We did this like mega analysis where we just analyzed every resting state
fMRI psychedelic data set, which resting state is a huge
psychedelic data set which resting state is a huge problem and unfortunately that's the way
everybody's been studying psychedelics and humans in terms of like the neurobiological effects
and it's largely garbage just want to put that out there i'm just all the default mode network
stuff did not hold up under the 250 subjects you know when we put everyone's data together
none of that default node network garbage held up um that story needs to go to hell
psychedelics on memory, an area we don't often hear discussed. He's very critical of the prominent
theories within the psychedelic field, as you've already heard, the decreased activity in the
default mode network, and Carl Friston and Robin Carhart-Harris' Rebus and Entropic Brain
Hypotheses. Das believes that many psychedelic phenomena can be better explained through the
manipulation of hippocampal memory circuits. More specifically, a selective dialing down of our
up of our familiarity memory. According to Das, this is why DMT entities feel real and why
we feel at one with the universe. Subscribe, like the video and comment below if you would
like to see me host a debate between giants Manoj and Robin. Every disorder
Manoj and Robin.
Every disorder has emphasized the default mode network, from schizophrenia to depression
to whatever, OCD.
They'll cite this one paper, where there's a meta-analysis of like a thousand participants,
you know, about 50% have depression, 50% of them are healthy.
They found that the default node network was more functionally coupled in depression.
So it's like, okay, there's like this like over, you know coupling and people are fixated
on their sense of self or whatever, right?
There's actually a paper that came out in 2021.
effect that patients with depression have less coupling of the default mode
network. So the point is is that it's not a good it's
not good biomarker. And the truth is that resting state is probably just a
garbage way of measuring people. This default
state is probably just a garbage way of measuring people this default mode network narrative became
huge because michael paulin wrote about it and what's funny is that when we actually put the data
together decreases the default network connectivity under the effects of psychedelics it was an
extremely unreliable finding
minoj das for the listener at home who's not familiar with you
or your scientific work, what's the most important context for them to understand?
Where does that story really begin for you?
Where the story begins, like my origin story.
Wherever you think the listener best understand your scientific
work,
what you've accomplished.
Um, I mean, yeah, okay.
So for more just my scientific work which I guess does sort of relate to an origin
story.
Um, so I yeah, started actually here at UT.
I did my undergrad here and I was working in a memory lab and I really liked all the memory stuff.
And I thought that it was, you know, it was very rigorous work.
And you know I learned about like a lot of the fMRI research that was kind of, it started in the field of episodic memory amongst other areas of cognition.
But I wanted my research to kind of have a little bit more applicability.
And so then I started studying drugs in my masters.
And then I then went back to studying memory research as like a post-SPAC research assistant.
And then I got to combine my interest in grad school.
So largely my research is about understanding the effects of drugs,
especially psychedelics on learning and memory processes.
Extremely cool.
The standard conversation with psychedelics is surrounding things like
neuroplasticity, mental health.
You don't often hear memory in that discussion.
So what is it that drawn you to connect those two?
Yeah, it's kind of funny.
obviously just kind of, it was like, all right, I study, you know, memory and I study drugs.
And I kind of combined it to without having, you
know, was rather exploratory in many ways when I started.
But there is actually a cool link.
Like actually memory researchers also talk about plasticity.
Although a lot of times we'll talk more about forms of plasticity like long-term potentiation,
which is, um, not necessarily you know the growth of like,
you know projections from neurons, dendrites or these, uh,
these little protrusions that come out called spines but rather the strengthening of like the
function of one neuron upon another so the influence one neuron has upon another is thought to be
kind of the building blocks of how memories can form or be disrupted even and so um yeah i mean i
was sort of you know like i said kind of just doing this like let's just see the effects of drugs
and just mac them out and maybe we can you know find something interesting and i think eventually
I did kind of come to that.
And, of course, it came in a way that I wasn't expecting.
And so, you know, typically, we think of like episodic, human episodic memory is like remembering where and when an event took place.
This is a more standard way of thinking of human memory.
You know, I went to, I don't know, what did I do yesterday?
I can't even, my episodic memories terrible.
But I actually went skating yesterday last night.
I went into this little spot, skated this ledge, saw a police officer.
He was actually really nice, etc.
So that's like episodic memory and that was initially what I was thinking.
I was like, this is what we need to target, you know, is is figure out ways to disrupt
people's episodic memories that are really maladaptive.
But later on, I think I've come to realize that it's not always the thing that's messed up,
but rather it's the semantic memory.
So this is these are things like facts, concepts, language, beliefs, life narratives.
And so for example, knowing that George Washington was the first president
in
of the country or for the Europeans out there, knowing that World War II ended in 1945.
We all know that as a fact, but we don't have the episodic memory for where and when
we learned those pieces of information.
And so it's the semantic memories that are hard to disrupt, and that can really, I think,
become maladaptive.
So of course, I gave you more of the kind of standard facts that many of us share, but
there's also forms of semantic memories, that not all the share.
And these would be kind of like, you know, kind of very idiosyncratic or self-relevant beliefs.
things like for some people they believe in god and god is the same level of like factualness as
like uh george washington being the first president of the country um there's also people who
you know after 10 years of depression feel like they're a bad person or the if they have ptsd they
might feel like the world is dangerous that's a fact to them and so how would you ever you know
you can't fathom how hard it would be to convince somebody that george Washington was not the
the first president of the country, or that World War II did not end in 1940.
five, right? It'd be very difficult. Likewise, it'd be really hard to, like, no matter how much
evidence you give somebody, you could tell them that, you know, you're a family person, you,
you, you donate your money, you volunteer, you help your friends out, and they still feel like a
bad person at the end of the day. To them, it's just a fact. It's hard to change that. So how can we
change that memory system, I think, is what would be quite important. Um, and it's kind of a cool
place where like my fields have come together, um, there's actually a really interesting sort of, uh,
this is a little, you know, tangential, but this guy Tolving who came up with this dichotomy,
Anvil Tolving, he's actually from Estonia where I just was. I kind of forgot about that. And then
when I was out there, somebody had to remind me. So he came up this dichotomy of episodic and
semantic memory. And to be clear, it's a spectrum. These things are on a spectrum, but you know
it's nice to simplify things. And so he actually referred to the experience of semantic memory
as being noetic consciousness, whereas episodic memory is auto-noetic consciousness.
consciousness. It's like the self-knowing. It involves placing the self back in time versus
noetic consciousness, just this sense of knowing. We just know this to be true. And what's interesting
is that the psychedelic world refers to the undeniable sense of knowledge people get while
they're under the effects of psychedelics, whether it's real knowledge or not, right? I think there's
a lot of garbage that people certainly get of the effects at these drugs. But they have a sense
of knowing, and they refer to that as the noetic quality. And so this was where my two worlds
And I realized I was like, oh, there might be something interesting here.
And then as it turns out, that's where we end up finding some of our effects with psychedelics
that no other drug does.
They seem to mess with this cortical dependent, noetic.
It's not quite semantic memory yet.
We haven't proven that yet, but that's actually the next step.
But we're getting closer.
This other memory system that's not necessarily episodic is where we're seeing effects of psychedelics.
And if anything, the episodic memory system, we just see impairments.
about that. That's super interesting. How do you figure this out in the lab? What exact experiment
did you run? I find this effect super interesting and not very commonly talked about. Again, it's
all about plasticity, but we don't really directly talk about the different forms of memory and how
psychedelics might affect different memory circuits in different ways. I think this is super
interesting. Yeah, yeah. So, I mean, I think part of it is that, you know, it takes,
you know. I mean just to like understand, I me, look, it took me a while to kind of come to this
studying memory for a while, and to really like wrap my head around these multiple different
memory systems. There's also like multiple frameworks of how to think about memory. And, you know,
it takes a whole PhD just to kind of get, get, wrap your head around some of this. And I also feel
like just to be clear that I'm falling behind in the like episodic memory world, because I also
have to pay attention to like all these other aspects of psychedelics and whatnot. And so,
you know. So I think that's one reason why is that people don't really, they don't study
or if they do they do it from a very shallow point of view but you know there's a lot of people out
there who could like could you know get dive really deep into perception and if they came into the
field of psychedelics they would like come up with whole new models of how psychedelics impact
perception that i think would be really useful to the field unfortunately i think people who get
in this field took a lot drugs and then they decided to study psychedelics without necessarily
having a strong background in like any specific aspect of cognition right right um and so i do
you feel
very fortunate to have had that background in episodic memory and then to be able to come in and
you know be one of the few people who can like understand psychedelics like what they do to
memory so yeah I mean I think that um you know in terms of yeah psychedelics and you know how
they impact so how do we how do you how do We study this the simplest way that we study psychedelic
the simplest we study memory um especially episodic memories is we can just show people let's say
bunch of pictures and then this would be what's known as the
encoding phase where you're kind of forming memories or encoding memories.
And then after a delay, we test your memory.
Um, and then we just show you all those pictures again, intermixed with
brand new pictures, and we ask you, did you see this picture?
And you should say, yes, I saw that, you know, this picture, no, I did not see
this picture etc.
And then we can actually collect, uh, confidence ratings and then apply
computational models to those confidence ratings that can extract
parameters that are thought to reflect this more
standard episodic memory, this recollection that's thought to, you know, be dependent on
the structure known as the hippocampus. And then there's also another form of memory, which is
kind of this, it's this cortical dependent memory. It's taught to be non-hippocampal
dependent. It'S not quite semantic memory.
It's almost like in some ways like a predecessor for semantic memories. And so that is a,
what we're seeing that gets enhanced. And to be clear, we're talking,
about when psychedelics are on board during the encoding phase.
So how you're forming memories under the effects of psychedelics, not how
you're retrieving.
And so people are forming memories that tend to be more driving kind of learning
into the cortex is the idea here.
And so it's actually, there is some evidence that, you know,
there's this recent evidence that maybe there is something about this
plasticity story that might be relevant here.
So what they found in this one study by Alex Kwan,
is that where the plasticity happens in the brain depends on activity that happened during
the acute psychedelic experience.
And so if people are forming memories that are like bootstrapping off of, you know, the
certain cortical circuitry, you
know, let's say I look at a picture of a frog and now my frog concepts are getting lit
up, as well as the things related to the frog concepts, lizards and whatever other amphibians
and reptiles are getting lid up, well, then there's going to be more plasticity
potentially in those regions that you can maybe subsequently drive more change.
And I can change your concept of, you know, maybe what a, maybe if it's like a bird,
I can changed your concept to be like, these are dinosaurs, even though most of us didn't
learn that properly in school.
We kind of learned that very peripherally.
Maybe I can changes your concept.
You need to consider those dinosaurs.
When you think dinosaur, birds come up there just as much, right?
They ought to anyways.
And if they don't, well, I feel bad for you.
