Marc Berman with Sean Carroll
Show: Sean Carroll's Mindscape
Cleaned and reformatted from published transcript or auto-generated captions — punctuation added, filler removed, restructured for readability. Not verbatim. For exact quotes, refer to the original.
Sean Carroll
Marc Berman, welcome to the Mindscape podcast.
Marc Berman
Thanks so much for having me on, Sean. I'm a huge fan and an avid listener.
Sean Carroll
Before we get into the specifics, it reminds me of a fact that's sometimes trotted out. We all live in environments that are changing very rapidly, especially compared to the timescales on which we evolved as a species. In a very real sense we're inheriting a genetic legacy that was put together thousands or tens of thousands or millions of years ago. And here we are — I live in a house. My genes are not adapted to living in a house. How do you think about this? Not so much a question as an observation I'd like you to reflect on: how dramatic is the difference between what we grew up to be and the environments we're in right now?
Marc Berman
It's something we have to pay a lot of attention to. And it's ironic, Sean, that even though we live in such arbitrary environments compared to the ones we evolved in, because humans have so much control over the environment, I think we believe we're almost immune to it. That's something we've thought a lot about in my laboratory — we aren't immune to the environment, and we're a part of the environment. People sometimes feel that humans are distinct from the environment, but of course we're very much part of the natural world.
Because we have so much control over it, it's a bidirectional relationship. Our brains have been shaped by the physical environment; now our brains are designing the physical environment, and that new environment is in turn shaping our brains. And it's happening very rapidly — we've only lived in these cubicle boxes for a few hundred years. It's kind of arbitrary, and our brains adapted to a very different environment.
I don't want to poo-poo architecture and design, but a lot of how we build the things we live in wasn't designed to make people more cooperative, or to give them better working memory or better attention. It was built to house people efficiently and move goods efficiently. In my lab we're trying to take a pause and say: these quote-unquote artificial environments we live in now may be changing our behaviour, maybe changing our brains, and we need to think seriously about what that means and what we might do about it.
Sean Carroll
There's an irony there. We have much more control over our own environments than we used to, but we don't always use that control for good.
Marc Berman
Exactly. And for a lot of years this would have been thought of as fringy science — not the most hardcore, rigorous work, in part because it's really hard to do carefully controlled experiments. In an ideal world I'd love to move people to different environments and manipulate different elements, but you can't do that easily. So it takes a lot of hard thinking, and you have to triangulate research from a lot of different areas to draw strong conclusions.
Sean Carroll
Give us the high-level message. You have a book out that came out last year — remind us of the title and what it's all about.
Marc Berman
The book is called Nature and the Mind: The Science of How Nature Improves Cognitive, Physical, and Social Well-Being. It talks a lot about a field called environmental neuroscience, which I helped to develop. It's interested in the bidirectional relationship between the physical environment and our brains.
In the book I focus on natural environments — contrasting people interacting with more natural spaces, like your local park, the trees on your tree-lined street, a lake you visit on vacation, with more human-made environments: cities, towns, office spaces, schools. And we find that when people interact with these more natural spaces it actually causes changes in their attention and working memory. People show improvements after spending short amounts of time in more natural environments versus more urban or built ones.
Sean Carroll
It's important to distinguish here. Lots of people say, 'Yeah, I like spending time in nature, it relaxes me, it puts me in a good mood.' But you're less talking about mood and more about cognitive abilities, which is a different kind of claim.
Marc Berman
Absolutely. I wasn't the first person to look at whether interacting with nature had psychological benefits. There had been studies where people interacted with natural versus urban environments and reported feeling better — maybe their mood improved, maybe they felt their attention was better — but they didn't use objective measures of performance. We did one of the first studies that actually measured objective changes in performance.
One of the ways we measured that was a task called the backwards digit span. You hear digits out loud at about one per second and repeat them back in reverse order. You might hear 8-3-7 and repeat 7-3-8, and we keep increasing the number of digits up to about nine. At around five digits the task is very challenging.
We gave people this difficult task, then had them go for a walk in either a more natural or a more urban environment — about a fifty-minute walk. They came back and did the task again, and a week later they repeated the whole procedure with the other environment. Everybody was their own control; it was a within-subject design. When people walked in the natural environment they showed about a 20% improvement on the backwards digit span — roughly a digit and a half — compared to the urban walk, where they didn't show any significant improvement.
