to actually program them.

So yeah, how do you actually stimulate a brain? Um, the brain is a plastic

brain? Um, the brain is a plastic incredibly complex network. If we even if we stick at the level of you know uh micro scale not going into uh lower

scales a large dynamical system. Um so

our approach my approach has been to look at this from the point of view of modeling mechanistic modeling. So um by leveraging um expertise in mathematics, physics, neuroscience um and develop

trying to develop models for for how the brain works. This has two branches. On

brain works. This has two branches. On

the one side you can see biohysical brain model that has to do with how uh physics um interacts in this picture.

How electrical fields are generated by the brain by biological systems and also how electrical fields interact with biology. Um and the idea is to bring

biology. Um and the idea is to bring these two points of view together uh to build what we call neurot twins uh which can be a um basically uh digital twin of

a patient or of a group of patients.

So a word about mechanisms of action.

The mechanism of action of transcranian electrical stimulation is the generation through weak currents injected in through the scalp uh of weak electric fields in the brain. These weak electric

fields have a neuromodulatory effect.

They cannot generate action potentials but they can modulate uh the probability of firing of neurons and they can alter the dynamics of the of the system and through these effects actually lead uh

to plasticity changes plastic changes.

So engaging brain plasticity in order to rewire the brain um uh towards health.

So one application area that we have started with uh which is very neurological very uh very biological and neurological is epilepsy. Um and here's a quick overview

epilepsy. Um and here's a quick overview of how that works. So we have developed pipelines now that start from a neurologist painting a target in the brain of a patient that has a focal

epileptic uh problem. Then we take the MRI of the patient and we build their personalized uh MRI uh model. uh it's a actually an electric conduction model

that allows us to simulate the uh flow of currents and the generation of electric fields and then we have algorithms to optimize that for the target that has been specified by the

neurologist. Okay. And then a cap is

neurologist. Okay. And then a cap is produced and sent to the patient. Um

this um uh we started working on this about 10 years ago now. Um and we had a first pilot study that was done in concurrently with teams in um in Boston

and in Mexico and we saw good efficacy in an open label study and we are now uh engaging in a larger clinical study but we have gone to the next phase which is

actually trying to model the biology and the physiology of the patient not just the the physics but the bi the biology and for this we use data uh that is available for epilepsy patients like SCG

stereotactic EG implanted electrodes.

And with uh this data, we built network models of of the the the epileptic uh network. And then we use again

network. And then we use again algorithms to find out how to um best stimulate this network as a now a dynamical system in order to reduce the

propagation of seizures. And um using this um we can better personalize we believe um the the solution. I should mention on on the

the solution. I should mention on on the right you can see that um there are different scales of modeling that we employ from the micros scale to understand how electric fields impact

neurons and then we go to the meos scale and represent cortical patches small areas of the cortex that are represented by so-called neural mass models and then

those are wired together uh by contos of the patient that can be inferred from MRI as well and this is combined with the electrical

model uh to give the whole package of the narrow twin for the patient. Um

so on their here if you want to see a bit more in detail the pipeline you can see on the right how we uh take the MRI we build the biohysical head model and this can be optimized or we can actually

use all kinds of data um that may be available for this patient if it's epilepsy if it's Alzheimer's if it's depression as I will talk about in a second and build this a whole brain

hybrid model combining physiology and physics Okay. And this just to mention that we

Okay. And this just to mention that we have now a clinical study ongoing with this technology. Uh it will take a

this technology. Uh it will take a couple of years. It just started a few months ago with colleagues in France.

It's a multic-enter study to study uh the for the first time the use of this whole brain modeling personalized treatment in epilepsy.

Okay. But now I want to shift to the part of the talk that is most relevant to you. Um and it's very exciting for me

to you. Um and it's very exciting for me which is how can we help patients with psychiatric conditions?

um and in particular major depressive disorder which is a a very prevalent problem. Um and we have some kind of um

problem. Um and we have some kind of um rational mechanistic rational even if it's very coarse about how to stimulate the brain of these patients also based on a lot of data and the idea is is that

there is some imbalance between the frontal uolateral prefrontal cortices and probably a network interaction with deeper brain structures that should be stimulated. There have been many studies

stimulated. There have been many studies targeting the left orolateral prefrontal cortex using uh trans transcranial electrical stimulation and there is some good evidence for this and in we have

actually carried our own studies now with personalized approaches optimized approaches at home uh with good results um and also by our colleagues using our

technology as well. So we believe there is already a path forward um for treatment of depression at home with this technology. The main advantages of

this technology. The main advantages of what we are doing is this kind of uh optimize montage approach uh to treat the problem.

But can we do better than this? Um this

is basically uh building on some loose theories out there about what depression is about. Um some trial and error that

is about. Um some trial and error that has been followed and it seems to be giving some interesting results. But

there is always uh responders and non-responders in this treatment. And

part of the problem is that MDD is not a single condition. The theology is

single condition. The theology is diverse. The causes of depression are

diverse. The causes of depression are probably multi- uh multivaried.

And we should uh re think about a crucial difference between neurology and psychiatry here is that MDD is characterized by a persistent firsterson experience of sadness, hopelessness, and a lack of interest of or pleasure in

activities. And this is what we're

activities. And this is what we're actually trying to treat a firsterson experience.

Um whereas neurology is primarily concerned with the physical and structural aspects of the nervous system and its diseases, psychiatry focuses on the mental, emotional and behavioral aspects of well-being. In other words,

psychiatry is more concerned with experience machine.

And this raises the question that we're all here for which is how can this experience and here I quote hypocrates that a long time already thought about this from the

brain and not from the brain only arises our pleasures joys laughter and gests as well as our sorrows pains griefs and tears.

So the challenge is that uh we are missing a principle unifying framework to define and operationalize what we want to model. what is depression and understand what its physiological

signatures are and how to measure it consequently.

So here I will um this starts um the colog 3 part of the talk in which I will give you a brief intro to the to the theory and in particular the role of

emotion in the theory and how that relates to depression and and the role of veilance in the theory. So just to mention that this is a long road that I

started uh many years ago. Uh actually

in the startup years in 1999 when I started thinking about this um thinking about reality as some kind of information based uh um thing and there

are many papers now that describe it in more detail. It's getting I think more

more detail. It's getting I think more interesting and it's also connecting now with with um neuroscychiatry.

So the at at the root chography of of consciousness says that there is experience. It's a postulate. We are not

experience. It's a postulate. We are not going to try to explain that we there is experience and this is taken as a primal fundamental u part of the of the theory that it's an axiom like a mathematical

axiom and the focus of the theory is instead on structure experience. So how

can structural experience arise if you take take it as a given that there is uh experience.

So I don't think I need to convince many of you that there is experience and that there is pure experience. And this

relates to the immediate subjective sense of what it feel like to be oneself at any given moment. Beautifully

described in this essay by Thomas Ngo.

What is it like to be you?

And one way to picture the theory is to say that we are actually um fields of exitation. Sorry, in the in a field of

exitation. Sorry, in the in a field of pure experience. So in the case the

pure experience. So in the case the movies are not starting or they will start by themselves. Anyways, so the idea is that there is some kind of uh field of experience and then there are exitations in this field that one can

think in analogy to physics. There's a

quantum electromagnetic field or there are quantum a few quantum fields and there are exitations in them that give rise to actually in physics to all that we see right now. But here we're talking

about an experience field and patches of structure experience which is uh what observers actually are in the theory. So

what is structure experience? It is the spatial temporal and conceptual organization of our first person experience of the world and of ourselves as agents in it. So this is what we are

trying to formalize in the theory.

Here's a picture of what I mean by this.

So um it's the background of the theory is algorithmic information theory and the notion of uh colograph complexity which is the length of the shortest program that can explain a data set

which is now also becoming uh very relevant in the field of AI. Um but this is what I mean by structural experience.