I'm just kidding.
in any case like i think that that's you know that might be you know for example one way we
can think of like how you know this learning in your semantic memory your concepts can be changed
and you know it's a pretty like um harmless sort of change right um or pretty at least yeah whatever
you know we all don't need to know that birds are dinosaurs you know most of us anyways don't
need to do that i don't even need to go that and i'm a scientist um but you know
it probably would be useful to change other sorts of concepts right there's also probably ways
in which we could, you know, potentially
change people's concepts in a way that aren't good right and I think that does
happen to people where they might you know start to change like the ways they
think about reality and everybody's against them and to people will turn
you know rather paranoid and have delusions after an experience but then I
think there's other situations where maybe we might be able to drive certain
changes in people's like self-representation or how they represent
the world and be like hey the world's not a dangerous place people aren't out
there to get you right and we could you know potentially drive
And we could, uh, you know, potentially drive those changes during the acute effects.
And then it's those post-acute plasticity changes where maybe that's how they're maintained more.
Yes.
But I mean, yeah.
I do a lot of research for every single guest that I have on the giant shoulder.
And since I'm just one person and I can record six, seven or eight episodes a week, I use
AI to help with that research process.
But something that I've learned the hard way is that ChatGPT and Claude just,
cannot be trusted. When it says a paper stated something, you just never have 100% confidence
that it didn't hallucinate. I got really stung by this.
confidence that it didn't hallucinate. I got really stung by this and it actually cost me
thousands of dollars. So let me tell you about this new tool I found called SciSpace. It's
different because it's built on top of the actual papers, 280 million of them. When you ask it
a question, it pulls directly from the source text and it shows you exactly which sentence in
the paper backs up that claim. So you can have 100% confidence. You can check it. That's what
The feature I like to use the most is deep review.
A topic that I've been fascinated of because of a few episodes I've recorded
is Michael Levin and Platonic Spaces.
So type it in and it runs a full structured literature review.
Finds the papers relevant, filters them and extracts the key data.
And it maps out where the field agrees and then crucially where the fields doesn't agree.
I just feel a million times better using a tool like SciSpace because I know that the claims are
grounded in the actual papers.
Use the code and the link in the description
to get 20% off a monthly plan or 40% off an annual plan.
I really could not recommend them more.
Now back to the episode.
So would an example of that be,
you know, a lot of people like to do things like yoga,
Pilates, stretching, you know.
Would an example there on a psychedelic
or even something like cannabis or some other drug
be that you're able to induce extra plasticity
in the motor system that
is encoding that sort of those positions those sort of bends those twists that flexibility
and maybe that even plays into why there's such a sort of overlap in those communities and why
those people like to do that i personally greatly enjoy you know doing something like psilocybin
and doing yoga and i just feel very in tune with my body but i also feel like i can improve the
positions very well and i am i think that would be something that a lot of people that do a lot
of yoga or sort of physical stretching or movement.
movement in the body feel went on psychedelics and there might be multiple mechanisms going on there
there's also rich embodiments but I wonder is there more as you say plasticity in the region that
you're activating if you're if you kind of purposely activating that motor cortex and doing those
positions are you more likely to even improve in yoga on 2x speed sort of while on the sort of
them things that are increasing plasticity I didn't maybe yeah so that's it I mean okay so you
cannabis uh all the things that i'm mentioning are
are strictly all these effects that I'm mentioning, whether it's the plasticity effects, whether
it's encoding into your cortex is strictly on the serotonin 2A hallucinogens.
So you know, psilocybin, DMT, LSD cannabis.
We actually did not see this effect.
We do not see or THC specifically.
We also did not it with ketamine.
If anything, they impair the encoding, the formation of memories in both memory systems.
Whereas it was really strictly the, you know,
I think we've seen it now three times at psilocybin, once with 2CB, and then once with MDMA.
So, you know, five different, three different data sets, five, or four different datasets,
five, three, different drugs, five differ times. Yeah. So, yeah. So I mean, I think it's possible.
I'm not sure with the, you now, motor, like, I know that when Alex Kwan and others have looked at
some of these plasticity effects, I mean like the main region that they see some of this is
it's where they just tend to look is like medial prefrontal cortex but i mean i think it's possible
that these plasticity changes could happen in many places in the cortex and part of me actually
wonders with for example like hppd hallucinogen persistent perception disorder um or even milder
versions of it where people just have visual distortions that linger for you know um even
after the trip is over i have to wonder under those situations was there some type of you know
in the cortex, in the visual cortex, which is like the back of your brain that processes
visual information. If there were some type of changes that happened there that then like
kind of persisted, and this is why people will still, if they maybe focus on the wall,
they can see like, I can kind of still get the visuals again, you know, the start of like
a trip or something. And then some people, they just get kind of full blown,
you know flashbacks, right? Which are interesting. Certainly more, yeah, problematic.
maladaptive plasticity. But yeah, I mean, could the motor cortex also be somehow, you know,
you do certain things while under the effects of the of the drug and then it persists maybe.
Yeah, it'd be very curious. No one, we haven't, no one's really tested that yet. So as far as I know,
I mean I think it would be really cool. Like, so this is what's known as like procedural memory.
So procedural memory is kind of more of your like, uh, like riding a bike or knowing how to play
the piano. Right. And it also actually
very much involves your basal ganglia. And your
basal ganglia is this like, you know, subcortical
structures that are thought to be like related to habit
formation and reward. So they're obviously very much
involved in like addiction and probably the main, you
know, structures that or study for addiction, but
they're also involved in, like, learning languages,
stabilizing memories, like including semantic memories.
And so, you
know, it's actually been shown now. So we did this like
mega analysis where we just
every, you know, resting state fMRI psychedelic data set, which resting state is a huge problem.
And unfortunately, that's the way everybody's been studying psychedelics and humans in terms of
like the neurobiological effects. And it's largely garbage. Just want to put that out there.
I'm just, all the default mode network stuff did not hold up under the 250 subjects, you
know, when we put everyone's data together, none of that default network garbage held up. That story needs to
default mode network garbage held up um that story needs to go to hell um but I need to be
unraveled more because we've talked about the Default Mode network a lot when you say it's
garbage maybe maybe lay out what the mainstream claim is and then what didn't replicate when you
looked at the data because that that's another yeah so you know there's this idea that oh
psychedelic shut down your Default mode network and by that really the main finding was that
there's a decoupling of brain activity. So like you can have like
two regions of the brain that, you know, their activity goes up and down together. That would
mean that they're highly functionally connected, and which might mean that there's a direct
connection between them. They're modulating each other's activity, or even there's
a third region that's modulating the activity. There's somehow communicating with each other.
And so, you knows, the brain will form different networks that tend to co-fluctuate together.
And so one of your networks is the default mode network, which is also your semantic and
episodic memory network. But I just told you how different those memory
systems are one is like very like you know episodic and involves reliving and the other is like
stripping away all the visual details and just forming concepts many of these concepts have
nothing to do with one sense of self like knowing what a pizza is right and of course there's also
you know i was born in texas there's self-relevant concepts but i don't have an episodic memory
for that right so these are very distinct memory systems they're both encoded in the default
network we should tell you that it's not like you know homogeneous
like you know network necessarily it has all kinds of things that it does and so um in any case you
know it was found that psychedelics during the acute effects would decouple the default
mode network especially the two hubs of the default mode network which is medial prefrontal
cortex and the posterior singulate and these two different regions of the brain now are not
communicating with each other as much so that was kind of like the big initial finding it was
found in a few other data sets but not every data set but people always emphasize the ones where
It wasn't.
was found and then everybody kept it kept emphasizing the default mode network and to be fair every
disorder has emphasized the defollow network from schizophrenia to depression to whatever OCD
um and the funny thing is with depression they'll they'll cite this one paper um from 2016 I
think JAMA psychiatry was a meta-analysis of like a thousand uh participants um where you know
500 let's say are you know about 50 percent or have depression 50 of them are
healthy. And they found the default network was less functionally coupled. And so it's
like, okay, or sorry, more functionally-coupled in depression. So it's like okay, there's
like this like over, you know, coupling and people are fixated on their sense of self
or whatever, right. Which could also mean something else like, oh, their episodic memory is good,
which it's not. People with depression actually tend to have a worse episodic
memory, but people will make whatever story they want, which is one of the problems with
resting state. You can make whatever store you want out of this, right? There's no behavior
or stimuli to constrain how you interpret those data.
I could just say that like, oh, your posterior cingulate is involved in like navigation.
Your medial prefrontal cortex is involved
in metacognition.
You're now really good at knowing where you are in space,
which I don't know if that's true of depression,
but I have a feeling that that's probably not true, right?
But I could make that story up,
especially if there's no behavior to, you know,
go against what I just said.
So in any case, what's funny is there's actually a paper
that came out in 2021 uh pnas another really good journal uh and it was like a thousand patients
with depression a thousand healthy participants they found the opposite effect that patients
with depression have less coupling of the defalmo network so the point is is that it's not a good
it's a good biomarker and the truth is that resting state is probably just a garbage way of
measuring people because you take a sample of people and tell them you know
whatever they came in with they're gonna like let's say if they're nervous about getting
me
their brain scan. Most of these people have never been scanned before. They're going to
have some patterns of activity that you're not accounting for because you're
not constraining that search space by giving them a task and essentially doing certain
subtractions that could like subtract out within a scan certain types of noise or variability.
And so anyways, resting state has this issue. And yeah, I think with psychedelics, you know,
there's all these studies that were going on with like 10, 15 subjects. And this default mode
Michael Pollan wrote about it.
And what's funny is that when we actually put the data together,
this includes the default mode network proponents, to be clear.
Yeah, we didn't find the changes in within network.
You know, we don't see these decreases in default
mode network connectivity under the effects of psychedelics.
It was an extremely unreliable finding.
And there was one subnetwork of the default node network
that actually involves a medial temporal load,
which is more involved in episodic memory.
and that was found to
be less coupled, uh, which, you know, now I can make up a story and be like, yep, that's my,
decreases in episodic memory that I see. You know, um, even though the proper study would be
like half people do a task in the scanner that involves episodic memorandum, and then I'll be
able to say that. But um, in any case, what we did see to go back to the basal ganglia story
is that um, there did seem to be increases in functional connectivity between the cortex and
like the first structural of the Basel game
ganglia that the cortex projects to before it projects to another structure on the basal ganglia
before then projects to the thalamus which then projects back to the cortex and then there's
like this loop um and then we actually did see some evidence although it was like I think it
depended on maybe the the pipeline that um there was multiple pipelines that we used that there
was also increases in connectivity between the thalamas and the cortex so it kind of completes
that loop a little bit um which is actually one of the first models circuit level models of
psychedelic drug action so Franz Volenviter and Mark Geier came up with this
you
of what's known as the thelamic gating model or the cortical striatal thalamocortical gating
model and so um i think if we are going to make any inferences about what this massive resting
state data set can tell us i think the evidence was far more in favor of this thalamic gating
model or this cortex to the the caudate uh cortex to the basal ganglia than any of the mainstream
you know the the i don't even want to talk about those other models but yeah
Don't you worry, I have a list of them that I want you to shred because I'm curious because
I know you're critical of some of these.