Sean Carroll
A listener might say people probably enjoyed the nature walk more, so maybe they were just more motivated.
Marc Berman
It was certainly the case that people liked the nature walk more, and we had mood measurements too. Mood did tend to improve after walking in nature. But we didn't see a strong correlation between the improvement in mood and the improvement on the backwards digit span. It wasn't that the people getting happier were the ones showing the most cognitive improvement.
What I found more interesting is that participants walked at different times of year. Some walked in June when it was about 80 degrees Fahrenheit — they said, 'Marc, I can't believe you're paying me to walk in nature.' Healthy mood benefits, healthy cognitive benefits. But others walked in January when it was about 25 degrees. They said, 'Marc, I was freezing out there, that was not enjoyable.' And yet the January walkers showed the same cognitive benefit as the June walkers. So the takeaway is you don't even have to enjoy the nature walk to get the cognitive benefits. You do have to enjoy it to get the mood benefits, but not the cognitive ones. That's somewhat counterintuitive — it isn't a hedonic response, like eating chocolate and feeling good. Something deeper is going on.
Sean Carroll
I have to say a lot of these psychological tests seem pretty anxiety-inducing. I'd feel very bad not being able to say the digits backwards.
Marc Berman
Absolutely. We try to tell people beforehand it isn't an IQ test, but it is anxiety-provoking — and that could be part of the phenomenon too. If nature relaxes us a little, maybe we perform better.
Marc Berman
Related to that, we repeated the same experiment with participants diagnosed with clinical depression. Depression is often characterised by negative rumination — repetitively turning over negative thoughts and feelings. We had an NIH grant to study working-memory deficits in people with depression, and we wondered what this nature effect would look like for them. It might be that if you go for a walk alone, that gives you more opportunity to ruminate.
So we ran the same walk-in-the-park procedure, but this time we had people think about a negative thought or memory that was bothering them before the walk — deliberately inducing rumination. We found even stronger results. People showed even larger improvements in backwards digit span after the nature walk relative to the urban walk. And I know this isn't a self-help podcast, but in the UK and Canada physicians are now prescribing walks in nature as a supplemental therapy for people suffering from depression and anxiety.
Sean Carroll
I have to get the question on the board right away. This is an interesting finding in psychology, and we've all been trained that interesting findings in psychology should be looked at with skepticism — small sample sizes, hard to replicate. This was a while ago. What's the state of the art?
Marc Berman
I'm pretty confident in these findings. No one has identified this effect as being non-replicable, and many other labs have replicated it. You're right that the sample was small — about 40 participants — but it was within-subject, so that's roughly equivalent to 80 participants between subjects. Not millions of people, but robust.
There have been other studies outside the laboratory too. Researchers quantify the green space around people's homes and schools and look at children's working memory and school performance. Children who live in neighbourhoods with more green space have better attention and working memory, even controlling for age, education, and parental income.
There were also fascinating studies in public housing projects in Chicago — not desirable places to live, and many have since been demolished. What's interesting is that families were randomly assigned to different apartments, so it was almost a controlled experiment. Ming Kuo and Bill Sullivan at the University of Illinois looked at the views out of the windows and used satellite imagery to quantify how much green space surrounded the projects. The apartments with views of nature had residents with better attention, lower reported crime, and lower reports of aggression — suggesting that even modest nature can have salubrious qualities.
Sean Carroll
You mentioned the theoretical background, which I wanted to get into. As a physicist I always love to say experiments aren't real until they're confirmed by theory. You talk about attention restoration theory. What is it, and how is it relevant to your studies?
Marc Berman
This theory was codified by one of my mentors in grad school, Stephen Kaplan, and his wife Rachel Kaplan, both environmental psychologists. One of the main tenets is that humans have two kinds of attention.
One is directed attention — sometimes called top-down or endogenous attention. This is where you, as an individual, decide what to pay attention to. Right now you're deciding to pay attention to me; your listeners are deciding to pay attention to what I'm saying. But that kind of attention is fatiguable or depletable. You can only control your attentional focus for a limited time before you become mentally fatigued and can't focus. We've all had that sensation at three or four in the afternoon, staring at the screen, and it's just hard to focus. I see it when I'm lecturing — the first five minutes people are nodding along, and 45 minutes in they're tilting their heads back, dozing off.
