"We are opening the black box between the brain and the environment"

From urban stressors to the benefits of green spaces, environments shape mental health in powerful ways. Simone Kühn explains how research has just begun to understand the mechanisms behind these effects.
 

The Center for Environmental Neuroscience opened under your directorship in 2024. What is the main question it pursues?

For decades, neuroscientists have treated the human brain as if it exists in a petri dish: isolated, free of context, in no relation to the world in which we actually live. The concept of neural plasticity—the ability of the brain to change in response to experience, something first discovered in mice that, housed in so-called "enriched environments," formed new neurons—had long been discovered. Entire generations of neuroscientists acknowledged this concept, and then continued working away in the lab as if the world outside did not exist.

And your approach is different?

We ask, which environmental conditions promote neural plasticity, mental health, and cognitive function, and which detract? Surprisingly, this is a question that hardly anyone has systematically posed. We are on the brink of massive environmental upheavals: climate change, urbanization, the destruction of natural habitats, air pollution. It is increasingly urgent to understand what all that does to the human brain.

When did you personally start to systematically incorporate the environment into your research?

The German Socio-Economic Panel, or SOEP, was a key trigger. Early on, Gert Wagner, long-time director of the SOEP and a Max Planck Fellow, had pushed for the home address of participants to be gathered as part of regular data collection. From a data protection perspective, that’s complicated, but it represented a necessary new direction. Applying address data as a variable was decisive. Coupled with brain data from the Berlin Aging Study II, we could finally ask, whether there is a connection between the living environment of people and their brains. We were able to show that urban residents who lived near forests had a physiologically healthier amygdala structure and were likely better able to cope with stress than those who did not.

What do we already know about the influence of the environment on the brain?

Psychiatry and epidemiology had for years repeatedly found that mental illnesses—anxiety disorders, schizophrenia—are more common in cities than in the countryside. Urban space itself seemed to be a risk factor. Psychiatrists long interpreted this finding as: the city makes you sick.

But—in one of the most interesting findings of our work to date—it’s not that simple. Our data consistently paint a different picture: it is not that cities themselves are harmful, it is that natural elements in cities are lacking. When we give urban dwellers even a bit of green—a park nearby, a walk in the forest, plants in the office—positive effects can be observed. Positive effects on mood, cognition, measurable brain parameters. The evidence for the salutogenic effects of nature is now significantly stronger than evidence for direct harm caused by urbanicity per se. This represents a major shift in perspective.

You want to prove more than just "nature is good for you" though.

Exactly. That’s where most studies end and where we start. We are interested in what actually happens between our brains and the world out there. We want to open the black box and understand how external impressions are internalized in the brain, so that we can intervene to mitigate the negative effects of the city in a more targeted way. In doing so, the focus shifts to very specific elements: colors (e.g., green), the pattern of trees, the quality of the air, the sound of water, or the absence of noise.

Just as important are the pathways via which these impressions operate. Some might work on a purely visual level, others through the interplay of multiple senses, for example when we see our environment as well as hear and smell it. We want to understandwhich neural networks in the brain—affective or cognitive are involved (see Figure 1).

That sounds like a somewhat unwieldy research program. Where do you start?

With a concept that I describe as the funnel (see Figure 2). We work on multiple levels simultaneously—from broad and observational to specific and mechanistic—and on multiple timeframes. At the top of the funnel are large cohort studies: The Berlin Aging Study II, the Hamburg City Health Study, the UK Biobank, the German National Cohort. Contained therein are the data of hundreds of thousands of people for whom we have address data and whose living environments we can in turn precisely characterize: mapping tree coverage, air and noise pollution, access to green space. We can link that to brain data and thereby identify which environmental factors might be relevant.

And yet, such correlations do not tell us anything about cause and effect. That's why we need twin studies. Monozygotic twins share nearly all of their genetic material, grow up together most of the time, and go to the same school. When they then move to different environments in adulthood and show divergences in mental health or brain structure, this divergence can be traced back to the environment with a much higher likelihood. With this in mind, in 2022 we founded GERTRUD, Germany’s first twin registry. In other countries such databases have existed for decades, but Germany was missing one until now. It is a highly valuable instrument for distinguishing between genetic and environmental effects.

Further down the funnel are intervention studies. An example: we put participants in the magnetic resonance imaging (MRI) scanner before and after a walk in either a forest or on a city street and we measure the resulting differences. At the lowest level—the level of mechanistic acute effects—we work in the lab, where we can isolate individual environmental variables, like staging the same room with plants or without. Everything else remains unchanged. That's how we get closer to mechanistic explanations.

Which methodological approaches are needed in a young, still emerging research field?

Environmental neuroscience is not yet an established standard discipline with a fixed methodological toolbox. We still need to determine which methods are most suitable for meaningfully investigating these complex questions. Our methodological approach right now is correspondingly broad.

A cornerstone is MRI, which makes it possible to capture both brain structure and function, for example, changes in grey matter volume or the activity of certain networks. When combined with virtual reality (VR), this can be used very precisely: people lie in the scanner yet experience realistic environments, and we can directly observe what is happening in the brain as they are exposed to a specific setting.