No, it's super interesting and it brings up a really, a really valid point.
I am curious.
really, a really valid point. I'm curious what percentage of brain imaging experiments are
follow this protocol of just resting state, contextless, low resolution snapshots in time.
Is this a, you know, symptom of all of neuroscience that we're following this protocol? And then
kind of scientists, as you say, are retrofitting a narrative that sort of works like, well,
this brain area is very loosely associated with this thing. We saw a very small effect size
decrease or increase therefore we can conclude this you know story which is what the human brain
does is it creates narratives and stories and coherence right so is that a symptom of all
of neuroscience and brain imaging and if so like what the hell do we do about it because
that's a pretty significant problem so what you just pointed out is actually quite intuitive
it's what's known as a um it sounds like you've done your homework um it's
what's known as a reverse inference and early cognitive neuroscience research
was making these reverse inferences in like let's say the early 2000s mid 2000s the funny thing is
early cognitive neuroscience research wasn't doing resting state they were using tasks and stimuli
but they were still going beyond the data um there was an there's a new york times article
that you can still find that um i don't think the data from this uh from what they were reporting
the new yor times article were ever published because eventually a bunch of scientists signed
a letter and sent to the new York times saying this is complete garbage but um they talked about
like they showed people pictures of
like politicians and they were like you know scanning people and you know it's like oh your
amygdala went off when you saw you know uh george bush and it was like oh that means
you're scared you know you're getting anxious about him as a president or uh you know
you get a picture of bill clinton and your insula goes off and therefore you're disgusted
you know yeah and so it was just like, you know complete nonsense your amygda can go off
for like many different reasons right you can go off for like seeing
happy pictures versus neutral pictures it could probably go off just from seeing like a checkerboard
versus a gray screen because there's some salience in the checker board versus the gray screen
your insula can also go off for many reasons that have nothing to do with you being disgusted
and so um people you know these these researchers kind of like went beyond the data
and they got a nice little new york times article about it but uh russ poldrack who's kind
of like one of the godfathers of neuroimaging if not a if not one of
of the OG Godfathers because there were people in the 90s although I
think he was part of that but he really became one of the like big like you know let's keep
everybody in check and so he wrote about he actually wrote a letter and like 30 different
cognitive neuroscientists actually signed it and you know it kind of tore up that whole first
new york times piece and he's wrote about this about this idea of reverse inference it's like
using um like abductive inference so you know kind of like um all you know uh lions are cats
but not all cats are lions right you can't and that's essentially what you're doing by trying
to infer mental function from brain activity by just looking at a picture of a brain when
there's numerous reasons that, you know, as you said, this region has like some, you
know, support of, let's say, anxiety.
And therefore, people are just automatically pinning anxiety to the amygdala.
It's just like absurd.
And so, what you have to do is you have make like educated reverse inferences, which
can involve, for example, tasks and behavior.
And so you know one example I can maybe give is like, this is, again,
would be a reverse inference but i think it wouldn't be like the worst one let's say um
under the effects of a psychedelic i'm doing a visual memory task but then i see like auditory
activity maybe that has something to do with like some weird synesthesia that people have and
they're able to like when they're to remember a stimulus they're using more like multi-sensory
you know integration or something that you otherwise wouldn't use this doesn't happen just to
be clear i don't know this to happen it would be really cool if it did no one's tested it but
you know that would be an example of like okay we can now make
that inference that like yeah people are now you know in order to remember things on a psychedelic
they're yeah using multi-sensory integration even though this is like a purely visual memory task
they're now using like auditory regions so that could be like not the worst you know and then
there'd be subsequent ways in which you could test that to be true it can be like a hypothesis
generating sort of you know device to have uh fmri there and so that's what people should be doing
and then then russ polterick made that point that like people are you know going beyond the data
But then the funny thing is resting state.
became a big thing shortly after that.
And a lot of clinical researchers aren't cognitive psychologists
or cognitive neuroscientists.
And developing tasks is not easy.
And resting state, you just throw people in the scanner for 10 minutes.
They let them do whatever they want.
And you got your data, right?
And then you can just throw a bunch of math at those data
and you're good to go.
You're not like having to create very specific design matrices
that when is this event happening in time.
And then, you have to convolve these like events in time
with what's known as like the human
dynamic response function because the because blood flow is slow and so you can't just say that like
at this event your brain's activity is happening it's really going to happen after that event
because that's what you'll you'll see blood flow in the uh with with fmi which is i don't know if
you knew that but what we're measuring a lot of times is blood flow not actual brain activity
it's a correlate and so it gets a little bit more complicated you know the types of tasks you
you have to have a shitload of trials you need like you know per condition 50 trials at least
And so you can have people
the scanner for like 30 minutes to an hour for one task and your inferences become a lot smaller
you can't make these grandiose stories about some freudian brain under you know resting state
like you can when you have like a simple memory task where all we can say is hey look did they
use multi-sensory integration for a visual memory task like which is a pretty minimal amount of
what you can infer um the funny thing though is you actually can analyze task data just like
people do this and then you can also make some grandiose inferences if you'd like you know
which I think more people actually should do because it keeps the kind of cognitive
operations constant across your subjects and your drug conditions and it might actually give you
a better baseline for comparison but in any case yeah that's what so the cognitive neuroscience
world has used resting state in various ways in fact my first paper I'm like middle author on it
involved a resting state scan after the memory encoding phase
to see if we can see reactivations of memories previously encoded that were associated with a high
reward versus low reward. And that's what we end up seeing is these high reward stimuli tended to
reactivate more during that rest period. So it's not like resting state's totally useless.
But yeah, the cognitive world likes to tether it a lot of times to certain types of behavior and
things that we can like quantify rather than just like willy-nilly, you know, scanning people and
and just making some bizarre reverse inferences.
So, yeah, so that was an issue.
And there's actually a lot of problems that the cognitive neuroscience world had, you know, these small-end studies.
You know, we were running 10, 15 subjects.
And now I feel like the psychedelic world, as well as the clinical neuroscience world, is also running into these issues.
I think the clinical neuroscience world has gotten better with running larger studies.
They'll have like 50 depressed subjects, 50 healthy people.
But the psychedelics world is still running into issues with not running enough subjects.
And then they're also making some really bizarre reverse inferences.
It's like a lot of people aren't getting this initial like history of like, you know, the
field of cognitive neuroscience.
They're just jumping right in and wanting to make some really bizarre claims that I just
don't think that we can do.
Right.
Yeah.
I'm super interested to get to those claims.
I can notice a few subtle jab comments about who I'm pretty sure you're talking about.
I'm curious if any of the listeners can, but do you think the low sample size in psychedelic
studies, I guess, is down to difficulty with recruitment, difficulty with ethics boards?
You know, you have to get a certain amount of people that have tried acid before under these circumstances and have relatively the same experience under psychedelics and they also like, is that playing into obviously I'm in agreement we need bigger sample sizes and bigger data.
But that's probably a pretty big constraint.
Yeah. So, I mean, I think it was and I mean it still is, you know, in the healthy people studies, it shouldn't be that much of a constraint.
Money is an issue.
You know paying a clinician is honestly one of the things that probably costs the most to do any of these studies.
They're not cheap.
And even if I'm paying, you know, somebody with a $250,000 salary, 10% effort, that's
a lot of money over the next few years, right, as well as for the research assistants, you
know, as long as for my own salary, et cetera.
So yeah, I mean, I think that the longer you want to study, the more salaries you have
to pay.
And then that can cost more.
But you know at least, for example, with the resting state stuff, I'm hoping that we make
these data publicly available.
There were some issues with it.
I mean, it's not up to me to make them publicly available.
But, you know, there is obviously like always the opportunity if anybody wants to like work with us.
I think in theory we can like make them available to that individual.
I still don't want to make any promises just yet because I'm not technically the owner of those data.
But we're about to have, you knows, assuming we do make it available over the next year, two years, whatever it might could be.
50 subjects with resting state. There's never a reason ever again that we need to be running
these small end studies. And I don't really, really see the reason to ever run resting
state research again. It's just like a waste of your 10 minutes, at least for the acute
drug effects. And it's worth noting that the post acute drug effects with resting
state are even more unreliable.
So I didn't even know why people are doing that. But you know, people, like I said, it's
like there's a little bit of a laziness that sometimes feel like where it's like, eh, you
No, we don't want to
tasks and program it and have like some real theory and like and again you can analyze task
data just like rest data if i have an hour worth of tasks i can just take the whole time series
and analyze it just like rust and it probably gives you better data frankly but people it's just
become kind of the standard of of especially you know the the clinical world in general and i
just don't think it's it's actually very driven by any good reason but you know people will do what
trying to make a daily routine of meditation, but probably for a different reason than most.
If you've been following the Giant Shogun for a while, you know that I'm fascinated by this
question of altered states of consciousness. I've interviewed top scientists who study psychedelics,
sensory deprivation, lucid dreaming, and they all say a similar thing, that these kinds of
states of consciousness can be achieved by trained meditation. Now, while I've always been
the top scientists, people who I really trust.
So I had to try this myself, but the problem that I've always
this myself but the problem that I've always had is that when I do meditation it's very hard to stay
consistent when I don't know if I'm improving making any progress or if anything is happening at all.
That's where Muse comes in. It's a portable EEG headband used by NFL athletes, NBA stars to train
focus during stressful moments. I've been using it to achieve deeper and more focused meditative
track my progress over time as I improve. Now, as you can see, it also has the added benefit
of making you ridiculously handsome. This has been verified by multiple sources.
This has been verified by multiple sources.
Muse measures brain activity, heart rate, and blood oxygen in real time and gives you live audio feedback that shifts based on your brainwave state.
When your mind wanders, you hear it.
When you drop into deep focus, the soundscape changes.
There's over 500 guided sessions that is sleep tracking and a structured mental health strength program built on top of the world's largest EEG data set, over a billion minutes of brain data.
my sessions for weeks now, and I finally have a dashboard that I can see my progress. I can
actually see I'm improving, and this has such a good feedback cycle for actually doing it every
single day, because I know that that graph is going up, I'm getting better. I'm going towards
something. Now, the device is definitely not for everybody, but at least personally for me,
it has turned meditation from this uncertainty of not really knowing if anything is happening
to an actual structured plan that I Can see something improving,
motivation to do it. And just how cool would it be to have an altered anomalous state of consciousness
and catch your brain in real time and see what's happening? I mean, that idea is so cool to me.
