What this covers

Sean Carroll talks with psychologist Marc Berman, whose environmental neuroscience work and recent book examine what natural settings do to cognition. The conversation starts from Attention Restoration Theory: voluntary attention fatigues with use, and soft fascination, like a waterfall that lets the mind wander, can restore it. Harshly fascinating places such as Times Square cannot. Berman separates overt attention (eye movements) from covert attention, and notes that the dangers of phone calls while driving come from diverted attention, not from holding the phone.

The evidence runs from Roger Ulrich's gallbladder-surgery study, where a view of trees went with faster recovery, to Berman's own walking experiments. In a within-subject design, about 50 minutes in nature gave roughly a 20 percent gain on backwards digit span, and an urban walk gave none. That held even for people who disliked a freezing January walk. Further results include stronger effects for depressed participants primed to ruminate, better attention and school performance in children in greener neighborhoods, and a Chicago public-housing quasi-experiment. Berman also lays out a "staircase" from real nature down to pictures and simulations.

The mechanistic turn is information-theoretic. Nature images compress more under a JPEG-style algorithm, and the Hurst exponent of brain signals appears to drop with task difficulty, age, and a breast cancer diagnosis. Carroll's own interests then pull the talk sideways, into supervenience, compatibilism, his doubts about Roger Penrose's Gödel-based argument, and the Many-Worlds interpretation.

Sharpest takeaway

Berman argues that natural environments improve attention and working memory through their perceptual structure, not just mood, and that modern built environments reshape our brains without being designed for cognition.

  • Objective digit-span gains after nature walks held even when participants disliked the cold winter walks, so the cognitive effect is not just enjoyment.
  • Nature scenes are more compressible and fractal, which may let the brain process them with less effort; picture, sound, and video studies show weaker but similar effects.
  • Attention Restoration Theory says directed attention is depletable and can be restored by softly fascinating stimuli, which natural settings provide and harshly fascinating urban and digital stimuli do not.

The argument · threads11 threads · 63 claims
0.81

Nature walks produce a working-memory gain that is not explained by improved mood.

6 pointscentrality 5/5
  • Although mood improved after nature walks, the size of mood improvement did not strongly correlate with the size of the backwards digit span improvement, so the cognitive gain is not explained by people who became happier performing better.

    “we didn't see a strong correlation between the improvements in mood and the improvements in this backwards digit span task”

  • In a within-subject design where each participant walked about 50 minutes in a natural and in an urban environment one week apart, participants showed about a 20 percent improvement on the backwards digit span task after the nature walk, roughly a digit and a half, while the urban walk produced no significant improvement.

    “when people walked in the natural environment, they showed about a 20% improvement on this backwards digit span task. So they improved by like a digit and a half compared to the urban walk where they really didn't show any significant improvement.”

  • Participants who walked in freezing January conditions and disliked the walk still showed the same working-memory improvement as those who enjoyed June walks, while mood benefits required enjoyment, so the cognitive effect of nature is not a simple hedonic reward.

    “the people that walked in January showed the same cognitive benefit, the same improvement on the backwards digit span task as the people that walked in June. So the takeaway message was you don't even have to enjoy the nature walk to get these cognitive benefits.”

  • Berman is confident the nature-walk cognitive findings are robust because many other labs have replicated them, although the original study had only about 40 participants in a within-subjects design, equivalent to roughly 80 between-subjects participants.

    “no one has IDed this effect as being something that's not replicable. So many other labs have replicated these effects.”

  • The nature effect should be read as a cognitive-performance effect rather than a mood effect, because earlier studies relied on self-report and this work used objective performance measures.

    “So we did one of the first studies where we actually measured objectively people's changes in performance after interacting with these different environments.”

  • The backwards digit span task presents digits at about one per second and requires recall in reverse order, increasing up to about nine digits, with around five digits being the point where it becomes very challenging.

    “you would hear digits out loud at about one digit per second, and you would need to repeat the digits back in backwards order.”

0.67

Restoration requires softly fascinating involuntary attention, not passive rest or harsh stimulation.

6 pointscentrality 4/5
  • Involuntary attention captured by softly fascinating stimuli, such as a waterfall that lets the mind wander, can restore directed attention, whereas harshly fascinating, all-consuming stimuli like Times Square prevent mind-wandering and do not restore it.

    “we think the kind of stimulation that captures your involuntary attention should be softly fascinating and not harshly fascinating”

  • Restoring directed attention is an active process that engages involuntary attention, not a passive one like sleep, so mental fatigue differs from physical fatigue and needs a task switch to recover.

