Jared Powell:
Dr. Mervyn Travers, welcome to the show.
Mervyn Travers:
Thank you very much. Thanks for having me, Jared.
Jared Powell:
You're very welcome. Merv, we've been, can I call you Merv? Is that all right? Yeah.
Mervyn Travers:
Of, yeah, it's Merv. Yeah. Yeah.
Jared Powell:
Are we close enough? Yep. Cool. I should ask you this before we went live, but anyway, we've been having a really productive conversation, so apologies for that. So, Merv, I know who you are, but do you mind introducing yourself briefly, both professionally and personally to the audience?
Mervyn Travers:
Yeah. So I am a physio by training. I trained in Ireland where I'm from and I qualified oh, 20 years ago now, I think, or even more. And did my, you know, bit in, in general hospital and my basic rotations worked on the side in some clinic and eventually transitioned into private practice and working in sports. Came to Australia to do my post-grad at Curtin University, and only plan on staying for 10 months, but here I am, I don't know, 15, 16, 17 years later. I'm a citizen and my life here now. So yeah, did my postgrad went on to work in, in professional sports and then in did my PhD here too. So I kind of sit on both sides of the fence, both. I'm a physio and an SSC coach, and I've kind of worked in roles on both sides of that fence in the past.
Mervyn Travers:
So my kind of professional interest, my research interest, my teaching interest is kind of trying to merge kind of high level exercise, rehab, and kind of an understanding of pain. And so the types of patients that I see are often, you know, patients who've been struggling with pain and like, compromised performance for a long time. So they're often athletes or more athletic populations. But their goals are very lofty. So my leg hurts and I cannot run even three steps, but I would like to run a marathon. So it's kind of bridge that enormous gap. So that's kind of what I spend my, or spent a lot of my time doing clinically over the years. And and, and still, still do a little bit of clinical work seeing patients like that. And yeah, on a, on a, on a, on a, on a personal level, I'm a, a, a proud father and hopefully decent husband. And yeah, just enjoy getting out. If, if, if you, if you, if you find me, you'll find me out in the dark early morning, running along the coast of Perth.
Jared Powell:
Beautiful. So running your major sport or physical activity of choice?
Mervyn Travers:
Oh, look, I'm, I'm no good at it. I can tell you that I didn't grow up as a runner and I, I don't have the body size or type for it, but like I, I'm happiest. I'm, I'm happiest in the gym, to be honest. But I, I do, I do pound the pavement a little bit.
Jared Powell:
Cool. And as an Irishman, how's your Aussie accent coming along?
Mervyn Travers:
Well, it's interesting. I think I'm as Irish sounding as they come when I go home, at times, my, my, my mates will pull me up in the pub and say, you sound Australian. I'm not sure any Australian has ever accused me of that . So yeah,
Jared Powell:
I, I heard a, I heard you say a word that was Australian ish before we started recording you said Ripper and that was . Oh, there you go. Yeah,
Mervyn Travers:
There you go. But it is funny, my my, my daughter, she was born in Australia and she, you know, we were trying to say, well, you know, you've got Irish family and all that kinda stuff, and it's like, where was I born? Australia. It's like, I'm Australian
Jared Powell:
.
Mervyn Travers:
Having said that, rugby is you know, having worked in professional rugby and been around it and played it and kind of, it been a big part of my life, rugby's kind of one of my major sports that I'm interested in. And she's seen some Wallabies games. She's seen some Marlin games, and she's, she's, she's crossed the fence when it comes to cheering for rugby.
Jared Powell:
I was gonna ask you the rugby question. Ireland's very good at being number one between World Cups and don't turn up at World Cups, Merv, so can, you guys can can take the number one for those few years. Oh, mate. Win the big tournaments.
Mervyn Travers:
I, I was in, in, in the World Cup in oh seven in France. I was there with my, with my, with my head in my hands when we were drastically underperforming . So I, I'm afraid yeah, our performances were very strong in the last World Cup, amazingly strong. And in our defense, the game we lost against New Zealand, I reckon is about the highest quality game I've ever watched. I thought it was, and maybe I was just caught up in the drama of it with the vested interest, of course. But you know the small margins out there, they're small margins,
Jared Powell:
So agreed. I mean, it was a tremendous World Cup. Anyway, the New Zealand Island game was amazing. The South Africa France game was amazing. Yeah. Any of those four teams could have won the whole thing, was it?
Mervyn Travers:
Yeah, it, it was that close, but, you know it's, it is what it is. And, and we, we, we rolled forward
Jared Powell:
The, the, the, the most important thing, the last point on rugby is we have a British and Irish Lions a tour coming over to Australia in a few months. Yeah. And that's gonna be massive. The Wallabies were, were written off a couple of years ago, and I, I think we're gonna do a little bit of damage, so that'll be . Oh,
Mervyn Travers:
May I, I'm much more nervous about the prospects of the Wallabies or playing the Wallabies now than I would've said 18 months ago. Unquestionably, 18 months ago, I think people would've said, this is gonna be a whitewash. I don't think that would be the case now. You know, not like the Irish team is a little bit in transition, the wealth team's very much in transition. You know, there's a lot of kind of rebuilding happening across the kind of teams that make up the lions, whereas the Wallabies have been rebuilding and are kind of a bit more on that ascendancy. Mm-Hmm. So I think, I think you've got we're in for a much more entertaining series than I thought it was gonna be. Yeah.
Jared Powell:
Yep. I agree. And we need it too. And that's because we have an Ex Island coach, Joe Schmidt at the helm. Yeah,
Mervyn Travers:
Yeah, yeah, yeah. Well, I mean, it's like, it's like all things, he's, he's not Irish, but when it comes to rugby, we're gonna claim him, you know? But, but, but yeah. Yeah. He, he, he did wonderful things with the Irish setup. He did wonderful things for, for my hometown team, Lester, which would be be the kind of provincial and and professional team. And the part of all I'm from, he was coached there as well. And, and he did amazing things. And, you know, you know, it's probably no coincidence that Australian Rugby's on the rise.
Jared Powell:
Agreed. We're gonna leave rugby. I'm, I'm, I'm aware I could talk about it for hours, but I'm sure we're marginalizing and alienating nine five. I thought
Mervyn Travers:
This was a podcast.
Jared Powell:
Yeah. join us for the Shoulder Physio Rugby PO podcast. All right, let's get into, let's get into the meat and the academic aspects of the conversation. So, Merv, I've read a lot of your work in tendinopathy. I've heard you talk at length about strength and conditioning, something that I know you're interested in, which maybe you have some works underway. I know, I know you've talked about it as well on podcasts before, is something that is broadly called Active Inference. Would you like to just, first of all, explain what active Inference is, maybe define it and, and just help all of us simpletons understand what it is that a very superficial level?
Mervyn Travers:
Sure. so I suppose I'm gonna go back and, and kind of, it's always nice to tell as a story here, you know, 2017 started writing this paper with one of my, a couple of my colleagues at, at University of Ana, especially with a friend, colleague, and mentor Ben Wand. And it was meant to be this kind of, look, it's 2017, where are we, we're not winning the war on, on back pain. Like, where are we at? Like, what are the promising treatments? What's the evidence base of where we're at right now? But also, does it align with how we think pain works? 'cause If, if it doesn't, then we're kind of barking up the wrong tree. And so in that reading, I started, or, or, or in that space, I kept coming across these concepts of kind of predictive processing at the time.
Mervyn Travers:
Predictive coding, predictive processing to try and explain conscious experiences. And so you kind of have to understand predictive processing to really get active inference. That's kind of why I start there. And so this idea of predictive processing, which has been around, like the idea that we have a predictive brain has been around since like the 18 hundreds. Alright, so it dates back to helm halts. And so, you know, this is an idea that we don't just sit and wait passively for stimuli from the external world, okay. That actually we are constantly making predictions about sensory information coming from the outside, but also from within our bodies. Alright? And so we're constantly making predictions about it and to try and explain what's causing the sensations that we have. And then you use that information to kind of successfully navigate our world. And so, you know, one of the, a way of thinking about this is if, if you, if you are, if you remember the days of CDs and, and, and, and cassettes even, and records that would play songs sequentially, and maybe we're too used to having playlists and stuff now that don't, but back in the day, if you were listening to your favorite record and it was skipping between tracks, you'd almost start to hear you'd the next record.