Use the link in the description for a special discount. Now back to the episode. So a drug that
I know you've studied that I've become super fascinated with is salvia. And I really want to
under salvia is just the weirdest thing I've ever, I've never experienced it. But so I posted a
It got loads of views and just have had hundreds, almost a thousand, I would say, comments of
just people talking about their outrageously strange salvia experiences, you know, people being
a sentient conscious piece of corn in the back of a pickup truck driving across in America,
in a bag of other pieces of sentient corn. And it's just like comment after comment of like
we experiences of people being sentient objects being folded
into chairs being books and each page being a part of their life bizarre and weird.
But you've looked at people's brains on Salvia.
You've compared them to other psychedelics.
I really want to know how do they compare?
Like how do the, what does a brain on salvia look like compared to a brain
on classical psychedelics, classical being five HD2A agonists?
Yeah.
So we qualitatively compared, we didn't like directly.
But there was a little bit of, you know, I want to say some tongue in cheek in that paper where here's a clearly distinct drug.
As much as people like to group these drugs together, I don't always find them to be as similar as people make them out to be.
You know, yeah, it's true.
People's salvia experiences are quite variable.
And it is really funny when people like talk about like, oh, like there's a salvia goddess.
And, you get to see her.
And people have the same salvia Goddess experiences.
And it's like...
our people had completely different experiences you know one person saw a mickey mouse cartoon
another person was transported to his grandmother house um another person felt like he was being
lifted up on like like as if like you know when you know like cheerleaders get like you
know um you're being held by like your your foot being held up um one kind of constant relatively
constant effect at least the initial uh effects before i feel like people just fall into salvia
is like there seems to be a pullback that people get you know people will report kind of falling
back into the couch becoming part of the couch etc um head of i knew a friend at a party one time
where he decided to hit some salivie he'd never done it before and he thought that people like were
like pulling his head back like all the way and he was like dude why'd you do that and i was like
i didn't do that like that was all you dude like your head didn't go back at all so that's um
chairs, I guess so many people talked about just like sitting beside an adamant object and then
suddenly sort of sliding into it and then the boundaries between them and the adamant objects
sort of just fused. Yeah, yeah, there is that seems to be like the most constant effect that I've
seen or heard about, you know, and it's kind of like the initial effect too. It's like eventually
people are long gone and you know that's that's in the past. The past 30 seconds. And but yeah,
Yeah, it's a very
it is a very interesting drug, but quite different than serotonergic psychedelics. It's not very
pleasurable. People don't tend to, you can get this in several states still. I mean, there's still
like 20 states that, you know, you get in buy it at a head shop. And for that matter, I think that
it's probably a legal. I think it's illegal to sell in states, like, for example, like here in
Texas, but you can probably order it online legally, and people just don't intend to use it
that way. And so it's a Kappa opioid agonist. It's a completely different
you know, mechanism of action, the capo opioid receptor is not the same opioid receptor as like
what heroin goes after. You know, you have three different opioid receptors and here's a different
one. And what's interesting is that, yeah, there's other drugs that actually also can activate this
receptor and produce hallucinogenic effects. There was a drug called enatoline that people were
studying at one point. And people have studied, at least in rodent models, like this capo
opioid receptor and seeing how it can maybe treat like cocaine or
immune addiction. Um, as it turns out, Ibogaine or its, and, or its metabolite, nor ibogaine,
um, might bind to this receptor non-negligibly. Uh, no one really knows, I feel like with Ibogain,
what its primary mechanism of action it is. Maybe it's multiple things. It's an NMD antagonist
like ketamine. You know, uh, maybe it's a serotonin reuptake inhibitor, whether or not that's
important to its, you know, effects. Who knows? Um, but at least it's subjective effects.
But yeah, it might be, it seems to be a capo opiate agonist.
And then there's a company called Gilgamesh that they're actually developing in Ibogaine
analog that's more selective for the capo opioid receptor.
And it seems where that's probably where its effects come from, assuming it's even psychoactive,
which it probably is.
And so, you know, at least with salvia, people could show that if you administer naltrexone,
so naltrekzone, naloxone, you, these are drugs a lot of times that can be used for like
opioid addiction.
So niloxone is what you give somebody in the street, right, if they're overdone.
And I'm not actually sure how long it lasts, but like 30 minutes, you know, maybe an hour
or something like that.
And in theory, people can, it goes away and then people can overdose again if they still
have a bunch of, whatever fentanyl in their blood.
There's naltrexone, which lasts a little bit longer.
And like suboxone is naltrexone plus buprenorphine.
buprenorphine is also capo opiate agnist, as is naltrexone.
So these drugs are all mu opioid antagonists, not bupremorphine.
but naltrexone, naloxone, but also capa-opioid antagonists.
So people did the study where you administer naltrexone
and this is actually people from Hopkins
as well as this group that was in Barcelona.
Administered nalterexone and they had people smoke salvia
and Salvadoran A, I think,
which is the primary psychoactive constituent.
And it didn't work.
People didn't trip.
And so I'd be really curious to see, like,
let's do that with Ibogaine, you know,
and there's some, like I've heard some stories
where people will take suboxone.
And you're supposed to like detox off of Suboxone before going to these Ibogaine clinics.
But I've heard, you know, if you read the reddits, it's a little bit all over the place.
Maybe I'm seeing what I want to see in there, you
know, but there's some stories where they're like, oh, yeah, I went to, you
know, the ibogaine clinic on while I was still on suboxone or I'd only gotten off of it
for like a day or two.
And I didn't get a very strong Ibogain experience, you knows, which then makes you think,
oh, maybe the primary mechanism of action of Ibogane is the capo opioid receptors.
So Ibograine, though, also not a drug that tends to be fun.
People aren't taking Ibogine, you now.
out there for fun and so this is by the way it all to say that i don't think that salvia is like
useless like i think that for all we know that maybe there is a therapeutic mechanism of action
maybe if you have to take longer doses of salvia you know if you could have a way to titrate it and
you don't have to blast off every time but just kind of sit there on it you know for the next two
hours taking puffs or something from a something yeah I don't know um or you could do the
the comments for for how weird the experiences were at almost every
single one and it's going to be granted a selection bias here for salvia almost everyone said it was
of the most positive experience they have in their life of salvia and that wasn't the case for
ibogaine and this is pretty big that video has 80 000 views i have other drug videos with 500k
there's a good thousand thousands of comments on drug experiences now and for some reason it
never they would always describe what didn't seem like a positive experience and then say
again. I'm like, did you read the dude? Did you read
the first paragraph? He was just said that you were like, you
know, living through the life of someone that you tormented in
school as a bully or something, it would just be some insane
experience. But they said, yeah, like it was profoundly
life changing. I began was different. People said I began
was hell. I am. Yeah. And you know, like, many other
experiences were variable, but at least and you know selection
bias aside, which could be all of this or none of this, who
knows I am
They were saying Salvia was ridiculously positive in their life.
Whether it was a positive or negative experience in the moment was different than if it was
a positive experience in their lives, which I think is also pretty interesting.
Yeah.
Yeah.
So, I mean, it is.
Yeah, that is actually surprised me because a lot of people I knew obviously in high school
or college, they tried Salvia and then they never wanted to touch that again.
Right.
There's also something to be said about, you know, preparation and whatever, right?
Even with Salvia.
And many of these people at high school or college.
like, oh, legal drug. They probably never tried any other drug other than maybe cannabis and
alcohol, right? And then they just go into it. And especially the idea that you have to blast
off, you know, is kind of like what people will say. Like, I hate when people say that
there's like, this is the only way to do a drug. Right. Like the only day to do DMT is like
massive doses. Right? And so it does kind of, yeah, like it's anyway, I think that's maybe one
reason why people have these negative experiences. But then you have the psychonauts, people who've
tell you uh one of the subjects from our salvia study because he wrote a vice article about it
um but i think in the vice article what he uh wrote about was that he had tried i think salvia in
high school or college or something and didn't really have a good time on it and then here he
like had like not a bad time you know i in fact maybe he would even say it was a good
time uh you can read the vice articles i feel like he was he was relatively positive about it
feel like he's probably done it again. You know, I don't know that's to be true, but like in the
sense of like, you know what, it's definitely like he wants to do Savia again. Yeah, yeah. I kind of,
I kinda got that impression. I don' think he ever told us that he did that. Um, and so,
but like it definitely, uh, yeah, and, and then there was another participant of ours that also
had a very similar experience that like you know, found it to be not as bad as the first time and
was kind of like, yeah, I would totally do that again, you know, like, and not like
again in like an addictive sort of way or anything like that but yeah so yeah i mean it yeah i don't
know i think it's it's interesting and so you know going back to that study that we did we basically
just found all the effects of the classic psychedelics at least at that time you know we found
it was a small study so we weren't powered but we did see changes in within network default
mode network connectivity that that went down on average if you just look at all the within
network connectivity which is like so essentially these networks you know there's
Let's say you can parse your brain into, let's say eight networks and then the
communication within those networks seems to go down.
So now the brains community, these networks are less, you know, uh, segregated.
And then what we saw was between network connectivity.
So the networks like are now communicating more with each other instead of within
each other.
Right.
That was kind of the story that was initially told, um, at least in that mega
analysis within network connectivity did not really go down the between network
connectivity went up, but.
very, very selectively. It wasn't like your default mode network is talking to like all the different
networks. It was like, I'm forgetting what they actually found. There was like some parts of the
default mode were talking to some sensory networks. Again, not everything was talking to everything.
It was very selective. That was the mega analysis with the classic psychedelics. But essentially
with Salvia, we were finding kind of the classic initial effects, at least that's where the trends
were kind of towards. But yeah, it's a very different drug, which would tell you that like, here's a
week.
completely distinct drug that like is still giving us the similar sorts of patterns that
people were talking about at that time right like this is i don't think that resting states the
way forward yeah and i mean with something like salvia blasting people off i'm not sure maybe
that's you know i don'T know what the other options are you can't necessarily give
somebody a task very easily but you can certainly do lower doses um you know if we had a way
to titrate things better i think that we could certainly do slower doses do certain tasks and
make better inferences or even you know there's other drugs
out there like anadoline is something that people have studied that's a capo opiate agonist where you
could inject that in a more kind of titrated manner so yeah bizarre drug for sure that i think would
be super interesting to study in more or even any capa opiate Agonist in more principled way but
it was the first pass yeah a really interesting question for me becomes does the brain have
several reality generating streams infrastructure mechanism and you know there's one way
into that infrastructure is the kappa opiate agonism and one way into that
infrastructure is 5-HT2A or is that sort of the top layer of the system and then
underneath where the chemical pathways kind of get obscured and we don't really
understand you know chemical cascade signaling pathways where we
don't fully understand the level of complexity that's happening at the molecular
level do they both come down into the same reality generating mechanism if you
get what I mean.
