Latest / The STA Clinician Podcast / Ep.273 'Understanding Pain in 2026 - Pt2/4 - Predictive Processing & Active Inference'

Transcript

Ep.273 'Understanding Pain in 2026 - Pt2/4 - Predictive Processing & Active Inference' · The STA Clinician Podcast

0:00
You're listening to the STA Clinician Podcast, connecting clinicians across healthcare.
0:19
The guests straight on the screen. Apologies for that. We thought we were live
0:22
already. So we're going to jump straight into it. If you have joined us,
0:25
then leave something in the comments.
0:26
And if you listen to the podcast, then I'm sorry, this is a very abrupt starting to the episode.
0:30
You are here, you're listening or either joining us live for the SDA Collision
0:34
Podcast, formerly the Sports Therapy Association Podcast.
0:37
And my name is Matt Phillips. If you don't recognize the voice,
0:39
creatorofunchetlive.com. And we are recording this episode live on the Sports
0:43
Therapy Association YouTube channel and Facebook page.
0:46
Like I say, if you have joined this live, then we'd love to see comments in
0:49
the section where you can ask our guests questions directly.
0:52
This is part two of our focus on understanding pain in 2026,
0:56
in which the topic is going to be predictive processing.
0:59
Which I'm going to define this now, and I'm very worried because with me is
1:02
Professor Dave Newell and Anna-Marie Matsieri, and this is their bag.
1:05
And I'm just going to say it, as far as I understand, predictive processing
1:08
is where your brain guesses what's happening and outputs pain when it predicts danger.
1:12
But also we're going to be looking at active inference, which is where the brain
1:14
acts to confirm or update those guesses, which shapes how you move,
1:18
feel, and interpret sensations.
1:21
So with me today to discuss that, as I say, Professor Dave Newell,
1:25
Head of Centre for Pain and Active Inference Research at Health Sciences University,
1:29
and also Anna Maria Matsieri,
1:31
who is currently a PhD student at Health Sciences University,
1:33
and also, as many of you will know and recognise, Director of the ST School.
1:38
Welcome, Professor Dave Newell and Anna Maria Matsieri.
1:41
Thank you for being right. Yeah, nice to be here, Matt. Thanks so much.
1:45
I really appreciate you joining us. Last week, if you haven't heard it last
1:50
week, basically, then just we had Dr.
1:52
Bonnie Lennox-Thompson and Ben Wybrow, who gave us a fantastic introduction
1:55
to understanding pain in 2026.
1:59
Tonight, we're looking at something which I think predictive processing,
2:03
I don't want to imagine or presume that our audience are aware of that.
2:06
So I'm very keen just to go over what predictive processing is,
2:09
and then we can go into active inference.
2:10
But before we do that, Anna Maria, a lot of people, like I say,
2:14
know you from the ST School and the Hanson Hub and for making Therapy Expo fantastic
2:18
over the last couple of years and so on, and the other conference which we put together.
2:23
But not so many people know you or know that you are currently starting a PhD.
2:27
So maybe you could tell us a little bit about that and how you met Dave Newell.
2:32
Yes. Yes, I applied for the PhD because...
2:36
I've seen it advertised, and I follow Dave Newell anyway, because if you don't
2:41
follow him, please do so on Facebook and social media.
2:44
Read some of his papers, because he's one of our brilliant minds.
2:48
And I already follow the HSU University, because they're very,
2:52
very progressive in what they're doing.
2:54
And the PhD title that came out really, oh, what can I say,
2:59
really matched what I was studying at the time, and what I was working with,
3:06
which was the trauma paper with Laura Case and the active inference and so on.
3:10
And I've been learning about active inference for the last three years and virtual
3:15
reality and so on. So it just came at the right time.
3:18
I've never expected myself to go for a PhD, but actually I am enjoying it very much indeed.
3:26
And I urge everybody to maybe rethink about their problems
3:34
pathway so it's for me also it's it's
3:37
not just going for the the phd it's about
3:40
collaborating and learning more and matt you
3:42
said you know you do not know how much much about predictive processing and
3:47
active inference we are all in the same boat this is why we're doing it we are
3:51
all learning this is so very new we're still all getting our heads around it
3:57
and this is very exciting coming from you because anyone who knows you knows
4:01
that you've been on the case,
4:02
working with people with pain and trying to understand pain for I won't say
4:06
how many years, just so I'm not going to be your age, but a long time.
4:09
Pain and touch. This is a lot.
4:13
Pain and touch. This is what for me, you know, viewing touch,
4:17
not talking about massage or maniotherapy,
4:20
touching the broader sense through a different lens within pain,
4:26
within therapeutic environment and so on.
4:29
Fantastic. Okay, looking forward to it. Before we turn to you,
4:32
Professor Dave Neal, just got a few apologies to make. The crowd have joined
4:35
us now. They're flocking through the door.
4:37
I'm so... Apologies, for example, Sarah Jones here from the ST School as well
4:40
says, oh no, all of my witty comments have been lost. I'm sorry,
4:43
Sarah. I'm sorry they're not around.
4:45
I'm getting Glenn Murphy. Well, Sir Glenn Murphy, so he can say anything he
4:48
wants, but he's telling me that we're 30 minutes late. Apologies, Glenn.
4:52
Sorry, we did start. Italian time, Glenn. Italian time. You should know us.
4:55
I'm very sorry. Yeah, we can blame it on Animated Matt's area. That's fine.
4:58
And Gary Benson, founder of the STA, is also here joining us.
5:01
Thank you for being patient, people.
5:03
Gary says, lots waiting on YouTube and nothing happening. Someone just messaged
5:07
to say it was okay on Facebook. Okay, so maybe you can drop it in the comments
5:10
whether we are on YouTube as well.
5:11
But if I have let down everyone on YouTube, I apologize.
5:15
There will be a recording, obviously, and I'll make sure it goes up on YouTube.
5:18
And anybody else is here. Nikki Mansfield is here as well saying nothing's on YouTube, man.
5:23
Okay, well, hopefully you can spread the word and make sure people can join
5:26
us on the facebook link cecily hisloppy as well good the old faithful,
5:30
people have all come through the door and they found us wonderful right then
5:33
so sorry about the introduction professor dave newell first time on the podcast
5:36
pleasure to spend time with you.
5:39
And yeah maybe you could tell us a little bit about yourself and and how our maria is getting on.
5:47
Yeah we'll leave maria to the last hour just uh i'll come to that in a moment And all joking aside,
5:54
it's very unusual and it's really great for us to have somebody like Marie who's sort of, you know,
6:01
already come in at 36,000 feet in a way, you know, because she's done a lot
6:07
of stuff before, she's published before, and so she's got a lot of expertise,
6:10
a lot of real clinical insight, and also obviously teaching all of this as well and setting it up.
6:17
So I think we're, you know, we feel mutually lucky, I think,
6:20
to have her at the center.
6:24
And we've got some, we're just going...
6:26
Through trying to figure out precisely where we're going to land at the moment.
6:31
I think we've landed somewhere much more than we were a few weeks ago,
6:34
which I think we're all sort of flying around like jet fighters in the air,
6:38
trying to, you know, come together around what we wanted to do.
6:41
But no, I'm really looking forward.
6:43
Anne-Marie is one of two PhDs that we have, or three PhDs if you include another
6:49
one that we've got that's slightly out of pain, but it's still in the center.
6:53
And we've started to build, you know, a real community. in the center around this particular area.
