Finding Nature

My latest research in the Journal of Environmental Psychology explores how everyday life conditions have shaped mental health discourse and prevalence over the last two centuries. Mental health difficulties have increased across many societies and this suggests that mental health is not just a condition of mind, but also a condition of life. It is shaped by a complex interplay of structural and cultural factors. Economics matter. Education matters. Urbanisation matters. So do technological change, social exclusion, and the physical environments in which people live. Over recent decades, psychology and public health have steadily widened their lens to recognise these broader determinants of mental health.

Yet despite this broadening of scope, one foundational aspect of human experience remains largely absent from how mental health is usually explained: the human–nature relationship. This omission is not trivial. For example, the American Psychiatric Association’s Task Force on the Social Determinants of Mental Health makes little reference to nature at all.

This matters as in this latest research across this long historical view, two foundations of good mental health stand out.

1. The importance of economic conditions—reflected in how much cultural attention is given to issues of work, income, prosperity, and insecurity.
2. A close relationship with the natural world—reflected in how much cultural attention is given to nature.

Together, these factors appear to explain long‑term mental health trends – a trebling since the 1950s and doubling since 1980 – more so than many of the determinants that are mentioned more often.

What did the study do?

The study took a long historical view, examining how often mental health‑related terms have been used in English‑language works over more than two centuries (1800–2020). Large‑scale word frequency data was used as a proxy for societal concern, stress, and broader mental health trends.

A carefully selected set of mental health‑related terms was compiled from five historical eras. Statistical techniques were used to ensure these terms reflected a coherent underlying concept, rather than shifts in language, fashion, or diagnostic systems. Each era was equally weighted so that no single period dominated the index, preserving long‑term comparability.

Is word frequency data a useful proxy for mental health prevalence?

Language is not the same as prevalence, but it reflects reality. What societies talk and write about indicates topics that matter. These language trends were therefore tested against real‑world indicators—namely mental‑health‑related hospitalisation and social security records—and showed strong alignment across time, see the chart below.

The use of mental health terminology matched actual prevalence well

The study reveals a marked and accelerating rise in cultural attention to mental health across two centuries, particularly from the mid‑20th century onward with a trebling since the 1950s and doubling since 1980. This pattern closely matches real‑world prevalence indicators such as mental health related social security claims. These results suggest that word use trends can serve as a reliable proxy for societal concern—and potentially for underlying prevalence—opening up new possibilities for tracking population mental health over time.

Which life conditions best explain the rise in references to mental health since 1800?

The core aim of the work was to test whether macro‑level life conditions, also measured using the word‑frequency approach, could explain the long‑term changes in mental health discourse.

Eight factors were included: six established mental health determinants, plus human‑centred values and the human–nature relationship.

1. Economic Environment, e.g. unemployment, capitalism, economy
2. Urbanisation, e.g. urban, cityscape, housing
3. Technology, e.g. computing, digital, social media, data
4. Environmental Stressors, e.g. pollution, landslide, global warming
5. Social Equity, e.g. exclusion, gender, discrimination
6. Education, e.g. university, curriculum, literacy
7. Human-Centred Values, e.g. autonomy, ethics, individualism,
8. Human–Nature Relationship, e.g. bud, dew, mosses, mountain

Discourse about everyday life conditions explains mental health discourse well.

A predictive modelling approach was used to see how the eight factors best combine to fit the mental health trend. The analysis showed that long‑term changes in mental health discourse can be explained by a combination of socio‑economic macro‑level factors.

Discourse about everyday life conditions explains mental health discourse well. After several validation tests two factors stood out as contributing most:

• The economic environment, reflecting the degree to which economic conditions such as work, income, growth, and employment are salient in everyday concerns
• The human–nature relationship, reflecting whether nature remains part of everyday cultural experience

The economic environment emerged as the strongest positive contributor to lower mental health discourse, consistent with extensive literature linking economic and employment to mental health outcomes.

Notably, the human–nature relationship showed the same protective pattern: when nature featured in our lives, mental health discourse was lower.

Urbanisation and technology showed substantial negative weights, meaning that as these factors grew, mental health discourse also increased. This aligns with evidence linking urban living to social isolation and the growing recognition of digital‑era stressors.

Human‑centred values, including individualism and autonomy, also showed negative associations with mental health discourse, echoing concerns about cultural disconnection and emotional strain in highly individualistic societies. Social equity and education contributed positively, but more modestly.

The six factors with the highest weightings are shown in the figure below, with the stone size proportional to the weightings found in the analysis.

Do the foundational factors get the most attention?

The economic environment and human-nature relationship stood out as foundational factors, but the results show that attention to both has been overtaken by the other six determinants in recent decades.

While cultural attention to technology (+11%), education (+66%), social equity (+77%), environmental stressors (+47%), human-centred values (+48%) has grown since the 1980s, attention to the economic environment has fallen (-45%), alongside urbanisation (-13%), while the human-nature relationship remains around half of its historical highs.