So I know they relate to it.
in the claustrum, they seem to have like both, both receptors seem to be highly populated
in the cloustrom, which is related to perception, but I'm curious of those distinct systems.
Then why do we have two reality generating systems?
Curious for you to even just riff on that.
Yeah.
So yeah, the classroom, I don't think we totally understand what it does, honestly.
Yeah.
And, you know, there was some research that actually found that when they stimulated
the classroom they didn't really get like, it wasn't the seed of consciousness.
the way that Crick and Koch had said, it didn't just knock people out, right?
It got some, some people got some bizarre experiences.
Some people got less bizarre experiences, you know, certain things, like changes in
perception.
But you can get that from like stimulating auditory cortex or stimulating visual cortex too,
right?
You can get actually from stimulating like anterior temporal lobe.
I mean, there's this one case study where this person's like, I'm back in high school
right now, you feel like it was stimulating anterior temporal load and it like brought this
person back.
There's, you knows, stimulating posterior cingulate and you can sometimes disassociate
association depending on the frequency you stimulate at so this shouldn't be super surprising that
some other region can also produce some bizarre sort of you know effects uh but it didn't seem to
knock people out and just to be clear there was a case study where they did stimulate the
clostrum and in that one single you know they're like it knocked this person out but I think
their electrodes ended up not being quite at the Clostrum where it should have been and then um
I'm forgetting his name, but he's over at Stanford.
I can't believe we're forgetting.
He's an intracranial researcher.
He's been doing this work.
I think it was him.
His name will come to me.
I can believe I'm forgetting it.
But he's the one who did the whole dissociation from posterior cingulate cortex.
So he, I think, then ended up having a bunch of patients where they had electrodes in
the claustrum.
And that's where they didn't, they weren't able to replicate this like loss of consciousness.
And instead they found that there was, you know, some people had some bizarre effects
with
maybe some people had less bizarre effects um okay so the idea that these receptors are somehow like
you know are like keys to certain type of realities is just not true because you can block these
receptors and you don't really get much of an effect if i block the serotonin 2a receptor
um alone without you know taking a psychedelic there's not really much that happens there um
you know people have done these studies with catanser
in which it obviously blocks a lot of receptors it blocks the serotonin 2a receptor blocks the
seratonin 2c receptor it blocks some adrenergic receptors nothing really happens um anti-psychotics
the second generation antipsychotics those uh do produce i think some sedation but i don't think
it's due to the block in the saratone two areas after they block all kinds of things but they
block the sarotone two receptor and i guarantee if you took a second gen antipsychotic you might
get a little sedated but you're not going to like feel like oh my like you know my my what's
reality is going away. Same thing with the capo opioid receptor. If you take naloxone and
altrexone, I don't think there's much of a subjective effect unless you have like opioid
addiction and then you go into withdrawals. But then like if so you're blocking those receptors
from doing any signaling. And in fact, there's other drugs too that can block the serotonin
receptor like myrtazepine, which, you know, I think probably makes you sleepy from it's like
an antihistamine as well. There's also trazadone. I actually should know what the mechanism is.
that makes you sleepy there but um it also blocks the serotonin receptor those drugs aren't like
changing the way you're thinking about reality it just seems to be that by activating them in
very particular ways and you can activate them in other ways and don't get a psychedelic effect
um so for example like you know you can activated with lisseri you can
activate the seratonin receptor that doesn't give you a psychedelics experience um there's also
a kappa opioid agonist that i think is approved in japan maybe for certain types of
Thank you.
I'm actually not sure, but it's not hallucinogenic.
And so it just seems to be that you activate these receptors in a very particular way in whatever intracellular cascade that goes on.
And it produces these bizarre effects.
But I don't really think of them as being like any different from like, you can, you know, bind to the cannabinoid receptor, CB1, and you can get a high.
You can also bind the cannabeid CB1 receptor with like your endogenous ananda.
mind and that doesn't make you high um and so it's like yeah you can just mess with these systems in
different ways and get psychoactive effects and some of them maybe we might consider to be more
interesting than others but you know it's not necessarily like some like channel into reality
or something i think that people might want to believe that that's like it's kind of like
sometimes i feel like psychedelics are people's like last hold out for magic you know
it's all right you know santa claus stopped existing all right we can stop you know believing
and religion, you know, whatever sort.
Like, all right.
You know what?
like all these times when my friends were talking about ESP and feeling connected, it turns out they
couldn't replicate any of those things. But you know what, psychedelics, that's the last thing,
you know, that that could produce some degree of magic in the world. And yeah, I don't know.
You know, it's like, I'm not sure they're any more special than any other crazy experience. I mean,
the patient with schizophrenia, we don't give them enough credit. We're going to say that
this is some channel to reality these drugs are. Well, then we'll have the patient
with schizophrenia who thinks that they're in some special reality and know, have some like
privilege.
knowledge like it seems kind of like a dick move that we uh don't give them any credibility but
these drugs we do yeah it's refreshing to hear that perspective i guess because a lot of
psychedelic researchers can be very um you know psychedelics of the holy grail of revealing the
you know ultimate perspective on reality the universe and everything um and it's interesting
to get a psychedelic researcher who is quite skeptical of psychedelics themselves and a lot
made. So I'm curious, everything you said there about receptors is super interesting, but I'm
curious what you think that teaches us or what's the extrapolation from that. What do we learn
from that? Are we placing too much emphasis on the receptor story because that's what we can
study? Because that's sort of the higher level stuff that we can point at and we can analyze and we
can build models around and build frameworks around and teach in textbooks. All of the
interesting stuff is really happening under the hood that we don't understand. And that's why we
dismiss it why we tend to summarize summarize it as signal cascade pathways all of this do you
think that's doing that's the case i think the receptor is super important i think that a very
it's a very unique way of causing certain downstream effects that would be very difficult
to replicate without just binding to the serotonin all the seratonin to our receptors in your brain
because it's now hitting very specific sets of neurons in a very specific sort of way it's hitting
all
You know, there's a lot of excitatory neurons that express the serotonin-2 receptor in your
cortex.
Many of them are in layer five, but I think the ones in visual cortex are in somebody's
going to get mad at me, layer two or three.
So it's in a different layer.
And depending on the layer, they project to different regions.
Then you have your hippocampus.
I think it's like 90% of inhibitory neurons express serotonine-2A receptors.
So now you're getting set up by inhibition.
You get a very unique pattern of effects by activating these receptors that can then
produce a very unique sort of
you know yeah experience so i don't know if we'll ever be able to replicate that unless we can
stick in like you know billions of electrodes across the brain and then activate you know
everything in a certain way that you know you're activating the serotonin to where a receptors
did right and so it does the receptor is important it's just that it produces a unique pattern
of activity that then produces a a unique experience i think and yeah i mean look you
block that receptor you're not able to get a psychedelic experience you know
so it certainly does um yeah I think I mean I do think it's important but I also I don't think
that there's anything special necessarily about like psychedelics or really any drug in terms
of like telling us something about like how the brain functions or you know any universal principle
of the brain of how the brains functions or any like you know truth about the mind or reality I
find that to just be like y'all are taking too many drugs you know um and I think there's this
knowing that we talked about the very beginning, this noetic quality that people get while they're
under the effects of these drugs. And that makes people like, you know, when they take them,
and they're also thinking about, oh, what can we learn about the mind? It makes people feel like
they're learning so much about the Mind. And yet when they come back from this experience,
they can't give me anything that's like tangible that's worth testing or anything that I can say
is like a principle of how the mind operates that we didn't already know, especially. And so,
Roland Griffiths would say, right?
Oh, we're going to learn so much about the mind from these drugs.
And he was like, Roland, what have you learned?
What have we learned about the mine from these
drugs, like, if anything, you know?
And it's been 20 years of doing this research.
This has been back.
And yet I've learned nothing about some, you
know, principle of how the mind operates.
I've earned more principles of the mind
operates from basic cognitive neuroscience.
And I've learnt about how these drugs work,
but I hadn't learned anything about like the mind or,
you know, some general principle.
the brain or consciousness or whatever. I mean, I think that that's, yeah, where people are kind of,
yeah, getting a little ahead of themselves. And like,
I think they need to check themselves about like, you know, it might feel like you're learning a
lot, but that doesn't mean that you are. And I do think there's certain things that we maybe
might be able to learn. It's kind of like how visual illusions can sometimes teach us about the
imperfect algorithms, um, about how like, You know, visual processing operates. There's these
rhythms that are have evolved over time to, you know, probably serve certain functions and,
you know you can kind of unravel them with visual illusions.
And likewise, there might be something here with psychedelics that, you
know, we can show that, hey, look, like, especially with some of these more basic effects, like
color constancy is the one that I always use.
Color constancy, is, I have this wall behind me, it looks gray.
But really if I were to take a picture and then blow up individual voxels or sorry, pixels,
up onto like a white screen, they wouldn't just be grays.
There'd probably be some purples in there,
depending on where the light's shining and how I took a picture.
There'd be some, I don't know, some reds, some browns, et cetera.
And, you know, yet we see this all as one solid gray wall.
That's what's known as like color constancy.
I think that might break down under the effects of psychedelics
and sometimes certain colors pop out more than others,
depending on the light.
Maybe depending on what the individual, I have no idea.
And I mean, that could be kind of a useful,
yeah maybe that maybe it does depend on the individual
individual maybe we can learn about the yellow dress versus blue dress from psychedelics but
no one's doing this research so they're just scanning people with resting state instead of studying
these like fundamental mechanisms of you know visual perception and I don't feel like we're getting
very far you know by doing by yeah that method um so I have a bunch about a visual person yeah
say web I've a bunch of thoughts on this so yeah super it's so super interesting I am there
produce unbelievably strange and reliably weird effects in people that aren't simply explained by
drug induces this highly specific inhibition and excitation of neural circuits in the brain
because what you described there is this extremely complex mix of neural circuitry change
that i totally agree with but then why do 90 of people see dmt entities and why do 25
percent of people see mantises. I know you're going to be skeptical on this, but it's super
interesting. Why do 50 percent of atheists have an ontology change and believe in God after
seeing a DMT entity? These are David Luke's figures from his field study that I spoke to two weeks
ago. Please go on this because that seems too strange to me and too weird to just be simply
explained by neurochemistry. I don't believe we have the neurochemistry to explain that.
So, okay, so first of all, I think that that
uh you know people's change in metaphysical beliefs i feel like sundeep i think did a study on this
and he wasn't able to replicate that i'm sure it's going to depend on the person i know loads of
people who take psychedelics and they're still you know atheists at the end of the day if not
atheist at least like you know it's like look we don't know and there's no point even trying to
speculate because like none of us are getting anywhere i've never heard any philosophical
argument that's going gonna change my mind you know like i there's nothing that there's
there's no new evidence and I feel like it's almost impossible to ever know and just why assume anything
and then live your life that way, being like, oh, I'm not going to have sex because the book told me not to, you know, or something like that, right?