You you can start with a very complex looking data a million pixels of in a picture but what your brain actually do is lossy compression. So it does a process of course graining and then

extracting useful structure which we may conceptualize as going from the left to the right and of course the limit is com complexity the shortest program but this is probably not necessary for the

purposes of agents as I will discuss in a second and this uh gives gives us a framework to talk about what is special about brains or indeed about life. Um and the

first thing is that um brains or living beings have high mutual algorithmic information um with the world. So we

share algorithmic information with the external world which is another way of saying that we have models uh in our structure that reflect what's happening

outside. They're compressed models lossy

outside. They're compressed models lossy and also that they compute. So brains

and life compute they they have to do this modeling and decision making. So

they need to work with these models and and uh and compute paths forward.

So in a nutshell the theory says that um evolution gives rise to a agents uh as we will see and and we are agents. We're

a particular form of agents and agents runs run models of the world and as a result enjoy structure experience.

Agents also have goals and this actually gives rise to what it's uh veilance and emotions.

Okay, what is an algorithmic agent? Um,

it's a computational system that interacts effectively with its environment by planning actions using compressive predictive models to maximize an objective function and all

these words are important. So objective

function compressive modeling and planning and also uh as a corollary using a model in entails as I mentioned computation and dynamics because dynamics and

computation are actually two words for kind of the same thing and this is the struct this is a more formalized version of what an agent is.

Um it has three main blocks. Well first

there is a world outside with which the agent is interacting. Um and here you must imagine information going back and forth between world and agent. It has a modeling engine which has the the goal

of creating compressive models of the world that are useful for the agent. And

for for this reason it has three elements in there. One of them is the comparator. The comparator is uh a box

comparator. The comparator is uh a box where external data comes in and it's cross checked with the model predictions to see if the modeling u is going well.

If there are errors, those are sent forward to this other box called the updator. Um, and then the model itself

updator. Um, and then the model itself is used to uh simulate uh scenarios and uh help the planning engine. And the

other big box in there is the objective function which is the which represents the goals uh of the of the agent.

Actually in the theory it's a it's it's a box which takes in a model and it gives you a number from you can imagine from 0 to one or minus one to one. It's

a scaler that says good or bad.

Um, now let's go back to models because they are so important in the in the in the theory. It's a model is a program in

the theory. It's a model is a program in the model in this in this framework.

It's a program that allows the agent to compress coar grain information. It's a

simplified but useful in the sense of the objective function uh representation of reality and it's a mathematical object in the theory.

So a model uh requires computation and it's also can be seen as a dynamical um it can see we can think of it as producing dynamics

and dynamics can be characterized by um geometry and topology as as I will try to explain in a second and just a disclaimer that um here computation

doesn't specify classical or quantum I'm agnostic about this but as far as I can tell quantum computation does not add uh hypermp computation um classical computation is already true

to incomplete but um there is lots of work to be done in this area I suppose.

So the dynamics that the models produce as you run them um then um are what give rise to structure experience and the structure of the model is represented in

the structure of these attractors in these reduced manifolds. If you take um each data or fMRI data from a subject and uh it's and it's a highly dimensional object because it has you

know you can get data from thousands and thousands of channels or voxels in an FMRI but it actually lies in a small dimensional manifold. This is called the

dimensional manifold. This is called the manifold hypothesis and it has to do with the fact that the data from the world itself is uh constrained to lowdimensional manifolds and these manifolds have topology and they have

geometry and the claim is that the structure of these manifolds is uh is what correlates with our structure experience.

And this can be formalized using uh tools from mathematics like group theory um that can capture uh the idea of simplicity uh in an geometric way. And

the as I mentioned the basic idea is that tracking the world uh forces the engine as a dynamical system to mirror the symmetry and the compositional aspects of data in the outside world.

And this forces the dynamics of the system into a reduced manifold.

Um and and this is what corates with structure experience.

So the central hypothesis of KT just to summarize is that an agent has structure experience to the extent it has access to encompassing and comp compressive models to interact with the world. More

specifically the event of structure experience arises in the act of running models.

Model structure determines the structure of experience.

And to connect with the field of presence, uh successful comparison with data leads to wakeful presence.

And a final caveat is that much structure experience is unreported.

Um so there may be a lot of structure experience out there that we will never know about including in our brains probably. Um and on the right you can

probably. Um and on the right you can see um kind of a logical kernel of the theory where structure is the main idea and this connects with algorithmic

aspects of structure um common of complexity compositionality recursion um with dynamics uh talking about symmetry complexity criticality

stability geometry and topology and with experience uh in which we have psychophysics phenomenology and then artificial intelligence experience presumably Okay.

So in practice um let's look at some practical implications of this uh abstract theory. So

abstract theory. So one is that it gives you requirements for structure experience and two uh that it gives you this idea of neurocoralates of structure experience.

So um as a practical tool it gives you guidelines on how to build computational brain models. Um and this for example if

brain models. Um and this for example if the brain is to be a computational system able to do all this fancy stuff it needs to be it's probably um it's very plausible it needs to be near a

critical regime because criticality gives you uh powerful information processing uh in in um in in in complex systems.

The second tool uh is uh the mathematical tool you can use is topology and group theory. And um again

I refer to the manifold hypothesis and the idea that um neural data lies in um lowdimensional manifolds and you can use

tools from topology uh to characterize these manifolds topological data analysis and presumably these are what gives you uh give you correlates to structure

experience. So ideally one would like to

experience. So ideally one would like to connect these two worlds in practice. The second part is uh

in practice. The second part is uh computational psychiatry uh where as I mentioned in the intro the first person and third person views meet and the framework here is that we're

going to start with algorithmic agent framework and we're going to try to translate it into a circuit view of the brain and see if we can make sense of these two descriptions to guide

treatment.

So you have on the left bottom left a depiction of the agent um with um the main boxes. The modeling engine is in

main boxes. The modeling engine is in pink, the objective function in yellow and the planning engine uh in blue interacting with the world.

So agents as I mentioned have an objective function that sets their goals. It quantifies how well or bad

goals. It quantifies how well or bad they're doing. Uh and this is in the

they're doing. Uh and this is in the mathematical version of what we call veilance, pleasure of pain. And the

theory assigns veilance uh first person veilance to this uh to this mathematical construct.

And as I mentioned the agent has these three blocks model goal and planning and based on this we are now in the position to tentatively define emotion.

So emotion is the is the sum um cartisian product whatever you want to call it is a combination of model structure and veance.

So those two things are what we are to um equate with emotion and depression uh can be characterized as a pathological state of low

persistent veilance.

You're just feeling low with low veilance.

And from this point of view you have multiple roads to get to this low veilance state. It could be that your

veilance state. It could be that your modeling engine is dysfunctional. you

have just the wrong world models. Um, it

could be that the objective function is persistently giving you a low veilance state even though the world is great objectively to others or subjectively to other agents.

It could be that your simulator is screwed up and it's giving you wrong simulations despite the fact that you have the right model. Or it could be that you're just planning wrong. you're

making wrong plans and this is actually impacting negatively in the world and the world is kicking back at you. Or it

could be objectively that the world is is a bad place. It's in a hostile and um this can also lead to of course a

low persistent low state of villance and this is a graphical description description of what I just said. uh you

have the three different the four different routes the agent types of depression as we call them they are not necessarily bio types but they are agent types that can lead to this low uh

veilance state.

Now the the challenge and this is what we try to do in a recent paper is to map all these circuits all these agent modules into the brain and we propose a

first map of this um where the planning engine is more in the periphal cortex in the frontal cortex um the the objective function it's um more in deep brain

structures nucleus acumbance amygdala and the posterior part of the of the brain and the and the h and the

cerebellum as well is involved in uh modeling and simulation.

Okay. And once you do this you can try to make sense try to make sense of agent types and the so-called bioypes of MDD which seem to be um clusters of symptoms

in patients and also network affectations in patients.