Attention to other mental health determinants has overtaken foundational factors in recent decades (stone size represents analysis weightings).

References to education, human‑centred values, and social equity overtook economic discourse in the 1990s, with urbanisation, technology, and environmental stressors doing so in the 2000s. Importantly, this fall in economic language does not mean the economy has been “sorted.” Rather, it indicates a declining focus on economic conditions in everyday cultural attention. The economy has not ceased to matter; it has receded into the background and been overtaken by other topics.

These shifts suggest that more visible or contemporary concerns, for example “screen time”, may have come to dominate attention, potentially overshadowing the foundational conditions of mental health. These determinants came to dominate attention, while unresolved economic conditions and long‑term losses in nature connection continued to shape mental wellbeing.

It is also notable that Western nations, which exhibit the steepest rises in mental health difficulties, tend to be trending to weaker relationships with nature, a pattern also linked to economic priorities and individualistic values – mental health and the human-nature relationship share foundational factors.

Can the use of non-mental health words predict real world mental health prevalence?

The next step was to explore whether the use of non‑mental‑health words—such as those relating to the economy, urban living, technology, environmental stress, social equity, education, human‑centred values, and nature—can explain real‑world mental health prevalence? The answer is yes, and with surprising accuracy.

When the model was tested against long‑running indicators such as mental‑health related hospitalisation and social security records, it tracked real‑world prevalence patterns across more than a century of data quite well, see below.

Use of non-mental health words can predict mental health prevalence.

Importantly, this predictive power was achieved without using mental health terms themselves, showing that the use of language about broader life conditions carries a strong signal about population mental health.

This matters because it suggests mental health trends are not solely the result of changing awareness or diagnostic practices. Instead, they are closely tied to underlying social, economic, and environmental conditions that shape everyday life. In that sense, mental health discourse acts as a readable signal of deeper structural pressures and protections.

What does this mean for the future of mental health?

By extrapolating recent trends in factors such as the economic environment, urbanisation, and technology, the model produced future scenarios suggesting that mental health burden is likely to continue increasing if current patterns persist. However, alternative scenarios told a markedly different story. When the model simulated a restoration of the human–nature relationship—or a renewed focus on economic conditions—predicted mental health discourse fell substantially. Of the scenarios tested, restoring cultural attention to nature produced the largest reduction, suggesting a powerful and under‑recognised lever for population mental wellbeing.

Future scenarios of mental health

These future scenarios reinforce a central message of the study: mental health trajectories are not fixed, but depend heavily on whether societies strengthen or continue to erode the foundational conditions.

What are the implications of the results?

The results point towards some interesting conclusions. In addition to focusing on individuals’ mental health, two structural foundations stand out:

1. A focus on the economic environment as this underpins the ability to materially provide for oneself.
2. Creating a culture of meaningful engagement with the natural world.

Sometimes simple evolutionary accounts provide a useful perspective. Humans are part of the natural world and like other great apes, there are core material foundations to living well. The opportunity to provide for ourselves in our natural habitat is fundamental. Like other great apes we thrive when we can provide for ourselves and be close to nature.

This perspective cuts through some conceptual clutter as lists of influences on mental health continue to grow. Of course, these factors matter. But many operate within constraints set by deeper conditions. This research suggests that good mental health is built on material foundations. The context of our lives can be seen to sit on top of these foundations. This does not deny complexity—it helps organise it.

From a policy perspective, the long view of this research highlights the need to ensure that current ‘hot topics’ do not overshadow the fundamentals. There is only a certain amount of bandwidth and policy should ensure it considers the most important levers, rather than focus overly on present day context. And scenario modelling suggests that restoring everyday engagement with nature, and re‑centring economic conditions as shared foundations of life, could meaningfully alter future trajectories.

This work also highlights how language use mirrors shifts in societal stress and wellbeing. The model’s predictive capacity also points to practical opportunities: language trends can offer early signals of rising mental health burden.

Finally, the shared structural drivers behind both the decline in the human–nature relationship and the rise in mental health difficulties—especially in Western nations should also be considered. These include economic priorities, urbanisation, technological change, and increasingly human‑centred value systems.

The findings suggest that unifying both human and nature’s wellbeing depends on creating economic and urban environments that allow people and nature to thrive together—reinforcing the need for integrated approaches that bridge mental health, social, and ecological policy.

Richardson, M. (2026). Nature, economy and mental health: Predictive insights from historical language patterns. Journal of Environmental Psychology, 112, 103078. https://doi.org/10.1016/j.jenvp.2026.103078

Back in January, we opened the call for contributions to Nature Connections 2026 on 16th July. Since then, the response has been both heartening and revealing. We were oversubscribed three-fold, pointing toward a field that continues to deepen, diversify, and innovate in how it understands the human-nature relationship.

Before looking into the themes of the day, a couple of announcements.