6:58
And I think active inference is, I can talk about that in a moment,
7:03
but active inference, I think, it's a profoundly sort of paradigm-shifting idea,
7:10
I think. And I've heard quite a lot,
7:13
a few people that I really respect that are out there and written books around
7:16
this who are sort of calling it a sort of paradigm shift that is almost Darwinian
7:22
in its impact in some sort of ways.
7:24
But as with most things, it takes something like 14 years for new stuff to come
7:30
into research and research mentalities and the public arena.
7:37
And you know and uh for some academics it
7:40
obviously takes about like 40 years for for things
7:43
like that for me anyway it takes me 40 or 50
7:46
years for this to sort of land so but i'm
7:49
we're really excited about about that and i i don't
7:52
know whether i don't know whether you want me to go on and
7:55
just give a quick pre-say of the difference between predictive coding
7:58
and active inference it's a bit of a counter intuitive that
8:01
i've actually i think you probably just already
8:04
you know taken my whole speech matt because i think that little bit
8:07
of beginning there it's pretty much justice but but
8:10
but i mean we could we could unpack it a bit more if that
8:13
that's helpful i think that'd be great particularly because i imagine
8:16
i don't want to presume or put words into our listeners minds or mouths but
8:21
i think let's just repeat what predictive processing is in in your language
8:25
and then maybe after doing that we can compare that to active inference and
8:30
show any similarities or differences or the relationship between the two would that be okay.
8:34
Yeah, sure. So this idea that, you know, if you look back at those diagrams,
8:41
you know, back in the sort of 1500s, 1600s, 1500s, I can't quite remember when Dacre was around,
8:48
you know, of the guy putting, yeah, 1630, thank you.
8:51
That's the diagram of the guy putting his hand in the fire and the wires going
8:56
up from the hand to the brain, you know.
8:58
And, you know, things have moved on a lot from there, obviously.
9:01
But I think still, you know, this experience that we have of somebody behind
9:06
the eyes, you know, it feels like somebody's here, you know,
9:10
looking out, like some sort of pilot,
9:13
you know, sitting up just behind my frontal lobes and, you know,
9:17
driving everything and so on.
9:19
And yet when you look at the brain there's nobody there you know
9:22
there's that you know you can't see it you can't record it
9:25
there's not like a little me on looking at a screen and
9:29
called the theater of the mind which is an interesting way of looking at
9:32
it and and so there's been you know
9:34
a lot of work over decades now you
9:38
know recording brain signals and so on trying to figure out you know
9:40
how you know what are the mechanism behind
9:43
how we perceive and if you
9:46
so predictive processing what came out of that
9:49
but started like in the 18 1880s with
9:53
helmholtz probably started before that with canton people
9:56
like that and with this idea that you that you guess what's going on so it sort
10:03
of inverted the way the idea that you know sensory signals come first and they
10:09
all travel up to the plane and then gets integrated in some sort of way and
10:12
magically out of that comes this experience.
10:15
And so predictive coding was sort of born of that idea and essentially is precisely
10:22
that. So if you think about it,
10:25
the, you know, photons don't go into the brain.
10:28
It's not like your eyes are windows or a light goes in. You know,
10:32
as we know, all that goes in is clicks and electrical signals,
10:37
all right, at varying frequencies.
10:38
And that's true of everything that arrives in the brain.
10:41
And so, you know, I think one of the things that I try to sort of get the students
10:46
to understand, and probably don't succeed a lot of the time,
10:49
is if that is true, if the brain is in this sort of bony black silent box.
10:54
How the hell do you experience what you experience when you open your eyes?
10:59
Because if that's all that's going on, there's a whole bunch of electrical signals
11:02
in there. This is an amazing experience. I'm just having, how the hell is it doing that?
11:06
And so, so if you've got the idea that, you know, you can't actually,
11:12
you know, behind this barrier, you're behind the barrier of your skin,
11:15
the brain's behind the barrier of the skull.
11:17
The system that is you cannot reach out into the world. It doesn't experience the world out there.
11:24
It can't reach out. It only has this stuff coming in.
11:28
And so the idea would be is that you have to sort of infer or guess what's going on.
11:35
And that's the typical side of predictive processing.
11:40
But you can't just guess randomly.
11:45
So you're trying to guess what the signals are coming from. So what is causing
11:49
this thing that's arriving, these signals that arise in my hand?
11:52
But you can't guess randomly. You can't be walking down a street somewhere and
11:57
see something out the corner of your eye and just randomly guess that it's an elephant, you know,
12:02
or randomly guess that it's an aircraft or whatever.
12:07
You know, you need to make an educated guess about what it is.
12:11
So you have to have experienced
12:15
the world over some sort of time period
12:18
in order to build this model of
12:21
the world that you can then guess against
12:24
so it's like i got a model of the world it may be good it
12:27
may be bad but it's better than just random guessing
12:30
because if you random guess you know
12:33
a black shape coming down the the the road is i don't know something benign
12:38
like i don't know fairy dust cloud and you step out in front of it it's a bus
12:44
you know you're toast basically all right so you have to guess in some sort
12:48
of way you know that that's accurate so you build a model.
12:52
And then you use the model to make educated guesses about what sensory information is going on.
12:59
And then you are either near to what's going on in the real world or you're a long way from it.
13:07
And so those things could be no gap between your guess and the sensory information
13:14
coming in or a very big gap.
13:16
And so the way that
13:20
you learn is that you constantly try to reduce the
13:22
gap and as you reduce the gap over more and more experience your
13:25
model gets better and better so you're sort
13:28
of using your sensory information and
13:31
what's coming in as a way of making the model better and better and better and
13:37
so eventually you end up with relatively good models and they're different models
13:41
visual models auditory models higher cognitive models of who i am what's like
13:46
to happen in this circumstance and so on and they get really you know better and better.
13:50
And so predictor coding is that idea, and that you can, if you get a gap,
13:57
you can close the gap between the sensory information and the guess that you think it is.
14:03
By either by changing the model so you
14:06
you you you make the model better you you update
14:09
the model in some sort of way so that's where predictive processing
14:12
sort of gets to but active inference takes another step and did
14:16
you want to go to i mean maybe anna marie can take it from that point
14:18
what's the difference between active inference and predictive coding i
14:22
i thought you i thought it was you going
14:25
for the active inference is tricky dave i'll never forgive you for that so well
14:33
i'll help you out what do you think have a go yeah so and this is where it gets
14:38
very interesting because predictive processing brought us something which is.
14:44
Was meaningful that gave us a
14:46
vocabulary and the way to understand the
14:50
behavior because here we're talking to clinicians so we
14:53
need you know clinician but with active
14:56
inference we are also adding the extra element
15:00
of changing the
15:03
action of changing that prediction
15:07
so what we will be doing what active inference
15:11
tells us is that in order
15:14
to match that prediction because active
15:18
inference comes from the underpinning thought that the brain or the system will
15:25
always want to minimize prediction error because when there is too much error,
15:35
the system has to work hard to remove that mismatch.
15:41
And therefore, there is a loss of free energy, which the system does not want.
15:47
So we will always try to minimize that error in order to minimize the error
15:54
we act and this is the beauty of active inference it is a,
15:58
It's producing, behaving.
16:03
Changing our behavior, our action in order to minimize that prediction error
16:11
That is in health and if we relate it to pain, that is in pain So actually.
16:19
We minimize in a prediction error of pain by predicting pain and actually stopping
16:26
moving in order to prevent that that prediction to be mismatched,
16:31
that is a way to minimize free energy.