And so, yeah, anyways, I feel like, okay, so now as far as that's one thing, the whole changes in metaphysical beliefs, I think, sure, some people have it, but they don't necessarily have the changes in the same way.
People are like, Oh, I became more Jesus forward.
Other people are like oh, i'm going to become a Buddhist.
Other people were like, i just don't know anymore.
I'm in a crisis now, you now.
Those none of those those things are not the same.
Let's be honest.
You know?
And then, okay, on top of that, the, you're talking about the DMT aliens.
Dude, I go a lot of people, people love to emphasize that.
I know a lot people who, you know, did DMT, and sure they see something, maybe things that
resemble people, but that also happens with psilocybin.
We are people processing machines.
I see faces and things all the time.
You know, you get a little drunk and your vision's a little blurry.
You see dot, dot line.
Oh, there's a face.
You know?
And so, yes, these things become more personified under the effect of
of drugs that are known to create better visual imagery, right?
But it shouldn't be much of a surprise that people run into entities.
And I don't think there's anything special about DMT.
DMT is, you know what psilocybin metabolizes to is psilocin.
Right.
You know what silocin is?
It's 4-hydroxy-DMT.
It's one chemical group away.
Right.
And so the one thing about DMT is that you don't have to do,
you can do massive doses.
And you can end up like being fine because it only, you're only tripping for 10 to 15 minutes, right?
Whereas you do a massive dose of psilocybin and you're going to have to sit there that whole
afternoon on that, you know, blast it off. And people do see entities. I mean, Roland talks
about this on, I don't know how many micrograms of LSD he took, but he said it on that New York
Times article. He saw his cancer, you now. That's an entity, right. It's an illusory social event,
right? That's what I like to call them. And so this is like people, so a lot of people don't
realizes, by the way, that DMT is actually less potent than psilocybin.
Miligram for milligram, less potent.
It's certainly more potent than LSD, which is active in micrograms.
But in milligrams, if you do oral psilocyanin, an oral DMT, which you have to take with
the molyminoxidase inhibitor, 50 milligrams of DMP is probably like your standard trip,
like maybe your museum dose, walk in the woods type thing.
I don't really like the idea of museum dose because the idea people going to museums on
these drugs, it just doesn't sound like a good idea.
lots of people. Let's just say the walk in the woods dose. Um, whereas like your walk in
the wood's dose of psilocybin is like 10 to 15 milligrams. You know, it's probably the equivalent
of like two grams of your average magic mushrooms, right? And so it's a very, yeah, I mean,
we were talking about much more potent. If you do IV psilocybin and IV DMT, IV psilsybin hits
within the minute. It lasts like 15 minutes, much like DMT. I think it's, DMT is a little bit
in two milligrams is enough to get you going. And I think there's one study where they went up to
three and people started throwing up. With DMT, there's these studies now. People are going up
to like over 20 milligrams of IV DMT. So milligram for milligram DMT less potent. So let's first
get that out, right? But it's the way that people will use DMT that they will use it at these
high doses. They'll inhale 30 to 50 milligrams because again, they don't have to sit there all
afternoon with it. So of course when you're on a huge dose, you're going to be
inclined to see things and like, you know, uh, and especially see we're, we're social machines.
So we're going to have these illusory social events. So then now on like,
you know psilocybin. If you, if you talk to people in these like clinical trials,
when they get 25, 30 milligrams or sometimes even more, which is a huge dose,
you know? We're talking like the equivalent of like five grams of dried magic mushrooms,
your average magic mushrooms. Um, people a lot of times will talk about like, oh, I faced
my cancer, like Roland said, or, oh, I saw my addiction, you know, or I saw something, or
like, you
know, people will say, Oh,
I saw like,
you know.
They get,
they get,
you know all like into the indigenous stuff.
They're like,
I saw some tribal people.
And it's like,
that's kind of racist,
but whatever.
You know,
it's,
like the people will see things.
They see entities.
They have these illusory social events.
So to act like,
you know everybody's seeing the same thing,
but I think part of it is like it's people's priors, right?
I hate using the term.
People's like knowledge that they come in with.
And I hate set and setting.
So I'm not going to say that one either.
But if you think about like, um, I think it was in, it was
in LSD, my problem child.
I can't remember if it was Albert Hoffman or if it
was his wife.
Um, they report like when they took psilocybin, they got
these like, you know, Aztec art visuals and then, you
know, they had the like kind of, uh, you
know, at least the, the foresight to say, you
know, or maybe should I say hindsight, they were like
saying that like, maybe I'm seeing this because I was like
biased to
see this, because I know that this, you know, the mushroom is coming from like Aztec use.
So, if you don't have that, you're not necessarily, I've heard people say like, oh, I saw like
a daff punk, you type, you guys with weird gloves, which is not really the same as the
mantis or the elves, you can try to like do some mental gymnastics to say, oh yeah, what
we're seeing is actually the same.
And it's like, no, it's not.
Daff punk and elves and mantises are all quite distinct.
you
So I don't know.
I think people really want to make, again, this is again, holding out in this like magic
that like they just really wanna hold out on this one thing being really true.
The last thing I was going to say is I do think it would be really cool to do a head-to-head
IB DMT, IB psilocybin, try to figure out like equivalent dosages, maybe
two milligrams psilocybin to be equivalent to 20 milligrams DMT.
I dunno.
And then like, do people really do report and don't tell people obviously what they're getting.
Do people really due report more like, you know, uh,
illusory social events i have heard people will say that there's something more like immersive
environmental with dmt that like it feels like you're moving through space more versus psilocybin
you know you don't get as much of that what seems like self-projected motion and maybe it's the
case that there is these self you know there's these motion generating self projected motion
generators in the brain that like are being somehow activated more by dmt i wouldn't you know
that that would be a good hypothesis that you could test with some tasks
Um, but you know, people aren't doing that.
They're just making all these conjectures that are like just based on doing a
bunch of drugs, which I'm not sure like I can, you know get on with.
Yeah.
I get on board with.
So I really, I appreciate your perspective a lot.
And I think I agree with it a lot, I'm just, I just don't know if it's all the way
there, like, so if I'm remembering my numbers correctly and you can be skeptical of
the data side, I think it was a survey of like 2000 DMT users, the team that David
looked at. And he said that somewhere between 15 and 25% experienced mechanical elves, I think
it was about 14% experienced mantis. And I'm like, that, that's one where I'm just like,
illusory sense, elusory social event. There aren't any mantis in Ireland. I didn't grow up
with any Mantis. I have no association with the mantis, they don't exist in my country as far
as I know. But yet people apparently cross culturally,
It doesn't matter if mantis exist in your country or not, we'll report mantis hallucinatory events.
And for some reason, the mantis are more often than not operating on you on a diagnostics table, doing some sort of surgery.
And look, I'm totally skeptical of the priors and absolutely these things are front loaded with belief.
I think one thing that we can say about psychedelics is that they ramp up the predictive machine.
machinery in the brain and the coherence making so if you believe you're going to have those
experiences I totally believe that that contributes to some of it I just don't know if it's a hundred
percent of the way there because these numbers seem too high and too anomalous like I'm with you
but couldn't we also say look what I'm what I think expectation and front loading and priors
maybe that's 80% of the explanation here but I don't now if we can confidently say that it actually
give an answer to these people's experiences because, you know, and apparently like in David
Luke's field survey, 34 people he interviewed, it was 94% experienced entities. And he assured
me as I asked him, is that a lot higher dose per dose of psilocybin? And he issued me yes.
Now I don't know if that exact study has been done, but I've asked several people this
question. Now, maybe they were biased on the DMT side. So they were still thinking DMT
entities are special but they prefer me that there was something irregular about dnt and its
entity encounters you can experience some of the other psychedelics but i believe there is a
quantitative difference it just seems to me that that that expectation and priors gets you a
lot of the way there and explains some of that but i just don't know if it explains all of it
and i have absolutely zero explanation for what explains the rest of it i've absolutely zero
you're you know experiencing i feel like is also what a lot of patients with you know certain
delusions schizophrenia feel right they're just like no evidence is going to convince them right
and that they feel like i don't know what else can get me there okay so you're saying 15 let's say
let's just say that's one of the numbers of the things that they saw 15 you don't think 15
of people who are filling out these surveys have read about like mantises and elves and read the
aeroids and read rick strassman's book the people who fill out these surveyes are massively
people.
I think there was actually a survey about surveys being biased that just came out.
And if you think about like, you're like, oh, that just seems like a huge effect.
Think about this one.
There was a microdosing study that came out, this is obviously a completely different domain,
but you can, it can really show you like how different things can be when you control for these conditions.
Right.
Um, there was a microdosing study.
That came out with open label for patients with depression.
And there was like a massive effect in depression that like persisted.
Even months after the micro doses ended, recently they did that study, double blind, randomized
control trial, and it didn't work.
It's not published yet, but you can actually see that there's a LinkedIn post from the CEO
who funded it.
And I saw it presented at a conference at ACMP in the beginning of this year and nothing.
So when you control for some of these conditions, they got like, because, you know, a hundred
reduction in the effect.
So likewise here, if we were to take, you know, some group of people like my mom, who has no concept of mantises or whatever, and we're to blast her off on some DMT, she's not going to see mantises and elves.
She might see some people.
She might say Krishna.
I don't know.
You know, like, I mean, this is what people ended up getting in that.
I think that religious trial, right?
They saw the, they saw the whatever God that they wanted to see.
Like, you knows, the Christians saw some Jesus or something.
i don't know if the muslims saw muhammad i
don't think they're supposed to but you know there's you know whatever i
don't if the the rabbi saw you know moses i i
know like i'm not really sure that it's actually as consistent as people
make it out to be and if people see this that was silasai been that whole
you know religious trial right people do see entities
like experiences, whether it's, you know, people that they know from the past, something that they
don't quite know what it is, daft punk looking things. These all do like, yeah, I think we're
sometimes trying to tell a story that's not quite there. And I think that it is so interesting
that the cultural trope of seeing mantises and elves has influenced so many people.
But I can tell you, I know a lot of people who've done drugs and who were, you knows, as skeptical
or their experiences like that.
The last thing I want to mention is that
when people do these drugs,
there is an impairment in episodic memory encoding,
as I just, you know, I've told you.
Hypocampal-dependent episodic remembering coding.
The more cortical-dependant thing,
the sense of knowing that something's happened,
and it can be useful on certain types of memory tests.
It can also lead to illusory senses of knowing.
But like if I show you a similar stimulus,
you might be like,
did I show your this water bottle
or a different water bottle?
And you'll be like...