That's different from involuntary attention — sometimes called bottom-up or exogenous attention — which is automatically captured by interesting stimulation: bright lights, loud noises. We don't have much control over it, and it's thought to be less susceptible to fatigue. You don't hear people say, 'I can't look at that beautiful waterfall anymore, it's too interesting, I'm too tired.'
So the idea of attention restoration theory is that if you can find an environment that doesn't place a lot of demands on directed attention while still having interesting stimulation to activate involuntary attention, you can restore or replenish that precious directed-attention resource. Natural environments are one kind of environment that can do this — not the only kind, but one.
Marc Berman
There's one more important thing: the kind of stimulation that activates involuntary attention matters. When I look at a beautiful waterfall it captures my attention, but I can still mind-wander and think about other things. It doesn't consume all my attentional resources. When I'm in Times Square in Manhattan — also super interesting, lots of stimulation — it captures my attention in an all-consuming, harsh way. I can't mind-wander there. So the stimulation that captures involuntary attention should be softly fascinating and not harshly fascinating. That's the kind of environment that can restore directed attention.
Sean Carroll
The idea that really stands out to me is this: you might have thought attention is a finite resource — we have a bucket of it and we use it up. But you're saying some ways of paying attention use it up while others restore it. Even as we're paying attention, we're gaining more ability to do so.
Marc Berman
That's right. I think of directed attention as a bucket or bank account that you can cash out or empty. But refilling it is not a passive process — it's engaging another kind of attention. It's a Goldilocks rule: interesting enough that it captures your attention, but not so interesting that it's all-consuming. Presumably sleep does something similar, but often when we're mentally fatigued we're not physically fatigued. It's almost like the body is telling us to go pay attention to something else, or switch tasks.
People sometimes ask, 'What about sitting in a dark room?' My answer is no, for a few reasons. Most people don't like sitting in dark rooms — they find it boring, and boredom is fatiguing. This isn't just a passive process like sleep; it's an active process of engaging a different kind of attention that can refill the directed-attention bucket.
Sean Carroll
This might go against everything you believe in, but for me I take a break from work by playing little cheap video games on my iPad — if you find that sweet spot where the game is engrossing enough that you want to play but doesn't require too much effort, it re-energises me. Is that compatible with this, or am I just weird?
Marc Berman
It's compatible, because nothing in attention restoration theory says it has to be nature. It only has to meet the criteria — not placing a lot of demands on directed attention, while having softly fascinating stimulation to activate involuntary attention. If those little games do that for you, good. But I'd note the games our kids play are much more engrossing and more harshly fascinating. A lot of these new technologies are very harshly fascinating — they're designed to keep your attention and not let go. They're addictive, and most addictions aren't good for cognition.
Sean Carroll
Is there a relationship between directed versus involuntary attention and the system one versus system two distinction Kahneman talks about?
Marc Berman
Yes. System one is the more heuristic, easier kind of thinking, and system two is the more arduous kind. If we mapped it on, involuntary attention is more like system one and directed attention is more like system two.
Sean Carroll
An example came to mind reading your work. Years ago, when cell phones first became popular, there was a controversy about talking on the phone while driving. There was a thought that as long as you had somewhere to put the phone, so you didn't have to hold it, it shouldn't matter — you're still driving with your eyes and hands. But in fact that's not true, because talking on the phone depletes your attention. Was that impression correct?
Marc Berman
You have it correct, and it's really important. There was a peripheral hypothesis: is the problem with phoning and driving that you're physically holding the phone and only have one arm to drive, or is it attentional? It turns out it has nothing to do with holding the phone and everything to do with where your attention is directed.
We've all had the sensation of driving on the highway and not remembering the scenery from the last couple of minutes, because your attention was elsewhere — you were mind-wandering. Even though light from the road is hitting your retina, the brain isn't attending to it. That's the difference between overt attention and covert attention. Where your eyes are pointing is overt attention; covert attention is what you're actually attending to internally. The problem with the phone is that your covert attention is directed to the call, so you pay less attention to the road. And the person on the other end doesn't know what conditions you're in — if you had a friend in the passenger seat, you'd tell them to be quiet in hard driving conditions, but that doesn't happen on the phone.
