The Neuroscience of Flow, Reconsidered: The Brain Has No Switch

There's no flow neuron. There's no flow brain region. Instead, flow, this really complex phenomenological, psychological, cognitive experience, requires this coordination between all these different parts of the brain. But here's where it confuses me a little bit. The people that experience the most flow actually have lower entropy scores for their brain responses than people that experience less flow. And that one tripped me up a little bit and I don't have a conclusive answer for what's going on there.

Hey everyone, welcome to Flow Radio. I am Dr. Michael Menino. I'm the Chief Science Officer of Flow Research Collective and I'm joined by my co-host Steven Kotler, Executive Director of the Flow Research Collective, bestselling author. And our guest today is Dr. Richard Huskey, an Associate Professor in the Department of Communication and the Cognitive Science Program at University of California, Davis. He's our— he's the principal investigator of the Cognitive Communication Science Lab and is affiliated with UC Davis Center for Mind and Brain and the Center for Neuroengineering and Medicine. He's also an associate editor at the Journal of Communication. So Richard's work sits at the— at a fascinating intersection: communication, cognitive neuroscience, media psychology, computational modeling, and human motivation. At the heart of his research are some very human questions. Why do we remain motivated when something becomes difficult? Why do certain experiences capture our attention so completely? And how do reward, learning, and cognitive control shape the decisions we make? To investigate these questions, Richard and his lab combine behavioral experiments with fMRI, EEG, reinforcement learning models, and AI-driven predictive approaches. Their work examines how media shape attention, beliefs, choices, and well-being, and why certain stories, games, and digital experiences become immersive, emotion— emotionally powerful, and deeply rewarding. A major part of Richard's research is focused on flow, the state of deep absorption in which attention becomes focused, self-consciousness recedes, time changes, and action can almost feel effortless. His research has helped move the neuroscience of flow beyond the search for a single brain region and toward a more dynamic picture, which we'll get into and discuss, one involving flexible, distributed, and constantly reorganizing brain networks. Richard has also been a valued collaborator of ours at the Flow Research Collective. Together with Steven Kotler, Scott Kelso, and myself, we actually co-authored The First Few Seconds for Flow, which proposed a neurobiological account of how the brain transitions into flow. More recently, we collaborated on a complexity science framework for understanding flow as an emergent relationship between brain dynamics and subjective experience. So today we're gonna step back from the technical language and explore the larger questions. What is flow really at the level of the brain? Where does the current science tell us— what does the current science tell us, I should say, and what does it not yet tell us? And what happens in the brain when we enter flow? And where the field is headed next? So Richard, welcome. It's great to have you with us. Thank you so much for joining us. And I am going to turn the first question over to Steven Kotler.

Richard, thank you so much for being here. And I realized Michael and I were talking about this. Somehow we've been working together a long time and I don't know what got you into flow research. Like, how did you get drawn to flow research? And I guess as a part 2, you've really sort of bridged the gap between media studies and flow research. So why the approach you take?

Yeah, well, first off, I just want to say thanks for being here. It's super cool. to get the chance to talk with y'all, and I like what you're doing on the podcast. Um, so do you want the, the honest answer or the answer that makes me sound as impressive as Michael's intro? Um, you know.

Well, let's go with both. I'm a gold digger, I'm sorry.

All right, so I'll start with the honest answer, and then if, you know, this is not— you know, this is tanking your podcast, um, then we— you can tell me to sound Better. I was a first-year grad student and I was at UC Santa Barbara. And at the time I really was interested in how all these people would come together online to do these big collaborative things. Everything from like, why do people come together to edit Wikipedia pages? To like, how do people come together in video games like World of Warcraft? 40 strangers come together and play for like 40 hours a week to figure out how to solve some video game thing. And I thought that was super interesting, how across all these different things people collaborate online. And so my advisor took me to lunch and he asked me to tell him about my research interests and what I wanted to study. And as I tried to explain these things, his face got into a deeper and deeper frown. And it almost looked like I was physically hurting him with my words. And at some point he's like, all right, Richard, enough of that. While you figure out how to do something interesting, I've got a problem for you. We've got this study on flow. You know what it is. And I said, sure, yeah, of course. And I'd read about flow. I'd read Csikszentmihalyi's work. I'd seen it all around me and in my personal life. And all right, like, let's try that. He's got something ready for me to go. And that was how I actually got started doing flow research. It was in some ways a total accident based on not having, frankly, a better idea, which is a little embarrassing to say, but, you know, here we are. And that was really the start. And so it was an initial set of studies that were trying to understand one small part of flow, which was this question of— Flow state. Does one of the things that for 50 years we thought causes flow actually cause flow or not? And that's this challenge-skill balance. One of the key causes of flow has been the challenge of the thing you're doing needs to be really, really high, and so does your skill at that thing. And only when— well, not only when those 2 conditions are met, but those 2 conditions are critical for the emergence of flow. And that's what this series of studies was designed to test, is this actually true or not? And we found some promising evidence suggesting that it was, and that was really the thing that got me going.

It's funny because they're both sort of the same question, right? How do these groups collaborate? And you're actually asking the same question from 2 different directions, but it probably took a while to figure that out.

Yeah, I mean, so the funny thing is, you know, I know you folks have been working a lot on group flow, and that's an area where I haven't dug into too much because I've already sort of decided that trying to understand flow within one person is already complicated enough, rather than bringing a bunch of other people in. But there's all sorts of coordination components, even internally within one person, that seem really critical for flow. And so that's— been kind of a fun thing to dig into over the last— gosh, it's weird to say, but last 15 years now. So

yeah. That's interesting. So like in the— does the challenge-skills balance play a role in group flow? Are the group flow triggers different than the individual flow triggers? And Steven, we can talk about this, but on the challenge-skills balance issue, so some of recent work suggest that flow can sometimes occur during relatively easy tasks. Like, we can just— I want to discuss Melnikov's paper on mutual information and his computational model of flow. Do you think that overturns the challenge-skill model or does it suggest that there's maybe multiple pathways into flow?

Yeah. Okay. So, I want to just start by saying I really admire what David Meldikoff and his colleagues are doing. I think it's one of the coolest breakthroughs to happen in flow in a long time. And it was one of those papers, you know, the big nerd I am, I was at an academic research conference, a neuroscience conference, and I got a notification that this paper got published and I left the conference to go read it because it felt like that important in that moment. So nerd alert, I realize. I'm saying all that to say I think this is— what they're doing is really important stuff. Am I convinced that it overturns 50 years of research? Not yet, right? Like, 2 manuscripts with about 9 studies is really a fantastic start, right? But I'm not so sure— you know, it takes a little bit more than that for me to be convinced that we need to throw everything out. And so, you know, I don't know if your readers are as big a nerd or your listeners are as big a nerds as I am. So like what's the idea that Melnikoff and colleagues have? Basically, they think that flow is an information theoretic problem. Very specifically, there's this idea called mutual information. And basically, oversimplified a lot, mutual information is how much does observing something in my current environment reduce uncertainty about a future outcome. Okay, and you could imagine, like they use an example of slot machines. So, you know, before I pull the handle, I have tremendous uncertainty about the future outcome, but as soon as the wheel stops spinning, I have tremendous certainty about that outcome. Right. Right, there's a pretty good relationship between when the wheel stops spinning and information. Now imagine— and like the reward I get, whether I lost my money or not. Now imagine the same slot machine, every time I pull the handle, I have a lot of uncertainty, but when the wheel stops spinning, the reward is also uncertain. Sometimes, you know, getting all the bells gets you a big jackpot, and other times you get nothing, right? There's not a good relationship between the initial state and its ability to reduce my uncertainty about an outcome state. Melnikoff and colleagues think this relationship between observing a current state and its ability to reduce uncertainty about a future state is really critical to flow. They think that relationship needs to be really, really strong. So in instances where I, you know, get information about my environment and that gives me strong information about My future outcomes, they think that's going to be a flow eliciting experience. And I think that's pretty freaking cool. They've got a new paper that's currently in a preprint. It's not been peer-reviewed yet, that shows that under certain circumstances, things that look like the challenge-skill balance for flow do elicit flow within this mutual information framework. But in other cases, Not so much. Actually, low difficulty tasks can also elicit flow. I am a little unsure. Like, I think their experiments are really clever and are designed to kind of, you know, really showcase this result, but I am having a hard time understanding how their really carefully designed but narrow experiments extend into all the different rich ways that we experience flow. You know, like, they are using target shooting games and like tile and memory type games. But like, that's pretty different from being on the mountain snowboarding or improvising jazz or, you know, composing music or playing an instrument or all the different places that we find flow in our everyday lives. So that's where I'm still a little uncertain, right? But I will say I think it's one of the coolest developments that we've seen in flow in a long time.