16:34
It's very unhelpful. It's an unhelpful belief.
16:36
From the system perspective, it is helpful because you are matching the original prediction.
16:44
So active inference and the work of Carl Fritz at the beginning of the 2000s is computational.
16:54
David, can we call it computational neurosciences Yeah,
16:59
it's computational It was actually, I think he started computational psychiatry
17:03
which is where we talk about that at some point around where it explains that sort of stuff too Yes,
17:09
so it's very mathematical for people that look into active inferences the base
17:15
is very mathematical based so it's but the theory.
17:22
It's a theory where then you can build up the framework but again,
17:27
that is not the beginning in and, you know, come in David,
17:30
but it's not the beginning and the end of all, because what we are trying to,
17:35
what we are personally, especially doing,
17:39
I am a little bit afraid that we are swapping a Cartesian dualistic reductionist
17:45
model for another reductionist model. We have to be very, very careful.
17:49
So some predictive processing, active inference, often from the theorists themselves
17:56
and those who operationalize those theories,
18:00
they're actually operationalized often as very brain-centric, i.e.
18:06
We are, like Dave said in beautiful, We are our brain. Are we?
18:11
We are not, you see. And this is, and I resonate very much with one of,
18:17
a paper from Mick Thacker, who was one of the strongest proponents of predictive processing,
18:26
and Jeanne Hartis, Jeanne, I can never pronounce his surname.
18:31
The paper on the predictive processing theory of pain, where it looks at predictive
18:38
processing not only as a brain computation,
18:41
but as a whole body perception,
18:44
which really suits and fits the embodied cognition that we have been seeing in work like lovely,
18:57
lovely, I mean, very interesting, Bronnie was talking about,
19:01
actually that comes, the sense-making that comes from embodied,
19:06
the sense-making of our understanding of the world comes also from our somatosensory,
19:14
sensory motor system, as well as our experiences.
19:18
So it's beyond just the brain or just the body. Yeah, yeah.
19:23
Okay let me just do a quick recap for our listeners then because i think oh
19:27
basically i'm going to tell you how i have what conclusions i've got from that
19:32
and then and then listeners can,
19:34
can ask questions or say but so essentially what we're talking about because i think.
19:38
I think the first surprise for listeners might be how old predictive
19:41
processing is i mean when we were talking off air i think as soon
19:44
as you say the brain is predicting things i think a lot of listeners probably think
19:47
that mosley and butler came up with that and that was the whole idea came from that
19:49
explaining pain that your brain is looking out for danger and um and
19:53
it's actually not it goes back to like you mentioned it basically goes back to
19:56
philosophy with Kant and then Wilhelm Wundt and and
19:59
you know when psychology started so it
20:01
goes back a long long time which is interesting because it probably adds more
20:06
credibility to it in the sense that it's not we mentioned it last week there
20:11
and you mentioned as well Maria there's this problem of of brain centric and
20:15
it's kind of mere logical fallacy as Mick Thacker said to me on Twitter I I
20:18
think in 2013 or something.
20:20
And I was like, oh my God, that blows my mind and made me think a lot about this idea.
20:24
It's all from the brain. The brain's looking after the body and we are the brain.
20:27
But so predictive processing, that's really cool that we've cleared it up.
20:31
It's actually very computational.
20:33
It's based on a lot of mathematics and stuff. So that's great. It's the way we operate.
20:36
And then also, I think what I'm taking from this is like you said,
20:39
it links in with what Ronnie Thompson and Ben Weibra were saying last week in
20:43
that with the active inference.
20:44
Suddenly we do have to understand that although it's the brain that's processing this,
20:48
we have to realize that there's all of this external stimuli
20:51
going on there this environment this context that we're living
20:54
in so although the brain's kind of like processing stuff we have to be very
20:57
aware when we're with our patients and clients that of all the whole sensory
21:01
input that's all around them that could be adding to what decisions if you like
21:06
this kind of brain is making is that where we are if i kind of active inference
21:10
is a little bit different than predictive processes is part of it.
21:14
Active inference is actually us actively.
21:18
Trying to minimize that prediction error.
21:22
Predictive processing tells us that we predict active inference.
21:27
We actively try to minimize the prediction error because we do not want a big prediction error.
21:35
And what does that mean in clinical practice or in pain?
21:40
I predict that going out for a walk will give me pain. I will feel pain after I walk for 10 minutes.
21:49
So active inference could be for better or for worse. It could be used positively,
21:52
depending on what the framework is, and that's where the clinician comes into it, hopefully.
21:55
Or it could be leading to kinestphobia and all that sort of stuff. Correct.
22:00
There are some other things I think that we need to add here because,
22:04
you know, that was great, Anna, for putting you on the spot.
22:08
I do apologize for you, but you did a great job.
22:11
And so I think you also have to think about there's another element. So we've got models.
22:18
Without models, you can't guess anything. You're just random guessing, all right?
22:22
Without models, in fact, no living things exist, but that's a sort of deeper story.
22:27
You have a sense of information coming in. You try to close the error gap between the two.
22:31
As Anna-Marie eloquently said, you know, you can either change the model or
22:36
if you act, if you act on either your body, which can turn body signals up and down into reception,
22:44
or you act on the outside world you can also close the the error so you can
22:50
you can either change the signal coming in to fit what you're guessing or you
22:54
can change your guess to fit what's coming in yeah and so that bit about acting
23:00
then becomes a very interesting thing because that means then that.
23:05
What the system is doing. And I think it's also useful to think about the brain
23:11
and cognitive bits of the system as being just particular examples of active inference.
23:18
So they are cognitive examples of active inference.
23:21
They are the things that allow us
23:25
to be outside of our bodies because they generate language
23:28
and we can commune with each other and we can watch body
23:31
language and so on and so forth and so that
23:34
that is this inactivism where where you know you've got
23:37
a system that's already embedded in the larger environment as
23:40
well but but but it's but it's important to understand that there are not just
23:47
one model you can actually see the whole entity the whole agent as a model of
23:52
its environment so so the idea is that the the physical things.
23:58
Things that are living they they build
24:01
a model of their niche they build a
24:04
model of their world the thing so you've got a spider tiny
24:08
spider is at the bottom of some vast tree in the amazon rainforest it doesn't
24:12
build a model of the entire rainforest because it doesn't really experience
24:18
very distant types of signals if you like that may affect but they tend to build a local sort of model,
24:25
The structure it is, how long its legs are, what it smells, the hairs on the
24:31
top of its head, you know, how big its eyes are, are all parts of the system
24:36
that are modeling the environment or the niche that it's in.
24:40
And so it is sort of reflecting in some sort of way the dynamics of its local environment.
24:49
So it's, and it stays in that environment to keep minimizing errors.
24:54
So as long as it doesn't move outside you
24:58
know something and the the classic one is
25:01
like a fish will find itself out
25:04
of the water as being very surprising so that's a very large
25:07
error gap and so fish tend to stay in the water to minimize the error gap all
25:12
right so so you can act in order to do that and the idea comes this this idea
25:17
this amazing idea called self-evidencing which is that the model that you build,
25:24
and these are multiple models of different hierarchies,
25:28
generally just reflect.
25:32
The sort of things you expect if you're in a particular part of your environment.
25:38
So what happens is it means that your action constantly makes you act into stained
25:45
bits of your environment that produce the least surprise.
25:49
So you can understand how then chronic pain is such a, you know.