Sean Carroll
Not to be too grandiose, but maybe there's a lesson here for the philosophy of mind, and even for AI and its relationship to intelligence and consciousness — namely that we're much more than a surface-level input-output machine. There are so many layers going on simultaneously, all interacting and feeding into the final output.
Marc Berman
Absolutely. That's why behaviourism failed in psychology. Skinner and Watson were frustrated with the introspectionism of earlier psychology — the idea that we could just look within ourselves and report our subjective experience, and that would be enough. They thought you can't build a science around that; it's not verifiable. So they went to the other extreme and said we can explain all behaviour just by mapping stimuli to responses. But that turned out to be far too simple. There's a lot happening in the brain, and you can't ignore it. Now we've settled more in the middle.
Sean Carroll
That was my next question. You talk about environmental neuroscience — how much can we say there are different levels or kinds of attention and associate them with different processes, or even different parts of the brain?
Marc Berman
We can, and I do a lot of neuroscience in my lab, trying to merge the environment with neuroscience. But it's challenging. We do a lot of brain imaging in MRI machines — functional MRI, essentially a flip-book of your brain to see the dynamics. But it's an uncomfortable, arbitrary environment; it's hard to give someone a nice nature experience inside the scanner. There's newer technology we're starting to use called functional near-infrared spectroscopy — a cap, a bit like an EEG cap, that shines infrared light into the brain and measures blood oxygenation, similar to MRI but mobile. People can wear it with a little backpack and walk in natural versus urban environments.
We know something is happening in the brain, since backwards digit span changes after the nature walk. But it won't be satisfying to just put people in a scanner and show them nature versus urban pictures. I'll find differences, but many features differ between the images — different fractalness, different entropy, different amounts of curved edges, different colour palettes. It's hard to zero in on what's causing the difference, and whether that's the thing causing the change in performance. It'll be more complicated than 'show nature pictures, show urban pictures, look at the difference, and there's your answer.'
Sean Carroll
That's exactly where I wanted to go next. What is it about nature versus the urban environment that's more restorative? The colours are different, the shapes are different, the sounds are different — but also the purpose is different. In nature we feel we're there to chill out; in the city we're on an errand.
Marc Berman
Great question. I'll back up to say we've also done studies where you just show people pictures of nature, or play nature sounds versus urban sounds, or nature videos versus urban videos, and still get some of these changes in backwards digit span. That suggests it's something about the perceptual features of nature producing the benefit, which gets away from it just being leisure time. I'll caution that these effects with pictures and sounds are not as strong as being in real nature, and not as robust — but the fact that you can get them suggests there's something about the perceptual qualities.
We've become really interested in ideas of information and compression. I had a student, Omid Kardan, an electrical engineer, and we wondered whether nature images get compressed into fewer bits than urban images, because there's a lot of repeated structure in a natural image. A few years later my current PhD student, Nakwon Rim, actually tested it. We took hundreds of images, had humans rate how natural they thought each was on a scale of one to seven, and ran a JPEG-style compression algorithm on them.
JPEG compression throws away information your eye can't easily detect, taking an image that might be 100 megabytes down to 10. We found that more natural images tend to compress into fewer bits than urban images — more information is being thrown away in nature scenes. And these JPEG algorithms are based on human visual perception; they work well because they've been tuned to how the human eye sees.
Sean Carroll
I didn't even know that — it's not purely mechanical?
Marc Berman
Not purely mechanical. It's tuned for humans. Then we asked another question: if nature images are easier to process, maybe they're also less memorable. Listeners might think being less memorable sounds bad, but it might be good — it might mean the brain hasn't exerted much energy to process it. So we did an experiment showing people hundreds of images varying in naturalness, where they respond if they see a repeated image — a measure of memory. People don't remember the nature images as well; they remember the urban stuff better. And statistically, part of the reason nature images are more forgettable is that they're more compressible.
Sean Carroll
I'm going to wander off the reservation and make a conjecture, because what you're describing sounds right next to what I'd suggest — thinking about entropy, but from the information-theory perspective rather than physics thermodynamics. Claude Shannon taught us that information comes from surprise. If you get the same message over and over, you're not getting information, even though you're getting a lot of symbols. So you can calculate the entropy of a set of messages, and the broader it is the more information it contains. It sounds like you're saying the information input from nature is low entropy — you're not surprised when you see the tree, so you don't need to do much processing. In the urban environment there's a sign, but which sign? You have to pay attention. Higher entropy visual input, more processing.