Um so you can you can try to conceptualize um the roots of the problem in this way.

Okay. The third part uh which we're going to apply KT is uh by going back uh to predictive processing theory because comograph theory includes predictive processing as you saw as part of it and

you will see uh applications in Alzheimer's and uh psychedelics. These

are computational applications you will see and also um how the framework can help you connect the whole brain model and there are some papers that I had

released on the right where you can learn more about it and now we go back to the computational modeling technology that I described before. So we have um this idea that we

before. So we have um this idea that we are going to build a brain using neuromass models which are connected with connetoons and we have created uh we have proposed

a model of uh neuromasses which is minimalistic but still is able to represent two types of uh oscillations in the brain. The slow oscillations and the fast oscillations and their phase

amplitude couplings. Um

amplitude couplings. Um and um using this uh neural mass modeling approach we can actually um address the question um of how to wire

the whole brain model because now we have a neural mass which is composite and it has slow uh or a slow osillator subsystems and fast oscilly subsystems and you need to decide how to connect

them. You can see on the right picture

them. You can see on the right picture which is actually in the office um how we think about uh the process right now.

You have feedback connections from one neurom mass uh to another and we have forward connections um and these are feed forward and feedback connections

that the brain uses when it's doing predictive processing.

And the question is how how do you connect them? Do you connect top to top,

connect them? Do you connect top to top, top to bottom? um what what kinds of um uh connections can you can you do and and and what is the right way and in in

particular it's not symmetric links that you need you need asymmetric links as you will see and um using this formalism uh we have

proposed a model for the comparator the comparator remember is this box where um predictions from the model and data from the world come together and you the first thing that you're supposed to be

able to get out of it. It's an error and it's an error signal.

And the and the proposal that we have made is that um data uh or prediction errors from lower hierarchies uh are encoded in the envelope of fast

oscillations and predictions come from top down in the slow in the form of uh lower uh frequencies in the signals themselves,

not in the envelopes but in the signals.

And with this we can we can actually show that you can uh wire if you wire properly uh these neural masses you can actually get prediction errors from the system which is the first step to do uh

to to creating models right you first you need to u be able to extract an error and then you can start building models and now using this um

this model we can study the effects of psychedelics for example. So we know that psychedelics um have um um they latch to serotonin receptors. We know

that they act as agonist of glutamate uh in pyramidal cells and we can model this in the neuromass model because it has all these um aspects represented

and for example what we see is that um in the healthy brain um you can see let's see if I can point here um yeah so here um you can see that the healthy

brain you have an input with an envelope and you have a prediction and you have um a reduced output because the signals are matching. So you're actually getting

are matching. So you're actually getting um you're you're suppressing the um the inputs because you're matching correctly. But if you're under the

correctly. But if you're under the effects of psilocybin, the error is not uh is not canceled. It's actually

amplified.

So I mean the model also predicts things about what happens to alpha power to gamma power. But the interpretation from

gamma power. But the interpretation from the point of view of predictive coding and uh and chromograph theories is that weakened top-down constraints with reduced precision of priors lead to

unfiltered bottom-map signals and excessive prediction errors which aligns well with the idea that uh in the rebus model of psychedelics of relaxed priors

uh leading to an anarch increase in uh in error signals increased complexity etc. We can also do the same model uh for

example in Alzheimer's disease or in other disease where you have uh PVC cell dysfunctions. These are fast inter

dysfunctions. These are fast inter neurons that are uh used for the generation of fast rhythms in the brain and and you can play the same game and

see what happens if you disrupt this connection here.

And um basically to give you the summary what happens. So at in the early stages

what happens. So at in the early stages of of uh Alzheimer's um gamma power increases and predictionary signals are amplified.

So you have a hyperexitable cortex with excessive surprise signals. In severe PV loss in the late stage gamma oscillations collapse into slow wave dominance. So you have a hypoactive

dominance. So you have a hypoactive cortex and error signals are basically greatly attenuated or absent. So the

interpretation is that the early inter neuron dysfunction uh in Alzheimer's causes exaggerated prediction errors overestimating surprise while in advanced dysfunction uh you

have a complete breakdown of error signaling. Um and if you remember what

signaling. Um and if you remember what they said before about uh KT and the the main hypothesis

this means that um your engine to compare models and data is broken. So

this u must lead to a derealization process um at least

conclusions of future directions. So

a summary of what I said um is that in this theory there is spirit um we can call it from by many different names experience and it's the mix of experience and mathematics. Mathematics

here has been represented with algorithmic information theory.

Mathematics gives rise to an algorithmic soup of possibilities and in this uh algorithmic soup there are persistent patterns. Uh do you can imagine a soup of interacting programs.

This has also been uh proposed by Gregoire Chatin. Um you have these

Gregoire Chatin. Um you have these interacting uh programs and given given the right conditions you end up with these persistent patterns. Um and those

we call agents and they are driven uh by natural selection algorithmic natural selection to um uh to telecom homeostasis which means um a persistent

of pattern. So it's not a persistent of

of pattern. So it's not a persistent of individual but a persistent of the program pattern.

Uh each of these is an observer. It's

what we call observers. Um agents coarse grain information to create compressive models. Um and this is this can be

models. Um and this is this can be linked to the idea of emergence that we talk about as well. And then we saw how this leads to the uh the subjective veilance uh structure experience emotion

structure experience and this is observer's reality um in this experience dream this experience mathematical dream.

So um key concepts pure versus structure experience mathematics which is the science of structure algorithms computation algorithmic soup chromograph complexity and mutual

algorithmic information persistent algorithmic patterns teleomostasis agents course grading for emergence useful compression

um modeling compression simplicity seeking simplicity model structure as valance is emotion and then I didn't discuss this too much but reported versus unreported

experience is a crucial conceptual element and I think what what we uh hear about when people talk about consciousness is really reported uh to

self or other structure experience.

So future challenges uh one of them is demonstrating how to computially evolve agents in an algorithmic soup and here I launch a conjecture which is under the right conditions which I don't know how

to define uh persistent patterns are unavoidable in a computational soup if we wait long enough question are there are there persistent

patterns other than agents um we should be uh based on this we should try to think about methods to detect an agent through its behavior, inner dynamics or structure

linking to the ideas of mutual algorithmic information, reduce manifolds etc. the challenge of associating the structure of dynamically reduced

manifolds with firstperson uh experience and then as I was uh as was discussing in the case of MD map the neurobiology of agent hood try to match these two

worlds and um the last thing is that we can imagine a sequence here all it's all an algorithmic sequence from an algorithmic

soup to agents that evolve in this algorithmic soup agents are really life. They have mutual information with the world but they can only learn through evolution. So a

single agent does not learn too much but through evolution natural selection these agents start capturing information about the world. The next epoch is brains in which agents can actually

start building models of the world within a single lifetime.

And uh the next layer I suppose is multi- aent systems, societies and um all the mysteries that this have to do like we each one of us is composed by

gazillions of agents of we have no um experience of how does that happen?

Okay, so the summary we focus on the requirements for structural experience.

Uh reality is a model. KT is well aligned with idealism. As you can see, tracking the world requires computation to run models. This constraints brain structure and also collapses dynamics to

reduce manifolds. The characteristics of

reduce manifolds. The characteristics of which are good candidates for the neural correlates of structure experience.

Emotion is defining the theory by combining structure, experience, model structure and veance.

Um and the idea is that this framework is to guide computational neuroscience uh bringing together the first and third person perspectives with applications for example in neuros psychiatry.

Thanks to uh some of my funders, the Hive project, uh the wonderful FET open program that is uh now being replaced by the EIC and the Galvani ERC Neurot Twin

project which just finished and my collaborators and and funders of course and I would like to announce that there is also uh a special issue in entropy uh dealing with the mathematics of

structure experience. If any of you are

structure experience. If any of you are interested, please uh submit your your ideas. Thank you.

ideas. Thank you.