25:56
How can we see people suffering with chronic pain in the clinic,
26:03
how they literally manifest their self-referencing, self-evidence because they
26:10
are trying to minimize the error so they continue doing those things because
26:14
that gives them the sense of self.
26:16
The sense of self has to be reinforced because if you're actually mismatching
26:22
what the predictions are to the reality of their environment and then they cannot cope with that.
26:30
So they have to always maintain the coherence in self-evidencing themselves.
26:36
So this is where, to me, in activism, it's really important,
26:42
speaks, I think, a better language than the biopsychosocial model,
26:48
because the biopsychosocial model is maybe just as it's operationalized,
26:53
as Engel meant it to, but it can be quite a tripod, kind of.
27:00
It's still divided. And also, those are factors and quite separate factors,
27:06
you know, the lovely Venn diagram.
27:08
What with an activism instead is not only you have those factors,
27:14
but you as a system, as an agent, you're actually changed.
27:20
Your influence are influenced by the environment you are in.
27:26
So you see things because those things afford you to do things.
27:30
So, you know, and that's where active influence and activism,
27:34
actually, they work very in active inference because you act to minimize the
27:41
prediction error and you act into your environment.
27:45
Your environment shapes you, but you shape that environment.
27:49
Yeah, so, and this is, sorry, this is where the cognition, the thinking,
27:56
call it, is not only in the brain. So the way you feel.
28:01
Your world, your proprioception, the way you feel your internal world,
28:07
your interoception, your exteroception, your somatosensory and sensory motor, they're all part.
28:13
They're not just an expression of the brain.
28:17
They are integral part of how you perceive and process the world.
28:27
Because then through that you change your hormones
28:30
change and so it does not have to
28:33
cognition does not only happen into the brain yeah i mean in some ways you can
28:38
so you can see that there are there's cognition all the way down in many ways
28:43
because if you think about if you you can imagine if you've got an an entity
28:49
that's that there's got some networking and it can.
28:53
Instantiate, store a modeling side of it, you can imagine that there might be
28:57
a model of some neurons in the dorsal horn that model their local stretch receptors
29:03
in the muscles that they're connected to.
29:05
So they're trying to minimize the error between their guesses of where those
29:10
muscles should be at any particular time, informed by multiple layers coming
29:15
down, which are all these predictions coming down, and all of these errors that...
29:19
So most of the time... So you've got active inference units,
29:23
if you like, at all scales, all the way from cells, all the way up through to societies.
29:29
And the beauty of the active inference framework is that it sort of explains
29:32
all of them by the same mechanism, that you're modeling those sorts of things.
29:37
So that's important to remember that. And cognition is just,
29:41
a particular example of that sort of thing. We mustn't forget precision.
29:45
We'll come back to them in a moment.
29:46
But one of the ways of thinking about how organisms stay in a set of states
29:53
in their environment that match their model, all right?
29:56
So fish don't explore what's it like to sit on a rock looking in the pond,
30:02
all right? They don't go that part.
30:04
I mean, it's a possible thing they could do, but they don't go into that
30:07
particular dynamic part of the landscape they could occupy
30:10
so they occupy these bits of the
30:13
niche the their model represents because
30:16
if they stay in that part of the niche then their model is minimizing prediction
30:21
error and maximizing its own evidence all right so that's where self-evidencing
30:25
comes from and one of the ways of thinking about that is imagine this imagine
30:31
this big house right there's this huge party going on in LA,
30:35
not that I've ever been to LA, but I can imagine parties in big houses. The big X-Math?
30:40
Yeah, X-Math. All with LA, yeah. Big, big parties.
30:44
So, you know, and you've been to those sorts of things where,
30:47
you know, it's packed in, it's upstairs, it's downstairs, it's really difficult
30:50
to get to the drinks. You have to push past people and whatever.
30:53
So imagine that, you know, there is an agent,
30:58
an active agent like us, or it could be a
31:02
bit of the dorsal horn or it could be the visual cortex
31:05
which makes a visual model or it could be an auditory model all
31:08
right so imagine that as a little entity a person that can move about basically
31:12
that entity stays in the space in the room amongst a small number of people
31:19
that are all yes men so they're all telling it is right all the time okay and
31:24
and they'll stay in that space.
31:26
So they act to stay in that space. And there's a lot of jostling going on so
31:30
they can be pushed away from that space.
31:32
And also the number of people joining that particular group of yes men,
31:36
you know, sycophants around this particular entity.
31:39
Are are changing so that that that external bit
31:42
is and and sometimes somebody really weird will come in there will be a
31:45
big error gap and whatever but but generally it sort
31:48
of stays the same so so that person is moving around
31:51
with that group trying to stay with that group so
31:54
that's this idea that you're self-evidencing yourself and you're
31:57
acting in order to stay within a part
32:00
of your niche that represents your model now what
32:04
you could do is you could you know and actually moving out
32:07
of that across the room to some other strangers over there you've never
32:10
really talked to is really uncomfortable so that's a lot of
32:13
uncertainty and and the systems don't
32:16
like the uncertainty they're always the desire to minimize error all
32:19
the time but it could be that for example you end
32:22
up with a bunch of people over here and you get
32:25
stuck in a group of people that are telling you really bad things about yourself
32:28
all right that and you get used to that as
32:32
as the sort of signals that you should
32:35
match so you start to match your model to that
32:39
information going on and you get stuck over there and so
32:42
and that could be a chronic pain belief or
32:45
a prior that's a threat prior or whatever you just expect pain or whatever because
32:49
you're in this group and so in a way you know you can almost see the therapeutic
32:54
encountering clinicians role as finding a way of persuading them to go through
33:00
across the room through this really uncomfortable bit where they're neither
33:04
in one group or the other.
33:05
Back to somewhere where the model that they are generating.
33:11
Is fitting with a much more benign set of circumstances that are minimizing the error.
33:16
Now, so just good with precision. So outside of the sensory information and
33:19
the models, and we talked about models all the way down and all the way up,
33:23
cognition being just a particular example, you also have this contextual approach.
33:29
Ability, and we'll talk about it cognitively because it's a bit difficult to
33:33
talk about it in terms of a dorsal horn, all right?
33:36
But you've got this very big model of the world, if you like,
33:42
outside your sensory models, that is about who you are, what you expect,
33:47
what you normally do, what this situation normally means.
33:50
If I'm experiencing this, this is going to happen next, that sort of stuff.
33:54
And that gives you a huge context. So for example,
33:57
if you're walking through a park in london and you and it's in the fog and you
34:02
get some sensory visual signals that are coming in it looks like it could look
34:07
like either a a dog or a lion all right in other words,
34:11
it's so noisy the data it could fit either a dog part of your model or a lion part of your model,
34:18
you don't see a lion because you know you're in london and it's really unlikely
34:22
that you're going to see a lion.
34:24
So what happens is you distrust the sensory information and you put much more
34:29
precision on your models because your models are informed by this wider context.
34:34
Now, if you're on an African savannah, then you're going to probably see a lion
34:39
coming through because it's more like you're a lion there. And also you've got this added danger.
34:43
So even if it is a dog, it's better to think it's a lion because you're not going to get eaten.
34:46
So there's that threat bit to it as world.
34:49
So you can then think of conditions that put people in ill health outside of the ones that just.
34:59
Mess the whole system up so you lose active inference you know like you know
35:03
you can't active inference your way out of jumping off a cliff for example so
35:07
you know those are going to be stuff that just breaks the system up but lots
35:11
of conditions you could then think of as just.