Mervyn Travers:
It's not like, oh, I, I know it's coming, but you could actually have the sensation that you're hearing it before it comes as an example of how predictive our system kind of is. And so from there, from predictive processing there's this idea that, you know, I'm making predictions about sensory information, but of course the sensory information that impacts my system is never gonna be exactly the same. So there's this difference or prediction error. Okay? It can be also referred to as Surprisal. And, and what active inference is, is active inference is, it's, it's based on this idea of ris, it's a grand kind of unifying theory of planning, cognition, memory, learning perception and action. So kind of how do sentient beings, how do humans work really, and how do they survive and navigate their world? And so it's got links to computational neurobiology and it's got links to philosophy and links to cognitive neuroscience.
Mervyn Travers:
It's kind of this overarching on umbrella framework. And the idea, or one of the key tenets of it is that we and all living systems are engaged in trying to suppress error all the time. So we're trying to minimize the RIS all the time. And so we use our bodies to minimize ris. And so what I mean by surprise, well, you have preferred states that you'd like to stay in, and there are dis preferred states. So for example, the, the common example that's often given is a fish is preferred state, is to be in water, a fish out of water that's a lot of ris, and it will stop being a fish very quickly. It will become a dead fish, right? And so that's a very extreme example, but we have, like, the whole concept of allostasis is based on this idea that we have preferred states is a temperature range in which we want to stay, you know, so we have systems to, if we get too hot, we'll start sweating, but we might also take a drink or we'll go and turn the air con on.
Mervyn Travers:
So there's this range from, from physiological to kind of broadly societal policies that we can use to try and suppress the error associated with being in a dis preferred state. And so active inferences is really all about that, that, that we have this biological imperative to minimize ris. And I know that must sound really abstract because I'm not pfi that example, but I suspect our conversation will go there. And, and, and, and we can discuss like what that means for, for pain and for clinical practice and for movement. But I think the fundamental people think that people need to understand is it's about surprise or minimization. So about having boundaries with which your system wants to stay in and using error and suppressing that error to stay within those boundaries as best as possible.
Jared Powell:
And so the active component is your sort of acting upon the world, as it were to minimize that error or prediction error.
Mervyn Travers:
Yeah. Okay. So we got this idea that we are agents in this world and, and we engage in kind of reciprocal feedback loops across multiple timescales with the world around us. So basically we're acting on the world and the world is acting on us. And so action can be as simple as, you know, the the rapid eye movements that you have to, to pick a stimuli when you're investigating, you know, you pick up a golf, golf ball and you kind of look around, look at the little dimpling. That's, that's a form of action you're trying to learn. Okay. Or suppress error.
Jared Powell:
Cool. So let's just define terms again. So we've got active inference, we've got predictive processing, we've got, I've even heard something called Bayesian inference. Yeah. Are all of these terms interchangeable, do you feel, or do you think they're different and should be demarcated?
Mervyn Travers:
Look, I think, I think Active inference kind of sits, like they're all steps along the way to Active Inference. I'm not sure if you've read the Active Inference textbook Carl Friston and Polo and Pars book. It's a, it's a wonderful book. It's actually available as a PDF online and it's a wonderful as a free PDF online. And it's a wonderful, it's a wonderful read. And, and they speak about like, the journey to active inference as a concept and kind of basic inference and predictive processing are kind of steps along it. So I don't think you can probably use predictive processing an active inference interchangeably, but I, I I, I, I would, I would have less of a problem with kind of people talking about the other ones interchangeably because they're more about this idea that, you know, what you see and what you feel and what you hear, et cetera, is not a direct reconstruction of what's actually happening.
Mervyn Travers:
It's in impaired from the interaction of incoming sensory information and your predictions. Okay. So, so the idea that you, you never truly, like I, I've got a coffee cup here. I never truly experienced that coffee cup if I'm looking at it because, you know, my cranium houses my brain, so it's separated from the world and, you know, photons hit my retina, it's transduced into kind of chemical electrical signals. They propagate along my optic nerve, but really all my brain receives is beeps and clicks and whistles, right? So in order to make sense of it, it's also using predictions, alright, from, from from past experiences. And so the idea of basing inference and predictive processing are really about interactions between your incoming sensory information and the top down predictions held within your system. So you have these generative models, you have these internal models of yourself and the world and, and how you interact with it.
Mervyn Travers:
And, and so you use that to make predictions. And I think the key thing is most of this is subconscious. It's not that you walk around all the time saying, I'm gonna predict this. This is just how we, we think from a, on a cognitive level, things work, but equally on a neuronal level. Like people try and map hierarchically how the brain works in this way. But you know, the, the key thing within that is active inference then speaks as an overarching framework to the idea of trying to suppress error within that. That's a biological imperative to suppress the error. And, and kind of you in our chat, earlier you mentioned ya Howie's work and, and so I think it was ya Howie you coined the term self evidencing. So you could, you could, you could, you can flip the script and say suppressing error, minimizing supplies is really, really important. Or you can actually talk, describe it in terms of constantly confirming your model that you are you and that you are still alive.
Jared Powell:
Cool. So is it fair to say that all of this is, is basically predicated on the fact that our brain is a prediction machine?
Mervyn Travers:
Oh yeah, a hundred percent. Yeah. Yeah. It gets away from that kind of kind of stimulus response kind of perspective on, on, on, on our brains. And it makes us more kind of goal oriented based on our predictions. Yeah. Cool.
Jared Powell:
And that seems to be a fairly uncontroversial as far as I'm aware in my reading. You know, in my listening to cognitive scientists, to philosophers all around the world, it seems to be like there is a consensus that the brain is predictive rather than receptive. Do you have any insight into that?
Mervyn Travers:
No, I like, I think I, I think there's been an evolution probably in the last 50, 60 years to kind of, for that to become more mainstream and certainly in the last 20 years from, from me trying to put dates on the things I've read. And I've certainly not read everything that exists on it, but I, I think it's, it's probably fair to say the pendulum has kind of swung that way for an acceptance for people that, that we are, you know, predictive machines.
Jared Powell:
Cool. Yeah. Okay. So let's take all of this and apply it to musculoskeletal pain. No, no. Small task, I'm sure. So, so how, how can an active inference lens, how can a predictive processing lens be used to explain musculoskeletal pain?
Mervyn Travers:
Yeah, look, this is really interesting 'cause it's funny 'cause I, I'm really interested in cooking. I really, I love cooking and, and again, going back as a clinician and reading pain stuff for years and years and years, one thing that always struck me was, you know, we can change your perception of smell and taste and your experience. So I remember reading these old experiments, and I think one of them was called Grape Expectations. I mean, what a fantastic name where they got a bunch of people. Oh no, I think great expectations was a, was a, was a review of these papers. I, I saw, I, I digress. But the, they, they didn't, there was an experiment that was done a number of years ago where they got some, you know, fancy wine people, people who can smell wines and get, I, I don't, I don't have such a refined sense of smell. I think we tell people, they, they get a glass of white wine, they put their nose in it and they can say, oh, that's, that's oaky, or that's,
Jared Powell:
They're called a, they're called, so, or something like that.
Mervyn Travers:
Yeah, yeah, that's right. So they got a bunch of those and they they, they brought 'em out of white wine and they asked 'em to smell and kind of describe what they smell. And the kind of consensus from most of these people was the kind of descriptors you often see from white wine, like citrusy and lemon and, and, and melons and grass and whatever else, right? And so they took that tray away and they brought them out a red wine, and they asked the people to now give their description of that. And they all cherry and chocolate and rich. And, and so the mahogany, like the, the, the smells, the descrip, it's all changing. It's different, right? Except what the people didn't know is that the red wine was actually the same white wine. They just applied a odorless flavorless red dye to it.
Mervyn Travers:
All right? And what that suggests is that their perception, their experience was influenced by the color which they see before they even to their nose. I think for me, that's a real example of how what we experience can be led by our expectations or predictions in our system. And so you've asked, okay, why, what does that matter in terms of pain? Well, it matters because for me, I don't see pain as any different than taste or smell or hearing. It's, it's another conscious experience. And I was always wondering how come the other conscious experiences are so malleable, right? We can, we can change them. So, so why is pain so difficult to kind of change? And so I started reading in that space and also doing experiments in that space to see how do, how malleable are these experiences and how predictive are these experiences?