Great. Thank you, Julio. We have time for some questions. Raise your hands.

This gentleman in the first row there, I see.

And then we'll get >> Hi, thanks for the talk. And I was just wondering why you didn't mention in the

way to define emotions uh the um arousal and dominance or presence uh parameters >> arousal or dominance.

>> Yeah.

>> Um those would be part of model structure in the in the theory I believe.

So I mean emotion is defined as uh model plus villains and model includes um

aspects of the world of the self. Um

so they would be included in in that aspect but it's a it's a good question.

Yeah.

>> Okay. This lady in the third row here please. Right there.

please. Right there.

>> Yeah.

>> Thank you very much. I wonder have you ever considered integrating the neural networks of the heart and the gut into this work? What would happen if you

this work? What would happen if you would not just work with the brain but with those two other sections?

>> Great question. So um I must say that when I talk about brain here, it's an abstract brain and in fact uh you can

think of the entire uh human body brain gut as an agent itself. So um

all those I as I mentioned each one of our cells is an agent as well and you have different uh meta layers of agenthood in you but you can think of the entire body as an agent and in fact

this is the persistent pattern that we are repeating across generation. So they

would be included in that and it's possible or the theory predicts actually that your gut has structural experience but it's not reported to you. There is

no communication between those two agent. uh subp parts in the same way

agent. uh subp parts in the same way that there is no um communication to you from each one of the cells in your body.

>> Okay, let's take one more one more over there.

>> Yeah, that now thank you. Uh your emphasis on the first

thank you. Uh your emphasis on the first person uh experience and also the no report paradigm seems to suggest that you are kind of targeting the fundamental consciousness and also quia.

However, you are also emphasis on the column uh complexity and so abstraction of the pattern uh seems to more emphasis on the you know u access consciousness

reportable kind of you know components or cognitive uh appreciation of the um consciousness. So I I'm wondering what

consciousness. So I I'm wondering what do you mean actually by the structure of consciousness? Is it more of the

consciousness? Is it more of the structure of the phenomenology or structure of the access consciousness or cognition?

So I think it what I'm talking about is closely related to qualia. So what I'm talking about a structure experience refers to structural qualia which may be a qualia about being in a room here with

a lot of people or um the qualia of a face in a face or of listening to a piece of music. That what the theory says is that all those uh the what makes

a qualia qualia is the combination of a layer of experience with a structure on it with some structure.

And it can be accessed or unaccessed. So

you it can be that you have for example that your cerebellum has is enjoying structural experience of some sort but you know nothing about it because it's unreported to you in the same way that maybe you have somebody by you who

cannot communicate and they are having very rich experience but they are not communicating with you. So access is another word I believe that you meant

meanant to refer to reported versus unreported to self uh to yourself model.

Okay, let's let's thank Julio very much for Thank you.

Our next speaker, uh, come on up, Jay. I

think you're ready. Is my colleague from the University of Arizona. He's

assistant director of the Center for Consciousness Studies, the head of SEMA Lab, uh, and also the, uh, founder, co-founder and CEO of San Mai. and he's

going to talk about from states to traits, how neuro how non-invasive neurom modulation with mindfulness can help shift consciousness to lasting well-being. Jay, take it away.

well-being. Jay, take it away.

>> Thank you, Stu. Gave you a long title.

Uh, so good to be here with all of you today. Um, I'm going to talk about work,

today. Um, I'm going to talk about work, as Stu said, that comes out of our lab at the University of Arizona, um, under the Center for Consciousness Studies, which hosts uh, this conference as well

as in the psychology department. And I

do have a company that's uh creating focused ultrasound systems. Um so you'll be able to see that device later today at the demos. Um but today I'm talking about our work um that's done in our lab

called Simolab. And I want to give you

called Simolab. And I want to give you uh one of my favorite quotes from William James where he says, "Our normal waking consciousness is but one special type of consciousness. while all about

it, parted from it by the flimsiest of screens, there lies potential forms of consciousness entirely different. Um, so

it's a little mysterious there. But what

our lab has been focusing on are these other potential forms of consciousness, how they're developed. So using

different tools like mindfulness, meditation psychedelics brain stimulation, and others, and how those can inform our understanding of consciousness and the nature of consciousness. And so the lab is called

consciousness. And so the lab is called Simma Lab, science-enhanced mindful awareness lab. Uh we founded this lab

awareness lab. Uh we founded this lab with a guy called Shenzhen Young who is one of the world experts in contemplative science um and contemplative practices and it's been a real joy to have someone with that depth

of knowledge in the lab informing the neuroscience. Um he's a mindfulness

neuroscience. Um he's a mindfulness teacher as well developed something called unified mindfulness um which is a system unifying all the different uh forms of mindfulness together.

and in the lab in the last few years, the core construct that we've really been focusing on is something that's developed through mindfulness practice, but actually through lots of different

practices, including sports. Uh I'll

talk about surfing in a minute. So, it's

not just mindfulness, but this is a core feature of the system that when developed to large degrees tends to shame change the shape or the nature of conscious experience itself. Uh this is

called equinimity which at one level you can think about it as a sense of openness and acceptance to all experience. So this is the good, the bad

experience. So this is the good, the bad and the ugly. And usually people talk about it in terms of a balancing between resistance and grasping. Um so let's assume you're at the conference and uh

you have your pet theory and you're in front of your adversary. You're having a nice consciousness fight. Uh what is consciousness? You you're sort of

consciousness? You you're sort of hunkered down and they say something that insults you. If you're resisting that insult, you might be resisting what that feels like in the body and pushing it down, which is going to cause some

trouble later. Uh, you could also be

trouble later. Uh, you could also be grabbing on to what they said, trying to think of just the most clever thing to zing them back. And you're going to miss what they're saying in the present moment and actually then maybe get frustrated that you missed another thing

that they said about you. And so

equinimity you can think of as this sort of balancing between the two. Um, and

sort of how much friction is in the system. And so you can think about the

system. And so you can think about the way I describe that is how much sort of stickiness or friction is happening in your sensory systems. Shenzen calls this sensory conductivity. I think that's a

sensory conductivity. I think that's a great term. And so the more equinimity

great term. And so the more equinimity you have, the more sensory conductivity you're going to have through the system.

Now what's really interesting to us is that's a developed skill or or maybe even a property of the system itself, but that also seems to modulate and moderate the health outcomes of

mindfulness. And so, as you all know,

mindfulness. And so, as you all know, lots of science on mindfulness leading to physical and mental well-being outcomes. The question is why? And for a

outcomes. The question is why? And for a long time, we were looking at concentration and attention itself actually doesn't seem to be the case.

I'll talk about that in a in a little later. It seems to be that equinimity is

later. It seems to be that equinimity is one of the core features leading to these health outcomes. And so, I wanted to give you one little example because often when people hear this concept for the first time, they tend to think

equinimity means that you're sort of dull. you're kind of cool and calm,

dull. you're kind of cool and calm, almost stoic in a way, and therefore you're not going to be moved to action.

You're not going to be able to go fight, you know, in my case, climate change, politics, all the things you want to do.

You're just going to be sitting at home feeling good all the time. And that's

actually the opposite of what equinimity is. Um, so I like I like the surfing

is. Um, so I like I like the surfing analogy where you're thinking about dynamically balancing on the surfboard in motion in time. So you have to have the time component here. And obviously

in this analogy, you can think of the wave as sensory experience itself. And

what you're trying to do is learn how to balance given the fact that you can't control the wave, right? But what you can control is your body on the surfboard. And if any of you have

surfboard. And if any of you have learned to surf, you know, and I love this picture because the moment you figure out how to stand up on the surfboard, it is a joyous experience.