Marc Berman
I agree with your conjecture. Though it's interesting — we've run grayscale entropy measurements on these images, and if I remember correctly it doesn't show as big a difference as we'd expect. Maybe we have to do something with the colour entropy or the different colour channels.
Marc Berman
Actually, when I was reading your book, The Big Picture, you have a figure about this that I took in another way. In the chapter I think is called 'The Universe in a Cup of Coffee,' one thing that's always bothered me is that the highest-entropy image would be white noise. But as a neuroscientist I wouldn't think of white noise as a very complex image — it's complex at one level, but it's just randomness. In your figure, entropy and complexity are highly correlated at the beginning, but at some point they deviate: complexity maximises and then decays as entropy keeps rising.
So for an image to be restorative, I think it has to have the right amount of complexity, and maybe not too much entropy. And there's some relationship to ease of processing, which is where fractalness comes in. Fractalness is repeated structure at different scales. Natural images tend to have high fractalness because they're built from the same recursive process. Think of a maple tree: the branching structure is echoed in the veins of the maple leaf. Zoomed in or zoomed out, it has the same characteristic shape. Those natural fractal patterns are ubiquitous in nature and not in the urban environment — a city looks very different zoomed in versus zoomed out, but a forest looks similar.
There's an ease the brain has when there's this fractal structure — you can get the gist of a scene very quickly. In an urban scene you can't. It's related to compression, but it might also have to do with fractal structure. I'm not a fractal-geometry expert, and it's not trivial to find the right way to measure it. Another thing that's interesting: we don't really have a language of nature the way we do for the urban environment. In a forest I might say trees, shrubs, grass — not very semantically complex. In a city I might say Volkswagen, Toyota, BMW, Gothic architecture, stop sign. So much more semantic complexity, which might make it harder to process but also easier to remember.
Sean Carroll
I'm going to start biting my tongue, because we should talk offline — there's too much to say. But this whole idea that it's relatively straightforward to invent quantitative measures that are small for perfect order and large for perfect disorder — that's not what you want. You want something that peaks in between. That's full employment for complexity theorists. Logical depth is one idea that's very relevant here.
Marc Berman
It's a classic problem, and it's great that you're suggesting it. If I can rephrase: just as with people trying to capture the complexity of strings or images, if you do the simplest thing — algorithmic complexity — a very long random string is high-complexity, in the sense that the best way to print it is just to print it, there's no shortcut. But that's not very helpful. When there's some structure, there's something interesting going on that's more than perfect order — structure the brain can latch onto, so it doesn't have to fret about every detail separately. Maybe that's our happy place, what we evolved to do.
It's a little more complicated, though. If we showed people a perfect statistical fractal like a Sierpinski triangle, I don't think we'd see restoration, because it would be too boring — too perfect. You've had Karl Friston and Tom Griffiths on the show talking about Bayesian models of the brain. There's a nexus where the brain likes predictability, but not perfect predictability. It's satisfying when a prediction is right, but a little jitter around it can be good. A lot of natural stimulation has that quality — imagine walking in nature and there's a curve in the path; you don't know exactly what's behind it, and there's something softly fascinating about that.
Sean Carroll
It makes me think psychology isn't finished yet. You don't have the theory of everything right around the corner.
Marc Berman
Absolutely not. I joke with Luis Bettencourt at SFI and Andrew, one of my former students, that I have this desire to make psychology more like physics — to get lawlike formulas. But it's very difficult. We don't even have concepts like a rigorous definition of mental energy; that's not trivial. We lack the kinds of principles a field like physics has. If we had them, we could do so much more — but we're certainly not there yet.
Sean Carroll
I'm going to ask you to do one more speculative thing. There are debates in neuroscience about criticality in the brain and power-law, scale-free behaviour. It all gets very heated in a way that seems a little weird to me as an outsider — they're fitting curves and arguing about whether it's a perfect fit everywhere or a pretty good fit in many places. But do you think any hay can be made out of whether structures and behaviours in the brain having similar architectures to what we see in nature can play a role in the restorative aspect?
