One of the reasons I like writing about city parks, and about landscape architecture, is that they have a long connection with the general health and welfare—but the reasons for that connection change over time. In other words, they're always for the betterment of people, but the reasons we understand them to better us change as we understand more about ourselves.

There's also something profoundly human about it. Within a built, urban, decidedly human environment such as a city, we dedicate spaces to "nature"—but a natural landscape planned to the inch by people. When I talked to the landscape architect of the 606, for example, he described how that abandoned rail track was rebuilt from the ground up to reflect its state when it was turned into a park.

In other words: it was originally nature; it was designed into an industrial corridor by humans; it was abandoned and nature began to reclaim it; it was then designed into an space that looks like an industrial corridor returning to nature. We take nature apart, then try to put it back together again. Why go to all that trouble?

These impulses—for natural environments for health and welfare, and the granular understanding of them—cross paths in the work of University of Chicago psychologist Marc Berman. He's pursued a breadth of inquiries in this field. Recently, he published a study, conducted in Toronto, that found "an additional ten trees on a given block corresponded to a one-percent increase in how healthy nearby residents felt," which may not sound like much until it's converted into socioeconomic terms: "For the people suffering from these [cardiovascular] conditions, an extra eleven trees per block corresponds to an income boost of twenty thousand dollars, or being almost one and a half years younger."

In that study, Berman was looking at effect, not cause (trees could create cleaner air, encourage people to exercise, or have subtle psychological effects suggested by research Berman and others have pursued). But he also studies cause, at an intensely fine-grained level, attempting to figure out what looks "natural" and why, using effects that will be familiar to power Photoshop users, like edge detection and hue and color saturation, breaking down nature into its most basic visual elements.

When I first read about Berman's tree study in the New Yorker, my mind went immediately to evolutionary psychology, influenced by having lived in the California desert while in college: trees grow because of water, people need water, etc.

But one explanation for the Toronto results could be another idea Berman has studied: attention-restoration therapy, which has its roots in the work of William James. To greatly simplify it: letting your mind wander is good, and a natural environment is a good place to let your mind wander.

At work, or on a city street, or at home, the things we do require active attention, whether it's talking on the phone, or crossing the street, or watching TV. Many things in nature, on the other hand, encourage "involuntary" attention, like the movement of trees or the light reflecting off water. You don't really pay close, engaged attention to how a river moves in the way you pay attention to a person or show or a traffic light. Involuntary attention, perhaps, restores the mental fatigue that comes from active attention.

It's not a matter of nature being more "simple"; what we recognize as natural tends to have more visual information, at its most basic level, than what we recognize as urban. It's more a matter of how we engage with that information. Which has potentially profound implications: if it's really just a matter of visual (and auditory) patterns, perhaps these restorative environments don't have to be natural at all, and it just so happens that natural environments tend to have those qualities.

Which, of course, would have profound implications for the built environment (it's not for nothing that some of the research in this field focused on green space in Chicago public housing) and how architects and planners make healthier spaces, shaping the city around how our brain perceives it. I spoke with Berman about this possibility, where attention-restoration theory comes from, and how he's pursuing his future research to get to the building blocks of what nature does for (and to) us.

What are the origins of attention-restoration theory? I understand that it goes back to the dawn of psychology.

It's a theory proposed by Steve Kaplan, but it's based a lot on some seminal work at the early state of psychology by William James, where basically James proposed that we have two types of attention. One kind of attention is called directed attention, where you as an individual person sort of decide what you're going to be attending to. For example, if you're listening to me talk, you're deciding to attend to me, even if there may be more interesting things in the environment to attend to. It's thought that that kind of attention is very effortful, and is susceptible to fatigue or depletion, that you can only focus or direct your attention for so long, before you become mentally fatigued, and you can't really focus so well anymore.