These techniques in the lab are complemented by methods that get us closer to the everyday. Functional near-infrared spectroscopy (fNIRS) plays an important role here: light is directed into the brain and measured to see how it is reflected depending on the oxygen content of the blood. The underlying signal is very similar to that of an MRI, but the advantage lies in its portability: participants can get up and move around during fNIRS, which enables totally different study designs.

We also record real-world environmental exposures via GPS tracking and link location data with high-resolution environmental information, such as on air quality, noise level, or vegetation cover. This allows us to map long-term environmental influences as realistically as possible.

You recently acquired a mobile MRI scanner. What does this enable?

We can use the mobile 1.5-Tesla scanner directly for field studies. This allows us, for example, to scan people immediately after a walk in the woods, without them first having to travel an hour to the laboratory—by which time the relevant brain activity we want to study has partly abated. The temporal proximity to the exposure represents a real leap in quality. Previously, we had to compromise between precise measurements in a lab setting, far from actual experience, or mobile measurements out in the world, with all the attendant artifacts and lack of comparability. The mobile scanner gives us both: proximity to the field and lab-quality results.

What is the aim of combining these diverse methods and approaches? Why is this necessary?

We rely on converging evidence. If the same effects are observed across diverse methodological approaches—using MRI and fNIRS, in highly controlled VR studies as well as in GPS-based real-world data—then we can speak of a robust phenomenon. The diversity of methods is therefore a deliberate strategy to do justice to the complexity of the subject matter.

In the long term we hope that this diversity will lead us to be able to identify the "active ingredients" of the environment: the individual elements and combinations of environmental factors that are actually relevant for mental health. Only when effects prove stable across methods, timescales, and contexts can we begin to speak of fundamental mechanisms or effects.

You do not just investigate natural environments, but built and digital environments as well. Do they really fall under the same research area?

Absolutely. We spend around 80–90% of our days indoors. How does a hospital's architecture affect the recovery process? What does it do to children when their school is bordered by a green courtyard versus a heavily trafficked road? Such questions are often far more relevant to people's everyday lives than the occasional walk in the woods at the weekend.

Meanwhile, the use of digital media has become a common part of everyday life for many people. With this in mind, we have begun to examine the effects of digital environments—particularly social media—in greater detail. As a first step, we are collecting data on the nature and duration of social media use, as well as on participants' self-reported mental health.

Behind this is the heated public debate on the potential effects of social media use on mental well-being, for which there is currently astoundingly little scientific evidence. While our previous findings have shown a clear association—intensive social media use is linked to greater psychological distress—it is worth pointing out that correlation does not equal causation. To reliably clarify cause-and-effect relationships, we need carefully controlled longitudinal studies, ideally using designs that account for genetic and environmental factors, such as twin studies.

How important are interdisciplinary cooperations for your research and how do they come about?

They are indispensable. The environment is a highly complex system that can only be understood to a limited extent from a purely neuroscientific perspective. That is why we actively seek collaboration with disciplines beyond our own.

We collaborate with toxicologists who can identify which environmental substances are detectable in blood and which actually reach the brain. Chemists help us to model how air pollutants spread in cities, and in the field of olfactory research we work with labs that use mass spectrometry to quantify which substances are inhaled during a walk in the forest.

Together with colleagues from the Department of Psychiatry and Psychotherapy at Charité—Universitätsmedizin Berlin, we have founded the working group Psychiatric Environmental Neuroscience (PEN). This enables us to transfer findings from healthy volunteers to clinical populations, for example to investigate whether urban environmental factors have different or stronger effects on people with schizophrenia.

These collaborations often come about astonishingly easily: from an email or a conversation. It is my impression that there are a surprising number of people who are all too glad to have their expertise called on when it comes to exciting research. The Max Planck network is, of course, hugely helpful in this regard. The combined expertise under one roof is enormous.

What do your findings mean for our living environments? For our own daily lives right through to the design of cities and buildings?

At an individual level, it is certainly possible to draw concrete conclusions. Measures such as a picture of nature on the wall, plants in the office, or a brief walk in the park are now known to have well-documented, measurable effects on stress and well-being. Such interventions are low-threshold and can be implemented relatively quickly.

However, when it comes to urban planning and policy decisions, I am much more cautious. There are studies that calculate very precisely just how many trees would be needed to statistically lower blood pressure. This is then quickly translated into the message: more trees, fewer heart attacks. That is not fundamentally wrong, but it is not thought through. Such findings do not help us to determine whether it would make more sense to allocate additional funds to green spaces or to invest this money elsewhere, for example in the health sector. For robust recommendations in urban and building planning, we therefore need more transfer research: research that systematically examines what specific environmental measures actually achieve compared to other societal investments.

What are the future directions?

The next big question is that of universality. Whether what holds true for our Western, urbanized populations is also applicable in other cultures, or whether there are interindividual differences; whether the environment in which I grew up fundamentally alters how I react to my surroundings. There are fascinating results on this: where we see square shapes, other cultures see circles, for example. This means that the same physical world literally looks different depending on the environment in which one grows up. If this applies to such fundamental things as perception, what does that mean for the effects induced by nature, architecture, and urban planning? This will keep us busy for a long time to come.