One of the things that I was listening to you talk, which is a really common experience on the ski mountain, on a snowboarder, is you're in a situation where the challenge-skills balance is you're up against it, right? You're skiing really hard stuff, you're snowboarding really hard stuff, and it's not flowy. You're actually scared enough that you're not— it's not flowy. And then you back off and you go ride some intermediate run and you can be really expressive because suddenly— And suddenly that will drop you into flow. And that happens all the time on the ski hill. And so one of the things I wonder when I read his work is prior states seem to have a big influence on present state. So how steep was the gradient, the challenge-skills gradient, you know, 10 minutes before you went and did this thing? There's probably some cortisol profile or norepinephrine profile that, you know, underpins some of this. I'm just freestyling here, but that's what I was thinking when I was listening to you talk.

Yeah, I mean, so that's an interesting point. And my gut is there is some anchoring effect, right? So that's often what scientists use, or they use this word anchoring when what happened previously influences how I think about something currently, right? And we haven't done— this has been one of those studies I've been meaning to do a little bit more carefully. So in the lab, when we test our various experiments, we're usually, you know, randomly assigning— we're good scientists, so we're randomly assigning everybody to every condition. We're randomly assigning people to all the conditions, so we get a lot of statistical power. And one of the things that we sometimes see is this like ordering effect. So if you get a non-flow condition, particularly like a condition that's designed to elicit frustration, that flow condition can amplify that.

Richard, just to back up a second. Your lab is usually using video games, right? So you're having people play video games. So you, you can adjust the challenge skills level. You can make it frustrating for them. You can make it easier. Okay.

Yep. Yeah. Yeah. So we'll bring people in the lab. We've got a variety of different video games that we use, including some ones we've custom made, uh, to rules at Flow. And, you know, in the, in the game, we'll have a, usually we'll have a boring condition where the game is super easy. Everybody's already trained on how to play it. They're doing really, really well, right? Their skill dramatically exceeds the challenge of the game. Mm-hmm. Other times we will have a frustration condition where the game is really, really hard. Even though people have practiced and trained, it is just too difficult for them to succeed at and they are experiencing frustration. The challenge is much higher than people's skill level. Right. And then we will have a flow condition where a little algorithm in the game dramatically or dynamically updates the difficulty of the game depending on how well people are doing. And that's designed to constantly keep this challenge-skill balance mapped. And so what we find when we do these sort of studies is if I put you in the frustration condition first and that's what you've experienced and then I put you into the flow condition, that flow condition tends to be a little bit more amplified. Mm-hmm. Than if you start in the flow condition And then you go to the frustration condition, or if you start in the boring condition and then you go to the flow condition, right? And so like, I think there's something there. I just, I haven't— it's one of those things that for a while now I've wanted to really dig into and I just haven't had the chance to do it. But yeah, I absolutely think there's an anchoring effect that's happening there. And like, we see these kind of anchoring effects all the time across the cognitive sciences, like When people are doing tasks that require performance monitoring, which is like certainly flow. You know, you talked about snowboarding or skiing. You know, if I'm going down the mountain and I mess up a turn or I fall or something like that, I'm, you know, that's an error monitoring experience right there. And one of the things that we see in cognitive science is that when people discover that they've made an error, they often slow down. There's like post-error slowing, right? Mm-hm. And so I've slowed, you know, I detected that I made an error, I slowed down a little bit, I tried to be I try to not make that same mistake in the future. I think we see that not only in the lab, we see it of course in our everyday lives, we see it in flow experiences. And that's a different type of anchoring as well.

In general, for flow science methodology in general, so you use video games and we just talked about skiing, different kinds of— all of that is built around the challenge-skills balance, like your entire experimental protocols, the frustration, boredom, and flow, 3 conditions that you're putting people in, are all based around that one flow trigger.

Yeah.

Do you think that there are others missing, that video games and certain protocols like that will fail to capture other ways of getting into flow?

Yeah.

There was a recent study, I think it was last year, Joydeep Baciarata, he had that study about neurophysiological conditions for experimental flow states and they were looking at dynamic ways with having different kinds of goals, autonomy, dynamic ways to adjust dynamically the challenge-skills balance and not personalize it, not just make it a general 3 thing for all of these subjects.

Yeah.

Which I think is really cool.

Yeah.

Can you speak to that a little bit?

Yeah, sure. All right, so I'm not familiar with that particular paper. Now I've got to read it, and I feel a little embarrassed for not having seen it. But if I'm going to come down on myself a little bit, I'm one of those ugly reductionists, okay? I'm an experimentalist by heart. I'm an experimentalist by training. And so the thing that I always want to do is get the tightest control over the world and be able to isolate one cause, right? This is what doctors do when they're trying to develop drugs or develop treatments, right? This is what I'm trying to do when I'm trying to understand flow. What are the very specific causes of flow, and do they map to what we know from the research? And so to do that, you're usually trying to hold everything constant that you can and then manipulate one little thing, right? Right. You know, and that's what I'm usually trained to do in the lab. That approach is kind of in tension with flow, right? Because flow is this rich phenomenological experience that has a lot of different experiential components to it. And by reducing it and trying to get it to something that you can control in the lab, you lose some of that richness. Okay, and so like that, that's just a trade-off I have to make. Do I love that trade-off? Not necessarily, right? I— why I use video games is because a lot of the other more reductive lab-based flow experiments were doing stuff that I just, I thought, I don't know, I don't know if that can really be flow.

You and I are in agreement. Those math equations never put anybody into flow. Right?

Yeah, so like there is one team and I really like their work a lot, but their flow eliciting task is doing math equations. And it is a German team and it is on German research participants and so maybe Their, you know, German heritage makes them love doing math. I'm not so sure. But like, at least for my American college students, doing math is usually not a flow experience, right? And so, yeah, so we've been trying to figure out like, how do we do something that is a little bit more realistic? And, you know, scientists call it face validity, but I'll just call it believability. Like that. Right. Believable than doing math. Looks not as realistic and real-world as composing music, playing music, skiing down the mountain, but gets you kind of close. And like, if you go to a video game conference, they talk all the time about designing games for flow. If you go to, you know, so players report experiencing flow While playing video games. And so that's, that's where we're— how we settled on video games in a lot of ways. What can I get a lot of control over, but then what also is a little bit more believable as a flow listening experience? I will say, and you know, we don't have to touch on this right now, but one of the things that is kind of cool is if you look across different tasks. So there have been people that have had people improvise music to elicit flow, play video games, solve math problems. There was a recent paper that had people do calligraphy while they were in the scanner. And we see— Sweet. Yeah. And so we see different brain responses across some of those studies, but then we also see some really common brain responses. And so what that tells me is there's something that is task-specific, and then there's maybe something that's common to flow across all of those tasks. Yeah. And so that becomes kind of the interesting challenge to start to tease apart. Like, if I see a weird brain response in one study that I didn't see in my own, is that because I messed my study up, or is that brain response a task-specific brain response? Those people were doing calligraphy. Well, I had people playing video games, but are there also common things between our 2 brain responses or not?

But that makes sense from an anecdotal point of view because intrinsic motivators are flow triggers, but different things can motivate us. Yeah, exactly. So that would make sense. Yeah. So before we get into the brain and some of our work and some of your recent work, Wanted to ask you a question. So we recently published an edited volume called Performance Neuroscience, a textbook. Yeah.