Mervyn Travers:
And so the first thing for me is that means that information you hold already, it doesn't have to be explicit information. It could be just built up through your past experiences, prior knowledge, social, cultural context. They influence what you experience. And so maybe when the, when your doctor shows you that scan or your physio says, that's the worst scan I've ever seen. Or you tell that 25-year-old, you've got the back of a 90-year-old, or you tell jar jar, you know, if you don't, if you don't back off from exercise, you're gonna end up in a wheelchair. Well, maybe that information influences what the person experiences, much like the color of the wine influenced the experience of smell. So that's the first thing for me. You're saying why is it relevant? So the first thing about being predictive for me is that prior held information may well influence what people experience.
Jared Powell:
Yeah. Cool. So, so we're saying that pain is an experience like any other conscious experience, as you mentioned, taste, smell, touch, et cetera. Mm-Hmm . And if it is that then it logically follows that it can be influenced by lots of implicit things that are going on within our system that we may not be aware of in terms of socio-cultural factors. What our clinician is telling us, what the general beliefs are out there that we're hearing, that we're inundated with, maybe on TikTok, maybe on Instagram, et cetera, et cetera, et cetera. Pain might be malleable by all of these factors.
Mervyn Travers:
Yeah. And what it suggests to me is that pain is about more than just what's happening in the tissues. And what's really important here is it doesn't mean that I'm saying pain is about, it's in your head, it's in your brain. It's not an active inference would suggest that your body is largely involved in these things and we'll get there. But I don't, I don't for a moment think that pain is just a concept that's in your mind or in your head. And I think what you're really careful about not laying that on on patients in any way, 'cause I, I think it's a, it's not only an unhelpful narrative for patients, I don't think it's academically honest. And then if so, so that's the first thing that it's about more than just what's happening in the tissues. And that opens up all sorts of possibilities around, you know, there are some kind of ways of treatment that have a kind of central concept around demonstrating modifiability.
Mervyn Travers:
You know, and if you can demonstrate modifiability of pain through true non tissue based things, you know, my pain was worse when I was doing my exams, it was better when I was on holidays. Maybe, maybe they're the moments that allow a patient to explore the idea that it's about more than just their tissues, rather than us saying to them it's about more than just their tissues. But my perspective in doing that is informed by knowing that we're predictive and therefore other factors may well influence the conscious experience of pain. But for me, the kind of obvious example you look at in that in the clinical world is the kind of common narrative around damage degeneration. You know, it means that we need to deeply probe our patients understanding of what's going on, what they've been told, what they think is happening, what they think the consequences of their pain is going to be. Mm-Hmm.
Jared Powell:
Yep. So I just wanna reiterate, Merv, you, you're saying that predictive processing in these series do encapsulate and include nociception and neuro, neuro immune contributors to pain.
Mervyn Travers:
Oh, I think what, what it, what it, what it encapsulates is sensory information from the body and from multiple sensory streams. And so what we're saying is that your, your pain won't be just, can't be just about your tissues equally can't be just about what's going on in your brain. And, and like, if you look at, that's kind of where the two caps have kind of been for a long time in paying researchers this kind of very, very neuro centric camp. And then this is very kind of pathology oriented camp. And, and, and I actually think active inference in these perspectives we're talking about replace those and kind of unify them and say, look, you're, you're right. And, and you're right. It's just that when Jared comes in the door, I don't know if Jared is 90, his pain is 90% caused by his tissues and 10% top down or 50 50 or, or vice versa.
Mervyn Travers:
But I think, I think the key thing is getting away from this idea that it's just about the tissues or it's just about talking to people 'cause in their brain and understanding that what we experience is a complex interaction of incoming sensory information and top down predictions. And then we've not got into suppress action and suppressing of, of, of, of the error associated with that yet. Right. But I suppose the key clinical concepts at this point are one, pain is unquestioned about more than just tissue damage. And it doesn't mean tissue bowel damage is irrelevant. And I think that's really important. I think tissue damage is relevant, is, is relevant. And so when we talk about predictions, people, you don't just predict your incoming sensory information. You predict what we call the precision, or you can call it trustworthiness, but you know, we prioritize information in our system based on its precision, right?
Mervyn Travers:
And so we select, we attend to our system, shines a spotlight on high precision sensory information. And so my contention would be that nociception would likely be a high precision signal. And so it's going to have a spotlight shown on it by its very nature. And so that, that means that this is not just about what's top down, right? It's also about what's bottom up. And so nociception is real, nociception is relevant. So with that activation of neuro immune system you know, all of those things are real. This, this idea of active inference or predict processing these kind of views, they don't, they don't say none of the biology is relevant. In fact, they're, I think an explanatory framework for encompassing both sides of that equation.
Jared Powell:
It's funny, Merv, as you're talking, it's, it sounds you know, this whole dualism thing that Descartes sort of invented three, 400 years ago, we're still stuck. We're still separating the mind and the body, aren't we? And it's a bit sad.
Mervyn Travers:
Well, yeah. Well we still talk about bio psychosocial as though we're encompassing it, but that term and it's perspective by its very nature kind of says, you know, there's a biological component, there's a social component and there's a psychological component. And what active inference says is they're entirely entwined. 'cause The experience is governed by all of those things interacting. Mm-Hmm.
Jared Powell:
Bit of a curve ball here of how does active inference and predictive processing relate to in activism.
Mervyn Travers:
Oh yeah. So you're like probably talking here about like the five E's perspective of pain and gosh, I've gone blank on the authors of that wonderful paper. Still,
Jared Powell:
Well,
Mervyn Travers:
Still, well, Harmon, Catherine Harmon. Yeah. Peter still and Katherine Harmon. I, I thought that paper was fantastic. Like, I remember reading going, this is wonderful. I wonder if it's too early. Like, is the world ready for this? And so I think it's entirely compatible with this idea of an, you know, that you are an active agent who is a, you know, embodied and within the world and interacts with the world. So I think I I I think they are, you know, very similar flavors of the same ice cream. I, I think that the, that five e or an active perspective, 'cause you got, you got Factors paper with Julian Verstein and Michael Koff as well that, that kind of goes into that space as well. And so I think that we are having a shift in our thinking around pain. There's certainly some stuff starting to come out the last couple of years that, that are kind of talking in this way. But I think active inference is basically explanatory framework for that. Yeah.
Jared Powell:
Yep. I agree. Yep. In my reading of both theories, it seems like they're equivalent in terms of what they're saying. So that's good to hear. Okay. So let's get into how, how can we use these principles of active inference to help help somebody with musculoskeletal pain? How do we apply it? How do we be clinical about this instead of just being talking heads in academia?
Mervyn Travers:
Yeah, sure. Well, I think, I think, I think there's a couple of things that, and, and they're concepts that we've not touched on yet that, and we'll, we'll get to it. 'cause It, it's, you know, the idea of suppressing error will come in here a little bit. And I, I think the first thing is that, you know, I think people's beliefs, their understanding, their social cultural context, their history, their expectations around will they get better, et cetera, are likely gonna influence their pain and their recovery trajectory. And, you know, as a clinician, that's really hard. You, I'm gonna say something that could, could be reasonably controversial. But the patients that I see and have seen over many years, generally by the time they end up with me, they've been on a long journey and they've read a lot of things. They've been told a lot of things, some of which is unhelpful, some of which is maybe unhelpful, but I'm not res their, their pain is still a function of all of those things bound up.
Mervyn Travers:
'Cause Their internal model encompasses all of those things they've been told, all that they've been seeing every experience they've had in their life, implicit knowledge, explicit knowledge, et cetera. What really matters there is I will never understand the fullness of that as a clinician, because the person will never understand the fullness of that. The only access I have to their model is by probing and asking questions about their explicit understanding. But that will, that will, you know, only get us a small piece of a larger puzzle. But I think we need to be deeply curious and get to truly know our patients. We need to really understand and know our patients. And that's really hard in clinical practice. 'cause You know, we have a model of care that is, you know, 30 minute appointments and charges this much, et cetera, et cetera. And, you know, people need a livelihood.