35:16
Aberrations in in precision you put precision too much on a model when you should
35:22
be listening to the signals or you put precision on the signals when you should
35:26
be modifying the signals with your models about what is most likely to be going
35:30
on in the world based on your experience.
35:32
And actually, Friston has talked about this, and there's been a few papers around this.
35:37
So you can look at, for example, schizophrenia as being something where you've
35:42
just got way too much precision on your internal models.
35:45
So you've got no sensory information correcting that. So you've got delusions
35:50
and hallucinations and so on and so forth.
35:52
And autism could be the opposite.
35:54
So there's too much precision on the sensory information. You've got no central
35:59
model that's dampening that noise down.
36:01
So you're experiencing all the noise and all the sensory stuff,
36:04
that becomes overwhelming.
36:05
Now, in a way, you could see chronic pain as being on that precision,
36:10
spectrum okay so you've got
36:13
precision around a threat prior
36:16
so you can start with acute pain you've got a good balance between between
36:19
the pain and your models and the central information coming in and as that pain
36:25
dissipates you know your model goes back to you know the guy where the yes men
36:29
where there's lots of little you know lack of errors and so on and you're back
36:32
normal but if the pain persists one of the and as Anna Marie was saying one of the ways of.
36:39
Reducing uncertainty around this pain persisting it's like why is this pain
36:43
here it's still here one of the ways of reducing that is because you've got
36:46
an error gap all the time i've got a pain,
36:48
and and i know that my model should be no pain but i've got this gap who won't
36:53
go away so one of the ways of closing the error gap and anna-marie said this is is to just create a new,
37:01
model a new prior that says i'm going to be in pain all the time.
37:07
The brain needs to know it doesn't like uncertainty we're going to fill it in
37:13
and we fill it in with what we got,
37:15
this is why those you know different factors
37:19
you know mental health anxiety traits you know lifestyle factors they all contribute
37:25
to create or they are priors or to create beliefs to fill to minimize the gap
37:33
but as you say And the inference is so beautiful because it brings everything together. Yeah.
37:39
But as you say, the problem then is that you generate pain to fit the model.
37:48
As the nociception dies away, you're now in a situation where you've got a threat
37:52
model that says, well, this is going to be generating pain.
37:55
And so you generate pain where there's no damage.
38:00
And that confirms the model. So the model digs deeper and deeper into this flat
38:05
shape, and you get stuck in this new hole.
38:08
And the problem with getting out of the hole is that as you climb up the side
38:15
of the hole, because these are – you have to experience more uncertainty.
38:20
So to get from this aberrant model, yeah, you've got to – We've got some questions.
38:26
We've got some questions. And what I think, what I'd love is we've spent 40
38:31
minutes on the theory behind it.
38:32
I mentioned a little bit of the practical, but what I'd like is I'm going to take some questions.
38:35
We've got some amazing questions in here. We've also got the great David Poulter
38:38
in here as well. So I'll get rid of some questions from David.
38:41
Poulter. Yeah, he's in here as well, Mr. Poulter. Mr. Poulter, good evening.
38:45
So we're going to look at the questions. But afterwards, I'd like,
38:48
if possible, for the last 15 minutes to look at what the listener,
38:52
the healthcare MSK clinician should be doing then in order to affect this active
38:58
inference and the errors and the priors and all the things we've kind of talked
39:02
about. Before we do that, we've got some amazing questions in here.
39:05
Thank you again, people who've jumped from YouTube to here. Apologies that YouTube didn't work tonight.
39:09
It's not a problem. It's just technology is getting worse and worse.
39:13
But anyway, David Poulter, thank you so much for joining us.
39:16
I'm going to bring these up in order, people, so we get them through.
39:18
I don't think they've necessarily been answered yet but david poulter says we
39:23
know that sound and vision can be quickly affected by predictive processing
39:26
i.e peter k misheard lyrics the chessboard shadow illusion but why can't pain
39:31
why can't pain be influenced quickly by predictive processing in the same fashion,
39:37
So do you want to go for this one first? So thank you for that. It's a great question.
39:43
So there are, Friston has talked about slow and fast models.
39:50
So some of the higher cognitive models operate at different temporal,
39:55
on a different temporal basis.
39:57
Things like vision, for example, they're very precise models.
40:01
So they don't have a lot of wiggle room. I mean, in our analogy of the party,
40:06
visual models are operating in a very small room with about six people.
40:10
You know, there's not a lot of wiggle room around there, and it works really fast.
40:15
But things like threat prize are very high-level cognitive models that include,
40:21
you know, my history. They include what I'm expecting. They include all this.
40:26
And those models tend to both form and.
40:32
Extinguish much more slowly and so
40:35
i think that that may be a reason for that and also
40:38
because pain is an emergent phenomenon so
40:42
vision and sound there
40:45
are some sensations pain is
40:49
a phenomenon which is emergent and coming
40:52
from multi-factorial system
40:58
wise and that's you know there is this hierarchy
41:01
that's really good that's really good question as well david because
41:04
i think the the everyone's kind of probably familiar with peter k
41:07
back in the day and now it's all over tiktok and that with misheard lyrics now
41:10
you're going to hear this and you do hear them and it's incredible and then
41:13
when they put the new words up you hear the new words coming up and it's that
41:16
is explained nicely by predictive processing yeah great question i'm sure you'll
41:20
be back david with a follow-up to that in fact i'll just put here some comments
41:25
brian huxley thanks for joining as Brian has said,
41:27
he said, yes, it can as it predicts the perception in the context of the original pain.
41:33
David is back with, I think it's a question here, if we provide positive environmental
41:37
input, safety and positive feedback, why does this not positively affect chronic
41:42
pain via predictive processing and or inference?
41:46
Is that a simple answer? Is it to do with that? Yeah, well, no, but I think it does.
41:50
So remember we talked about precision and precision is dependent on this broader,
41:56
wider, contextual sort of place that the cognitive bit of the system is in.
42:03
So positive environmental input, so that's why therapeutic alliance is one of
42:08
the few predictors of outcome in musculoskeletal pain, because it provides a safe environment.
42:15
In that safe environment, if you imagine the pit that we were talking about,
42:21
that the chronic pain model is in and.
42:24
To experience uncertainty, you're more likely to have a go at doing that than
42:28
if you're in a really safe environment and you trust people.
42:31
So therapy of the alliance is really important in that. So I think it probably
42:34
does do it, but again, I think,
42:39
as Anna-Marie said, the multivariable aspect
42:42
and the N equals 1 effect of patients'
42:46
idiosyncratic experiences and
42:50
models means that it's it's it's
42:53
quite difficult to do so and i think that's probably why there's
42:56
been there's been no success looking at you
42:59
know generic inputs to say help low
43:02
back pain or musculoskeletal pain it's been very difficult
43:05
to find any clear predictors you know that are transferable across
43:10
different settings and so on and the reason for that i
43:12
think is because in the end this is a very
43:15
n equals one type of thing so actually despite all
43:18
the research it still really comes down to the clinician in the spot in the
43:23
place where the patient is and um yeah and they're gonna use the word the placebo
43:29
in a very positive way in terms of well david you wrote actually a beautiful
43:35
paper fairly recent i think it was last year.
43:39
Contextual factors, so we know the intervention, no interventions is superior.
43:44
To another for chronic pain. Contextual factors, they affect positively outcomes.
43:51
So I would say where possibly changing the context that the positive environmental
43:58
input might have an effect.