Just came out actually like a day or two ago, right? Yeah, yeah.

Very recently with Steven and Marcello Bigliassi, our psychophysiologist neuroscientist colleague at FIU, Florida International University. And in that Textbook, you have a chapter, one of the best chapters.

Oh, give me a break.

But thank you. Which is on flow and performance. So we wanted to ask you a little bit about talking about the relationship between flow and performance. How correlated are they? Can someone perform exceptionally without being in flow and feel deeply in flow without actually performing better. So talk a little bit about your research with your student, amazing, your amazing PhD student Rachel Key, right?

Yeah, yeah, Rachel. Yeah, yeah, so we're, we're both really interested in this question, and this was something that she really ran with. So I just want to say any good idea in that chapter came from her, by the way. But, you know, we talked a lot about this. And I think one of the— if you read some, you know, blog posts or articles about flow, there's often this assumption that flow is always peak performance all the time, right? And like, I think flow and peak performance are definitely related, okay? But what I think one of the questions is, what do we mean when we're talking about performance a little bit more generally? Think about a basket— 2 basketball teams, right? Any given moment, one team scores, another team scores, one team's up, one team's down. You know, there's these moments where there's a dip in your performance, right? Your defense suffers, the other team scores, but you rebound, you get, you know, the next basket, you're up again, right? Both teams are playing at a really high level. They're probably experiencing flow most of the time. It's not like when the basket happens against them, like all of a sudden, you know, you can tell by the way I'm talking about basketball. I'm a total fan. Uh, but, um, it's not like as soon as you get scored on you're knocked out of flow, right? It's not necessarily if you're on the mountain snowboarding or skiing and you mess up one turn you kicked out of flow, right? You might maybe have a heightened sensitivity to your mistakes, right? Like we actually kind of talk about in our paper together, we talk about error monitoring as this really important component Of either eliciting flow or what's happening while you're in flow, right? So I don't necessarily think flow is synonymous with peak performance. And here's another intuition for that. If I, you know, I'm pretty good with numbers, so if you ask me to count to 1 to 10, I'm gonna have really great performance on that, right? But that's not interesting. Right. That's not a moment where I actually care. Like, it's not going to elicit flow. Me counting 1 to 10 is not a flow-eliciting experience. It's also not a terribly interesting high-performance experience, right? Because it's trivial. It's trivial challenge. It's trivial moment where I don't really care. I think characteristics of flow that matter is one, they can actually help us pursue peak performance, right? Like when you are struggling to You know, hit a more difficult jump, something with a little more distance on the mountain, if you're trying to get into that better yoga pose, if you're trying to get, you know, down a really difficult piece of music, you gotta practice. You know, that just doesn't happen. Well, for most people, it doesn't happen naturally, right? And there's all these moments of what can often be really frustrating getting up to that, like, perfect— Point. moment where you nail it, right? And I actually think flow is the thing that helps us get there. It's the thing that helps us push past the difficult experiences and frustration. It's the thing that actually helps us achieve some of those higher peak performance moments. It helps us push past the parts that are difficult. It helps make it motivationally relevant and interesting to get to.

You know, when you were talking, I recently We've been writing about group flow. Yeah. So I went back and I reviewed Bill Russell's famous quotes from Second Wind, which I haven't looked at. So Bill Russell was a basketball player. His biography was Second Wind, and he wrote one of the most famous passages about group flow ever. But I went back and reread it. I mean, I quoted it, I think, in West of Jesus, my first book on flow, and it's actually like 3 pages long. I remember it as a paragraph, but it's like 3 pages long. And one of the things he says, as I was listening to you talk, is He would— he felt it was all the players on the court would get him to flow together, and he found himself cheering for the other team while he was in flow because he, like, everything was so perfect. He wanted their shots to go in too because he felt he needed them to compete at their best for him to be at his best. Yeah. And it was really interesting, like, in reading it again 20 years later, I was like, Oh wow, there's actually a lot more going on in this basketball story than I first noticed.

Yeah, I mean, yeah, absolutely, right? Like, so if we think this challenge— so like, okay, if we are thinking about what are some of the 3 causes that are pretty well documented for flow, right? We're talking about challenge-skill balance, which we've already discussed. Immediate feedback on how you're doing, right? Clear goals. Clear goals. And I think you can extend those very clearly into, into group flow experiences, right? Am I— do I have a clear goal? Is it basketball? What do I need? What are my goals? Am I in a band playing music together? What are the goals of the song we're about to play, right? Like, how do we do this well? Immediate feedback. Oh, the other team just scored. Oh, you know, my bassist hit the wrong note, or the drummer's out of tempo, or something like that, right? I'm getting immediate feedback about how this is all going. Yeah, all of those things, you know, you can think about that internally as just monitoring your own goals and how well you're pursuing them and instances where you're messing up. But you can also, of course, monitor those in group situations as well. Those extend not only to people on your team, but of course people that you're competing or playing against.

Yeah.

This is a weird thought, but it came out of a recent conversation. I was talking to my friend Sekou Anders, who's a performance poet. He's a spoken word artist, 2-time national spoken word artist, slam poet. And he's been in individual flow Right? Done doing slam poetry, hip-hop, whatever. And he's been in Group Flow.

Yeah.

So here's where this story gets interesting. He does corporate poetry performances. So he'll go and like write a poem for Nike about like their stuff and present it. And then he'll sort of give a workshop using improv techniques and whatever. But he was talking about how he likes to focus, like he really likes to write the poem the day before. He'll get general ideas, but he'll wait Till like the challenge goes up, right? It's the day before, there's pressure, and then I'll write it. And he was talking about how much he thinks about the client, and I was like, Sekou, tell me something. You've had individual flow experiences and you've had group flow experience. When you're focused on the client's need and writing a poem for the client, you're talking about hearing their voice in your head. Is that group flow or solo flow? What does it feel like more to you? And he said it feels more like group flow. Now, total anecdotal, but it made me start to wonder, are we looking at a continuum based on really internal things, less— maybe not less external? It really sort of like— it's been in my head for like 2 weeks now. I haven't even told Michael about it, but I was like, oh, wait a minute. This is really interesting and weird.

Yeah. Yeah, that's super cool. Hmm. Yeah, I haven't seen anything that has tried to study that systematically, right? But that makes sense, right? These are not binary things. You're not only paying attention to your internal state or what is happening with you internally or externally. You're of course paying attention to the world around you. And like, just like we can dial up, you know, different parts of our stereo to make it sound better or worse, we can dial up or down how much we're paying attention to those pieces of information, right? And from our external world, then we can probably dial up or down those pieces of information based on how informative they are, right? Mm-hmm. Is this piece of information about somebody else giving me immediate feedback about what's happening or not? If it is, probably something we ought to dial up a little bit, right? That might actually help us achieve flow in group settings a little bit better. If, you know, and this is all speculative, right? But like, I think it would be pretty cool to look into. Like, we talk about in our paper together, the 3 of us talk about how people selectively attend to information in their environment during flow experiences. And I think this is exactly that case, right? In our paper, we are thinking about it within an individual and how they selectively attend to information very specifically within their environment, within individual flow experience, right? But you can surely extend that into group flow experiences as well.

So let's get into that a little bit, our research together, right? That was a good segue. Yeah,

okay.

Let's talk about the neurodynamics of flow in individual brains.

Yeah.

So Stephen and I started thinking about the— historically, I think Stephen and I started thinking about this simultaneously right around the same time you were thinking about this. You've published some papers on network neuroscience and how the brain reorganizes itself for flow rather than just one part of the brain lighting up or shutting down or something like that. And then we published, we got together and we published this 2022 paper paper for how the brain transitions into flow where we sort of first asked the question how the brain does reorganize itself during flow. Uh-huh. And then more recently, we published this paper simultaneously with a paper that you came out with too, looking at this concept of metastability among other different ways, metrics of looking at how flow emerges in the brain. So I want to talk about— we want to talk about those papers now and those results and how those results speak to each other, how they advance the field of flow itself, how they might complicate or bring a more nuanced picture to flow neuroscience and what that looks like. So let's start off with our paper that we published almost the same day, I think, or same week. Yeah. You released this paper also on metastability. The title of the paper is called Neural Symphony, I think.