Mervyn Travers:
So I'm, I'm not criticizing that, but time constraints are a barrier to truly knowing and being able to explore those concepts. But I, I think we need to really get a deep insight into the person's understanding of their pain and their problem because that is part of the equation. And so that's one kind of thing I think really matters within clinical practice. But the second thing probably more controversial about it is that we're not responsible for the person's model. As in, I I will do my best for every patient I see. And I am deeply invested and I want them to get better and I want the best for them. But I recognize that for some people, maybe their model's incredibly fixed and isn't for changing and they may not resonate with what I have to say or what my approach or what I'm doing.
Mervyn Travers:
And, and I, I can't take on the responsibility of things that I can't control. 'cause Clinician burnout and fatigue is a real thing. And, and, and, and one of the things that understanding the system, understanding how much of this I can't control and I'm not responsible for, has allowed me to lift that burden. And probably as part of what has allowed me to stay in the profession, you know, to have gray hairs is to be, not that I'm okay with the fact that I can't help everyone is, but, but to accept that that is the case.
Jared Powell:
Yeah. Yeah. I mean acceptance in, in coping therapy is a thing for patients. I think it should be something for clinicians as well, but it, it's it's implausible to think that you can take on all of these burdens, all of these problems and be responsible for them. That's not a sustainable attitude or mindset.
Mervyn Travers:
No, it's not. But I think, I think recognizing that, you know, some people's positions are fixed, that their, their internal models may become what we call in our group sticky. Mm. Yeah. And, and, and, and, and hard and hard to change. The next thing is, I think I've said it's about both sides of the equations that top down and bottom up so that our treatments need to consider the cognitive behavioral aspects of pain, of course, but also consider this idea of, of, of, of sensory information from the body. So I'd like to touch on that if we may. Sure. And so I said before that we, we deal with error and we effectively deal with error. And this might differ from some of the things that you've read, Jared, on it 'cause 'cause a lot of the books and papers we'll talk about dealing with error in two ways. Which one is by updating your model and learning and the other mev,
Jared Powell:
Can you just define prediction error again? Is it just the difference between expectation and
Mervyn Travers:
Experience? Yeah. Prediction, error, or surprise. Yeah, sure. Prediction, error or surprise is basically the difference between incoming sensory information and the and your predictions. Another way of looking at it is also not being in your preferred state. That is an error. Yeah. And so that we are a biological imperative to suppress that. And we do it in a couple of ways. You know, you learn, alright, that's why you update your model. So you can change your internal model to fit the narrative or fit the sensory information that's coming in. So I tell you you go my back, my back is hurting, I've got this pain. I go, yeah, that's 'cause you've got a bulging disc and I end up in a wheelchair. Well, you know, you update your model and you learn, right? And then you think about how that's going to influence the predictions within your system.
Mervyn Travers:
Another way to, to suppress error is to use your body. Okay. And I'll, I'll give a pain example of this. Like if you were floating around the physio landscape, I'm gonna say like from like 2005, like 2010, like Pete Sullivan had that wonderful work on back pain at the time where he was kind of trying to sub classify and there was the passive extensor subgroups and the, you know, active coactivator groups and all of that kind of stuff. So, so think about this and I, I think this is really, really important and you've asked me about exercise and movement. I think this is in part where this comes in. So like, let's take one of those examples and probably lots of people have seen this back then we'd say, we would've said, look, these people are sore because they're coactivator and bracing this, this co activation pattern people.
Mervyn Travers:
But actually an activation or, or an active inference perspective suggests that you're constantly making sensory predictions and you move in ways. You use your body in ways to confirm your predictions. So you're trying to suppress error by confirming your model, right? There's less of a, there's less of a model less of an error if you move in a way to change the sensory information. So perhaps those people, they've got a system like on a cognitive level, they're expecting things to hurt on a neuronal level. They have a system that's primed to shine a spotlight and expect nociceptive information. You could argue that the changes in the dorsal horn from central sensitization, for example, are just that second order in your own modeling what is expected from the first order in your own. For example, when I say you could argue, I do argue that, right?
Mervyn Travers:
And so, so you've got a system that in that way shines a spotlight or tends to, or expects or assigns a high priority to no susceptive information. So what do you do? You move in a way that confirms those predictions within your system. So by coactivator and increasing the load through the tissues, are you actually fulfilling those predictions? Not on a conscious level saying, I'm gonna make myself worse. And this is where you have to be careful. I never use this terminology with patients. This drives what I do and my thinking. These are words that never speak to with patients, but active inference suggests not that they're sore because they move in a maladaptive way using the terminology of that time. But actually the opposite, that you're moving in this way to confirm the predictions of your system. And so when Pete does his live demonstrations, he gets the person to breathe through their belly and let go and move forward.
Mervyn Travers:
What you've actually created is a positive surprise on right? And that, oh, that feels better. I can go much further, et cetera. You're actually creating, you're actually hijacking that system of surprise that you're trying to create positive surprise on. And so for me, the whole concept of movement experimentation is extremely important. Trying find ways a person can move or do a task that they wanted to do in a way that is in some ways better than I thought it would be a positive surprise that could be less fear laden, less anxiety laden, less painful. But the idea is understanding that they move the way they do in order to fulfill the predictions of their system. Okay. So if we can violate those predictions in a positive way, then you lay down a new model and then you're trying to strengthen that over time. That's where I see exercise is a way of strengthening models. 'cause You're constantly self-evident in your system when you move.
Jared Powell:
Yeah. I love it. Can I just jump in for a sec? So you said violate the do is is prediction error and surprise the same as ex expectancy violation in the psychological literature?
Mervyn Travers:
So I suppose in some ways yes and no. I, I'm, yes in terms of like cognitive expectation, but I'm also speaking here about your, your, if, if, if, if you change this experience, the person has, then you change the model in some way. So you may well have changed how information is weighted at a neuronal level. So I suppose it's not just about cognitive expectation violation. I suppose that's, that's I suppose a distinction there.
Jared Powell:
Cool. So prediction error would go all the way down perhaps, whereas expectancy violation is more of a higher order cognitive phenomena.
Mervyn Travers:
I, I would, I would see it that way, but I can also see how, and I probably do people just use the terms interchangeably. Yeah.
Jared Powell:
Cool. Yeah. Okay. So
Mervyn Travers:
I I, I, you know, I suspect there'd be a million different ways people could phrase that. Yep.
Jared Powell:
Yeah. Cool. I really want to sort of hammer in on the exercise component here. So, and I think you were just getting into it. So we can use exercise as a vehicle, as a vessel, as a method, as a way of, of positively changing expectations or beliefs to how, however deep seated they might be. Yeah. And updating their sort of model or worldview of, of how this should feel or, and how it actually feels.
Mervyn Travers:
Yeah. A hundred percent. A hundred percent. And you're laying down new models, right? And that, that's really what it's all about. And to me, that's what pain education is. I like, I I think someone learning, they don't learn that well, like by being told it's not just about your tissues, but if you're find able to find a way for someone to do movement or, or, or, or act in some way and use that body part that's better than they're expected, then they lay down a new model, right? And that, that really, really matters. I think people learn experientially and the way that we learn about our body is to use it. That's how we learn about our body. I mean, there's some visual information you can look down and see if your arm is crooked. Yeah, it's probably broken, but largely it's by moving. Like you, what happens when you, you bang your elbow off something, you kind of, you do this, what are you doing?
Mervyn Travers:
You're, you're updating your model about that arm, right? And so we're hijacking that system. It's the flip of that. And it's, it's a really challenging thing. 'cause You know, if you look at the exercise, like, I'm gonna use back pain as an example, but use this for any real body part. You look at the data on exercise as a treatment, it's not very strong, right? It's like from back pain. You've got Jill Hayden's amazing Cochrane review on low back pain exercise there. It's maybe a couple years old now, but it was just a, it's just a stellar and enormous piece of work. But you look at that and, and, and basically the summary, the highlight reel is exercise is just meets the MCID when compared to basically doing nothing and is no better than any other treatment for improving pain and doesn't seem to do a whole lot for function for back pain people.
Mervyn Travers:
All right? And that's, that's not a great premise on which we should be basing a great evidence base on which we should be basing our profession. Like having said that, there's a real challenge and a real problem there because I don't think the way we execute exercise often in clinic and certainly in in, in exercise trials, is consistent with how we understand how pain works. And this is where this whole evolution of thinking of mine came from way back going Well, is what we're doing matching how we think pain works? 'cause If our solution doesn't target the problem, what are we doing? So if you look at most back pain studies and, and Leanne Woods has done a lot of work in this space. There's a number of other people who've kind of analyzed them. If you look at what most back pain interventional trials, they'll start off all the same.