    “this isn't just like sleep. This is sort of an active process of engaging maybe a different kind of attention”

  • Sitting in a dark room does not restore attention because most people find it boring, and boredom is fatiguing, so restoration requires engaging stimulation of the right kind rather than passive rest.

    “most people don't like sitting in dark rooms. They find it boring, and boredom is fatiguing”

  • Light video games that are engrossing but undemanding could be compatible with attention restoration, but modern kids' video games and digital technologies are harshly fascinating, designed to hold attention and addictive, and most addictions are bad for cognitive functioning.

    “a lot of these new technologies, I would say, are very harshly fascinating. They're designed to keep your attention and not let go, and they're addictive”

  • Attention Restoration Theory holds that humans have directed (top-down) attention that is voluntary but fatigable, so after sustained effort people become mentally fatigued and cannot focus, and that taking a break in nature is a good way to recover.

    “this directed attention ability is really important because it shows that humans are not just stimulus driven... we can really only do that for a limited amount of time before we've become mentally fatigued”

  • Involuntary (bottom-up) attention corresponds to Kahneman's System 1 fast, heuristic thinking, while directed (top-down) attention corresponds to effortful System 2 thinking.

    “involuntary attention is more like System 1 and directed attention is more like System 2”

0.57

Restoration depends on moderate visual complexity and fractal structure, not maximal entropy.

7 pointscentrality 3/5
  • Grayscale entropy measurements of the nature and urban images show a smaller difference than the entropy hypothesis predicts, so color channel information may be needed to explain the processing difference.

    “we've run grayscale entropy measurements on these images, and that doesn't show as big of a difference as we would think. Maybe we have to do something actually with the color entropy”

  • Our language for nature is semantically simple (trees, shrubs, grass), while urban environments have many more named objects and categories (brands, architectural styles, signs), which may make urban scenes harder to process but possibly easier to remember.

    “We don't really have a language of nature like we do for the urban environment. If I go and see a forest, I might say trees, shrubs, grass. It's not very semantically complex.”

  • Restorative images likely need the right amount of complexity rather than maximal entropy, because the highest-entropy image is white noise, which is not a complex image in the psychological sense, and complexity peaks before entropy keeps rising.

    “for an image to be restorative, I think actually it's gotta have this kind of right amount of complexity and maybe not too much entropy”

  • Perfect statistical fractals would be too boring to restore attention; restoration likely comes from a balance where predictions are mostly met but with some jitter, as in a curved forest path that hides what lies beyond, which captures attention in a softly fascinating way.

    “the brain likes predictability, but maybe not perfect. There's something interesting when something deviates a little bit, not too much”

  • When human-rated nature images are run through a JPEG-style compression algorithm tuned to human vision, more natural images compress to fewer bits than urban images, meaning more information is discarded from nature scenes because they contain more repeated structure.

    “more natural images tended to get compressed down into fewer bits than the urban images, suggesting that there is more compression, there's more throwing away of information in nature scenes than urban scenes”

  • Natural scenes have high fractalness, meaning the same structure repeats across scales, as in a maple leaf's veins mirroring the tree, so viewers can grasp the gist of a natural scene quickly by looking in or out, whereas urban scenes look different at each scale and are harder to get the gist of.

    “there's some ease that the brain has when there is this fractal structure where you can get the gist very quickly of a scene”

  • Simple measures of complexity are small for perfect order and large for pure randomness, but the useful quantity peaks in between, and logical depth is a relevant candidate measure because structured patterns the brain can latch onto are neither trivially compressible nor random.

    “that actually peaks in between. That's a whole full employment for complexity theorists to invent different ways to do that. Logical depth is one idea that I think is very relevant here.”

0.53

Field evidence links nature to health and attention, though school cost-effectiveness remains unknown.

6 pointscentrality 3/5
  • In the UK and Canada, physicians are now prescribing walks in nature as a supplemental therapy for depression and anxiety.

    “now there's actually, in the UK and in Canada, physicians are prescribing walks in nature as sort of a supplemental therapy for people suffering from depression and anxiety”

  • Studies quantifying green space around homes and schools find children living in greener neighborhoods have better attention and working memory and better school performance, even after controlling for age, parental education, and income.

    “children that live in neighborhoods that have more green space have better attention and working memory, even when you control for things like age, education, and income of their parents”

  • In a quasi-experiment in Chicago public housing, where families were randomly assigned apartments, residents of units with views of nature (measured by satellite imagery of surrounding green space) showed better attention, and the area had lower reported crime and aggression.