Yeah.

Okay. So, let's talk about our paper first.

Yeah.

And what we did.

So, where do you want to start?

By the way, I'm your co-author and I'm not sure exactly which of our papers you're talking

about. Sorry. So, we published this paper I forget the title of— now it's not in front of me.

Complexity Science Framework for Flow.

Thank you. Complexity Science Framework for Flow, looking at brain-body dynamics and neurophenomenology. So complexity science means— it was more of a methodological thing. We just took these different metrics that have not traditionally been used to look at flow in the brain and use them to look at some of your data, neural data, and what did those metrics show us? And they come from complexity science.

Yeah, yeah. So I think, I think if you want, I can jump right into what we found, but it might be useful first to kind of like talk about complexity science a little bit. One of the key ideas that we're looking at, which is metastability. So like, where do you always start?

You always start by defining metastability and then—

yeah.

Go a little bit into the framework.

Okay. Yeah.

Let me actually back up one second for the— like, for somebody who's listening, when we talk about brain dynamics and everything we're talking about here, we're asking a really simple question, which is what happens in the brain over time? So what happens in Richard's brain between 10 AM this morning and 10:05 this morning? That's a neurodynamical So, one of the things that complexity science is focusing on is what's happening in the brain over time. Over to you, Richard.

Okay, cool. Yeah, so let's see. So, I'm going to back up and I'm going to bore everybody with a little bit of history of the brain and a little bit of philosophy, and I'll try to make both of those like really fast and not too boring. Okay. Early on in the brain—

We need a flashing nerd alert. We do.

Yeah, like, yeah.

Danger nerd alert, right?

Yeah, but I think it's important, like, I think this is important to, like, contextualize why, like, why we do what we do, right? And so, like, there were some early philosophical ideas that were called, like, modular theories of the mind, right? Every part of the brain was responsible for very specific cognitive processes. So there's a part of your brain involved in attention, there's a part of your brain involved in memory, there's a part of your brain involved in motor responses, these sort of things. That's definitely true, right? But this modular view of the mind struggled to scale up, right? I can understand attention really well by looking at parts of the brain involved in attention, But how do I deal with something a lot more complicated, which requires attention, requires memory, it requires motor responses, it requires planning and error monitoring and all these different things, right? Like, how— like, there's no module in the brain that does that, right? Uh, there's no flow neuron, there's no flow brain region, right? Like, instead, flow, this really complex Phenomenological, psychological experience requires this coordination between all these different parts of the brain to get pulled off, right? Like, if I put you in the scanner and I'm like, I want to see the part of the brain that lights up when you're in flow, and that's the flow region, like, that's not principled science, right? Like, we need to think about how these are— these regions are interacting. So this modular view of the brain doesn't extend very well to Things like flow. Okay. And so that, that's the brief philosophy and history of the brain is now over. Okay. I hope that wasn't too bad. So what metastability does in these complexity science frameworks is try to understand how do brain regions interact dynamically over time, right? And so there's a bunch of really complicated math that you have to do to figure that out. And for the paper that the 3 of us were on, there's an excellent postdoc, Fran Hancock, that really took the lead on that, and I would just be remiss if I didn't mention her.

Oh, Fran killed it.

Yeah, so fantastic stuff. And, you know, she's a statistical physicist, and I'm just like in awe, frankly, of what she can do.

Which is another way of saying, in conversation with Fran, it's likely she can go off the rails and into another language that nobody in the room speaks besides her very quickly.

So, like, you know, if I'm needing your— if you need your nerd alert sign for me, like, you know, Fran will need that sign as well. But I mean, she's also super cool. Like, you should talk to her about flow also. She lives this crazy cool life. But anyway, So metastability is trying to under— one way of trying to understand those brain dynamics, right? And the brain doesn't operate in just one perfectly coordinated way where every region is kind of busy talking to all the other regions of the brain all at once, right? Some regions kind of go off and work together for a little bit and then come together into this bigger coordinated whole. And I think the simplest way to think about this is actually like music and bands. Okay, so I'm a kid of the '90s and the early aughts, like my first Warped Tour was '97, so I'm gonna put this in like a punk band experience, right? So like sometimes the drummer, the guitarist, or bassist, like completely locked in, they're all playing the same rhythm, uh, all together, right? But then like something in the song shifts, the drummer speeds up a little bit, starts Smashing the hi-hat and the snare a little bit harder. The bassist kind of goes off and does their own thing. Everybody like breaks off into their own little bit, but then they come back together. Or maybe like the drummer and the bassist are together while the guitarist is off doing their other thing, right? It's not coordinated chaos. It's actually just people kind of coming off to do their own thing and then coming back together, coming off to do their own thing and coming back together, right? We can think about the brain As doing something exactly like that. That would— what I just described in the brand— the band is a metastable experience. Instances where the band is all together all at once, but then they kind of go off and do their own thing, comes back all together, comes back and does, you know, does their own things again, right? That kind of shifting in and out is a type of metastability. And so where we needed Fran is there's like 26 different ways you can calculate this. And so metastability isn't like one specific thing. It's actually a variety of different ways that you can calculate and statistically describe that difference between integration and segregation, right? Doing everything together, kind of coming across and doing your own thing, right? And that's what our paper was trying to figure out, is trying to handle— okay, when we say the brain is metastable, When this rich flow experience requires regions of the brain to coordinate together but then to kind of split apart, what do we actually mean when we say metastable? So that was the key question. Did I explain that okay, or am I— That was great. Perfect.

Yeah, that was good. Yeah, that was great.

Okay, cool. Okay, so— So what do you find? Yeah, so what do we find? I don't know, you tell me. No, so—

let me— okay, let me— Because one of the things that we found that I think is like wildly interesting and wildly practical for anybody interested in hacking flow, and it wasn't one of our principal findings, but it struck me as very clear in the data, and I've been wanting to ask you about it so I can do this all at once, which is it looked to Michael and me that you can go from frustration, a condition of frustration, into flow far more easily than you could go from boredom into a condition of flow, which I thought was really interesting because forever when we talk about the challenge-skills balance sitting between boredom— there's not enough challenge here, I don't care— and anxiety— whoa, way too much challenge, right? I don't have the skill to meet the task. We've always sort of thought about boredom and anxiety as neutral in a sense around the flow experience. But one of the things that I saw in the data was, oh wow, if your choice is you're depressed or you're really pissed off, which one is going to get you towards optimal performance faster? It looks like really pissed off is actually going to be an easier gateway in, which is sort of, I guess, a little wild. But like, it sort of started to make sense with like talking about athletes and bulletin board material and We used to always talk a little bit about small-s spite as a motivator, as a really useful motivator for flow.

I have big-ass spite all the time. Yeah, totally. I mean, we've talked a little bit. We write about this a little bit in our First Few Seconds for Flow paper as well, right? We talk about the distinction between fight or flight, approach or avoid, right? If you are thinking about the distinction between am I pissed off and angry or am I depressed and sad. Pissed off and angry tends to be an approach motivation, right? Yeah. That's something where you're actually going to approach the experience, whereas depressed and sad tends to be a little bit more inward. It tends to be a little bit more of an avoidance experience. People pull themselves, retreat from the world a little bit, right, when they're experiencing those emotions and psychological states. And so, yeah, it seems to me quite natural that if I am in frustration, if I am upset, it might be eliciting this approach response that might make flow more likely. And kind of tying— before we talk about the metastability stuff, I know we are off on a little bit of a tangent here, but I mean, I think it is kind of interesting and I would love to dig into this a little bit more because, um, earlier I mentioned that if I frustrated you in the lab first, then that flow experience is amplified when I put you in flow second, right?

Yeah, I remember you said that earlier.