Marc Berman
This is something I'm really excited about. Let me take a little detour. I became interested in how to quantify mental or cognitive effort. As a postdoc at the University of Toronto, working with a grad student, Nathan Churchill, who was doing interesting things with nonlinear dynamics, we applied some of these measures to data sets in the lab. I'd mentioned fractalness in space — but you can also think about fractalness in time. You measure one voxel in the brain over time and ask how fractal, or scale-free, that signal is: if you look at it at one second, ten seconds, sixty seconds, two hours, does it have the same characteristics? You convert the signal into the frequency domain and look at its power spectrum.
Signals whose power is proportional to one over frequency — this one-over-f shape — suggest a signal with repeated structure. More precisely, power is proportional to one over f to the H, where H is the Hurst exponent; when H equals one, the signal is perfectly fractal in time. We measured how fractal people's brains were during different cognitive tasks. When people did a very easy task, or no task at all, the signal was more fractal; when they did a very hard task, it was much less fractal — it deviated from that exponent of one.
Sean Carroll
Is it too cheeky to conceptualise that as focusing all our energies on one thing?
Marc Berman
Exactly. Maybe some frequency characteristic of the task comes to dominate. We saw this for other things too. Older and younger adults, titrated so they had equal performance — say both at 80% accuracy — and the older adults' brains were less fractal than the younger adults'. People under distress — who've just been told they have breast cancer — have less fractal brains than same-age people without that diagnosis.
All that was measured with fMRI, a slow signal, about one cycle per second. There's also EEG, with much higher temporal resolution — up to 50 or 75 hertz — but poor spatial resolution, so you know something's happening but not exactly where. I have colleagues at Chicago, Ed Awh and Ed Vogel, who use EEG to quantify how many things people hold in working memory. One way they do that is through alpha power, around 7 to 12 hertz: as alpha power changes, it tracks how many items are in working memory.
They do tasks where you're shown coloured squares that flash for about 200 milliseconds, then flash again with maybe one colour changed, and you report whether a change occurred. It's much easier with one square than with three, four, or six. They've found humans really have a working-memory capacity of about three items — show more than three or four and performance falls off. Here's what's interesting: if they show you six items versus three, in both cases you can only remember three, so your performance is the same. But with our Hurst measure we can distinguish those two — the signal is less fractal when someone tries to remember three out of six than three out of three. It seems like fruitless effort, since you can't grab the fourth item, but people are trying harder.
Sean Carroll
It's fun to conjecture about the brain, but there's still the empirical finding that attention is restored rather than depleted when we interact with nature. What can we do with that? Does it have therapeutic value, and implications for how we design our urban environments?
Marc Berman
Wide-ranging implications. There was a classic study by Roger Ulrich in the 1980s on recovery from gallbladder surgery, which back then was an inpatient procedure. On a single hospital corridor in Philadelphia, patients were in different rooms — some with a window onto modest nature, some trees and grass; others with a view of a brick wall. Ulrich found patients recovered about a day faster with the view of nature, and used less pain medication, than those with the brick wall. It was a quasi-experiment, since patients didn't pick their rooms — they were assigned to whatever was available. I don't think it was about air quality; there was something about the aesthetic of nature that had healing qualities, maybe partly through elevated mood.
There are tons of implications, but psychology as a science hasn't had as large an impact on society as you'd want. Economics has had a profound impact, partly because to make hay in society you have to talk in terms people can grab onto. It's hard for me to say what a 20% improvement in backwards digit span means economically. People will be more productive at work and kids will do better at school with more green space, but is it worth it? These things aren't inexpensive. As a field we need to quantify these effects in ways a policymaker could act on. Ulrich's study did have impact — people design hospitals now trying to incorporate more nature.
Sean Carroll
I wonder whether it's literal nature that helps, or whether there are aspects of nature we could incorporate into office or apartment design to get some of the same benefits.
Marc Berman
I think of it as a staircase. Nothing beats real nature. Windows or views onto nature can be good, but not quite as good as being out in it. Pictures of nature can work; even fake plants have been found beneficial; designing with the textures and patterns of nature, mimicking them without actual nature present, can have benefits. So: if you can't get into real nature, can you look at real nature? If not, can you bring some real nature inside? If not, can you look at simulations? It's a staircase like that.