And actually, it continues to be. That's

why people continue to surf. And then

maybe they grab onto that a little bit and try to surf bigger waves, right? So

maybe they're not fully aquatous on the surfboard. But the idea is that by

surfboard. But the idea is that by learning how to be dynamically flexible, you're sort of being able to adapt to the current sensory situation you're in, or on the wave in this case. And that's

allowing you to fully respond in the moment to what's happening. And

ultimately, that tends to feel good.

There's the sense of almost freedom that you experience when you first stand up on the surfboard because there's a lot of frustration before you get there, a lot of pain, um, a lot of sort of being pushed under the water. And so hold that

analogy in mind as as we go forward.

When we start looking at what systems in the brain and the body are starting to change when people practice mindfulness and acquire this trait or this skill of equinimity, uh we see lots of changes in the brain and the body. Someone asked

about the heart. Actually, we're

interested in heart and breath rate as well. So there's lots of dynamic changes

well. So there's lots of dynamic changes in the system. We take a systems approach. But when you look in the

approach. But when you look in the brain, you tend to see large changes across many studies in the default mode network. This is a network in the brain

network. This is a network in the brain that's kind of your default state. So

studies would say that half of you right now are in your default state and you're thinking about what you're going to have for lunch or something like that. Uh if

you weren't now the other half of you will start thinking about that. And

totally nothing wrong with the default mode. I want to say it up front. You

mode. I want to say it up front. You

need a default mode. It's used for mind wandering, creativity, and all kinds of useful human features and functions of your mind. But what can happen if you

your mind. But what can happen if you think about this analogy of balance, it can tip in the wrong direction in certain cases, depression, anxiety. So

if you're caught in worry and rumination, what we'll see is an overactivation of this system. And so it becomes a sort of interesting

neurological system that we might want to act on to increase equinimity and free people from some of those inner habit patterns. And of course, if you

habit patterns. And of course, if you look in the meditation literature, there's all kinds of beautiful studies.

This is from Jud Brewer's lab back in 2011. What happens if you put long-term

2011. What happens if you put long-term meditators and then new newbies, people just learning to meditate in the scanner and they're trying different forms of meditation. So the colors are just

meditation. So the colors are just different forms of meditation. long-term

meditators will massively reduce the default mode network um specifically the posterior singulate whereas uh newbies actually activate it a little bit and if it's sort of like learning to surf right it's a little frustrating in the

beginning for most people and totally fine but as you start acquiring certain skills particularly equinimity we think then you see the downregulation of the system so the opposite sort of

depression anxiety all of that and u as Julia mentioned in the last talk relating the phenomenology ology, the experience itself to the neuro that's called neuropenomenology. Uh that's a

called neuropenomenology. Uh that's a really powerful use case here. Um and

actually Jud's lab back in 2013 published this really cool paper. I

highly recommend it where they're actually feeding back the activation of the posterior singulate. So that's the main hub of the default mode network in real time to meditators while they're in the scanner. So they're seeing this

the scanner. So they're seeing this little fluctuating activation. Blue

means you're deactivated below baseline.

Red means above baseline. And look what happens when you put long-term meditators in the scanner over 30 minutes. You see massive reductions in

minutes. You see massive reductions in the posterior singulate. Right? And that

that makes sense rel relative to what I just told you. But what's really cool is that maps their experience. So as

they're sinking into the experience or sinking into their meditation, becoming more open and accepting to the fact that it's uncomfortable or whatever, uh their experience is more positive. And you see

that blue line, look what happens to the controls, right? So, these people are

controls, right? So, these people are just learning to meditate and they're experiencing what I did when I first learned uh frustration. Your sort of discontent, suffering ends up increasing. You're in a scanner. You

increasing. You're in a scanner. You

can't get out. Now, you're not riding the wave anymore. You are a victim of the sensory experience that you're in and you can't escape it. And that was what these subjects reported.

So, this gives us a really interesting place to look at what's going on um in terms of developing these skills of openness. The way that Jud talks about

openness. The way that Jud talks about it is that this is really the system of inner stickiness and the less sticky you get the less activation uh you see there. Now I also love the work of Robin

there. Now I also love the work of Robin Cart Harris which I'm sure most people here are familiar with. Um so Robin somehow convinced people to take psilocybin so magic mushrooms and get inside of an MRI. Um so they probably

already have a little bit of equinimity going into this. And again what you see is in that square down reggulation of the posterior singulate as you take uh

the psychedelic. And what's really

the psychedelic. And what's really interesting is this is related to the subjective sense of ego disillusion. Um

so the more you downregulate the posterior singulate, the more likely you're you are to report an ego disillusion. And if I if I showed you

disillusion. And if I if I showed you the other graphs, all of that actually predicts long-term health outcomes or longer term health outcomes. So the

people out there at the far right of the graph, they're having sort of a unification experience. Sense of self is

unification experience. Sense of self is very not present at the moment. They're

sort of they are the surfboard. I guess

in this case, they're not even on the surfboard anymore. And that then

surfboard anymore. And that then predicts those longerterm health outcomes.

So it leads to a question for us. um is

the PCC the neural substrate for equinimity right and you know obviously the answer is no this is a system of networks uh the brain isworked in the body and the body's in a human social

structure uh and then that'sorked in time and space um and so it's got to be the PCC is a crucial piece of this whole puzzle but if you think about the concept that Shenzen gave us of sensory

conductivity it's conductivity through a system through a sensory system it just seems that the PCC is highly relev relevant as a central hub. And so with neuroiming, it's hard to tell if these

brain regions that show up in the MRI are functionally relevant. So is that where equinimity is happening or is this a correlation to something happening in a much bigger network? It's kind of like you're looking at uh the shadows in

Plato's wall. Are we looking at the

Plato's wall. Are we looking at the shadows? Are we looking at the actual

shadows? Are we looking at the actual thing happening inside the cave? And we

don't know with uh with fMRI. And so

what we've been doing in our lab is trying to use methods to causally disrupt or modulate those hubs of those networks directly and then we can look at the effects on conscious experience

itself. This is obviously a very old

itself. This is obviously a very old idea. So Wilder Pinfield uh did some

idea. So Wilder Pinfield uh did some very awesome studies back in the 1950s.

If you take an electrode and electrically zap people's brains while they're having brain surgery, you change their conscious experience. You can

elicit conscious experience. And so

obviously the brain has something to do with consciousness. I think we can say

with consciousness. I think we can say that as a group. Um but it also shows that the causal manipulation can change uh both the content as well as the shape of consciousness in real time. And so um

I don't like drilling holes in the brain. I didn't uh like working in the

brain. I didn't uh like working in the O. I worked on deep brain stimulation

O. I worked on deep brain stimulation for a while and was really looking for a way to get that type of effect non-invasively. And I happened to run

non-invasively. And I happened to run into Dr. Hammeroff uh who's an anesthesiologist as you know and he said, "Hey Jay, have you heard about ultrasound?" Ultrasound? Yeah, of

ultrasound?" Ultrasound? Yeah, of

course. You know, use it to image mothers when they're pregnant and so on and so forth. And Stu said, "Well, you can actually get those effects that you're talking about non-invasively. You

don't have to do all this brain surgery and all this." And I said, "You know, I I knew about Stuart from documentaries and things." So, I was I was sort of

and things." So, I was I was sort of like in awe of getting to meet Dr. Hamarov, but also, you know, this set my whole life and my whole career on a path to trying to figure out how to do this non-invasively with ultrasound. Stuart

actually just published the first study.

So besides, you know, publishing one of the most famous and well-known theories of uh quantum consciousness, Stuart also set the field of non-invasive brain stimulation with ultrasound on its path

while working in the O full-time. So I

have no idea how you did all of that.