That's counter to another form of attention, called involuntary attention, or bottom-up attention, and that's the kind of attention that's automatically captured by interesting stimulation in the environment. Loud noises, bright lights, those automatically capture your attention; you don't really have any control over it, and it's thought that kind of attention is not fatiguable or depleteable, or is more immune to fatigue. So you don't often hear people say, wow, I'm exhausted looking at that beautiful waterfall, or I can't handle staring at that beautiful painting anymore.

The idea behind attention-restoration theory is that if you can find environments that don't require a lot of vigilance or directed attention, while simultaneously having stimulation that captures or activates the involuntary attention, that you might be able to sort of replenish, or restore, this directed-attention resource.

How did that theory end up being translated into the idea that natural environments would be such an environment?

Right—the theory doesn't say that the environment has to be natural. But natural environments that we tend to encounter on a more or less daily basis tend to meet those criteria. Parks in urban areas tend to not require a lot of directed attention—you don't have to worry about getting hit by a car, bumping into people, not a lot of advertising, things like that, that would require directed attention, while simultaneously having interesting stimulation to softly capture involuntary attention.

It's important that the type of stimulation be what we call "softly fascinating." Walking through Times Square, it also has a lot of interesting stimulation that will capture your attention automatically, but it does so in an all-consuming way. Whereas we think a lot of natural stimulation captures your interest, your fascination, in sort of a soft way, so it's not all-consuming.

We're always told to let your mind wander. And that sounds like what it's actually all about.

What we're interested in now is to see, when you interact with nature, does your mind wander more than when you're interacting with urban environments?

One thing that was interesting to me is that when you're outdoors, the information out there in a natural environment—visual and auditory—it is actually a lot of information. It's not like it's not very busy. So it's not a matter of reducing information.

We also find that if we show pictures of nature, versus pictures of urban scenes, we find some similar benefits, although the benefits aren't nearly as strong as going out into real nature.

So one of the things we've started to do is to say, okay, what are the visual dimensions that distinguish nature scenes from urban scenes. And there are many different things that differ. For instance, nature scenes tend to have curved lines, urban scenes tend to have more straight lines. Nature scenes tend to have higher color saturation, or more pure color than urban scenes.

And also, if we grayscale the images, and sort of try to quantify the amount of information in the scene, nature scenes actually tend to have  more information than urban scenes. What I mean by that is that urban scenes tend to have more redundancy. If you know one pixel value in an urban scene, you can kind of more easily predict other values. That's not the case in nature scenes.

So your intuition may be right—there may actually be more information in nature scenes. However, if we take it a layer up, and not just look at the pixel value, and say which scenes have more objects, urban scenes have more nameable objects than nature scenes. So that's sort of a different form of information, and that kind of information may actually be somewhat fatiguing.

Let's say, when you're viewing a nature scene, there's lots of information there, but you can just survey the scene, and get the gist of it. When you see an urban scene, we might start engaging in this process of visual search, identifying all the objects in the scene. And that may be somewhat fatiguing or tiring.

One of the things we're interested in looking at now is to see, if we look at your eye movements, when looking at a nature scene versus an urban scene, what differs? When you look at an urban scene, and we don't tell you any goals, do you automatically engage in a visual search process? Whereas, when you view a nature scene without any instructions, do you engage in eye-movement patterns that look more like just trying to figure out if you like the scene or not?

I've experienced this when walking with my two-year-old in the city. There's practically an infinite number of things I can point out and name for her.

Right. And our brains may be sort of automatically geared to try to identify all of those things. And that may take energy.

In your recent study about Toronto, you found not only that cardiovascular health influenced by the presence of trees, but also that the  perception of health was influenced by it. What's the relationship between the perception of health and actual health?

First I would say that there's a strong correlation between mental health and physical health. They're related to each other. When we ask people how healthy they feel, that correlates pretty well with their objective health measures—do they have diabetes, do they have a stroke, do they have heart disease. Those are related to each other, but they're not perfectly correlated.

So certainly there can be people who feel healthier than they are, and people who feel less healthy than they actually are. What we're finding is that having more trees on the street in your neighborhood actually improves both subjective health and objective health.