Mervyn Travers:
The introduction says back pain is a big problem worldwide, and it's the global burden of disease index means it costs us a lot of money and lots of absenteeism from work. BA pain is a biopsychosocial experience, meaning that it's got a biological component, psychological component, social, they're all the same. We wanted to see if strength training would help back pain. Like, well, why? And you read the method section and the method section already says, 'cause it's about increasing tissue capacity or doesn't, most of 'em don't really list a mechanism what they're actually targeting. So what are you try, why do you think this would help? What, what are you, what are you trying to do here? Right? Because we've just said pain's about top down predictions and incoming sensory information. In what way is your intervention gonna pull either of those levers? There's a third lever, by the way, which is differential weighting of sensory information.
Mervyn Travers:
We'll get to that in a moment. But, but the idea is like, I don't, I don't see, it's not, it's no surprise to me that trials that are geared that way and have a philosophy that way give, you know, at best modest effects because they don't target the problem. They're not geared towards the problem. So I don't see exercise in pain for people in pain as being about tissue capacity as being about, you know, flexibility or some change in the tissues. I see it as a form of of learning, of learning about the body, a form of self evidencing a form of laying down and reinforcing new models that my body is capable. So I don't know if you've read that Fitness for Purpose model paper that Ben Juan, gem Orange, a couple of my colleagues at the LAR as well a number of years ago.
Mervyn Travers:
It's, it, it, it outlines some of these kind of concepts. But I think the exercise is about you demonstrating the fitness, the capacity of the body, not actually about getting stronger. It just so happens that, you know, I was lifting 30 kilos last month, now lifting 45 gives you objective measures, and that really helps with that learning process, right? So I think it actually, I think exercise can be enormously useful. I think the evidence base is not strong for it, but I think the way it's applied is non-compatible. What I'm arguing for is, is applying it in a way that's more compatible with how we understand pain works. And, and it's, it's, it has been done in part, if you look at the RESOLVE trial, which I think is, I'm gonna say this, I declare my conflict of interest. I wasn't on the trial, but a number of my friends, colleagues and mentors were part of this trial.
Mervyn Travers:
But I think it's the most important trial in back pain. Certainly this come out in the last kind of five or 10 years. This is where they looked at graded sensory retraining for low back pain. And it was compared to a truly credible sham. And so for the first time ever, they, they, and they outperformed the, the sham For the first time ever, we've signaled above the noise, as far as I'm concerned, we have true evidence of a intervention non-surgical, non-pharmacologic intervention that shows important and long lasting improvement in pain and disability in a cohort of patients that are very unlikely to get better on their own. These are patients I think the average pain duration of them was about five and a half years. You know, these are patients with back pain that's probably not going anywhere in a hurry unfortunately.
Mervyn Travers:
And so they outperformed a truly credible sham, which is very different than most trials where head-to-head trials and you don't know if people have actually truly recovered. We've got signal above the noise above natural history regression to the mean contextual factors, all of those things. And so they actually went on a journey along the lines of consistent, what I've just described, this idea of learning using the body, okay, rather than three sets of 10 with 40% of your one rep max and adding 5% a week or whatever else. And so we have high level evidence randomized control evidence demonstrating a meaningful outcome using a treatment that is basically aligns with an active inference perspective. One critical aspect though that's missing that they had and they started their training with was this idea of rated sensory training. So should probably fill that gap in for people, I think.
Jared Powell:
Yep. Cool. Okay. So, all right. I love it a lot to get through there. I got a lot of questions. How, with that, with that resolve trial that you just mentioned, it, it sounds like it's consistent with the theory of active inference, but, but how do we know, so how do we know that underpinning the effectiveness of the resolve trial, you know, was this sort of updating of beliefs like the causal mechanisms underpinning improvement? Like I, I know that they did a mediation analysis on that study, I'm pretty sure, and it showed that a change in back beliefs almost explained the entirety of the effect. Is that
Mervyn Travers:
Right? Yeah, it was, it was change of back release reduction, pain, catastrophizing, and if I remember off the top of my head, improvement in in self pain, self-efficacy, I think they were the three main things that came outta the mediation analysis. Yeah,
Jared Powell:
Yeah, yeah. And that's, it's the usual suspects when it comes to mediation analysis of, of anything for musculoskeletal pain, really, it's never strength, it's never biomechanical variables. It's typically psychological variables, self-efficacy. Yeah.
Mervyn Travers:
I I'll say that though, very often, psychological variables that are measured, right, true. So that's what comes up in the mix in the wash, right? But having said that, the data that exists on correlation between changes in physical capacity and improvement in pain is doesn't, doesn't support that idea. Yeah. Yeah.
Jared Powell:
So this is, I guess I'm getting to a wider question here and, and tell me if it's too early. But there, there are criticisms for active inference and predictive processing, most notably, how do you know, how do, how do you test it? Like how do we Yeah. You know, it's like the theories of quantum mechanics, the many worlds theory of quantum mechanics. How do you know there's all these other worlds out there? We can't go and see them, can't touch
Mervyn Travers:
'Em.
Jared Powell:
Yeah. It's, it's just a, it's just, if you follow the logic, that's just Yeah. What you have to accept. So how do we, how do we deal with these criticisms?
Mervyn Travers:
Yeah. And I, and I think the criticisms are fair. 'cause If you're gone to that true kind of philosophy of science world, like you're dealing with the a system or a claims and assumptions that can't be falsified, right? So how do you really test it the best, the best thing? I don't think that's a problem that we can't overcome. I think the key thing is that we don't hitch our wagon to active in infant and say that's everything. I think what we do is we go, this seems to make a lot of sense, and across a whole series of domains and worlds that aligns and makes sense of some of the existing problems we have. Okay, that's a tick. It doesn't prove it's a, a thing, but it go, okay, that's something. It, it, it has, there's some plausibility there to it. Okay, well, what are the assumptions that you can, that this makes in terms of pain, for example, 'cause that, 'cause that's the world we're, we're talking about, well, what are the assumptions that are brought about by this framework?
Mervyn Travers:
And can we test some of those assumptions? So for example, where I'm saying that I think your internal model would likely influence the pain that someone experiences. So that would suggest if we could alter influence someone's internal model via information, for example, that we could alter the person's pain. So we've done a series of studies on patients. We chatted about 'em before, before we came on Jared, but obviously they're, they're not published yet, though accepted some but not published yet where we, where we tested those assumptions. Even things like, you know like condition pain modulation, these ideas that, you know if, if I hit you with a hammer on your hand you that would be really sore. But if I shoot you at the foot at the same time, then you won't be too worried about your hand, about your hand and you'll, you won't feel it so bad type of thing.
Mervyn Travers:
We've done some experiments where we've looked at the predictive nature of that, where we've used deception to make people think and really believe something really bad is going to happen. And we've examined how my set about shining a spotlight and prioritized the sensory information in that moment, tested the acuity of some of sensory information and, and, and, and demonstrated a predictive component to it where people make themselves more sensitive, where they expect, expect there to be a, some kind of a damage or harm and completely suppress sensitivity in areas where, you know, it's not a priority. Right? Then again, suggesting that predictive nature, right? So, so we can, we can test and we are doing that some assumptions around it. We can also develop treatments that seem to make sense and align with that perspective and test their effectiveness. Could we ever test that on a your own level?
Mervyn Travers:
We've changed the kind of waiting and hierarchy within your nervous system to reflect? Probably not. And, and, and I'm, I'm, I'm very comfortable with that. I think all of these ideas are an evolution. You know, we're talking about active inference now, but if we'd done this podcast five years ago, it would be about predictive processing. And if we'd done it 10 years before that, it'd been about predictive coding. And so, and if we do it in five years time, there'll be something else we'll add to this and understand. And so it's an evolution. And so if that evolution allows us some testable hypothesis that, that we can see and does it stand up, that's really important. If it allows us to develop treatments and direct us in ways that have demonstrated efficacy in trials that we can move forward in a clinical world from, like, I, I, I'm willing to accept that it's, you know, there's no way we can truly, truly test every aspect of it. And you can't touch it. You can't see it in the same way of many worlds. You can't reach out and touch 'em.