Yeah, we see that there. Um, when I bring people into the lab and frustrate them, the thing that kind of always surprises me is very infrequently they're like, They are not giving up. You know, they— and part of this might still be a video game. People expect video games to be winnable, and so they kind of try to persist and hang in there. But then that they have persisted, they have done this frustrating thing, and then we actually change it up on them a little bit, put them in an opportunity where the game is configured in a way that should elicit flow. It really seems to like, okay, that payoff was there. I put in this frustrating experience and now I get that flow. flow as a consequence. You know, again, this is a little bit speculative because we haven't done all the careful controlled studies that would get us to, you know, a claim that I'm really willing to hang my hat on, but I think there's something there, there. Yeah.

So this paper that we're talking about, the complexity science, we'll talk about what we found. And this was, by the way, using your data, some of your data that was produced from your lab, the fMRI data. And then, the other study that we're talking about that you just came out with the same day, that was EEG data. So, I want to get back to the differences there. But, the motivation for this study was trying to look at whole brain dynamics and how the brain does reorganize itself and produce more possibility. Yeah. possible states that it can explore to get into flow. And so, we use this word neurophenomenology. I think the 3 of us and Fran and your researchers and collaborators are bringing the field of flow neuroscience into neurophenomenology. Hmm. Which means the word neurophenomenology, we're just trying to connect what's happening in the brain And the dynamics and the felt subjective experience, which is why the word— Stephen, you can talk about this better than I can— flow, flowy, why it's— why Semihai named it that. That's the phenomenological experience. We're trying to capture those, the subjective and the objective. Mm-hmm. And it's more complicated than it turned out to be. be because the brain tries to reach this difficult-to-reach state, right? Going from 1st gear to 6th gear, and you don't go from 1st to 6th, you go from 1st to 2nd to 3rd to 4th, you know, and you reach that difficult-to-reach state.

Especially because most cars don't actually have a 6th gear, which is— We're only in Ireland. A handful of Italian sports cars have a 6th gear. I just want you to know that our metaphor, like, it puts you in a Ferrari. And, you know, like, it's not gonna work in my Toyota, bro.

Well, maybe it's going in reverse. Maybe it's going—

Well, depending on which Toyota you have, you— well, you might. All right, so I was— my dad was a mechanic. I grew up working on cars, and actually one of my first jobs was with Nissan. So, um, if— so long as you don't have a CVT transmission and you're not looking at like a manual transmission, a lot of modern cars are actually having 6 to 8 gears because it helps achieve fuel efficiency standards. So depending on what Toyota you have, Steven, you might actually have like 7 speeds in your transmission. You thought this was a conversation about flow, but now this is going to turn into a car podcast.

Yeah, no, I'm gonna keep it— I'm gonna keep us back on flow. Let's, uh, let's talk about what Holbrain and I did next in flow.

Yeah, we found a bunch of different results. flow is related empirically to metastability. And then you also found that as well.

Yeah.

So what did these metrics from complexity science say about what happens in the brain while it's in flow? And then transition into how that relates to your— this paper that you came out with as well on metastability.

Okay, cool.

Speak a little bit because I think one of the things that most people don't realize is how much this cerebellum is involved, which was another thing that sort of popped up.

Yeah. So what— so yeah, so when we, when we look at the history of cognitive neuroscience, there's been a pretty big focus on the cortex, right? And there's a lot of reasons for that. There is less, or there has historically been a little bit less of a focus on subcortex, all these like deep brain structures that are involved in reward processing, and then even less focus on this big part at the back of your brain, kind of lower here, called the cerebellum. It's actually, as far as brain regions go, it's a really big mass of tissue, and it's largely been understudied. Mm-hmm. But one of the things that tends to be involved in is really fine motor movement and motor planning. You know, that's oversimplifying a lot of things, but that makes a lot of sense in flow tasks, right? Mm-hmm. Like flow requires, so long as you are doing something that is quite active, requires a lot of motor precision. It doesn't matter if you are a musician strumming away on a guitar, you are a painter, Whether you're working on your next masterpiece, you're a snowboarder or skier on the mountain, or in my case, you're a participant in the lab playing a video game in an fMRI scanner, there's all this fine motor movement that you're having to do. And we definitely see the cerebellum involved in our work there. And so that's one of those things also, like, I really need to dig a little bit more into this. I'm not a cerebellum expert. I need to bring somebody in that actually is. Because they could probably help me understand what the next question is even a little bit better than this. But one of the cool things we find related, kind of going back to this like metastability question, is one way you can characterize metastability is how unpredictable the brain becomes, right? And so we have a technical word for that. It's called entropy, right? It is a way of quantifying how unpredictable the brain becomes or any series of information becomes over time. And one of the things that we find is that the brain actually seems to have the highest level of entropy when people are in flow. And I think that is pretty cool, a pretty cool finding. What basically what that means is by observing the brain at any given moment, how much information does that give me about what's going to be happening in the brain in the future. Higher entropy means I'm having less ability to predict what's going to happen in the future brain based on what I'm observing right now, right? And so, like, what's going on with that? Maybe it's task-specific. We're having people play a video game, and so they're having to respond to all these different things in the game that are kind of changing over time. Could be, but that's kind of true about flow in general. Very infrequently is a flow experience something where you're not responding to the people around you, whether it be group flow or you're responding to things that are happening in your environment. And so, you know, flow eliciting tasks tend to be fairly unpredictable tasks. Mm-hmm. And so if the task itself is not terribly predictable, then it makes a lot of sense to me that your brain is also a little bit less predictable. So that was—

I got nerd alert. Here's a geeky downshift.

Do I need to pull out like a chalkboard and do some equations? No, no, no.

I think it's all over the board. The question is, because video games are unpredictable, because unless I program the game, I have no idea what's coming next, right? So there's They differ from certain real-world tasks in that, I mean, I guess if you played the game for a little while, you're expecting it's going to stay asteroids and there's not going to be a Ms. Pac-Man walking through the screen. Perhaps. You know what I mean? That was what, as you were talking, I was just wondering, is it like, how much does that correlate with anything in the real world where we really could predict the outcome versus a video game? Where, you know, it's literally unpredictable by design.

Yeah, I mean, I think a lot of real-world tasks that elicit flow are pretty unpredictable. Um, even so, I'll give, uh, you know, surfing is a fantastic example of this. Every wave's different, right? Sure, like a break, each break breaks a similar way, but that break changes even as the tide changes. Yep. You can be out there when the tide is high, and as the tide starts to let out, you know, it starts to go out a little bit, that break is going to dynamically change one wave to the next. It's going to change if it's a set or not, right? So even though you might know the break pretty well, there's an element of unpredictability there. You're at the X Games and it's slopestyle. Every time you hit the course, you're responding to an all— in some ways an all-new course, right? The sun has changed, part of the snow might have melted a little bit, somebody went just before you and the lip of the jump that you're about to hit is fundamentally a little bit different because the groomer hasn't gotten back there and cleaned it back up yet, right? So everything in our environment has some element of unpredictability to it, and I think flow tasks certainly have this. And this is sort of the key insight. We don't have to go back to David Melnikoff's work too much, but like, this is a key insight into Melnikoff's work, is there is unpredictability in flow experiences. And so I think that flow experiences have a level of unpredictability and that we see similar brain responses. That seems like a pretty tight mapping to me. Yeah. Now, of course, we have only done this in one study with fMRI only playing one video game. The best thing to do would, of course, confirm this across multiple studies. See if this is one of those task-specific things like we talked about earlier, or if this is one of those task-agnostic things. When I have people do a bunch of different flow eliciting tasks, do I see highly unpredictable brain activity? Brains, high entropy brains over time, right? Now, go ahead. Sorry, Michael.

No, go ahead, Richard. Continue on. Well,

I was going to say, yeah, so here's, here's the interesting wrinkle, right? Is I just told you a really cool story that I definitely believe, and I'm not being sarcastic when I say that. I think the evidence is pretty decent. But here's where it confuses me a little bit. The people that experience the most flow actually had lower entropy scores for their brain responses than people that experienced less flow. Uh, and that one tripped me up a little bit, and I don't have a conclusive answer for what's going on there. But like, there's a couple— what's that?