It goes back to the first thing we discussed — humans underestimate how much the physical environment affects their behaviour and cognition, in part because we have so much control over it. Imagine we took that control and really designed buildings, cities, and schools with this in mind. It could have profound consequences. But I'd say a big part of what we're fighting against is that people view nature as an amenity, not a necessity. They say, 'That's nice if we had the money,' or 'That's for rich people.' We're trying to make the argument that it's actually a necessity — to achieve human flourishing, to be our best selves, we need to incorporate more of these natural elements into our lives. How much, and what kind, are still things we need to work through and quantify. If someone said, 'Marc, here's 20 million dollars, design a school that optimises kids' attention and working memory,' I'd have ideas, but I'd feel a little uncomfortable — I'm confident it would work, but whether it's worth that much, I don't yet know.
Sean Carroll
Well, like we said, psychology isn't done yet. That's the exciting part.
Sean Carroll
We chatted before recording that you had questions for me based on previous podcasts. I wanted to give you a chance to ask them — this is the least I could do.
Marc Berman
Great. When you were talking about many worlds, you said it's silly to think about the amount of energy many worlds would take up — that that's not a good argument. Could you explain why?
Sean Carroll
I've really struggled with this one. It's one of the rare questions where, if you know what the equations are saying, there is no question — the equations are perfectly transparent, and the only difficulty is translating them into natural English. When you say you're making another universe by doing a quantum measurement, in a very real sense you're not doubling the size of the universe; you're taking the universe you had and dividing it in two, slicing it into copies that are almost exactly the same. Both you and, say, a dumbbell you're about to lift, and the Earth and its gravitational field, all get sliced by exactly the same amount. There's less overall energy in your branch now, but relative to each other everything has the same amount as always. So even though the whole universe has half the energy as far as your visible world is concerned, it's completely unnoticeable, because you're embedded inside.
Marc Berman
Got it. Here's another. I've become interested in free will and consciousness. It's interesting that physicists seem quick to dismiss free will but find consciousness more mysterious. As a psychologist I think the opposite — consciousness isn't that interesting to me; the free will bit is more interesting because it brings up existential ideas. If we're really not in control and things are just following Schrödinger's equation, it almost seems like, what's the point? And yet we all feel such a strong sensation that we're choosing what we do — we don't have a Dr. Strangelove sensation where the universe won't let us do what we want. I tend to think of free will as an emergent property, though you had David Krakauer on who said you can't use emergence as a get-out-of-jail-free card for everything you want to be true.
Sean Carroll
For me there's a close connection. I don't understand at the detailed level how to account for free will or consciousness, but I have a strong conviction that whatever the account is, it'll be 100% physical. And it'll be emergent, because these things aren't fundamental — nowhere in the standard model are there consciousness and free will. That's enough to convince me that's the framework to work in, even though the details still have to be worked out. It surprised me talking to Christian List, who did a wonderful, detailed examination of what it means to say free will is emergent — the ingredients you need, how you connect them to what's going on in the brain. I said, 'It's kind of like consciousness that way,' and he said, 'Actually no, I'm not a physicalist about consciousness.' It's the same road he's walking down, but any good philosopher will sometimes convince themselves of counterintuitive things, because they think through things and don't start with the answer and work backwards.
Marc Berman
One thing I wonder is whether there's a conflation between determinism and causality. There are causes to my choices, but the part that would bother me most is if it's completely predetermined.
Sean Carroll
There's an old joke — I think John Searle told it — about people who don't believe in free will: when the waiter asks what you want, should you just say, 'Give me whatever the universe has determined I'm going to have'? There's not even a very solid door to push against here. I think it's confusing two different levels of description. At the microscopic level everything might be determined — of course it's not, because of quantum mechanics — but it's still mechanistic. Quantum measurement outcomes aren't determined, but they're not created by our willpower either; they obey the laws of physics even though they're not deterministic. The language we use to describe people, choices, desires, and goals is almost inescapable. I don't know any hard determinist who, the moment they stop talking about free will, doesn't instantly use the ordinary language of making choices, having reasons, and assigning responsibility. You don't have to deny any of that; you just have to figure out how it's all compatible. Maybe we're determined, but I don't know what the determination is, so it doesn't matter to me.