Jared Powell:
Yeah. Yeah. I'm at peace with that as well. You know, as long as there's a sort of rigorous pursuit to explain as much as you can and test as much as you can and not just evade and explain away, you know, and that's what I think it sounds like you are suggesting as well.
Mervyn Travers:
Yeah, a hundred percent. And it's, it's the same as, you know, like my perspective will shift and, and so it should 'cause scientific endeavors is based on going where the river of evidence flows. And so one thing that people will use to dismiss active inference, for example, or any other philosophy that doesn't align with their beliefs, they'll say, well, that's not testable. And I'm like, okay, well demonstrate to me the evidence of your framework as well, and that that's what it is. That's what frame, that's for me, that's what frameworks are for. Like, when you put forth a framework, it doesn't necessarily need to be based on all of these, like every assumption having been tested, we're saying tested, we're saying at this point we think this is the best explanatory framework, let's go test these things. And that's kind of where we're at. But with, with certainly with some degree of promise,
Jared Powell:
The reason why I like, or this I find this model so appealing is because we, we did a study, a qualitative study where we interviewed a bunch of people with shoulder pain and just asked them about their experience with exercise. And also we see this in our clinical practice as well. A lot of times they report something that they say I I just know it's going to hurt when I lift my arm up. You know, that painful arc movement. And then a therapist tells them, they, they reassure them. They may be experiment with movement, they may be change the movement or the context of the movement, the lever arm, something, they change the movement, and that person then has a different experience. The pain is less, or it's positively improved. And for me, that just is, it's such a neat fit with the active inference model, and it just, it just works really well when patients are, are just reporting these things so explicitly, you know, it's really hard to get away from, like, how can that pain be so reduced via distracting them or via, you know, going from a two kilo to a one kilo or just simply changing the context, doing it in a slightly different way.
Jared Powell:
You know, like you can't, you can't minimize or completely abolish the pain just by doing those little tweaks of movement. I don't think, do you have anything to add to that, Merv?
Mervyn Travers:
Yeah, like a hundred percent. But that's what movement experimentation is, you know, like people would say, oh, scapular, dyskinesia. Remember when that was really popular? And we'd facilitate, you know, upward rotation when someone lifts their arm up. Like for me, that's like just, just doing it with a sensory information that doesn't necessarily match the predictions of the system, but creates a positive experience. Mm-Hmm. It's not retraining where their scapula tracks, it's not moving the position, you know what I mean? It's just some modification. Mm-Hmm. And if you think about with exercise, I often think about what are the, what are the ways that you can modify? Like, I mean, you've mentioned some of like you can change the load, you could change the, this velocity, you could change the type of contraction. You can ta change all sorts of things. One of the things that I often play around in experiment with patients is to reverse the origin and insertion, right?
Mervyn Travers:
So a patient can't bend forward, hurts too much, but I can get 'em to do a reverse hyperextension, which is just reverse the origin insertion. They're like, oh, that's much better. And I'll video them. I'll put real time video so they can see it, right? And like we have all that data on visually induced simul much easier where it actually, it hurts less when people can see their back moving, right? You've got Ben Juan's proof of concept study where they, they made used VR to make someone's back seem artificially extra muscular. So suddenly they've got like the rock's back or something. I'm saying they're holding a box that's hurting them. That hurts less. 'cause You know, my back is, so when they, like you asked about embody, when they embody that illusion, then it hurts less, right? So like that predictive nature, that interaction between top down and bottom up I think is kind of unquestionable.
Mervyn Travers:
But, but going back to what I was saying, going to reverse origin and insertion, it's the same movement. It's using the same body part through the same range and it hurts less. And that allows us to reflect that and go, well, why do you think it hurt that? 'cause That disc fold would be the same, or that degenerate would be the same whether you lean this way or lift your legs up. So it allows you to reflect and, and, and help develop those models and then reinforce it with, with with kind of practice and repetition. And that's where your structured exercise kind of comes in.
Jared Powell:
Yeah. I, I agree. So we've touched on it already briefly. So where do you see the future of all this going? We, we seem to have a camp mostly on social media that think there's gonna be some miracle pharmaceuticals that are gonna come out that's going to cure persistent musculoskeletal pain. And that's great. I hope that happens. I'll, I'll buy all the medications when they do come out. Then we have a different camp that suggests that it's gonna be more of these cognitive behavioral, higher order strategies that are gonna, that are gonna change the game. And we have proof of concept there with the restore trial and the resolve trial and a couple of others as well. They seem to be at OS with, at, at e with each other. We've discussed though that you think that active inference and predictive processing might be a way of sort of integrating and combining the two. Give me your five minute TED talk on this
Mervyn Travers:
. I think it's really hard. Look, I-I-I-I-I-I, I think it's the idea that you're describing about kind of pharmacological interventions, particularly ones that have kind of an effect on kind of a neuro immune system. Like I think they have enormous merit and I think that is so worthy of research interests and research money as well. I think, you know, 'cause what it's saying is it's both sides of the equation. I think we should be exploring both. And I think, you know, there will be a role for both, for treatments that target both sides of the equation. Because as I said before, I like, we don't know if yours is driven 90% by really high signaling nociception from your system. And we can't measure that in real time in humans, right? And we don't, or we don't know if yours is driven largely via, you know, top, top down influences that cause you to highly, highly, highly weight nociceptive information, right?
Mervyn Travers:
So we don't know if it's both side or which side of the equation it is. And so I think there's merit in pursuing both. And I hope both are successful. But, but on touching on this 'cause 'cause 'cause your suggestion is around treatments that are very tissue based. And, and I've mentioned this idea of sensory information and I said that we suppress error to learning or by using our body. I put this in this category. When it comes to pain, though it might differ from some of the things you've read in the space. I think that we have to think about it and discuss the idea of, of differential weighting of sensory information. Okay. It's a very, very fancy way of putting it, you know, all these studies where like, I remember them, like, I remember reading Penny Moss's work years ago saying, Hey, you know, people with neo a have cold hyperalgesia and people with, you know low back pain, they're not as good at detecting postural sway and, and all of those things that suggest there's some kind of sensory disturbance that or, or, or difference or change the way the person represents their body.
Mervyn Travers:
And I remember 10 or so years ago, I said, well, hold on, they're cold hyperalgesic, is that causing, is it cold hyperalgesia causing their pain or is that just another symptom of this? If I, if I could make them non cove hyper, would that resolve their symptoms? Like what do I do with this? And I, and I vividly remember the conversation with, with one of my colleagues again was actually with Ben W and in his beautifully beer and simple way of just putting things like Murph can't, you see, you put all of those studies together and you kind of look at, there's this consistency that people in pain seem to be very, very good at detecting noxious information and not very good at detecting non noxious information. Penny drop moment for me, it's like, yeah, that's it. It's differential waiting. Okay. So prioritizing information from different sensory streams. Okay? So if you've got a system that's incredibly good at prioritizing noxious information, what do you think it's gonna use to confirm its models?
Mervyn Travers:
Okay, so that's where greatest sensory motor retraining is of enormous value as far as I'm concerned. 'cause It's about training. I think the person's attention to non noxious sensory information from the area. Also why I think when we go to the movement experimentation we spoke about before, and I get the exam example of the co activation and I said, Hey, if someone's coactivator, they're likely increasing load on the tissues and, and, and tissue ischemia and causing more nociception likely. The other thing they're doing is if you stiffen and you're not moving, you're depriving your system of competing non noxious inflammation. The sensory information that would come from your body part saying, I'm okay, I'm actually moving. This is normal. So you're shining that spotlight further on nociceptive inflammation. Again, I hope this makes it clear how much I think nociceptive information is relevant and important in the concept of context of pain.
Mervyn Travers:
So is the top down and I think they're inseparable. Okay. And so I think so I think you're saying about treatments in the future. I, I think the most promising treatment is greater sensory motor retraining right now. And I know the groups behind it have large grant funding to explore it in different, different centers and different body parts, et cetera. And different pain types. I think they're doing A-C-R-P-S one as well, I think. And so I think it's got enormous promise because it deals with treating, if you will, treating probably the wrong way. But understanding and trying to modify someone's internal model about the brokenness of their body and their tissue capacities and their, their, what their body can do. They're looking to train and retrain the differential weighting of sensory information and then learn that the body's capable of movement through training or to, or capable of, of, of recovery and movement through exercise, should I say like that is entirely compatible with active influence.