Task design.

Task design could be— it actually could be, um, expertise, right? It might be these are like our— in our study, it might be these are the really high skilled video game players. Uh, it might be, um, that they had, you know, practiced a little bit more, something like that. I, you know, I don't know. Uh, but that one, that one was kind of surprising to me. It might, it might reflect some type of mastery, something, something there. And I'd love to dig into that problem a little bit more.

No, and that we should do that for sure. And I think we can do that. That's a testable—

I also wonder if it— as you were talking about, like, Mastery is one way to look at it, but with expert video gamers, you're going to get more motivation. There's going to be more intrinsic motivation as well, right? Which raises weird questions to me about dopamine and entropy levels, but I'm not going to go here.

Very interesting. Yes.

Yeah.

It does raise— it did raise those questions about dopamine because The other thing, as you were talking, you could look at it as expertise and that totally makes sense. You could also look at it as motivation because that would— one of the things that I always say is if your short-term goals are pointed at your long-term goals, are pointed at your purpose, you get a very clear pathway, much greater reduction of uncertainty, lower entropy. Right.

Yeah, it could be. Like, it's an interesting problem. Like, how the brain manages uncertainty is, you know, a really richly studied thing. This particular paper was not— and sort of set of analyses, I should say— like, it's just not really well suited to address that exact problem, right? Because we're just— Sure. When we're talking about entropy here, we're, you know, we're talking about how much does information that I have about the brain at one given moment tell me about what the brain is going to be doing at some different moment, which is kind of a little bit different from this, you know, almost free energy-like principle, you know, about uncertainty reduction in the brain, the whole Bayesian brain hypothesis, right? Which is super interesting. I know you guys have been doing some writing about that with Friston. Yes. But, unfortunately, this entropy analysis that we've done, I don't know if it says too much about that particular idea. Okay.

2 things to say here. One is timing is very interesting and important here, not only timing for how we measure things in the brain. This was with fMRI research, which tends to be slower because it's measuring blood responses. Yeah. Hemodynamic response function, which is slower than the electrical activity of neurons. And then, EEG data, which I want to— I want you to talk about your paper as well on metastability, which is much faster and has much better temporal resolution, but also timing of how things happen in the brain itself, not only how we measure, but how things happen. And, I just want to plug here, Stephen, our paper, which is now Under review. It's going to be coming out soon on— yes, hopefully— on synchronicity. So coincidental experiences, purpose, and then flow. Synchronicity, purpose, and flow. Interesting. And we offer an argument based on the free energy principle and metastability for timing, for being a temporal thing that happens on different timescales, but it's the same underlying neurodynamical process. But I want to say, I also believe this story too, and about the unpredictability in flow experience. I mean, this is the whole idea of metastability just in general. Forget about flow for a second. Life is unpredictable and the brain needs to be able to handle that unpredictability.

Yeah.

For survival. And that is what metastability Meta-stability is. It allows— meta-stability allows the brain to explore different possibilities and different coordination states to be able to handle and not settle into one fixed action pattern. Exactly. Like instinct. Exactly. 100%. That's what fixed action patterns are, instinct. We are beyond that, right? Like evolved. And so— I

have to ask a really geeky tangential question as you were talking. This is a non-answerable question, but does it evolve after habit and instinct? Right. Instinct is, you know, you hit my knee, my leg kicks automatically. It's going to happen 100% of the time. I don't have any option. There's no metastability. It's not like my leg can kick to the left or to the right. It kicks out front, right? Because it's a hardwired So, do we have instinct and habit laid in evolutionarily before we get metastability? Is that even possible? This is a totally stupid question. Sounds like a little bit of a stupid question, but—

No, no, I think, I think so. I think metastability is the mechanism that, yes, allowed us—

Yeah, that enables it, right? We're completely habit-bound creatures until the brain figures this out.

Yes, that's what I think. But then, so with, yes, and flow being, and you found this in your paper, right? So it seems to reason that life is unpredictable, so the brain needs metastability. Flow activities are very, very unpredictable, even more so. So it needs to be even unpredictable. a better coordinated brain. Yeah. And so, you did find in your paper, now you could talk about these results and how they add on to and supplement this complexity science paper, that more metastability is a predictor of better flow. And you guys call this the Global Dynamic Model. So, correct me if I'm wrong, but what you guys found was that, yes, higher metastability correlates with higher flow rather than frustration and boredom. And number 2 is the brain gets even more higher dimensional. So it expands the repertoire and space of states that it can be in to be able to get into flow.

Michael, would you explain dimensionality one— like take it one step further and put it into English for people?

Yes. See if I can do this.

By the way, I passed it to you because I didn't even want to take a crack at it. I was like, yeah, Michael, you do this

one. So I was going to use it. I'm going to— I am— nerd alert. Oh my God, now it's me. OK. So I'm going to use another term called degrees of freedom. So the low-dimensional systems don't have many— ways or variables to explore. Whereas a high-dimensional system, there's so many degrees of freedom and there's so many different possible states and configurations and permutations. And the brain is a very high-dimensional system because you have each neuron which can be on or off, right? It can be off. It can be— neurons are all or nothing when they fire action potentials. But then you have— 86 billion neurons and you have such a possibility of ex— you know, that's what I mean by high dimensionality. Richard, can you, can you help me out?

Yeah, I mean, I think you're good.

I think we got us there. Let's keep going. That was good.

Yeah, but I mean, I think you can even bring it back to like this distinction, like what, you know, what is the difference between a kind of hardwired response and flow, right? So you mentioned, you know, if you hit your knee the right way, your leg is going to kick out, right? That's a low-dimensionality response. We can map those circuits really well. Right, exactly. Like, that pathway is actually pretty well understood, how that all happens, right? That's a low-dimensional response. Whereas when we're talking about flow, we're talking about super high-dimensional, where there's all sorts of different ways that the environment around us can be configured, right? And there's all sorts of different ways that our brain can also be configured to respond to that environment. And now if you really want to get into weird nerd alert conversation, I'm going to take it back from Michael here. Now we can talk about the problem of multiple realizability, but we can also kick that can down the road a little bit. But anyway, So when we did this EEG study, we did something a little bit different. Okay? We had people— we had some commonalities with our other fMRI study. We had people play a video game. The video game had 3 conditions, right? Boredom, flow, frustration. But we had them play a different video game. This time it was Tetris. And so they're doing something a little bit different here. And we're measuring their brain activity with EEG rather than fMRI, right? And so Michael a moment ago said EEG is doing something a little bit different than what fMRI is doing, right? EEG is gathering really rapid electrical activity associated with neuronal activity here. Okay, and we're not talking about the response of one neuron. We're actually— EEG is measuring like hundreds, if not thousands, of neurons. neural responses, but really, really rapidly, like hundreds or thousands of observations per second, right? And basically, when we're using EEG measures, we can think about it in 2 kind of classic ways. How does a popular, you know, a group of neurons respond in a very fixed window of time? Or how do those groups of neurons respond over time, right? And that second analysis is what we did in this study. It's called a frequency band analysis. So, you know, you can think about different frequency bands. There's alpha, theta, beta, these different ones. They're basically describing the periodicity of neural response. Some are really rapid, right? Some are really slow. Slow, and this kind of gets back to this idea of the punk band, right? Maybe it's the drummer smashing away really, really fast on the hi-hat and snare. It's really fast, high frequency, or we might have some slower wave responses as well. EEG lets us capture that really, really well with a ton of temporal precision. fMRI is doing something totally different. It's measuring, like Michael said, blood flow associated with brain response, right? We're getting information about what the brain is doing rather than like a few hundred to a few thousand times per second. We're getting information about what the brain is doing once every 1 to 2 seconds, right? Mm-hmm. So like we're talking about a really different time scale here. And not only are we talking about a really different time scale, but we're also talking about a different brain mechanism. EEG is measuring information about what those neurons are doing. fMRI is measuring information about the metabolic cost of what those neurons are doing. So it's like really kind of abstracted away in some ways from the direct neural computations, right? And so any differences that you see between fMRI and EEG studies sometimes can be explained away by that, right? Resolving those differences is like a whole other body of research. Okay, so if you look at those 2 studies together and you see different results and you go, oh, something's messed up, not necessarily. It could just be that, like, you know, we have 2 very different measurement modalities here. But anyway, what did we find? You mentioned the dimensionality result already. The other thing we found is a different kind of classic way of calculating metastability. And when we calculate metastability in a more classic way on EEG data, we actually do see that the brain is metastable, particularly in frequency bands associated with memory and cognitive control. And so this actually could just be that EEG is better at capturing this characteristic of metastability. We didn't detect it in our fMRI data, but maybe that's because the fMRI data is actually just not changing rapidly enough for us to catch the more rapid changes that you can capture with EEG, right? With fMRI, when you take a measurement every 2 seconds, it all gets kind of washed out. It's all kind of averaged up in that 2-second observation. But when you get the faster temporal resolution at EEG, maybe you're able to capture it. And we definitely see that metastability component here. in our Tetris EEG study as well, which I think is pretty exciting.