44:01
The problem in science that we are not phenotyping and we're not yet as good as not testing,
44:08
studying contextual effect individually
44:12
or you know against and so i
44:15
think that's where you know who does respond better
44:18
to certain things so possibly that very good
44:21
i like it emmy tester has got a comment here um emmy
44:24
tester says hi to me how you doing emmy from the sd school as well
44:27
love this it's connecting different things i've been reading about like identifying
44:30
with different groups changing likelihood of behaviors and of course the
44:32
whole field and awareness of uh field and grace
44:35
and awareness of challenging habitual patterns it costs energy to change the
44:38
pattern what's the payoff very good nice comment there well i think the payoff
44:45
is that you if you're lucky as a clinician you can i mean i i think for musculoskeletal
44:51
pain in particular but i think you know.
44:54
The therapeutic encounter is a very rich, richly full of affordances and contextual stuff.
45:00
And I think, you know, in many ways, musculoskeletal professions have seen themselves
45:07
too much as mechanics in a way, and should actually see themselves as gardeners.
45:15
In the sense that what you're doing is you're creating a landscape across which
45:21
you, with trust and safety and empathy,
45:25
persuade the patient system to move back to where it really should be.
45:31
So the payoff is that you can arrive them back to a place where their system
45:38
is in a place where there's minimization of error without having to stick with
45:44
a very aberrant type of prior.
45:46
Because it's all about self-regulation. That links a little bit.
45:51
And that is what possibly differs from you, David.
45:55
That I would say I don't think that we actually create the landscape I like
46:01
here to always refer to my.
46:05
Son's talk on the liminal space in health he's an anthropologist and he's
46:13
You know, in plant science, they talk about this liminal space where there is uncertainty.
46:22
And one day the healer and the person, what they do, they co-create.
46:28
And I love the word. It's about making sense.
46:30
It's the co-creation of the new transformative, not going back to pre-diseased
46:38
state, but they co-create together equally this new state.
46:43
And also the liminal space. And this is what I believe that me as a manual therapist,
46:48
so, you know, that's where touch and the encounter,
46:52
the therapeutic alliance, we create the liminal space, which the space of in
46:58
between, of a little bit of uncertainty,
47:01
of where the prediction error can be safely mismatched.
47:08
Because you are safe. We create the safety. And then for them,
47:12
and this is what for me and activism is, we help them sense making.
47:20
And this is for me touch by creating a little bit more of a precision within the touch.
47:28
And for them to feel this is an area of pain, but I'm still intact together
47:36
with everything else a touch brings.
47:38
But in general, the therapeutic encounter is where the liminal space takes place, but safely.
47:48
So that when they go out of it, they both co-created, they shared decision-making,
47:53
a new transformative space of the person is new.
47:58
They are changing their environment in a new way. They're going to change their
48:01
habits, behavior change.
48:04
Behavior change because they update them eventually, their model.
48:08
That's very good. Scoob, we've got the answers are creating more questions,
48:12
so I'm going to take a couple more questions.
48:13
Apologies if I don't get through all of them, but we will get replies to you.
48:17
So I've got one here from David, who's come back. David Poulter says,
48:20
does predictive processing preclude the necessity for nociceptive input for pain?
48:27
Preclude. Preclude means includes it?
48:31
It's not needed. Not needed. Not needed. I think you touched on that earlier
48:34
on, didn't you, David? Yes, it is needed.
48:35
Sorry, Dave, when you said that even if there's not a nociceptive we
48:38
are a system oh in chronic pain i mean we are a system the system includes includes
48:45
pain if you look for example i don't know if you read the paper that mick tucker
48:49
and jean paper on predictive processing for pain where they they talk about embodied selfhood and.
48:59
Cognition or coming the predictions and coming from also from the stimuli somatosensory
49:06
but actually then affects the neurological,
49:10
the endocrine system.
49:12
So you cannot really separate the two. Yes.
49:18
And I've done it. Yeah, I think that's right.
49:22
The other thing to think about is that I don't know that anybody has a really
49:29
convincing idea about what pain is yet.
49:32
Still, some people call it emotion. Some people call it a stack of errors.
49:37
Know, some, you know, is it, can you have pain without nociception?
49:43
The other, I don't know. I think you can probably experience much more pain
49:48
with little nociception and you can experience lots of nociception and very little pain.
49:56
So, so I think the thing to remember is that those two things are not necessarily correlated.
50:02
But the other thing is that if you've got higher threat prize,
50:06
you know, these ideas that, that bending over is dangerous, that,
50:10
that, you know, I'm always going to have this pain, you know,
50:13
those higher level cognitive stuff,
50:15
then because of this hierarchical effect of predictions filtering down and errors
50:20
coming up and somewhere around there,
50:22
the error should get extinguished those predictions can
50:25
start to change the number of receptors for example in the in the the the nerve
50:33
endings that generate nociceptive pain so you could just turn the gain up by
50:39
things that you're thinking or where you are so in a way you can.
50:46
Where do you turn the gain up? Are you getting more nociceptive pain?
50:50
In other words, pain is not just a bunch of signals coming from some nerves in your body.
50:55
The pain is an experience generated, created, constructed by a whole bunch of
51:01
different things, including all your cognitive ideas about whether you should
51:04
be in pain, whether it's really dangerous or whether it's not.
51:07
And so I think nociceptive signals are in many ways, particularly as you become
51:12
more nociplastic, which I know some people don't like that term,
51:15
and I agree to some extent,
51:18
is that the nociception is just one of the signals that are coming in,
51:26
amongst many others, that then construct the experience of pain.
51:30
And I also think that maybe that if we're referring a little bit to the recent tension between pain,
51:41
Should the question be, can nociception be driven from a top-down mechanism?
51:49
Not can pain just be driven from a top-down mechanism? Because actually.
51:55
You know, in the moment in which we perceive, you know, if we have a threat
52:03
attention bias and we're always into this sense of lack of safety,
52:10
and our allostatic system starts working, eventually the allostatic load becomes
52:15
inflammatory, low-level chronic inflammation.
52:18
Would that activate nociception? Yes.
52:23
Is that nociception because of tissue damage? No.
52:27
So, you know, I think we need also to expand.
52:32
And actually, I'm okay with the fact to know that chronic pain can be driven by nociception.
52:39
Exception doesn't change my approach, doesn't change the fact that.
52:45
Psychosocial factors that manifest into pain.
52:50
But hopefully, yeah, hopefully if there is an inflammatory driver,
52:55
then maybe some pharmacological sometimes marker might help.
52:59
I mean, one of the ways of looking at this is if you've got multiple signals,
53:03
multiple systems from cognitive systems to immunosystems to whatever, In a way,
53:09
they're all individuals that have hands that are able to turn the volume knobs
53:17
up or down on the other systems.
53:20
And so, you know, they're all fiddling with each other's, you know.
53:25
Volume stuff, whether it's down or up.
53:28
And then out of that comes a sort of overall average that then generates,
53:34
you know, if it's above a certain amount, then it generates a pain experience if it's below.
53:39
You know, so everything's fiddling with everything else, you know.
53:42
And that's why phenotyping.
53:44
To me, this is why phenotyping. We're finding who, why, let's go back a moment
53:51
to manual therapy, a touch.
53:53
Why people choose to come to manual therapy, places where they also receive manual therapy.
54:02
They're already self-selecting. Again, McParley and George Estevez,
54:06
they did a beautiful paper on touch, synchronicity, active inference,
54:11
where touch is the active inference because we already predict,
54:15
we already recognize as something very positive.