And there's probably other things that he's done that we don't know. But now

this is a huge developing field. There's

labs and companies all over the world developing this. And the basic idea is

developing this. And the basic idea is that you can focus a beam of sound, low inensity ultrasound. We're not doing

inensity ultrasound. We're not doing damage to the brain, but we can focus that beam basically anywhere like the posterior singulate and we can modulate that structure in real time. And so lots

of studies I could show you. I'll just

give you one of my favorites from Sunu's lab at Harvard. So they focused a beam of sound into the visual cortex and they induced visual phosphmes. So if you right now if you just kind of push very

gently on your eyeball, you'll see these little geometric patterns. And actually

every person has different geometric patterns depending on how your sort of sort of visual cortex is laid out in the visual system. And so they were able to

visual system. And so they were able to induce these patterns with ultrasound.

So very clearly ultrasound can reach uh into the brain very focally and modulate conscious experience. What we wanted to

conscious experience. What we wanted to do is start testing the ideas about the relationship between PCC and the default mode network to equinimity itself to see if it's really related in the way the

fMRI studies showed. And so this is Brian Lord, a postoc and previous postoc in our lab who did most of the studies I'm going to tell you about. Um so next year Brian should be on the stage talking about this. Um but the basic

idea is we get an MRI. We focus that beam into that individual participants PCC and then we look at pre-post measures of brain change as well as reports of equinimity. And I will say

equinimity is really hard to operationalize in the lab. And so we're doing our best uh right now, but it's it's it's quite hard. And so these are non-meditators. They're just going in

non-meditators. They're just going in the scanner. They're having ultrasound

the scanner. They're having ultrasound uh they're having MRI pre and post and then ultrasound right in the middle focused on the PCC for about 5 minutes.

And the idea is to temporarily modulate the default mode network and then see if we change the network connectivity in the default mode network and do we also increase states of mindfulness,

equinimity and other outcomes. And uh we did this in about 30 people. It's sham

controlled. Half of them get placebo, half get ultrasound. And relative to sham, what we find is that we can for the first time non-invasively modulate the default mode network itself. So we

see massive changes in the connectivity of the system and it's a down regulation. Um so less activation or

regulation. Um so less activation or less connectivity at least in the default mode network. Um and that actually related to the subjective report. So relating this to what it's

report. So relating this to what it's like for these people in the active condition getting inhibition of the PCC we saw an increase in state mindfulness

as well as equinimity um as well as we can measure it with this task but um and this work is actually being replicated now at IAS by Nico's group and Josh Kaine so you should look for those

papers coming out sometime this year um but also if you sort of tune in to the phenomenology and we tried our best to measure what's the shape of their experience what's it really like to have this done. How long does it last? So

this done. How long does it last? So

we're interested in the temporal dynamics. What we found is that um so

dynamics. What we found is that um so basically the way to read this is the middle circle is zero on a liyker scale.

So you're just moving on a liyker scale.

It's very simple. And what you see is in blue in the stimulation condition there's a significant difference in things like merging with the surroundings, sense of ego loss, profound um sense of inner peace. So it

moved the shape of their experience in a way that if you're changing equinimity and unsticking these people who are inside the MRI and you know they have to sit there for an hour uh the experience

is overall positive and they were more open to the experience. Um and the the good news is if you look on the right side they're not afraid. We're not sort of making them suspicious or lose fear of losing control of their mind all

these types of things right you wouldn't expect that to happen. Um and so when you actually start talking to these participants and asking what was it like for you to be in the scanner many of them were reporting you know thoughts

were still coming which is great we don't want to make your thoughts go away but there was less stickiness to them so these are many undergrads they're they've got a lot going on in their life and when these thoughts come about oh gosh I got to do this test it just

wasn't sticking in the system then of course you ask well what's that like for you oh I'm not my thoughts that's what a lot of them said wait a minute I'm having a first level insight or they're having a first level insight that they

can have a little bit of separation from their thoughts and that's a really good experience. So this is kind of like

experience. So this is kind of like standing or sort of wobbling on the surfboard for the first time.

So that type of effect tended to last for about an hour to maybe 24 hours for most people and then they're right back to their baseline. So we found that we can modulate the default mode network.

The next question for us is can we actually boost the acquisition of equinimity as a skill and do that by combining it with meditation training itself. And the ultimate question is

itself. And the ultimate question is does that lead to health outcomes? Does

that change in the shape of conscious awareness actually moderate the health outcomes that we're looking at? So in

Shenzen system um you've got these different components. We're using

different components. We're using ultrasound to enhance equinimity while people are training in a mindfulness course um which is two weeks and then looking at the longer term outcomes. Um

and this is all based on some studies by David Creswell. He did this really cool

David Creswell. He did this really cool study where he's asking what feature of mindfulness training is actually moderating these health outcomes. And

long story short, he found that actually if you only train concentration or uh training your attention. So focusing on the breath, focusing on the breath, and you're not teaching people how to open

to the discomfort that may arise from sitting. Uh maybe you should do it in a

sitting. Uh maybe you should do it in a loving stance, for example. Be nice to yourself uh when that's happening. So

those are the equinimity pieces. If it's

just attention training, you can actually increase emotional reactivity.

You can make them worse off. Um, and if you have anxiety, for example, and you close your eyes and you focus on your body, that anxiety can actually increase. And so, it's the emotional

increase. And so, it's the emotional equinimity piece that's actually changing the emotional reactivity component. And the types of meditation

component. And the types of meditation training that add that in should boost it much faster. Um, so he's got a series of really cool studies that you can go look at that sort of dismantling these

two pieces. And we use that same

two pieces. And we use that same paradigm. So, we used a two-week uh

paradigm. So, we used a two-week uh meditation training course. So they're

actually learning on an app. There's and

there's no teacher actually involved in this case. They do the training at home

this case. They do the training at home and then they come into the lab and they get four sessions of ultrasound attempting to inhibit the posterior singulate um temporarily and then they're meditating while that's

happening. And then we're doing pre-post

happening. And then we're doing pre-post fMRI. We're doing some stress tests and

fMRI. We're doing some stress tests and we're looking at health outcomes. And

again what happened is we replicated the basic effects. So this is Brian Lord

basic effects. So this is Brian Lord again. Um and I'll say that he is here.

again. Um and I'll say that he is here.

So, if you're interested in all this, uh don't ask me questions, go ask Brian because he's the one who ran all of this. And what we found is that there's

this. And what we found is that there's another uh downregulation, a change in the connectivity of the default mode network. And that actually correlated

network. And that actually correlated with all kinds of interesting outcomes that we were looking at. I'm most proud of this one, which actually we came up with, which was basically after the

ultrasound in the clinic, we had a a meditation pillow and we said, "Go sit on the pillow. sit for as long as you want. You're going to get your payment

want. You're going to get your payment or your class credit either way, but just sit there. What we're going to record you for how long? And what you can see is in the active condition, they

sat much longer. And some people at the top here, what you see in the red, so those are all people in the active condition, some people sat for 30 to 45 minutes, right? That's a long time for a

minutes, right? That's a long time for a first-time meditator to start sitting on the pillow. And what we started asking

the pillow. And what we started asking was, why were you sitting so long? What

was that like? A lot of people said it just was easy. There was an ease. I just

sat. I was doing the practice. I was

sort of in it. I was sort of getting it.

Whereas in the sham condition, you know, a couple people said that because they're they're learning this meditation for the first time, too. But typically,

it's what you expect. It's like, uh, yeah, I sat for a minute and then I got to go do my test. Uh, so this is a really cool clue that something deeper is going on in this study. And so we

really tried to look at what uh in the networks in the brain are really changing to explain what's happening here. And I got to give you a little

here. And I got to give you a little primer on the different networks. So

when you're focusing your attention on your breath, for example, you'll have something called the salance network and the central executive network. These are

task positive. So the task is focus on the breath. And when you're really in

the breath. And when you're really in that, you'll see these other networks, the red and the yellow networks activated. and you'll see the down

activated. and you'll see the down activation of the default mode network.

So there's actually a segregation of these networks that can happen. Um but

that segregation tends to happen sort of when you have more long-term meditation.