My colleague Jenann Ismael is the world's expert at this. She's explained carefully how we can't predict the future — not just because there's too much information to be Laplace's demon, but because we're in the universe, so it's easy to interfere with what's happening. If you modelled the universe and predicted one outcome, you could then act to prevent it, which stops you from ever fully predicting the future. To me that's a bit of a technicality where free will is concerned, but it's another reminder that we're not Laplace's demon, nor will we ever be.
Marc Berman
A guy like Penrose wants to say the brain is a quantum computer, so the Schrödinger equation can't just tell you how his brain activity will unfold.
Sean Carroll
What Penrose actually says is interesting and, to me, completely unconvincing. People — including, I think, Penrose — relate it to consciousness, but it doesn't even relate to consciousness. It's really just about how the brain works. Whether a certain quantum effect is happening in your brain doesn't help with the hard problem of consciousness at all. Penrose is concerned with cognition; he thinks human beings can reach conclusions computers can't, and that's where he brings in quantum mechanics. One of my disagreements is that he puts a lot of emphasis on Gödel's incompleteness theorem. Gödel says that in a sufficiently powerful formal system, if it's internally consistent — which, by the way, you can never prove from within — there will always be statements that are unprovable but true, statements analogous to 'I cannot be proven in this system.' Penrose says, therefore there are truths that can't be reached by computation. The response is: that assumed you knew the system was consistent. Penrose says he does know that, and the rest of us ask how. I think he's taking the rules of mathematics and logic a little too literally and applying them to the brain to reach very sweeping conclusions.
Marc Berman
Can you say more? If consciousness and free will are emergent properties, does that mean the quantum level can fully explain them, or can't?
Sean Carroll
The word 'explain' is doing a lot of heavy lifting. 'Consistent with' is the best you can say. Philosophers invented the word supervenience to handle this — it's fallen out of favour a bit, they've moved to grounding and other things, but the idea is: even if I can't derive the higher-level description from the lower-level one, I can still specify what it means for them to be compatible. Supervenience says I can't change the higher level without a difference at the lower level. Something like that is true for quantum mechanics and everyday experience. I don't know how to explain how the table in front of me is solid, or how to calculate the boiling point of water — I have friends who can, but I can't — yet it's doable, and I'm 100% convinced of that. The same is true for all of human behaviour. I have no idea how to do it, no one does, but I'm convinced it's doable.
Marc Berman
Did you read Sapolsky's book, Determined?
Sean Carroll
I did, and I had him on the podcast before the book came out. He emailed me a couple of chapters to comment on, and I thought they were bad. The psychology parts are fine, but the philosophy parts aren't good, because he can't wrap his brain around compatibilism — the idea that determinism at the lower level is compatible with our conception of free will. He tries to argue against it by insisting the lower level is really deterministic, and everyone else says, 'But we're compatible with that — that's the whole idea.' You can't argue against compatibilism by arguing for determinism; that's just a misunderstanding. The same goes for the Libet experiments — that you could predict what someone would choose before they choose it. That doesn't negate free will either. A compatibilist thinks that what we call free will is a higher-level collective behaviour emerging from real physical things happening in the brain, so an experiment illuminating those physical things is in no sense a rejection of the view.
Marc Berman
Could free will or choices play into many worlds, or would that be crazy?
Sean Carroll
My opinion has changed by a tiny amount recently. The short answer is no, because in many worlds you're not creating another world by making a decision. You're saying a quantum system in a superposition, becoming entangled with its environment, makes another world. That sounds much less sexy, but it's what's going on. The other way around could be true, though: it's not that decisions branch the universe, but that when the universe branches, that could be interpreted by us as making a decision. It's conceivable that if you're making a genuinely difficult decision — you truly don't know which way you'll go — there ends up being a substantial quantum-mechanical probability of going either way, because there are small quantum fluctuations in your brain, and part of brain behaviour is chaotic and amplifies tiny fluctuations. That wouldn't say your decision branches the universe, but it would say that when you make a difficult decision, there's another branch of the wave function where you chose differently.
Marc Berman
Super interesting. I hope it's okay if I bug you again sometime — I've been thinking a lot about this. I had a good talk with James Evans, who said we should write an article about it, and I said we'd need someone like Sean to help.
Sean Carroll
That sounds good to me. You know where to find me. I hope everyone buys your book, and I'm going to go outside and touch some grass, to follow your advice. Thanks very much for being on Mindscape.
Marc Berman
Thanks so much, Sean. It was a pleasure.