54:19
That he has some physiological and emotional, and therefore we kind of helps
54:26
us to minimize the prediction error.
54:28
And that people actually choose to come to us because they already know that
54:34
they respond better so that maybe they are more embodied,
54:40
they understand the world more through the embodiment.
54:44
That should not be the beginning and the end of all, but maybe those people
54:48
that they understand better.
54:49
It's like when we teach, we need to understand how somebody responds better.
54:54
Why don't we do that with our client?
54:56
Not everybody will respond to the same approach.
55:00
And that's why for me, an activism and active inference really speaks to me. It's a model that helps.
55:09
Okay i wish you had another hour to do this but what i'd like to do now
55:12
as we've only got about five minutes left are you both okay it's 9 0 8
55:15
are you okay for another kind of five minutes no we're good just checking yeah
55:18
i wish we had another hour to do this but i'd like to spend the
55:20
next five minutes a little bit longer maybe on
55:24
on the take-homes for i mean first of all i think the take-home is listen to
55:28
this episode again the dissectics i think it's quite heavy the information we're
55:31
given it's going to take some listening and that's the joy of this podcast is
55:34
you can listen to it over and over again make some notes and test yourself a
55:37
little bit but once people understand what's going on here and and the mechanisms
55:42
of action that we're talking about,
55:44
How will this affect clinicians in how they treat patients who are in pain?
55:50
What are some examples maybe of what you think, how they should be tweaking
55:54
maybe what they're doing already, connecting it to what we'll be talking about
55:58
with predictive processing and active inference?
56:02
I'm going to leave this to Anna-Marie, but I'm going to make a very broad statement first to book.
56:05
I'm not a clinician, so Anna-Marie is the person to do this.
56:09
For me this new framework
56:13
should allow people to re-see what
56:16
they're doing through a different lens all right and so
56:19
so you know it's a little bit like a paradigm shift you know we saw
56:23
because of einstein's space time
56:25
we saw you know building rockets and flying
56:29
watches around at high speed through a different lens we
56:32
didn't change newtonian physics but but we
56:35
saw physics through a different lens in terms of movement and
56:37
so on and and that you know people say well that's really you know what's the
56:41
point of doing that so because it's like well now I'm doing the same stuff but
56:45
I'm just explaining it in a different way the thing that I think is most important
56:49
is when you start to get your head around this very profound framework it starts to,
56:55
I think should allow you to start to realize that some of the things that you
56:59
thought you were doing that were peripheral in your care suddenly become really important.
57:05
And some of the things that you thought were really important maybe are not
57:08
as important for everybody in that particular care.
57:11
So for me, a very broad statement would be the take-home message is thinking
57:16
about this differently. It really makes you look at what, you know.
57:21
It should make you look at what system you think you're working with.
57:26
If you think you're working with a patient in a traditional sense,
57:29
then this makes you think, well, you're working with something quite different
57:34
here to what you thought you were working with.
57:36
So I would say that would be my take-home message. Anna? Oh, absolutely.
57:41
Oh, so much. So in clinics, someone's thinking, what am I going to do differently?
57:46
Yeah. Yes. So no, no, no. I know exactly. First of all, it's just having a different
57:51
narrative is just not enough because it's.
57:57
You're just, you still could be doing the same things, but having the different,
58:01
it's having the mindset shift.
58:03
Because when you have the mindset shift, your clinical reasoning changes,
58:08
we, like David we said we are not technician anymore we
58:12
are not operationalists we are there personally whether
58:15
my advice or how to introduce an active difference
58:18
is that the the most important outcome
58:21
is to help our clients in
58:25
sense making for me the sense of self
58:28
is really important so that making sense of
58:31
that pain okay and and and once
58:35
they feel safe because once they make sense of the
58:37
pain they can feel safer even into the
58:40
moment of and the other thing is so
58:43
your treatment call it an zone or your encounter everything is part of creating
58:51
a space where the challenge is that the prediction error needs to be whether
58:59
it's the mismatch of the prediction error and you can be there.
59:04
Providing the safe environment.
59:06
And what you do is actually changing, helping them change action.
59:13
Without action, there is no active inference. So your touch has always to be
59:19
as a communication tool in this case, or as a sensory input for them to,
59:27
For the client to then act themselves in order to change their space, their environment,
59:37
their behavior, because they will feel safe to actually go into the space. So what does that mean?
59:45
Just put them, you know, you do your therapeutic alliance is the framework.
59:52
Use manual therapy whichever way you use it, but not to change anything.
59:56
Is to provide maybe more precision onto a pain-free input or as a symptom modification
1:00:04
in order, again, to then expose them to that error and feeling safe.
1:00:11
So gradually expose them to something, but make sure that they first got the
1:00:17
sense-making sorted out.
1:00:20
Yeah, because if, and we know that from literature, that especially in frozen
1:00:25
shoulder, that education, once they understand about the natural history, they feel safer.
1:00:32
So if there is that mismatch or prediction error, they still know that they're safe.
1:00:36
So they can breathe safety into that potential, you know, previous that it could have been a fear.
1:00:42
So exposed to them, but within a safe, but it has to be coming from them because you are not doing it.
1:00:49
It's active inferences, them acting within that environment.
1:00:55
That self-efficacy, it has to be at the basis of everything.
1:01:00
Because in my case, as one of my PhDs, the sense of self is sense of body ownership.
1:01:07
And sense of agency. We don't have those ones. We don't have a coherent sense of self.
1:01:14
And without that, we lose our coherence within the world.
1:01:18
So you just step back, co-create with them, expose them to a different movement,
1:01:24
a movement that before might have been a little bit more.
1:01:29
But they have to be ready for it.
1:01:32
Okay. This is great. This is going to take so much. I think, yeah.
1:01:36
That's great. people agree i'm hoping that in
1:01:39
the links we can put in some reading material maybe
1:01:42
for people to go over this and we'll definitely need to continue the conversation
1:01:46
once i think people have had a little read and taken on board what the mechanisms
1:01:50
of action that are going on here to to like you say just look through different
1:01:53
lens i think that's really important messages is with this evolution of thought
1:01:57
and understanding we're not very rarely we saying totally change everything.
1:02:01
With all the different jumps in history and the way we've kind of found out
1:02:05
that doesn't quite work. We're never really saying, stop doing that.
1:02:08
Maybe apart from sink plunger on the ITB to try and lift up,
1:02:11
that maybe we did stop doing.
1:02:13
But apart from that, I can think of offhand. Yeah, we're just,
1:02:16
but even then, if that's what the patient wants in their world and they're thinking
1:02:20
that's helping them, then you would still get a sink plunger and put them on
1:02:22
their thigh and just give them that new kind of exposure and that sensation.
1:02:26
That's fine. So there's nothing we don't do but yeah i think
1:02:28
we are one matt that we are
1:02:31
an integrated system we are not the brain we
1:02:34
are not the body we are not just our thoughts there is
1:02:38
the bbc actually did a fabulous in in
1:02:41
our time melvin bragg in august with a fabulous podcast on merle pointy on phenomenology
1:02:50
and talks about embodiment oh honestly that link in there you sent it to me
1:02:55
yeah yeah it's really Really, really interesting,
1:02:58
again, talking about phenomenology and embodiment, that actually we are more
1:03:02
than our body and we are more than our brain.