    “the apartments that had the views of nature, people had better attention in those apartments. There was lower reported crime, lower reports of aggression”

  • In Roger Ulrich's 1980s study of gallbladder surgery patients in Philadelphia, patients randomly assigned rooms with a view of trees and grass recovered about a day faster and used less pain medication than those facing a brick wall, likely through aesthetic or mood effects rather than air quality or exercise.

    “patients recovered about a day faster if they had the view of nature out of their hospital window, and they used less pain medication compared to the patients that had the view of the brick wall”

  • People often treat nature as an amenity for the rich, but Berman argues it is closer to a necessity for human flourishing and cognitive functioning, and the field is working to make that case.

    “people view nature as an amenity and not a necessity. So they say, "Oh, that's nice if we had enough money," or, "Oh, that's for rich people."”

  • Berman is confident that more green space, outdoor access, and natural patterns in schools would help, but he is not sure whether the benefits justify large costs like $20 million relative to alternatives, because the cost-effectiveness is still unknown.

    “It'd be hard for me if you said, "Okay, Marc, here's $20 million. Is it worth it to do that versus doing something else?"”

0.49

Environments shape brains bidirectionally, and built environments are not designed for cognition.

5 pointscentrality 3/5
  • Humans tend to feel immune to their environment because they control it so much, but people are part of the natural world, and the relationship is bidirectional: brains were shaped by the physical environment, and the environments people build now shape brains in turn.

    “we aren't immune to the environment and we're a part of the environment”

  • Environmental neuroscience, a field Berman helped develop, studies the bidirectional relationship between the physical environment and the brain, and his lab finds that short exposures to natural rather than urban environments change attention and working memory.

    “this field called environmental neuroscience that I kind of helped to develop, which is very interested in the bidirectional relationships between the physical environment and our brain”

  • Much modern architecture and design is built to house people efficiently and move goods efficiently, not to promote cooperation, working memory, or attention, so the artificial environments people now live in may be changing behavior and brains without deliberate design.

    “It's kind of just built to house people efficiently, move goods efficiently.”

  • Humans underestimate how much the physical environment affects their behavior and cognition, partly because they have so much control over it, and if people deliberately designed buildings, cities, and schools around cognition the consequences could be profound.

    “humans underestimate how much the physical environment impacts their behavior, impacts their cognition, in part because we have so much control over it”

  • Environmental tendrils, including embodied and extended cognition, reach into people in crucial ways, so it is folly to treat the mind as a brain in a vat or imagine uploading oneself without considering environmental effects.

    “There are tendrils of the environment that reach into us and vice versa in crucial ways.”

0.35

Brain fractality tracks cognitive effort and strain during tasks.

5 pointscentrality 2/5
  • Functional MRI is too uncomfortable and artificial for nature experiences, so the lab is using functional near-infrared spectroscopy, a mobile cap that measures blood oxygenation, to study people walking in natural versus urban settings.

    “This is mobile. People can wear this with a little backpack, and we can have them walk in natural environments versus urban environments”

  • Although people can only hold about three items in working memory, the Hurst-based fractality measure distinguishes trials where people try to remember three items out of six (less fractal, more effort) from trials of three out of three, showing the brain works harder even when the extra item cannot be retained.

    “We can tell if somebody is trying to remember three things out of three or three things out of six because the Hurst exponent... is actually less fractal trying to remember three things out of six”

  • At matched task accuracy, older adults show less fractal brain signals than younger adults, and people who have just been told they have breast cancer show less fractal brain signals than same-age peers without that diagnosis, suggesting fractality tracks strain and resources.

    “People who are under distress, they've just been told that they have breast cancer. The brain is less fractal than people who are same age but don't have that kind of diagnosis.”

  • Measuring the Hurst exponent of brain activity over time shows the fMRI signal is close to fully fractal (H near 1) during easy or resting tasks, but less fractal (H around 0.7) during hard tasks, so cognitive difficulty reduces the temporal scale-free structure of neural signals.

    “when people were doing a very hard task was much less fractal than when they were doing the easy task”

  • Colleagues Awh and Vogel use EEG alpha power (7 to 12 Hz) to estimate how many items people hold in working memory, since alpha power rises as more items are held, though EEG polarity conventions can flip the sign.

    “they find that alpha power, which is like 7 to 12 Hz, that actually tells you how many things people have in working memory”

0.27

Real nature restores more than images, with effects weakening across exposure steps.