And so we started asking does the segregation of those networks actually predict what we're seeing? And this was actually Brian's idea. This is a really beautiful finding. And basically what

beautiful finding. And basically what you see so the slope in red is the active condition. And so the more you

active condition. And so the more you get a segregation of those networks, the more that predicts how long they were sitting on the pillow. And actually

those people out there and you know at 20 in red, those are the long longest sitters. Whereas in the sham condition,

sitters. Whereas in the sham condition, it's what you expect. You're learning to meditate. Uh you've got this default

meditate. Uh you've got this default system. That default system is coming

system. That default system is coming online and probably trying to trying to control and navigate the other systems in the brain because that's normal for that brain. What happens when we look at

that brain. What happens when we look at our real measure of equinimity? So we're

using the equinimity 16 scale. Again,

it's not a great measure of equinimity, but the best we have. And we see an even bigger relationship. So those people who

bigger relationship. So those people who reported the highest levels of equinimity are also the people who show the biggest segregation in these networks uh relative to the sham people.

So this is really exciting because if you look in the meditation literature and you ask what's going on when people are meditating for long term and really acquiring equinimity, you see the same

type of long-term segregation. So down

at the bottom left in a where it says trait, uh that's the same type of effect. What I just showed you in people

effect. What I just showed you in people who just learn to meditate for two weeks is what is looking like in longer term meditation. That's typically, you know,

meditation. That's typically, you know, 5 to 10 years of practice, but it depends on how you define it. Um and

probably it's looking like that sort of state to trait. So they're learning the state of equinimity, which is how do I balance on this surfboard of sensory experience? and they're integrating that

experience? and they're integrating that so they don't have to apply it as a skill anymore. And we're probably seeing

skill anymore. And we're probably seeing something like the very early stages of that transition. Uh so that's quite

that transition. Uh so that's quite exciting and it tells us that it's not it's not just the PCC that's involved here, but it's how these uh networks are

organizing and organizing in unison with each other. Um to you know, you sort of

each other. Um to you know, you sort of think about the balance analogy again.

If you've got one system that's overactivated, it's going to dominate that system and likely increase the suffering because you you can't be on task. You've got a lot of tasks that you

task. You've got a lot of tasks that you need to do and they're sort of organized by your goal set. And so what's happening is by segregating out these networks, you're sort of allowing the sensory conductivity to flow through

these systems in an appropriate way for that task. And actually uh yeah, I put

that task. And actually uh yeah, I put this in. So I I really love sort of

this in. So I I really love sort of analogies of ecosystems because when you look at ecosystems that are dynamically balanced again there's a lot of like murder happening in this ecosystem right

this is not a cool calm and collected system but you know the the sort of predators and the sort of fungi they're all in this unison of harmony that's actually definable by thermodynamic

principles that allows it to be in balance and that same principles are likely playing out in our brain I mean the most complex thing we know in the universe iverse is a natural phenomenon and it probably follows these

principles. Uh we also know them from

principles. Uh we also know them from all of our religions of course and when we look in in the brain at something like depression, we see this asynchrony.

We see this out of balance thing happening in depression that's sucking up all the energy, all the metabolism in the system so that you can't be on task and then you're just going to wobble right off that surfboard, right? You're

not going to be able to sort of deal with the next incoming input relative to the non-depressed brain. It's just a more balanced, more symmetrical system.

And so sort of what we think is going on is that as you're acquiring increased levels of sensory conductivity, it's changing the shape of the way these networks are organized and balanced

together. And that's then allowing the

together. And that's then allowing the system to more fully express whatever is present uh in the sensory experience, which then we would predict is changing sort of the nature of conscious

experience itself and the access to uh joyfully riding that surfboard.

Um so just in the last two minutes that I have uh we've showed the ultrasound modulates the DMN uh for the first time.

There's causal evidence now that the posterior single is playing a functional role in this process and that it's really the reorganization and the balancing in my view the sort of symmetrical balancing of these networks

um that's allowing you to ride that wave in a useful way for your task. And what

we're what we're proposing now is that that segregation in those networks is very useful for the type of training that we want to do. So you can acquire equinimity as a skill or property of the system.

Um but one thing if you're following very closely talking about well-being, it's in my title and the last study we actually did not find a change in well-being. So that twoe course uh

well-being. So that twoe course uh relative to the sham condition or the placebo condition, their well-being was about the same. There's a lot of reasons for that. Uh we didn't have a sham or

for that. Uh we didn't have a sham or placebo control for the meditation condition. Actually, there is such a

condition. Actually, there is such a thing. David Creswell developed

thing. David Creswell developed something called my time. Uh where

people think they're meditating, but they're actually not. And relative to that, in his studies, you see the change in well-being after 2 weeks. Uh but for us, this is okay because we think, you

know, really it's about training this stuff up on the pillow and then acquiring that skill for life. And so we want to go from the state of equinimity

sort of the the capacity to apply this as an attitude to really integrating it as a trait in the system. And then on a there's a more fundamental level of equinimity that we're not even talking

about here which has to do with how the self expresses itself in the system. And

so we want to help move people there and see if that's actually predicting what we think are the health outcomes. And so

um for the first time we did uh the first 10day silent meditation retreat with my good friend Tucker Pek. He runs

something called the ESA down at the Dharma um Diamond Mountain Retreat Center in Tucson. And uh people are on an immersive silent retreat. These

people um are sort of intermediate or advanced meditation practitioners themselves and we wanted to know in this sort of immersive environment where people are really training the system

does uh multiple doses of ultrasound actually increase the acquisition of certain sort of equinimity type experiences. I'm not going to give you

experiences. I'm not going to give you these results. You have to go see the

these results. You have to go see the grad student from our lab, Sebastian's talk today um in the afternoon in C6. Um

but basically in talking to these folks, so they came in, we got IRB approval, and I'll say this 100 times, if you're going to do this, you have to have ethics committee review. Don't go build one of these systems at home. You're

going to hurt yourself or your or your friends. Uh this has to be done in a

friends. Uh this has to be done in a lab. And so what we did is got the

lab. And so what we did is got the university to approve this through their ethics committee. We then took the

ethics committee. We then took the equipment to the meditation center and we had Brian and other folks who are experts run this study and in that context uh we did find uh the

self-reports from many of the experienced meditators found it highly useful but um I'll let I'll let Sebastian talk about this and then on my

final slide um how do we really get to these longer term health outcomes that that we're looking for? Well, retreats

are important. Meditation in general is important. The ultrasound can be a

important. The ultrasound can be a catalyst to all of that. But really, it takes human support. Obviously, it takes a lot of time and training. And one

thing that we're really curious about is integrating conversational bots, which I know a lot of people are thinking about LLMs and meditation right now. And so,

Shenzen's creating something he calls bright bots. So, this is like having a

bright bots. So, this is like having a very deep experience practitioner and teacher in your pocket. And what we're going to do later this year is we're going to use Shenzen's first Britebot,

which he calls Ramona, during a Shenzhen retreat. So people will be talking to

retreat. So people will be talking to the bot about equinimity and openness, about how that can be applied in their life and what it is for them. Then

they'll go on retreat. The whole retreat is structured around concepts of equinimity and openness. And then they get the ultrasound multiple times through the retreat. then the bots's talking to them the whole time as much

as it's useful during the retreat but especially after so that we can continue to integrate the effects because the state of equinimity being on the surfboard for the first time can be

lifechanging uh but that doesn't mean you're always going to be able to surf right so you want to acquire this as a trait and so the idea is the breitbot will be helping them to integrate but also allowing us to collect data on the

back end that we can feed back into this protocol and so it becomes something that hopefully helps us really start predicting and enabling well-being broadly and deeply defined in our participants and really understanding

the temporal course as well as the network dynamics in the brain for you know when is this happening when is it appropriate to accelerate that for folks uh so that you know we can help them uh

really surf the wave in a useful way for them uh so lots of people worked on this I want to thank everybody and thank you for your attention [Applause]

Thanks. We have time for some questions.