1:03:05
That's really cool. And that will link into our subjective and the way we're
1:03:08
talking and communication, which obviously we haven't talked about at all,
1:03:10
but the questions we're using are open and questions. This is what I want people to understand.
1:03:16
For some people, they feel better when they talk once they are touched.
1:03:23
Because Zoe Parlin and George Estevez said in the beautiful,
1:03:28
and Ceritelli said, touch tells you you are like me. Yeah, we are co-specific.
1:03:35
So people are ready for those people that come through you. Don't force touch
1:03:38
on people that don't want touch, obviously.
1:03:41
So can I just do this? So there's a great book called The Self-Evidencing Agents.
1:03:46
It's just come out by a philosopher called Jacob Howey.
1:03:48
And if you want to dig into, like me, because I'm a complete geek and there's
1:03:54
not too many of us around.
1:03:55
Otherwise, it would be an even worse place than it is. But the...
1:03:58
Part of the three on the screen at the moment. If you're really...
1:04:01
Yes of course the if you're into
1:04:04
the profundity of how this might affect things this
1:04:07
is a tour of force actually that it's that so
1:04:09
you don't even see that can you yeah no
1:04:12
you can't i'll put it in the notes but it's a great book if
1:04:15
you get out of it if you really want to dig into the profundity of
1:04:18
this get that and i think i'll
1:04:21
also put a link in there this isn't just a self thing but a lot of what
1:04:23
we talked about particularly the links between the philosophy and the psychology and
1:04:27
the different models and behaviors and thoughts are actually
1:04:30
in my communication course i've enjoyed so much listening
1:04:33
to this and it's so links in with things like self-actualization even
1:04:37
plato's cave i was going to bring up when you're mentioning the den and all sorts of
1:04:39
things are covered there's so many links i'm so nice to
1:04:42
hear you say names like wilhelm kunt and
1:04:45
things and just thinking wouldn't and just thinking wow these are so applicable so
1:04:48
there will be details on my communication course which kind of links in
1:04:51
very nicely to this in the show notes as well for your benefit
1:04:54
not for my profit just because it links in very well right we've
1:04:57
reached the end i'm going to have to let these two wonderful geeks go but i'm
1:05:00
feeling very much there needs to be a follow-up once people have a time to digest this
1:05:03
if people do want to follow you and and look for
1:05:06
some more references on these topics dave for you
1:05:09
obviously we've got the hsu.ac.uk which is the website for health sciences university
1:05:14
so there'll be information on there and links to the health normally it's health.sciences.university
1:05:18
across social media and then also for yourself maybe most active on linkedin
1:05:23
and facebook facebook facebook maybe Yeah,
1:05:27
I'm an old dinosaur now on Facebook.
1:05:29
But yeah, and LinkedIn as well, yeah.
1:05:32
Fantastic. So I'll make sure they go in the show notes. For you,
1:05:34
Anna Maria Maciori, that's wonderful for you. It's just at the ST School,
1:05:37
isn't it? A lot of the time.
1:05:38
Again, I am a little bit on Facebook. I started LinkedIn again,
1:05:41
but as you know, you can reach me always and ask questions and,
1:05:46
you know, I will always try to reply.
1:05:49
Amazing. Fantastic. Thank you, Mark, for inviting me. It's been a real privilege.
1:05:53
Thank you so much. Oh, that's so kind of you to say. Really gracious.
1:05:56
You're a fabulous host, Matt, I always say. You're very gracious.
1:05:59
I'm just a reflection of you guys. So that's all I do.
1:06:04
So I'm really excited. This is week two. This is so cool.
1:06:07
And I believe actually, before we just go, thanks everyone who's joining us.
1:06:10
Obviously, this is not a personal conversation. This is all still relevant for
1:06:13
the people who are still listening to us.
1:06:14
I believe that next week, Dr. Rachel Dewa-Haggart, you've crossed paths.
1:06:19
She says she's crossed paths with you, Dave, when she was at Southampton.
1:06:23
And yeah, you might have to look her up and see, but she's crossed paths a few a couple of times.
1:06:27
She is currently at the
1:06:30
medical sociology and health experiences research group at the nuffield department
1:06:33
of primary care health sciences university of oxford but she was down in southampton
1:06:36
for a long time as well working with primary care so that's gonna be next week
1:06:39
and then the week after just so i don't forget we've got dr rachel zofnes pain
1:06:43
scientist international speaker science writer and she's got a new book coming
1:06:46
out tell me where it hurts coming out march,
1:06:49
26th which i've had the joy of pre-reading and we'll be able to talk about that
1:06:52
book in a lot of detail, which again links in quite a lot with what we've been
1:06:56
saying in week one and week two, and I'm sure it will be week three as well.
1:06:59
Right. That's it. Thank you so much for people who have stayed with us listening in the comments.
1:07:04
I know there's still four or five there. I'm quickly going through there.
1:07:08
Sarah Jones, I think has summed it up. That's all I'm thinking.
1:07:10
I need to listen to this again. So much fantastic information to assimilate.
1:07:13
Great chat. Thanks. Okay. It is one of those, but that's the joy of the podcast.
1:07:17
And if you are listening to this and you've got more questions,
1:07:19
which in theory you should have, I mean, I'd be very surprised She went,
1:07:22
yeah, I understand that.
1:07:24
Just could probably just either send me
1:07:27
an email i will go on youtube i'll make sure there's a version of
1:07:30
this on youtube put some questions in the comments comments underneath youtube
1:07:33
yes unfortunately this hasn't gone onto youtube at the moment as we
1:07:36
speak but i will be i'll upload the video to youtube nikki mansford said yes
1:07:40
please the reading links yeah nikki thank you for staying with us um i hope
1:07:43
your head is not feeling half as bad as mine at the moment a bloody brilliant
1:07:48
discussion says nikki uh thank the lord for touchy-feely nerds in the least
1:07:52
dodgy way possible. Thank you, Nikki, for that.
1:07:54
And Inri Testo also says, fantastic episode. It is a, was a fantastic episode.
1:07:59
Yeah, Inri Testo and Christ, so all of this should make very much sense.
1:08:02
Oh, hugely with Phil and Christ.
1:08:03
And in fact, Todd Hargrove is just on Substack, if you are on Substack,
1:08:08
still knocking them out.
1:08:09
I remember his first two looks, playing with movement and that,
1:08:11
but Todd Hargrove has released some really nice recent things on Substack to
1:08:15
do with these questions of nociception and the sensation versus perception and
1:08:19
all these sort of things.
1:08:21
I mean, again, it's not something for everyone but
1:08:24
personally i love the way he puts things and really philosophical as
1:08:27
well which is a bias of mine but anyway right let you
1:08:29
lovely people go and that includes my guest tonight professor dave
1:08:33
neal thank you very much dave really appreciate your time you do and also
1:08:36
thank you ann maria massieri and a great good luck i'm sure you'll take i'm
1:08:40
sure i'll talk to you very soon but good luck with your phd as well it sounds
1:08:43
so interesting thank you um until that everybody i hope you've taken something
1:08:47
away today and you are going to join us for the journey of understanding pain
1:08:51
2026 i think this episode again like last week just encapsulate that
1:08:55
understanding pain 2026 is a great title for this so until then if you can join
1:08:59
us next week next tuesday then do please do that eight o'clock it will be on
1:09:02
youtube he said crossing fingers and facebook so we'll see you then until then
1:09:05
take care of yourselves and obviously continue taking care of each other we'll
1:09:10
see you again very soon bye-bye.