The science of flow

When members of Phish and the Grateful Dead rock out in a neuroscience lab

The first time I’d ever heard the term ‘flow’ was in 2016 at a human-performance conference in Santa Fe, N.M., and I remember feeling exhilarated. It seemed the perfect word to capture an almost divine hallmark of performance, something we’ve all experienced — that moment of clarity while doing something we love, when time melts away and things become effortless and easy. We all seek that flow.

The fact that there was science, evidence of biomarkers and neurochemistry, to back up the existence of these ‘flow states’ was marvelous. But eight years later, I still struggle with making the leap between the flow we might experience in our daily lives — driving to school, listening to music, typing out this newsletter, etc. — and whatever transcendent psychological state that a Stephen Curry or Roger Federer might reach while achieving their peak performance. Are they flowing like me? Or is theirs a somehow different echelon of flow?

Rather than attempt to unpack a warehouse full of conflicting studies on this, I thought I’d zero in on some very interesting applied flow work being conduced by my friend, Greg Applebaum, who heads the Human Performance Optimization lab at the University of California-San Diego.

Greg’s work recently caught my attention because of whose flow is being measured. It’s Mike Gordon, bassist for the band Phish, and Bob Weir of, you know, the Grateful Dead.

Applebaum, Gordon, and a number of colleagues have been developing what they’re calling the XenboX, a Brain Computer Musical Interface (BCMI) used for defining and promoting musical flow states. In essence, the XenboX is a “flow meter” that can be activated by the musicians themselves to indicate when they are in or out of flow.

In the clip below, which Applebaum granted me permission to share, you see Mike (left) and Bob grooving in the lab-studio, donning EEG headcaps, and tapping the XenboX foot pedals.

I called Applebaum recently to ask him about the work, which is ongoing. I’ve typed up our conversation. Hope you enjoy.

This interview has been edited for clarity.

How would you define what ‘flow’ is?

I don't have a really satisfactory definition. And part of the cool but challenging part of the science is actually getting a convergence of definitions.

In the attentional control literature, we oftentimes think about flow as 'doing an executive function that requires a lot of top down control with a lot of automaticity.' So, the analogy I give is when you're learning to drive a car. First time you get behind that wheel, you see everything, you remember everything you notice the color of the other cars. Today, when I drove I didn't pay attention to anything. My driving is automatic.

The other important element in, I don’t want to say the definition of flow, but the value of flow, is who is doing what? Me swimming laps at the pool and feeling like I'm in a flow is not the same thing as Bob Weir at a concert or LeBron James in a basketball game. Everybody at some time feels like they have a lot of automaticity in their activities. Driving a car, for example, I may be flowing. But Mario Andretti driving a car and experiencing flow has a different value and utility for the flow state.

How is flow different from heavy concentration or focus?

For me, my kind of conceptualization of flow is that it is when heavy concentration becomes effortless. The thing that you otherwise would have to really concentrate on to do successfully becomes effortless. That involves some combination of automaticity and relaxation.

My conceptualization of flow is that it is when heavy concentration becomes effortless.

Flow also doesn't have a constant utility function over people or even within a person. It’s one thing for Steph Curry to be flowing during practice versus flowing during a game. It's a totally different thing for my son to be flowing during a high school volleyball game versus a professional athlete to be flowing during a game that has real meaning.

It still sounds like you’re hedging your definition a bit.

I think everybody has trouble putting their finger on what flow is. It's a fairly loose term. I think it's like pornography. You know when you see it.

How did this project come about?

Mike Gordon [bassist for Phish] has really been interested in flow for a while. He's a very smart guy, and he has a lot of scientific interests. He's the one who's been leading the charge.

What was your approach?

Well, a real challenge of trying to measure flow in the person who's flowing is that the realization of flow might knock you out of flow. The reporting of flow almost definitely knocks you out of flow.

Right, because you’ve lost your train of focus.

Right. So how do we get past that? Our idea was to create a foot pedal. Musicians, especially guitarists and bassists, they use their feet all the time to signal things. They thought it's very natural to signal flow with their feet with this switch. So we started doing some practice sessions, and it kind of quickly became, ‘Hey Greg, when I clicked the button, that means I was just in flow.’ Not that ‘I’m in flow now.’ It’s that, ‘I had the realization that I was in flow.’”

Retrospective reporting.

It’s a retrospective realization that I've been achieving something that's usually really, really hard for me, but I was doing it effortlessly.