Mervyn Travers:
And it's the most, and it's, it's shown promise when control compared to a really credible shot. I mean, it, it's a single center trial, so you can criticize that. There's always criticism you can make, but it's incredibly rigorous. Like you mentioned the, the resolve or the restore trial, the CFT trial, there's unquestionably amazing things in that and amazing promise in it. But e equally it's not being compared to a credible sham, right? So it was, it's as a, a really important effect size. We don't know how much that effect size is residual bias and how much of it's a true treatment effect. So if you're asking me on a pure evidence where the future is, like they both need to be pursued further. The third treatment, I think that has enormous promise and again, would need to be done in a more rigorous way and compared to a, with a better control group and larger numbers, but its effect size was enormous, was the pain reprocessing therapy treatments.
Mervyn Travers:
And there's, I know people have been critical of them and, and, and I'm critical of the, of the, the rigor of the study, but the effect size is so large. This is the paper that was in JAMA maybe a year or two ago I, their effect size is so large, it warrants real future investigation. So they, they, this treatment called pain reprocessing therapy is a very top down kind of reframing and understanding of, of, of pain and trying to modify and, and your, your pain type of treatment. And it was compared to an open label placebo. So they, you know, there, there's, there's problems. There's only about 50 people in each group. There, there are problems with the design. Do not get me wrong. And, and I think trials that compare to usual care, that's a problem as well.
Mervyn Travers:
'Cause Usual care can also be labeled as denial of care. You know, so they're set up to show naturally a large effect in favor of the intervention. So that's why with the PRT trial pain for therapy, I'd love to see it done again against a truly credible sham as it stands. We have one treatment that has done that. And so if you're asking where the future is, the future's there for me or, or at least the present is there and developing treatments around that. And I'd love to see CFT more CFT work, particularly against credible sham. And I'd love to see PRT tested more rigorously alongside the pharmacological interventions and all the other things that are happening too. 'cause Developments in that space really matter.
Jared Powell:
Yeah, we've got a lot of work to do. That's I think it's quite exciting really. I think a lot of people are cynical and nihilistic and have a bit of a doom and gloom mentality, but I think the next 10 to 20 years are gonna be incredibly exciting in terms
Mervyn Travers:
Of, oh man, I totally agree. I, the landscape has changed. I mean, I think if you'd asked me five years, so when I started writing the paper I'm referring to that hopefully will submit this week my, this active inference and low back pain paper I've been working on since 2017. If you'd asked me then, what are the hot treatments? What are the ones that are showing real merit and, and, and, you know, long lasting effects on and disability for, I'm using back pain as an example. I said, we don't have any that, that, that's really hard to say. Do we have any rigorous good quality trials that show a meaningful effect of a let's we, let's say a physiotherapy kind of based or psychological based intervention. I'd say I would've said then we don't, in the last two, three years, we now have three that I think show enormous promise I'm putting resolve at the front of that list simply at the greatest sensory motor retraining simply because of the rigor of that study.
Mervyn Travers:
And the design of that study answers that question is, does it have an a true treatment effect? The design of the others doesn't quite do that, but they have enormous promise. And I, I, so I'm not in any way poo-pooing kind of CFT for example. I think it's amazing work and done by amazing people. But if, you know, the it, it ask research design asked a slightly different question. And likewise, pain reprocessing therapy, it's got some flaws in the design. So I'd really like that done again. But I can confidently say there's three, there's three treatments that I feel show promise. And so that's an exciting time to be a clinician and even more so an exciting time to be a patient.
Jared Powell:
Unfortunately, in the shoulder where I I live, it's it's not as optimistic. It seems to be just high load strength training and all motor control exercise and changing the way the scapula moves. So we've got a little bit of work to do, but I think we'll get there.
Mervyn Travers:
Yeah. But I think like pain isn't region specific, so agreed. Like as a clinician, like I, I've got case studies that I present at conferences and on courses and stuff of, of applying these approaches to knee pain and to achilles tendon pain. Mm-Hmm. The, you know, of course, like if you want funding for research, say that you're studying back pain, right? So then you roll it out in back pain first, you know, . But like why would a CFT approach or, you know, greatest sensory motor retraining approach be successful in the back, but not for knee pain? Not for shoulder pain. I like, I I, I don't think that this is gonna be in any way region specific.
Jared Powell:
A hundred percent agree. There's just my knowledge of the, in the research area is just that nobody's pursuing these things yet. And we're sort of, I mean, we're, we're still struggling. We've only ever done one placebo controlled trial of exercise in shoulder pain. So we're, we're currently involved in designing one at the moment. So that's gonna be fun to, to do. And then hopefully other people, not myself, will start to pick up the, the slack and start to do these, these trials as well.
Mervyn Travers:
Yeah, no, it'll be, I i it's so important that we do that. Mm. And I think in some ways, back pain and knee pain kind of get to lead the way a little bit simply because it's probably an easier condition to get funding for
Jared Powell:
Just the burden as well. Yeah,
Mervyn Travers:
Yeah, yeah. Well, that, that's why Yeah, yeah, yeah,
Jared Powell:
Yeah.
Mervyn Travers:
Costs so much. The government's more likely to give you the $5 million for back pain than it is for, you know, a thumb pain or whatever. .
Jared Powell:
Yeah. Sorry to all the thumb people out there with chronic pain. No, it,
Mervyn Travers:
Yeah, we're not. No, but it's, it's, it's, it's the it's the societal burden.
Jared Powell:
I know what you're saying. Yeah. Key resources. Merv, there's a lot out there. Where, where should people start? Is there a good entry point? We've discussed Andy Clark and he's got a new book out. Where, where should people start?
Mervyn Travers:
Yeah, look, I think I, before delving into Active inference, I would probably, it's probably an easier way to dip your toe in with predictive processing, I would dare say. And so there are some really well-written books. Jacob Howie's book, the Predictive Mind, Anna Set's book being you Andy Clark's book surfing Uncertainty are probably really good ones to start in that space if you are. Then gonna go from there. There Michael Koff and Julian Verstein have a wonderful book. It's, it's a little bit expensive and it is quite technical, but it's a wonderful book. If you've read the other ones, then it's Predictive Processing The Third Wave. That's a wonderful resource. Really, really wonderful. I, I, I could have read that about 10 times I reckon, just, just to try and get my head around it. I reckon you'd probably figured out much faster than me, Jared.
Mervyn Travers:
The, then from there, Andy Clark's book, the Experience Machine, is a wonderful one. Mark Alms book, the Hidden Spring, that's really excellent. And then there's the Active Inference textbook, which is written by Paolo Parr and Friston. I do feel with that, I, I do feel this to get our head around it, there's an evolution. It helps to start reading predictive processing first. I, I, I do think so before going to Active Inference, but maybe other people might have more familiarity with it than I had at the time and, and, and, and, and, and could launch straight in there. That book's only about a year old, so it didn't exist at the time. Or maybe it's two years old, but that it's rather new. In terms of other resources and things that are out there look, if, if I may be some bo so bold like I, I fly around the world teaching clinicians how to merge paint science and exercise.
Mervyn Travers:
And, and the stuff that I do is in my courses, my pain on Stuck course is heavily influenced by this 'cause It's really important that it's not about telling and explaining to patients about predictive notions and, and precision weighting. It's not, I, none of that narrative comes across to my patients. It's a, it's a framework that's influence and determine what we do, rather than us pain explaining to people about this stuff. I think it's 'cause 'cause 'cause what we're saying really is like, you are an active en, you are an active agent in the generation of your pain, and it, it's caused by your predictions as much as by what's happening in your tissues. Now, could you imagine saying that to a patient, you are causing your pain because you're predicting it. 'cause 'cause Technically on a neuronal level, that's what's happening not on a conscious level. Right? And so you have like, I don't, I like, I so could not be more strongly in my statement of like, don't express sentiments like that to patients because it's not really true and it's, you know, likely very unhelpful. Pain education for me is experiential through both our physical examination process and through our, our interventions. It's not, it's not a talking point for most people, I think. Yeah. But the,
Jared Powell:
The quote that sort of comes to mind there is information is to behavior as spaghetti is to a brick, it doesn't change a thing.