Well,
familiarity, the like is ramped up, it must be the same water bottle that I saw. And it's like,
nope, it was a similar one. Like that's going to potentially, but it still allows you to remember
something from that experience. But the hippocampal dependent episodic memory, you're truly like
detailed, tends to be more veridical episodical memory. That gets impaired. And so now people
also going off of like, you know, they're kind of not very good memory. And they're like, yeah,
I saw something like a mantis. And I'm like, nah, you saw the daft punk guy actually, you
You know, you're just, but Daff Puck guys can kind of look like Mantises too.
And
So I don't know, I'm a little like, I dunno, people, like I said, really hold out.
Like, did you see the most recent thing with the DMT and seeing the like codes
to reality if you stare at a laser?
I don' don't believe that.
And I'm like, how many people?
Yeah.
So I think your explanation actually does go 100% of the way there
for the DNT laser experiment, because these are what I just not sure
it does on the grand scale for the mantises and the elves.
Maybe.
I think I think the laser thing might be more, there might be something more there,
there's codes to reality, but rather is there some weird distortions that, you know, lasers
are a very particular type of light, concentrated light, right?
Is there some really weird distortions that you combine with psychedelic, like, is there
a consistent distortion across people that looks sort of like symbols and that, given that
we're reading machines, just like we're social machines, at least those of us that we're
taught to read that you know that we will look for symbols just like we look for faces you know so
Yeah, totally reality, man.
It's the code to the simulation.
No, it's the real.
Yeah, I don't know.
You don't see it.
It's clearly the code because you shine a laser and you do DMT and clearly the
code to this simulation emerges.
It just, it is obviously the case.
I don'know how you can draw on any argument.
This might be a case where DMT, though, is special because think about this.
People take LSD at festivals and lasers and things like that.
I don'T.
Don't know anything.
Hey, look, I'm not a physicist.
I donno my lasers.
So maybe the lasers at, you know, those festivals aren't.
aren't the same as the lasers from a laser pointer right they probably do lose more there's more
dispersion um but you know people don't report on lsd and silo fibon when they go to these festivals
to CB that they see the codes to reality from those lasers I mean nowadays people do have the
DMT vape pens so maybe you know we'll see we'll hear people see coats to reality at these festivals
but I don't know maybe DMT really is special and you only get it with DMT if that's even a real
And again, I knew that's a selection.
pretty good critique done where a guy, you know, used like, use like computer vision software with
like OCR and kind of got the OCR software, which is like object recognition on the laser to
basically get these sort of quite close to these Sanskrit, Arabic, Japanese type characters that
people claim. So apparently the DMT isn't required. He got just like a computer vision
software to sort of decode the same letters. It's just sort of the optical
diffraction of the red laser at 560 nanometers or whatever it is does it tend to show up more under
the effects of yeah i think that's probably the case well i mean that that might tell us something
interesting about how the drug works i don't know what i know how like i mean it's like there's
like a um these supposedly there's these like uh uh what do they call like these cryptograms or
whatever where you watch these videos on youtube and it looks like nothing and they say if you
watch an lsd you can see the word love you know and then actually you can just slow i think you
down or you speed up the video i think you slow it down you can see the word love as well and it's
like okay so that might tell us that lsd is like slowing down the sampling rate you know right um
but i'm not sure what that tells us with the laser if you really do get a more likely you're
able to see some of these like symbols what looks like symbols anyways yeah under uh the effects
of dmt that's harder to do with the naked eye i mean again it doesn't tell us anything about
the codes to reality but just tell us a little bit about maybe how the drug works which again
Again, nobody's doing like
the proper research. And there are some people. There's this guy, Marco Akeel, who did a really
good visual perception study, looked at the Ebbinghaus illusion and how it gets larger
under the effects of psilocybin. And he went into the neural computations. People are starting
to do this, but not enough people are doing it. And we're not going to get anywhere with this
resting state nonsense. So, but hey, you know, I did want to point out something you mentioned
about, okay, so these drugs are like kind of, can kind of facilitate some of our priors or
saying you know i go completely against um the kind of leading idea of how the drugs work where
you know the the leading idea meaning the unfalsifiable idea that i didn't really want to talk
about let's get into it it's see this yeah yeah it's here it's hear i was gonna ask about rebus
let's let's talk about robin i i i had a great conversation i i think he's a gent i had
a great conversation with them but can you explain rebus and your issues with it what does it stand for
I'm why have you been jabbing at it all conversation I have yeah I mean I have it totally been that
wasn't honestly I really wasn't trying to go in on him or his model in much of the things I was
saying um I do think he sort of started the resting state movement when a lot of there was there
was actually fMRI and pet imaging studies with psychedelics that were done without uh that
were not done during rest they're done doing tasks and there's a really cool quote from one of the
They were studying MDEA, psilocybin, and I'm forgetting which stimulant.
Maybe it was methamphetamine.
They're kind of comparing them.
And they had them, you know, they actually said in the discussion, like, if you try to
do, you have people take these drugs, you're going to have completely different, you
know, diverse cognitive operations, especially compared to placebo.
And it just makes everything incomparable.
It's just a mess.
Keeping kind of cognitive operations constant allows us to see like, okay, when the brain's
in the same mode, how is it, how does it still operating differently?
the effects of these drugs that was like in 1998 this person had this insight uh i'm not going to
pronounce her name correctly i think is gazali's may frank um but he had his insight and then like
15 years later resting state fmri got big and so that's what robin kind of latched on to and then
he really made that the kind of mainstay of how we are studying psychedelics just to be clear i
spoke to robin very briefly on saturday night i was with david oritzso and
the netherlands and um we he david ritz i was like one of robin's really good friends they worked
together in imperial he's now david ritzo is now the uh head of um imperial college london and
dude david richard's great um you know amazing and anyways he calls robin at one point and then
you know i just i'd had several drinks in at that point i was just like look robin regardless
of what i write like i'd be more than happy to get a beer with you anytime you know and like i've
I've known Robin since like 2011, I think is
when I met him. I lived in London for a little bit. And so like, I have no, you know, bad blood
towards him, but I also don't have, I don't tend to mince words. You know, like,
I don's, I Don't like this idea of shortcuts in logic and in science that he tells nice stories
just for that purpose. Right. I want to actually get, understand something new and learn something
that I wouldn't otherwise know about these drugs and like, make sure that other people are
doing the same thing and not producing more and more like just echo chamber,
garbage, right? And so I think there were some good first passes, but then they kept going
with those first passes over and over again. Like to give you an example, there are more
publications than participants from some of Robin's data sets. So that is a huge problem.
Like there's nothing wrong with re-analyzing your data, but I don't know what the limit
is, but that's probably one of the limits is publications to participants. So, um, so
I'll talk about Rebus. So Rebus, let's go into
So yeah, this idea that you are reducing like the influence of high level
concepts or priors, if you will.
Um, and then now you allow this like influx of bottom up information
flow sensory information that somehow produces a bunch of entropy.
Okay.
So lots of problems with that.
So first, you know, and the high level concepts are especially
is going to be like coded in the default mode network.
So first of all, I just told you that there's situations in which we can get enhanced
sort of function of high-level regions that essentially you can get this what seems to be
enhanced encoding of memories in the cortex, at least, you know, to some degree, which is not
what you would expect. There's also semantic priming was shown to be enhanced under the effects of
psychedelics, which again would be a higher-level cortex, probably sort of a process. And at the
same time, episodic memory is also involved in the default mode network, and we get impairments
in the encoding of episodic memories so the other thing though is that
that although rare, it's not impossible that, I mean, in fact, an entity is a hallucination.
You can get full-blown hallucinations under the effects of psychedelics on really high doses,
especially when you close people's eyes and then you let them kind of go into it.
People can hallucinate their mother.
You have to have a high level concept of who your mother is in order to impinge on the
lower level sensory regions, which is the complete opposite of what Rivas predicts, where you
get relaxation or disruption of high level concepts and then an influx of sensory information.
So great.
greater bottom-up information flow okay so that's you know already from just a purely just conceptual
view something doesn't make sense there and then from some of my data it doesn't
make sense but people have actually shown including some of Robin's studies that you don't get
strictly bottom up information flow increases that there's actually a lot of like top down
increases there's bottom up increases, there's bi-directional increases it's probably an
artifact of resting state in what your sample just so happened to be doing depends on what you
whether or not you get increased bottom up or top
information flow i think you just saw a paper that was published that found decreased bottom-up
information flow um into the default mode network which is the exact opposite of what robin predicted
i think it was in pnas and it literally like came out like this week so again problematic right
um okay so a little bit more concrete for the listener just sorry real quick on
like what's the right way for the listeners to understand sort of bottom up top down these
are sort of phrases thrown around.
Like what's a good model
for people to kind of understand this simply what's an idea of a high level prior a lower level
sensory signal kind of just maybe make this a little bit concrete a high-level prior would be like
for a concept for your mother for example you have a concept of your mother or that you know maybe
the way that um robin explains it is that you have you have concept that walls don't breathe
and then none of the effects of psychedelics that gets relaxed and therefore now you're able to
see walls breathing rather than being like this is just a low level
sensory effect you get a bunch of binding of serotonin two receptors in your visual cortex
which they exist there and then it causes these visual distortions which is probably a more
parsimonious you know sort of explanation the truth is you can also stimulate visual cortex
and get various distortions people can get that so um so you know you might have a high
call level concept for your mother and so then to be able to hallucinate your mother it wouldn't
be a relaxation it'd be a strengthening of that high level concept that is now kind of forcing
down onto low level sensory areas.
So you think about visual processing, the way visual processing tends to work.
If you look at like what happens in visual, early visual cortex is your visual
cortex is like kind of like, um, it's what's known as like retinotopic mapping.
It's like different parts of visual cortex.
Let's say up here is responsible for like, uh, representing like vision down low here.
And then vision, you know, on the right side of your visual cortex
is going to be mapping your visual field to the left.
And so if you know you, um see a point of light right here, you'll
get some activity down here. You see some point of light down here, you'll get some
activity up here. And so it's just points of light. It seems to be what the early visual
cortex is responsive to. As you move up the visual processing hierarchy, you get edges
and then shapes and then even high level concepts where you'll have like certain regions,
like as you move more interiorly, more, more forward in the brain. It's like this region's
only responsive to cats. This, you know, this neuron, these neurons are only responsive
to cats. These are responsive to dogs. And so, um,
That's not to say if you got rid of that neuron, you'd forget what a cat is, but they have some
selectivity for certain types of concepts. And so it becomes more kind of abstract as you move
up and you can have these like higher level concepts such as your sense of self or your mother
or, you know, walls breathing, which I find that one to be a little absurd. Um, and so,
this idea that they get relaxed is a little weird when you can get full blown concepts.
There's also like strengthening of high level beliefs that happen during the acute effects.
say oh my god i'm god i can fly right that's a pretty crazy concept a high level concept to have
that people sometimes retain after they leave the experience which is far from relaxation um
i also find it kind of ironic that the relaxation of beliefs guy can't relax his beliefs um but
you know that's you know maybe a different story maybe he's uh not doing enough psychedelics and
No, I'm just kidding.