All the time that this is going on, we have them wearing an EEG system [that’s monitoring brain activity]. We also have a system that’s measuring all kinds of physiology, heart rate, accelerometers, gyroscopes, blood-oxygenation.

In addition to this system all hooked up to the musicians, we also wanted to get other people's judgment of the flow. Let's us have some people in the audience rate it. We’ve found that, every time we do it, the audience has five or 10 times as much flow as the musician, because they're subjectively enjoying it.

The real ace in the hole is we're also working with their sound engineers. The sound engineer is the person who is responsible for laying down the tracks for the recordings. They're the ones who have this really trained ear, and they have a lot of experience. If an engineer works with a musician or a band for a long period of time, they have a lifetime of experience of that musician trying that really hard thing and failing and failing, and then succeeding, and then succeeding kind of effortlessly and transcending and making it even better. So we considered them to be kind of like an expert rater of flow in the musician, who could potentially clue us in to when the performer is in flow without interrupting the performer.

Why is this information of value to the musicians?

One way it’s valuable is as a kind of a training tool. It’s neurofeedback. I can signal when I'm in a flow state, so now I know what a flow state feels like. I get feedback from my own physiology, and I can kind of recreate it through the feedback.

The other thing they’re doing is using the XenboX as a controller. So, you know the person's flowing, and now it can just act as a switch to turn on a sound effect, or turn on a light show or something. That's what they've really been doing and enjoying. In a perfect future version of this, there would be an algorithm that says, okay, the ratio of these brain activities that we think differentiate flow from non flow has crossed a threshold. The algorithm detects it, and that starts a sound effect or a light show or whatever. But what they're doing is they're just looking at it in real time, and the sound engineer is making changes, and they feel like this is fantastic, because it gives a sound engineer and the musician another lens into the performance.

Are there other ways you are trying to investigate flow in experts?

We’ve created a similar system that we’re using for measuring performance in surgeons doing robot-assisted surgery. What we’re trying to do there is link the physiology to the performance, and the performance is typically defined by errors. We consider errors to be like the opposite of flow. We're trying to look at what are the physiological indicators of when someone is performing at a high level, and when they're either about to commit an error or they’ve just committed an error.

Sounds like that information would be quite useful.

What we’re looking at is, ‘How do you find the features of performing at a higher level? How do you find the physiology associated with that?’ It may not be flow. It may just be error detection, or it may be the lack of error attention.

We're trying to discover things that people could use. In the surgery space, it could be an early warning system for error detection - that would be like the dream situation. It could be a warning signal. Everything we do now is so quantified. We have mobile monitoring of everything we're doing. So how do we use it? Like, I use my Apple Watch to remind me to go exercise more.

Will performers respond to internal feedback like a ‘flow meter’?

We're all looking for that little edge. I do believe that the insight that we can get into our own abilities helps us be better. A lot of what we do is based on external insight: Your coach telling you what you're doing. You might watch a video of yourself or listen to a recording of what you're doing. But more and more in the world now is we’re also getting an internal signal of what we're doing. We measure our physiology [through devices like the Apple Watch] and say, ‘Hey, I want that physiological state.’ I want to recreate that physiological state because I feel it associates me better with higher performance.

For more on Applebaum’s work, please visit: https://www.optilab.net/

Links

Japan halts gene research of athletes over privacy concerns … Kang “will develop and incorporate a training concept unprecedented in women’s [soccer], based on the specifics of the female body” … Athletes swear by red-light therapy, should you? … The performance-boosting effects of baking soda … Tulsky left career in science to become Hurricanes new GM … Amazon says its AI can predict blitzes … Why some say tennis is the world’s healthiest sport … Variety is the spice of learning, memory study suggests

Shameless plug

I wrote recently in the NYT about a sports memorabilia goldmine that had been hidden from the public for years in Virginia. Here’s the link:

https://www.nytimes.com/2024/09/14/business/sports-memorabilia-collection-discovery.html

References

D'Ambrosia, Christopher, et al. "The neurophysiology of intraoperative error: An EEG study of trainee surgeons during robotic-assisted surgery simulations." Frontiers in Neuroergonomics 3 (2023): 1052411.

D’Ambrosia, Christopher, et al. "The physiology of intraoperative error: using electrokardiograms to understand operator performance during robot-assisted surgery simulations." Surgical endoscopy 37.6 (2023): 4641-4650.