Yeah. It was a great paper. And I think these 2 papers really add to— so maybe let's speak a little bit to that, right? These papers add to the complexity and the picture that's painted for how the brain gets into flow and what is flow neurobiologically and phenomenologically. Phenomenologically, what do you think are the next steps in some of this direction of flow research?

Yeah, yeah, that's a cool question. You know, if there was 100 people in my lab and 100 of me, we could really make some progress here, right?

But I was thinking, actually, if you had— what is the perfect flow experiment? If you had unlimited funds, people, what is the next—

Yeah, I mean, I don't know if there's such a thing as the perfect flow experiment. I'm not one of those guys. I'm like, I'm the kind of guy that sees an experiment, including my own, and thinks like, oh man, this is messed up in 100 different ways, right? And, you know, it's always kind of thinking about those trade-offs, right? We talked about the trade-off between EEG and just an fMRI a moment ago. We talked about how I trade off realism a little bit to get a lot of experimental control, right? And so I'm not sure there's ever going to be like the one perfect flow experiment. Okay. But what I would say that I think might happen is there are a couple questions that to me feel really important. One is characterizing what neuron— what brain responses Are unique to flow versus task-specific. And the only way to do that is going to be getting a bunch of flow-elicitating tasks and comparing commonalities and differences across brain response in a really principled way. Not like this study did it on this group of people using this task, but they kind of changed their measurement. And so it all gets hard to compare these studies. Do it all in one really clean study. Just out

of curiosity, how many tasks would you— 3, 5, 10?

Oh gosh, I mean, so what I would really try to do is come up with tasks that are, um, letting us vary all the alternate explanations for brain responses that are not flow-specific but are task-specific, right? So, you know, in a lot of our work we motor response, it's people are playing a video game, right? Okay, what about flow listening experiences that don't require so much motor response? Maybe it's like singing or something.

Yeah, I've been designing a mystery game protocol that is— I want to— what I wanted to— I wanted to see if you could measure uncertainty reduction and flow So like literally like you could track the clues and things like that and you could get it cognitive flow instead of, instead of, and I was, I've been literally like I've got a, I've got a plan sort of. I'm at the front end of, it needs some vibe coding so I need Michael in on it next.

I've been looking at that one,

Richard. So if you want, like if you're looking for something like that. I've been poking at it because I wanted the same thing.

Yeah, this is super interesting. Yeah, I think, um, yeah, I, you know, so that task is one. The other is, you know, getting into how is it that we actually enter an exit flow. And I think this, this, this, this idea that you just expressed, Stephen, gets at that really well, right? Yeah. For, you know, from his earliest writings, Csikszentmihalyi talked about flow as something kind of binary. You're either in it or you're not, right? And this maps pretty cleanly into complexity science because complexity science has this idea of phase transition. You know, H2O is a liquid, a solid, or a gas, right? And it Kind of in each state, you know, the molecules can bounce around at different speeds, but at some point it goes from being a liquid to a solid or a liquid to a gas, right? That's a phase transition there. I think we can think about flow in really similar ways. You're either in flow or you're out of flow, and there's this phase transition that happens really rapidly between the two. And I would be really curious, like, I— we know what to do if we want to get water to have a phase transition from ice Or from solid to liquid, right? We apply heat. If we want to know what do you need to do to elicit that phase transition from not flow to flow, I think that becomes a second really cool set of problems to tackle. Like, we've been kind of fiddling around with that, trying to understand the role of challenge-skill balance, right? But that's only one, one component of the phase transition. That are one component that might be driving this phase transition. So that's the second thing. Yeah, I think this idea is super cool, Steven. I think the last thing is getting out of the lab a little bit more. Okay? Yeah. You know, I mentioned teams that have been doing math problems, calligraphy, music improvisation, but I'd love to get people actually out of the lab into the real world. A lot, a lot more in future flow work. And this is starting to become possible. Wearable EEG is a thing. It is increasingly possible to do wearable, portable, wireless EEG in complicated settings. Actually, one of my colleagues, Peter Giunotta, here at UC Davis, is doing this on mogul skiers, trying to actually understand how they decide in the moment. So he's got people, expert mogul skiers, they go out and do fieldwork in Tahoe. They're all hooked up to 100 sensors and they're trying to figure out what their brain-body connection is doing.

Did he put— I've been wanting to put an EEG in a ski helmet for 10 years. Did he do this?

No, he doesn't have this, but he's got eye tracking. But like, this is the thing that you could do. Like, and I mean, talking with him, he'll tell you about the billion problems you have cleaning up the data to make any use of it. Yeah. Right? But like, this is getting more and more possible by the moment. And like, my dream, like, if I, if I could, if I could find the money to do this, I would love to get people on the mountain wearing EEG and see what's happening, uh, as they're skiing or snowboarding and feeling flow or not. Um, I think that would be as, as wireless Portable sensors and biosensors and neuroimaging technologies become more available and our data cleaning procedures become more robust or reliable, I think this starts to become within reach, right? And now we move from math problems to listening flow to video games to listening flow, which all have all, you know, lots of problems, mine included. I'd say video games because that's usually what I'm doing. There's problems there, right? Now we can get people out of the lab and into the real world. Yeah. Which introduces its own set of problems, but that's also kind of getting back to that triangulation issue. If I can find something where, whether it's math problems, video games, or skiing down the mountain, I can find common brain responses when people are in flow. Now I've got what starts to look like more bulletproof Proof evidence for what the brain is doing when people are in flow.

So, let me,

uh, since we're running into our— into a time limit, let me ask a final question, which, um, we started with, uh, sort of what got you into flow research. To like bring us back there, how has all your research Changed or not changed how you seek flow in your own life?

Yeah, I love that question. One of the things that's actually done is helped me be more patient. I'm naturally a super impatient person and we talked a little bit about flow as being something that motivates Persistent performance training to reach a goal. Okay, that requires some patience. And so I'll give you a real-world example. For a long time, I tried to take up a yoga practice and I couldn't. It was— I'm not good at it. I'm not naturally the most coordinated person you'll ever meet. I also am like just kind of tall and gangly and weird, and flow was never happening. for me in yoga. But a few years ago, I tore my meniscus in my right knee, and I was starting to do yoga to help as part of like a rehab process on that. And what my research on flow helped me understand was that I needed to give myself some patience to just suck at yoga for a while and not get the poses. And have a hard time and feel frustrated, but if I could persist through that, I knew something awesome was coming, right? And so that's the, you know, recognizing that feeling flow is something that takes hard work and practice, and you need to be patient. You need to suffer through the setbacks and the frustration and the tedious parts. until you can get it. Like, that's, that's what my research has helped me understand. It's helped me build out a patience for knowing what is to come.

Can I ask— I gotta ask one follow-up because this—

Yeah.