What cues can planners, designers, and landscape architects take from what we're learning about what the brain identifies as natural?

We're starting to work with urban planners and architects, and there's already landscape architects that work in this area, such as Bill Sullivan at the University of Illinois. We have a session at the American Society of Landscape Architects conference where we're going to be talking about our research on the natural environment, and how our research can inform the design of built spaces, to improve people's psychological and physical health. That's kind of the goal—we want our research to be able to inform design.

I think we're getting there. The trick is that there's still more research that needs to be done. For example, in the tree study, I feel pretty confident that planting more trees on the street should have health benefits, but we don't really know the mechanism as to why the trees are related to improved health. Is it better air quality? Is it that having the trees on the street encourages people to exercise more? Is it just having this more visually beautiful environment has health effects? Do the kinds of trees matter? We're seeing some evidence that you might get a little more of a boost from trees that retain their leaves versus not… why is that?

I think that the research we're doing now, where we're trying to break nature scenes and urban scenes down into their low-level components, is going to have a lot of implications for design. One thing we know already is these low-level visual features, and how they relate to people's preferences for the scene. We know, for example, that people tend to like images that have a lot of curved edges; they tend to not like images that have a lot of straight lines. They tend to like images that have a lot of color saturation. Those are some of the design features that make people like nature scenes more than urban scenes.

One thing I thought of is that, in building in these aspects to the city, we tend to think in two buckets—the city, and then maybe a park in it, so that people can get out of the city in some small way. But maybe with this knowledge, there could be more of a spectrum.

We don't want people to move out of cities; there's lots that's positive about cities. And we don't want to destroy the existing nature that's already out there. What we're suggesting, what we're most interested in, is actually incorporating natural elements into the city, to make the city more livable, more healthy. Now the trick is trying to identify what kind of nature do we want to incorporate into the city, and if we make different design choices, how does that impact people psychologically and healthwise.

And then we have these tough decisions: planting more trees costs money. Is it more cost-effective to do that, or more cost effective to do something else? We need to know the answers to all of these things.

Are we starting to learn more about the neuroscience of this? That's something that we're really interested in doing, but there's a couple things about that. One is that we don't really have the technology now to have people go into nature and record brain activity while they're actually having a nature experience. 

The technology that I primarily use to look at brain data is FMRI—you've got people in this tube, and it's very noisy, and it's an artificial environment. It's complicated; we need to be able to simulate the nature experience in this artificial environment. We can get some nature effects just by showing people pictures, but there's going to be a little bit of a problem with that, because we don't know what's going to be driving the effect. Is it that because these nature images and these urban images differ on all these visual properties, so are we getting differences in brain activities because people are just seeing more green versus something else? Is it because they're seeing more curved edges? Or is it just that the mere thought of nature brings the benefit?

So what we're planning to do, in a more sophisticated study, is where we isolate these different things. So maybe we just show people nature words; show people images that are scrambled, that they don't know if it's a tree or a building, but it has all the low-level visual features of trees and buildings. We want to do a sophisticated study to do this; we don't want to just throw people in an MRI scanner and show them nature pictures and urban pictures, because that's going to be somewhat informative, but there's going to be a lot of questions still to answer. We're still trying to figure out what's the right kind of design. We've even started to do studies with virtual reality, that make the experience more similar to an actual nature or urban experience.

I'd assume that movement would be a part of it. When I think about being in the woods on a windy day, you're not just seeing trees, you're seeing leaves in motion.

With a static picture, you get some of the effect, but it's not as strong. That's an important thing—what is the movement's contribution? That would be something really interesting to know. There have been studies for looking at cubicles, and whether the cubicle has a window to nature, versus an LCD screen that has nature pictures being displayed on it. And you get some benefit from the LCD screen, but it's not as strong as a benefit as actual views of nature. And maybe that's because the actual views are more dynamic than these static images. There is some indication that your intuition may be right.