So, so yeah, I mean, I think that
that's it's already promised no top down versus bottom up so you can think of like you know one way
is attention you have bottom up capture which is like something will i hear a noise and then it's
like oh it directs my attention alternatively there's like top down attention where it's like
i'm gonna look for the red thing you know i'm going to look for where's waldo i have an idea of what
waldo looks like i'm just gonna like look for him there so you know you can also have like
literally the way information's processed where things are more uh uh driven like sensory
information's driving high-level regions.
versus high-level regions driving like sensory areas right right so um that's essentially the
idea there and so i think that both of these things can happen there's obviously seems to be kind of
like for example bottom-up attention people like visual capture oh something shiny on a psychedelic
you know but alternatively people also hallucinate these high level concepts so i'm not sure that
there's anything about this bottom up top down that it doesn't seem like it's getting us very far
least when it comes to these resting state studies where people are seeing bi-directional
information flow increases just depending on the study some studies that show top down some
of those are bottom up some that show both and so yeah that's that's all that's so that's
problematic the default network story we already went over that was also part of rebus that one's
mixed you know default network is not the primary action of psychedelics at least in these resting
state studies. The entropy story, massively problematic. We just put a page
paper out on this. And I think the thing that I find, you know, in addition to like all the
different measures of entropy, there's no such thing as like one measure of entropy. And they're
all technically measuring different things and very dependent on like what you consider a state
of some sort, uh, can really change what you're in a finding. Um, but one of the things that
I find really problematic is that if you have somebody do a task under the effects of, you
know, a drug versus on a placebo, your change in entropy goes away. So Robin would say that,
because when you have people do a task on a psychedelic, it grounds them.
And if that, if entropy were explaining something unique about the effects of a drug, well,
then that would mean that entropy should go down under the effects
of a psychedelics when you would have them do a test.
So let's say, you know, resting state, you
know, people are like all in their head and getting some crazy visuals and
entropy is really high on a psychedelic.
You then force them to watch a movie or do a tasks and they're like, okay, I now
have to fixate.
So you should get a reduction in entropy.
um that's not the reason why you remove the effect compared to placebo and then on placebo it's
like okay you're you know resting there and then you know you're like maybe this is the way the
curve would look it's like, okay you rest there and you do like nothing and you know entropy is
low maybe you're falling asleep in the scanner which tends to happen and then maybe it goes up
a little bit if you have people watch a movie because now they're more engaged with something
right and then, on a psychedelic it goes like this right you have like a curve that looks like
this where it goes down. That's not what they found. What they found actually was that if you
If you have people watch a movie or do a task on a psychedelic,
get a numerical increase and then on placebo you get a massive increase so really the reason for
the reduction entropy is because if you have people on placebo conditions falling asleep in
the scanner versus engaging with a movie yeah your brain becomes more complex it goes through
more numbers of states and so the way my friend sunde you know at hopkins said he's like wait a
minute wait a minute so if you do in the scanner shit happens in the brain i was like yes that
That is what they found.
And then you just think about the general concept of like, it's just a redesign.
description sometimes of like the experience of people get on psychedelics. Of course, people's
experience with psychedelics are far more variable than their experience in normal, you know,
waking consciousness, right? Like, so what psychedelics just scramble the brain? That's
essentially what like entropy is trying to say. It's not very, it's not a very descriptive. It
doesn't tell us anything new. And it's trying to use terms for physics to give it validity,
but it doesn't quite, it'S not capturing anything new, I don't really know what we're gaining
from this. So it's really like captured an audience in a very sort of simplistic way that,
know, give makes people feel like they're giving some credibility to like these studies of
psychedelics. But I think it's actually doing a massive disservice by acting like we're learning
more than we really are. Super good breakdown. Really interesting. Just by chance, 10 minutes
before I jumped on this call, Robin tweeted, and he has an update to like it was, I think
it was brain, entropic brain today or something titled like that. And he said that increases in
predict mental health outcomes and breath of subjective experience.
He only published the abstract.
The full paper wasn't released.
Breath of subjective experiences such as arousal.
And he was basically saying that's the reason the model has validity is that entropy measured
in this way predicts mental health outcome.
Do you think that if the model entropy measured in whatever way he's particularly measuring
it, if it does predict mental outcomes as a model, wrong or right, it's
useful do you think there's an argument for that so i think he's basing that on a study that he
recently did it was published in nature communications um word on the street is that he thought that
i was the one who kept rejecting it um that's what i heard from somebody uh before it was
eventually published which is not true i will say that i'm not going to say whether or not i reviewed
it at all but it i i don't tend to reject things actually um i actually will give like i think
give a lot of like points and then an editor will be like oh he has 30 points to say about this it
must be that the paper's bad and then they'll reject it based on that but um yeah so he's talking
about that paper i think um where they had eeg and isn't healthy people and um during the acute
effects they had eg and then i want to say that you know i haven't read its latest iteration but
was a pre-print for a while too so you could see it previously um i want to say that they
It wasn't as clear as just the Lempel-Ziv complexity during the acute rest, probably
eyes closed, I would assume predicting benefits in well-being, but rather it was like they
did some type of like path analysis or mediation where like, it was, like the Limpel-ziv predicted,
you know, breakthrough, which then predicted, you know emotional breakthrough, whatever
that is.
is, which then predicted well-being improvements. The problem is a lot of these things don't replicate.
I've, you know, look, the modularity findings, those don't
replicate. I was just at a conference, ICPR in the Netherlands, and people were just saying
that they couldn't replicate the modularity findings. And I actually heard that from other
people prior to that. I would just be surprised if it did. And if it
did, okay, think about this. Okay, yes, you have people resting and typically falling asleep under
eyes closed conditions when they're on placebo and now a drug that wakes you up more and blasts
you off into all kinds of different you know weird places i do think that the subjective experience
probably plays some role i don't think it's probably the only thing but i mean we sort of
talked about this like what you're activating during the acute effects might dictate where some
of those plasticity effects come or forget even the plasticity stuff it's going to dictate what
like burned into your cortex, what you learn, you know, if there's certain like driving of
change in these high level concepts, like, what I was sort of talking about, whether it's,
you know changing your concept for what a bird is and saying that it's a dinosaur or changing
your sense of self.
And saying that, okay, you're not a bad person.
Like driving that type of learning, doing the acute effects, which is going to change the
experience, probably has some influence on your subsequent, well-being or what you think
afterwards.
I
don't deny that um and is it the case then that like is there something that you can capture
from just lemple ziv that you know which is you know essentially a form of a compressibility
of a signal it's like g-zip on your computer how compressible is this signal uh that you
know is maybe like a little crude but can tell you a little bit of something about but we we kind
of already have this like we can without need we don't need eeg for this people have shown this
this with like
Mr.
mystical experiences or oceanic boundlessness or whatever, emotional breakthrough, all these
just basic subjective effects, which is like, did you have a bizarre experience?
Great, because that's probably going to predict improvements in your well-being.
Like did we really need this brain signal that I don't think is probably going
to be better?
And just to be clear, you know, mystical experiences don't always predict improvements, right?
It's like some of the time in some studies they did, some of them they didn't, sending
the emotional breakthrough or whatever.
But I'm not sure that the brain signal, it's still quite crude.
and probably not as good as certain subjective effects.
I think at the end of the day,
if people just got a really bizarre experience,
that's probably more predictive of improvements
and well-being, as long as the bizarre experience
wasn't like super traumatic itself.
Like I know that the challenging experiencing thing,
something will say like, oh, it's not necessarily bad,
but I think if you have a really challenging experience,
I could be off on this one, but I'm always positive
that if you've a really, really challenging
experience, that can actually be worse.
It's honestly like gonna necessarily predict
like you're going to, you know, be worse.
off but it might be less of an improvement um this is kind of true of like exposure therapy
exposure therapy a lot of people go through some hell doing exposure therapy and i think the
ones that go through more hell um they still get better it's just less improvements and so
there's probably some truth to that here with there's a degree of how challenging things are
we're like obviously if it's super super challenging you can re-traumatize somebody right but um if
it's some-if it's really challenging they might still get betters it's not going to be as good as
the person that had a somewhat challenging but like actually kind of beautiful
and also just really weird and whatever, you know.
But yeah, we don't necessarily need the Lempel Zip,
and I'm not sure that that entropy thing is predicting.
I'm sure that there was probably a bunch of other measures we could look at.
Like how impaired was their episodic memory encoding?
That might predict, during the acute effects.
Right.
That might protect, you now, emotional well-being,
although I'll also say like doing a task that's really like trivial,
it's also going to, yeah, take people away from thinking about the things
that they probably should be thinking about while they're under the effects.
Maybe there's a way we can do sometimes
of episodic memory tasks that's embedded in like priming people to think about their problems
and then show that hey look we primed them to think
about their problem but they can't remember anything that we did
oh that was associated with improvement in their mental health you know i think
that would be the way to probably do something like that but um i think it was
katherine preller somebody did a study and they'd scanned people at like an hour
and i think i don't know if it was alcohol use disorder depression and they
didn't get much of an improvement in there um much of it i
don't think they got a significant clinical improvement it might have to do with
like, yeah, throw people
into a claustrophobic environment when they're under the effects of a drug and yeah they're not
going to get better you know surprise let me yeah i don't think they were worse but yeah i would
love to host a debate between the two of you because i think that would that would be that
would be highly entertaining debate conversation however you want to frame it i think
that would really useful really interesting i've heard somebody actually maybe threw that idea
at him and um david aritzzo told me he was like i would tell robin do not do that
so i don't know today i wouldn't be i robin was awesome but i'm not too i would not sure i'd be
on his bedding corner either after hearing you today i'm Not gonna lie it is funny like i
don't actually think i think fast enough to ever want to debate anybody like phil corlett that
guy you know when i got in his good graces i was like oh my god like i felt really good about
myself because i wrote that you know critique of one of robin's papers like within you know several
hours and then you know phil hardly added it he just he's
them the language which is really nice you know it's that nice added that nice english touch but um
but yeah i don't think i think fast enough you know to debate but um after robin's response to our
critique i would i would go in on a debate with him for sure well i'll set it up mino has just been
so much fun i i really want to get all perspectives on the channel and i really do want the very
skeptical one i am i think your work is super interesting i i hope we have many more conversations
because you are so insightful on the farm.
I love how you can just riff off drug names, mechanism, receptors.
Everything is super, super cool.
Awesome conversation, man.
Thank you so much for your time.
Thank you.
So much for having me.
Have a good one.
The Giant's Shoulder mission is to explore radical ideas in biology, neuroscience, and consciousness
and elevate those stories to the highest possible level while keeping them accessible to everyone.
If this interests you and you want to support independent science, then please consider subscribing to the Clips channel.