It keeps coming up. You just circled it a little bit. I don't know why this is or how to prove it, but what I've noticed consistently is whatever the task is, There's very no flow when you're a beginner. Then when you're an advanced beginner, going from like advanced beginner to intermediate, there seems to be a ton of flow. Intermediate is this long, sucky, flowless plateau that goes on and on. And then once you get to expertise, it's almost anytime you show up, there's flow. It becomes incredibly reliable, but the, the advanced beginner, like, Period is really common, right? You see people like they get you to the like, oh, I'm going to suck here for a while, and once they sort of accept that, then there's a massive amount of progress and then it sort of shuts off. I'm trying to understand more than anything, why does it shut off? Like what is that second transition and why based on like— This is the thing that makes me challenge in my own brain the expert paradigm around flow. Because like, yes, it's totally true, as an expert skier, an expert writer, I can— I know that when I sit down to write, 90% of the time I'm going to be in flow. But man, I remember like periods of time when I was first learning how to be a poet, first learning how to be a journalist, or all those people, oh my God, they were crazy flowy and I really didn't know what I was doing. And you could make the argument that maybe my sheer arrogance was a way of reducing uncertainty. Like, I was really confident in my ability even when it wasn't there, and that might have been an uncertainty-reducing metric, and we could talk about it that way. But I just wonder, what do you think about that question? Because I'm sure you've had this.

Yeah, so I mean, first off, the, the person that you really need to talk to is a video game designer. I'm serious about this because like this describes the problems they encounter Right. If they create a game that is super frustrating early on, people are gonna quit and they're never gonna come back and the game is gonna be a flop. How do you elicit replayability in a game so that people keep coming back to it even though they beat it over and over? You do this via a variety of design tasks that are constantly adjusting the game's difficulty. Yeah. Based on how well people are doing. There's other tricks to that as well. I'm not a game designer, right? But like, if you talk to Jane McGonigal or a bunch of these people, they'll give you— or Ralph Koster— like, they'll give you a bunch of really good advice on how to do this, at least in games, right? And those provide some pretty good hints at how you bring it into your everyday life. But here's, here's my feeling about this, you know, and again, it kind of goes back to this challenge-skill balance a little bit. Which I know we talked about, maybe it's not everything it's made out to be. But, you know, I think different tasks have different challenge curves, right? Um, we've, you know, we've got some, uh, friends that their daughters just got old enough to start skiing, and they wanted to pick up skiing. They're like, should we ski or snowboard? I'm like, oh, start skiing, because like the challenge curve, at least initially on skiing is way flatter than the challenge curve to pick up snowboarding, but then it shifts, right? I think snowboarding is like really hard to learn, but then you're in this flat, everything is kind of like not progressing, there's a really long time where skiing is sort of the opposite. It's like, okay, you can go down the mountain pretty straightforwardly initially, but then rapidly the difficulty ramps up, right? And so I think A bunch of different tasks have different challenge-skill curves, and where your skill is relative to that curve is going to determine whether or not the experience is flowy or not, right? Mm-hmm. If you're in this vertical part of the learning curve for the task, but your skill is still like way down here and there's this clear mismatch, I don't think that's going to be a flow experience for you, you know. The second thing is we just get better at picking Unless you're a

masochist.

Yeah, I mean, well, again, you gotta talk to the video game player designers because like there are players that choose these really impossibly punishing games. I've never understood it. I've always like wanted to do some like interviews on these people. But generally speaking, one of the, one of the research findings is that people are pretty good at guessing and advance what is going to be a flow experience for them, right? Like, I'm— if I'm at the train park at Northstar, a ski resort here nearby me, or if I'm, you know, looking at video games I might play, I can make a guess what one is going to elicit flow for me, what versus what one is going to be boring versus what one is going to be, you know, a frustrating experience. And so I think this is an element of that type of mastery as well, knowing where you're at and selecting the right challenging task to elicit flow or not.

So Stephen, can I ask my final question?

Final question, Michael. Bring us home. Bring us home, baby.

Thank you. All right. So I got to ask you as a neuroscientist. So we're studying this thing flow. And we talked a little bit about how flow is correlated to performance, but I want to ask you, what are your thoughts about how important flow is to society and being human and dealing with all of this information? Stephen, in your book, We Are as Gods, you write about people in flow, for example, collaborating with AI are going to own the future. And the ability to collaborate with AI and how flow is, is so important to mitigating burnout and how it can fight resili— be— make us more resilient, make us more purposeful. And what I really want to ask you is at the level of the brain, how correlated is flow to meaning and well-being? Right. And how important it is for getting into flow for society because we're studying this thing but it seems to be such an important thing for us to practice on a daily basis.

Yeah. Yeah. So, I'm going to start actually with a little bit of skepticism, OK? I'm a scientist. It's what I'm supposed to do, all right? Yes.

I love it.

And then I am going to tell you, you know, after I listen or share a bunch of skepticism, then I will kind of bring it back to, you know, what I think is the hopeful, helpful story as well. Okay. So one thing that has bugged me about flow, we've been studying it since the '70s. For 50+ years now, most of the work that's happening in this space is correlational. There's all sorts of reasons for that, right? Like, it's a hard thing to study. If I ask you, you know, if I, you know, if you are a talented musician and I give you a guitar and you start playing and I say, hey, are you in flow right now? I knock you out of it right away, right? And so there are all sorts of reasons why we have a long correlational history of flow research. But that kind of bugs me a little bit because, yes, experiencing flow is shown to be associated with resilience, you know, buffering during stress, resistance to burnout, all these sort of things. A lot of that work is still really correlational. Could be the other way, you know, people that are really resilient are more likely to experience flow. I don't know, you know, I'd love to do— I'd love to feel like— but how do you tease that apart even? Like, how do you experimentally manipulate people's resilience, right? So like That's a challenging thing to uncover. Okay. So I just want to say, like, there are a lot of really big claims about flow. I think some of them are pretty well evidenced, but always having a little bit of a healthy skepticism is not the worst thing that you can do. Right. That said, I think to me, like, flow is Ultimately described as the optimal experience. It's intrinsically rewarding. Flow-listening tasks are things that are worth doing in and of themselves, right? Um, we don't have to seek out a reward for them. The thing itself is rewarding to do. Flow is remarkably egalitarian. I don't need the most expensive skis or the most expensive base or, you know, to be the richest person in the world to heli-ski. Alaska to experience flow, right? I can have a yoga mat and my floor and I can experience flow for very low cost, right? I can have a pair of running shoes and get out and go running and experience flow for very low cost. Flow is one of these things that everybody can experience in their everyday life. It's not a rich person's thing, a poor person's thing. It's not something— That is exclusive or it comes with, you know, something that only a limited number of people can have. That to me is one of the most important things. Life's hard. Things feel really hard right now. And the more that we can experience flow in our everyday lives, I think the better we can get through it. And that, I think there's something pretty critical and important to that as well. So to me, that's the most exciting part about flow research. It's egalitarian. Everybody can do it, and it brings meaning and purpose to our lives.

That's a great answer. I don't know if you know this. If you look in, uh, some of the work is actually in Beyond Boredom and Anxiety, but early Chicks at Mihaly, his Italian collaborator, and I'm gonna forget the, uh, Massimo, I want to say was his

Yeah.

Looked— they did a bunch of research on class and flow, and they found that poor kids— they were doing it with high school students— poor, less well-advantaged kids had a much easier time getting into flow than wealthier kids. It's never been duplicated, but it really— 2 different studies, and it really caught my attention, and I was like, oh, that's interesting. This, uh, this tells, this tells you something. I'm not quite sure what, but I found it really interesting.

Yeah, I'd love to, I'd love to follow up on that. But I mean, that, it's just one of my favorite parts about Flow, you know.

It's great.

As we're busy, like, every premium experience comes with some extra fee or something like that. Flow's something everybody can get, and I really love that.

I agree. Richard, thank you so much. Well, thank you so much for being a collaborator.

Thank you, guys.

This was a ton of fun. Thank you for coming on Flow Radio and talking to us about your work.

Thank you so much, Richard.

Yeah, awesome. We'll have to do this again and continue our collaboration for sure.

Yeah, absolutely.

Awesome.

Thank you everybody for tuning in. We'll catch you next time on Flow Radio.