Mervyn Travers:
Yeah. A hundred percent. A hundred percent. And I, I do reflect on a patient. I remember I just finished my postgrad and I'd learned all of this wonderful painting neuroscience from, from, from max Usman Curtin, who Max is no longer with us, unfortunately, but, but a, a wonderful educator in that space. And I remember I had this patient who came into me and she was like chief of anesthetics at, at a local hospital chief, an anesthesiologist or whatever it was at, at a local hospital where I was working at the time. And and I thought to myself, here we go, I'm gonna talk to her about her pain and that's gonna just fix her pain. Like this is a long time ago, right? But we've all, I don't know, I certainly fell into that trap at the time. And like, I had this long discussion where she and I were talking about nitric oxide feedback loops and glutamate and blah blah.
Mervyn Travers:
We were talking about all of this, right? All, all the kind of heavy neurophysiology and like, after about 20 minutes, like, mev, this is really interesting, but what has this got to do with my pain? And I remember thinking to myself, man, if like, if she doesn't draw the link between the kind of neurobiology and her pain, given what she does and her knowledge on this, I need to change what I'm doing here a little bit. You know, it, it, it really was. And so, you know, for me, one of the things that I really like is, is shifting. And again, it's about, you know, changing people's models. But you know, in university we're so often thought to try and use our physical exam to detect pathology with a bunch of tests with very low sensitivity and specificity I might add. And so of course we need to make sure the person doesn't have anything sinister. And if we do suspect a serious pathology, we should kind of act on it. But we should recognize that our physical exam are really well handled, really well communicated, and really well conducted and thorough physical exam can be an opportunity to demonstrate that the system is better than I thought it would be. Yeah. Right? Like, you do a neuro exam and it's clear. That's awesome. I was just checking that all the nerves were working. That's the really sinister stuff we don't wanna happen. You're completely clear on that.
Mervyn Travers:
Like that's, that's how you do it. Not like, okay, okay, okay and moving on. Mm-Hmm. We so often get into the routine. For me that's pain education. Mm-Hmm. Right? You're clear of that. Mm-Hmm . You know, you don't have any of that. You don't have that sinister sign that's completely normal. I really like that. That's stronger than I thought it would be. That's, you know, all of those things influence the person's model there and then, and I think are probably more powerful. Look, you never have a control and how do you know? But for me, with my biases, I think that's more as, as powerful as, you know, telling someone about, you know, G-Protein couple receptors. Mm-Hmm.
Jared Powell:
Yeah.
Mervyn Travers:
Neuros
Jared Powell:
Mer I'll let you in on a little secret. My physical exam is my treatment. Yeah. Almost all the time. 'cause I just see patients on telehealth now. So this, I'm like, show me what you can do. Move your arm around, let's experiment with it. And then, okay, let's put this into some sort of program.
Mervyn Travers:
That's it. Well, that's it. It's like, it's funny, you know, I know like if you watch the movies, Lord of the Rings, and they're having that big siege outside the castle and one of those big orc guys comes along and he blows a hole in the castle and they file all the troops in then. Right. It's just like that. Mm-Hmm. We are experimenting. You found something that was positively surprising. Mm-Hmm. I'm gonna funnel all my troops down there and that's gonna become your program. Mm-Hmm . Right? And so I agree, and it's really interesting, like if, if if I, if I can give you an example, right? I had a, a patient that I often presented at, at conferences and, and courses and that kind of stuff with really, really severe and debilitating long history of post-surgical knee pain. And before he came to see me, he, about eight weeks before he came to Perth to see me, he had had a, a scope on this knee that he'd had repaired previously.
Mervyn Travers:
And the scope showed that like it was, his tissues were all, his, his femoral condos were all degenerative. And he'd loose flaps of cartilage here and there. And he woke up from the surgery and the surgeon basically flashes these awful images of his knee and tells him he's gonna need a knee replacement by the time he's 30. That's pretty much it. And he comes to see me and goes, I just don't wanna have that knee replacement. You know, I'm 25, I don't wanna be having knee replacement when I'm 30. I've got young kids like, you know, what can we do? And during my physical exam, he's really nervous when I'm going to go and do some of the orthopedic tests of the knee. Now I've got the inside line, but I know that his physical ex, I know that his ACL is intact because he's had a scope only a few weeks ago.
Mervyn Travers:
Right. But I'm like, I'm gonna do this physical exam anyway. And I said, well, would you feel less nervous if I set you up? And he goes, yeah. And if you could see your knee, yeah, that would be really helpful. What if I do everything on the opposite knee first so you know what's coming? Yeah, that would be great. What if I start off really gently and like only build up the pressure as we feel more confident? Yeah, definitely. That would make me feel better about this. I'm like, isn't that how we taught like in first year uni how to do an exam when I'm communicating that and making that pop clear and explicit to him that that's how we're gonna do it. So then anyway, I'm doing the orthopedic test of the knee and no, it's no surprise they're coming up with very little right.
Mervyn Travers:
And he's and I'm, but I'm, I'm pulling gently and I go to his, his effect side. I'm pulling gently and I go to the hole really hard and like, you can't see me, but at the time I was about, it was about 109 kilos worth of Merv pulling on this guy's leg. So he's, you know, he probably feels like he's been mulled by a bear. But the key thing is I'm saying to him, Hey, look, can you, can you feel that stopping point? Sit up there. Can you see that stopping point? And can you see how it's stopping over here? That's normal. That's the same. Did that for every ligament, every meniscal test that I could even think of to show they were the same. And he sits on the edge of the bed and I finished my physical exam and I remember the time I was like, I was talking away to him like this as I almost always am.
Mervyn Travers:
And I realized old mate's not paying attention to me at all. He's looking down at his knee straightening and bending it, turning it, looking around at his knee. And I just, but I'm gonna sit here for a moment and let him update his model on his knee that his knee. And he, and he goes, I, I, he goes, I thought my knee was gonna fall apart in your hands. I said, mate, that knee is stable and strong. It's much better. I thought it would be. And he's like, yeah, it is right. Now you tell me this. Was that pain education? Was that treatment or was that assessment? I don't know.
Jared Powell:
Yeah.
Mervyn Travers:
You know, and so, so this is the thing, you know, it, it's, it's, it's, it's not assess somebody talking about their pain, then deliver some treatment. That's not how you deal with like the top down and bottom up components of, of pain. And so an active inference view understands that like your approach by deeply understanding the person and understanding and, and trying to create positive surprisable. Mm-Hmm.
Jared Powell:
Beautiful. I love it. Merv, I think that's a really solid, beautiful place to, to wrap up. I think we could keep talking for hours and hours and hours, but I, I think people are under their third cup of coffee now. And I want to let them go and, and look up all these books that you've recommended because I, I highly encourage people to pursue this line of inquiry and thinking. And more specifically though, where can people find more about you? What are your socials and what's your website?
Mervyn Travers:
Yeah, so I'm, I'm, I'm optimized rehab. That's me. I'm, I'm in that very forced position where I do get to fly around the world and, and, and, and work with different clinicians and on, on how to implement these things. So I, I I, I have some courses on strength and conditioning. And, but in particular my pain on Stuck course is about you know, implementing treatments based on this, on this framework, based on the evidence and really optimizing exercise rehab outcomes for people. So I'm optimized rehab, I'm on Facebook, I'm on Instagram. I've got my website. You can check out course dates and that kind of stuff. Got a, a a ton of, had a paper published last week there in European Journal of Pain where we looked at the influence of, of words and information on, on people's experience of pain. And we've got a whole series of, of papers coming out in the future for that. So keep an eye out for, for those. And once that active influence back pain paper gets over the line, jarred, I'll be sure to to send it your way.
Jared Powell:
Yes. And I will force you to come back on the show and talk about it as well, especially the, the paper that we chatted about before as well. I really want to get my teeth stuck into that too. But in the meantime, good luck to the the Irish rugby team and good luck to you , and I'll chat to you soon.
Mervyn Travers:
Yeah, mate, thanks so much for having me and it's a real privilege and well done on everything you do.
Jared Powell:
Thanks mate. Cheers.