Thanks. We have time for some questions.

Uh where's over there?

Who's got the mic?

>> Thank you. Appreciate that.

I was struck in your description of mechanity the similarity to that what we experience in MDMA therapy and I'm wondering if you had done any studies if you've done the psiloc assignment if you've done any MDMMA studies with

default mode network.

>> Um so I I studied a neuroscience lab giving MDMA to rats. So all I can speak to is what the rats experience is like on MDMA or mice actually. Um but I would say I think I made a comment that I think equinimity is showing up all over

the place. It shows up in your lives. I

the place. It shows up in your lives. I

mean if if you're sort of uncomfortable right now and you you decided I want to really hear what Jay's saying you're applying an equinimity practice. So it's

not just you know for meditation but I think the psychedelics uh used in the right context can really help the person sort of deeply have the experiential knowing that it's possible for them in

that moment. I think that's part of the

that moment. I think that's part of the transformative power and that's why I said you know I was talking almost surface level about equinimity because there's a much deeper piece of this that if a person can really experience that

you know it in our studies with ultrasound for example that also happens not not to the level of psychedelics but in our studies on anxiety for example you know the person will go actually

some people cry sometimes getting the ultrasound to the amydala and what's happening is that their anxiety is not going away they're sort of opening up to it in a way where they realize, oh, I

can open to my anxiety. Holy

Because my whole life is I'm an anxious person. These are anxious people. So, I

person. These are anxious people. So, I

have anxiety and I've identified with anxiety. So, there's the identity piece

anxiety. So, there's the identity piece and anxiety makes me very closed. And

this can kind of go, oh, it's over there. It's right here. It's right

there. It's right here. It's right

behind me. So, that that's what the hopefully the bot will help with. And

then you have a human on top of that.

And now you have the whole sort of integration of that. And actually I would say the Ramona bot will be very useful for your types of studies too. So

the hope is you know it's useful for all of these >> over here.

>> Have you compared the effect of the ultrasound to the effect of for instance the singing balls that we had this morning?

>> Yes. Uh very interested. I have a friend called Raja uh who's um a sort of condundalini teacher and he was supposed to be here to do some of that. So what

we've been trying to do is ask does adding the bowls in the space actually matter because I think set and setting is as crucial as any of these interventions. Um we've also been trying

interventions. Um we've also been trying to do something a little more weird which is recording the frequency and playing that right back through the transducer. So you'd be hearing the

transducer. So you'd be hearing the bowls through your ears and you'd be sort of hearing through your brain uh directly. Um, we haven't gotten that off

directly. Um, we haven't gotten that off the ground yet. So, so anyone who wants to do that, we'd love to partner.

>> I'm not sure I heard it in the in the presentation. Have you looked at the um

presentation. Have you looked at the um the longitudinal effects of the entrainment um over time? So, in the altered states of consciousness scale, have you looked

maybe a month later? Are they are these state level changes or are they trait level changes? It's a

level changes? It's a >> great question. Uh we do look out one month and three months at this point. Um

that's kind of where our funding is limited to unfortunately. Um and it's mostly state level changes. And so

that's why you know when I gave that graph it's like how do we like where's the inflection point? And you know we think there's an inflection point to trait changes but then Shenzen likes to

use this phrase exponential happiness.

So there's also another inflection point that long-term meditators, folks in long-term psychedelic studies, they tend to report it's not just that equinimity is helping me open and and ride the wave. It's deeply changing who I am in

wave. It's deeply changing who I am in this world and my relationship to the world and then deeply changing how I understand myself in that relationship.

There's some point in that inflection where that starts taking off and that's really where the deep well-being stuff is happening. And so we're interested in

is happening. And so we're interested in that whole time course. Um but I need like 10 million bucks. So if you got some money in your pocket, send it over.

>> Can you tell us more about just the dangers of using ultrasound, not not in a professional setting like you because I recall some dangers around cavitation and so on.

>> Yep. Yeah. And it's it's great to bring those up. Usually in these talks, I'll

those up. Usually in these talks, I'll have a slide that goes over those. So

we're using low inensity focused ultrasound. Um, if you go about 10x to

ultrasound. Um, if you go about 10x to 100x above that, that's called highintensity focus ultrasound. You can

use that to ablate kidney stones and things like that. It's used all over the world. You don't want to do that in your

world. You don't want to do that in your brain. I mean, that will damage the

brain. I mean, that will damage the brain for sure through heating and cavitation, which is violent shaking.

So, there's non-violent shaking and there's a violent version where it actually explodes. Um, and you can use

actually explodes. Um, and you can use that to open the bloodb brain barrier.

So, the nice thing about this is we have a nice map of the safety landscape. Um,

and that's true for obviously any any energy modality. So if you use

energy modality. So if you use transcraanom magnetic stimulation, you also don't want to crank up the power.

But for us, you know, to really even know if you're doing that safely, you have to have a really sophisticated microphone called a hydrophone. And

they're about 60,000 bucks to really have the whole tank. You have to degass the water. So that's why, you know,

the water. So that's why, you know, people write me all the time. They see

these talks, I want to do this at home.

And it's like well if you are an engineer and you have a hydrophone all the above and you can go to an IRB maybe you can do this safely but really you need to wait for some of the companies

there's a bunch of companies working on these devices they will be certified so CE 601601 certified then you'll have someone testing it to know and I would

highly suggest waiting until something like that happens my company's going through a clinical path so we're not releasing consumer products but I think you will see consumer products on the

market in the next 5 to 10 years. Um,

then you know there's the other side of just like with psychedelics like how much is too much? So, you know, if I'm uh watching YouTube shorts too much, I'm

I'm stimulating my brain with inputs and too much can be too much. Definitely the

same is true for brain stimulation, psychedelics, and all the others. And

that's actually why in the last slide, I think really to do this safely at scale, you need the whole container, right? You

need to build the set and setting. You

need the bot and humans and probably psychiatrist in the loop because stuff's going to happen. And I think that's true for psychedelics as well. So once you have the support container, then I think

the safety is there.

>> Okay.

>> One more one one more right there.

>> Uh yes. Um please I would like to ask you uh what is the explanation for the high efficiency of transcranial uh ultrasound stimulation? What are the PR

ultrasound stimulation? What are the PR principle behind it? Are those standing waves in the fluid or how do you explain that on a on a physical level?

>> So, so like what's the mechanism? Is

that what you're asking?

>> Yes.

>> Um I have a slide that I didn't put on this one that was at the talk yesterday, but there's multiple potential mechanisms and actually this is why we got into this because microtubules have a resonant frequency somewhere between

one and five mehz, somewhere around three mehz for some of them. So that's

why we got in. So there's likely a potential direct modulation of what Stuart's talking about. Ion channels are sensitive to stretch. There's actually

po electric ion channel. The PO1 PO2.

The Nobel Prize was given for this a bunch of years ago. So we know that if you mechanically force onto an ion channel like that, you can open it. Um

but also, you know, your brain's this sort of jelly mush and fluid. So there's

fluid dynamics that are happening as well. And some of you might remember

well. And some of you might remember headbangers uh back in the day. um you

know if you just mash your brain against your skull you're going to change your consciousness right so so that's not happening with ultrasound you know but we're we're doing that on a very low

intensity sort of jiggling the cells and then there's a sort of higher level which is reducing the inflammation you're also changing blood flow in the system so there's all these studies on Alzheimer's and uh even for depression

and anxiety I think inflammation and mitochondrial disorder is a big piece of those so ultrasound's also acting so long story short, it's multiple, you know, mechanisms, but if if this guy is right, then that's probably the only one

that really matters for this and everything else is following from that.

So, help me test it.

>> Yeah. Uh, thank you, Jay. Uh, uh, let's give Jay a round of appro applause.

Julio also back at 2 o'clock. 2 o'clock for the afternoon session.

[Music]