3 pointscentrality 2/5
  • Showing people nature pictures, sounds, or videos produces similar but weaker and less reliable changes in backwards digit span than real nature, suggesting the perceptual qualities of nature, not leisure or vacation thoughts, produce the benefit.

    “something about the perceptual features of nature that might be producing these benefits. So then that kind of gets away from it just being leisure time per se.”

  • Showing people nature versus urban images in an MRI cannot reveal why nature restores, because the images differ in many features at once, including fractalness, entropy, curved edges, and color palettes, and it is unclear which feature drives the performance change.

    “There's many, many features that differ between the images. The nature images have different fractalness, different entropy, different amounts of curved edges, different color palettes.”

  • Nature exposure works on a staircase: real nature is best, then views of real nature, then bringing real nature indoors such as plants, then pictures and simulations, with each step producing weaker but still useful effects.

    “So I kind of look at it as a staircase like that. nothing beats the real nature... Pictures of nature can work. Even having fake plants”

0.27

Psychology lags physics and economics in quantified, cumulative, policy-usable theory.

3 pointscentrality 2/5
  • Psychology has had less societal impact than economics because it has not quantified findings in terms policymakers can use, such as the economic value of a 20 percent working-memory improvement or of green space in schools.

    “psychology as a science has not had as large of an impact on society as you would want. Economics has... we need to really quantify these things more in ways that I think a policymaker could really take them and do something with it”

  • Behaviorism (Skinner and Watson) failed because mapping stimuli to responses ignored the substantial processing occurring in the mind and brain, and the field has since settled on a middle ground that recognizes internal processing without relying on introspection.

    “that's why behaviorism failed in psychology... that turned out to be way, way too simple. There's a lot that's happening in the mind or the brain”

  • Psychology has not yet reached the level of law-like formulas that physics has, and even defining a concept like mental energy is not trivial, which limits the field's ability to build cumulative theory.

    “We don't have concepts like even defining mental energy is not trivial, very complicated. So we're lacking those kinds of principles that a field like physics has”

0.22

Nature may help cognition by calming anxiety and reducing negative rumination.

2 pointscentrality 2/5
  • When clinically depressed participants were induced to ruminate on a negative memory before walking, the nature walk produced even stronger working-memory improvements than the urban walk, suggesting nature benefits may be greatest when negative rumination is high.

    “we had people think about a negative thought or memory that was bothering them before they went on the walk... they showed even stronger improvements in backwards digit span performance after the nature walk”

  • Berman suggests that nature may help cognitive performance by calming the nervous system, since participants found the digit-span task anxiety-provoking and nature could reduce that anxiety.

    “if nature is sort of relaxing us a little bit more, maybe you can perform better.”

0.17

Free will is an emergent property not reducible to determinism and deliberation.

4 pointscentrality 1/5
  • If choices were purely determined like a physical solving process, there would be no reason for deliberation time, such as deciding between salmon and spaghetti, so the existence of deliberation is evidence against a naive input-output picture of choice.

    “why is there this time delay if everything was just sort of determined? ... it would just solve it and you wouldn't have this deliberation”

  • Berman finds free will more existentially interesting than consciousness because having conscious experience is unsurprising, while the sense of choosing raises questions about whether determined behavior still feels chosen.

    “consciousness is not that interesting to me... The free will bit is a bit more interesting to me because it brings up these existential ideas.”

  • Berman asks why other Many-Worlds branches cannot be detected; Carroll answers that good explanations exist, and that Many-Worlds is hard to accept mainly because it asks much of listeners metaphysically.

    “There's good explanations for why we can't, but I think I'm absolutely sympathetic to people who find Many-Worlds hard to swallow.”

  • Berman argues that free will is an emergent property that does not need to be fully fixed by quantum-level facts, and that the unpredictability of choice and deliberation time indicate a process not reducible to a simple input-output solution.

    “I do think of free will as sort of being this emergent property.”

0.09

Covert attention, not eye or hand contact, determines whether processing occurs.

2 pointscentrality 1/5
  • The danger of talking on a phone while driving comes from attention being diverted away from the road, not from physically holding the phone, so hands-free calls are still dangerous and the 'peripheral hypothesis' is wrong.

    “it turns out it has nothing to do with holding the phone and it has everything to do with attention”

  • Eye movements (overt attention) do not guarantee that the brain processes what the eyes receive; covert attention, the internal focus of processing, can be elsewhere, as when a driver on a highway forgets the scenery they passed.

    “the light from the road and from the environment is all hitting your retina... the brain is not attending to it. So that's the difference between what people call overt attention and covert attention”