Firstly, our keynote speaker for the morning will see the return of Tony Juniper CBE. Tony is a prominent environmental figure who has been active in the defence of nature for more than 40 years. He has run global campaigns, written many books, and advised at the highest levels. He has led a number of major organisations and is currently Chair of Natural England. It will be 10 years, 1 month, and 1 day since Tony opened Nature Connections in 2016, and it will be interesting to reflect on how things have moved on since then.

Second, in addition to the talks, we’ll be hosting a panel discussion: Rooted in Evidence: How Nature Connection can inform Policy and Decision-Making. We’ll have more details on the panel members soon.

And, of course, you’ll hear from me, I’ll be closing the day this year.

Back to the themes of the talks on the day. Across submissions from researchers, practitioners, artists, educators, health professionals, and community organisations, a few patterns stand out.

One is an increasing emphasis on belonging – to place, to landscapes, to local ecologies, and to the communities that form around them. Another is a widening commitment to inclusion, with thoughtful work exploring how nature connection can be made tangible and meaningful for people who face barriers to access, whether physical, social, cultural, or structural.

There is also a renewed attention to the textures of experience: the sensory, aesthetic, and emotional qualities of being in nature, and how these shape identity, memory, and wellbeing. From creative methods to long-term learning programmes, many contributors are examining how sustained, repeated, or carefully supported encounters with nature can lead to deeper shifts than one-off experiences alone.

Innovation features strongly too – digital, artistic, place-based, and co-created approaches that invite people to encounter nature differently. Whether through gardening, the night sky, citizen science, therapeutic programmes, or new forms of design and infrastructure, the work shared points to nature connection as a living, evolving practice.

Perhaps most striking, however, is the continued movement toward relational thinking: seeing nature connection not simply as an individual outcome, but as part of wider cultural, ecological, and social processes. Many of this year’s contributions explore how connection can support stewardship, creativity, ecological responsibility, or collective wellbeing, and how people and places shape one another over time.

All of this promises a rich and thoughtful programme in July, one grounded in evidence, creativity, care, and a shared desire to understand how people can live well with the rest of nature. And if you’re joining us, we look forward to gathering to explore these ideas together – bookings are open!

If we want to repair our relationship with nature, we first need to understand the system that has been steadily eroding it – my latest research paper.

Responses to declining nature connection often centre on individuals or on providing more green space. But these approaches can overlook the deep cultural, historical, and regional forces that shape our relationship with nature. Without a system‑level view, policies risk being piecemeal, short‑lived or ineffective. To understand why nature connectedness has been slipping, we need a wider perspective, from individual moments of noticing nature, out to regional histories of urbanisation, and further still to national cultural shifts driven by science and industry.

My latest research paper, (a first Richardson & Richardson with my son who’s completing a Masters in Data Science), goes into the engine room of that system, showing how cultural change, local environmental conditions, and psychological processes interact to power (or stall) our relationship with nature. This is the first model to link cultural change, regional environments and psychological pathways into a single system capable of explaining two centuries of decline and today’s regional differences.

Built from real-world data and research evidence the model simulates individuals and families over two centuries in eight cities and counties. It shows how national trends and local environments interact to shape nature connectedness, placing everyday behaviour and attention to nature within the wider system policymakers seek to influence.

It’s the most complex work I’ve attempted and the culmination of a year using new approaches such as machine learning and agent-based modelling to study long-term trends and global differences in recent papers in Earth, Ambio, Sustainability Science and Biological Conservation. As such, the paper also offers a theoretical synthesis of the emerging macro-nature connectedness field, shifting the agenda from individuals and green space to strategic interventions that address the forces driving long-term disconnection and the realities of local places. Put simply, nature connection isn’t viewed as an individual trait, but a system shaped over centuries—this latest research starts to reveal how.

What is the Nature Connection Engine?

This study introduces a hybrid agent‑based model – conceptualised here as a Nature Connection Engine – that unifies macro‑level cultural dynamics, regional socio‑environmental context, intergenerational mechanisms, attention to nature and behavioural feedbacks within a single systems framework.

The conceptual framework of the model

So, putting it a little more simply, the model includes three key drivers of nature connection:

(1) macro‑level cultural context over time – outer oval

(2) contemporary regional environmental conditions – middle oval

(3) psychosocial processes, including attention and intergenerational transmission – inner oval

Even more simply, it’s a computer simulation of people and places over time.

Using UK cities and counties for training and testing, there were three research questions:

1. How well does the agent-based model reproduce current regional nature connection while reflecting historical national trends in the human-nature relationship?
2. How do macro-level and regional-level factors influence regional differences in the human-nature relationship?
3. Can the model be used to generate future human-nature relationship trends for scenario exploration?

Bringing the Idea to Life

 The Nature Connection Engine runs as a yearly simulation from 1800 to 2020, modelling how people, families and places evolve together over time. Each year, the model updates five core processes: how towns grow, how nature changes, how individuals age and form families, how their connection to nature shifts, and how this connection is passed to the next generation.

An individual’s nature connectedness changes through a balance of influences: the nature around them, their attention to it, and wider cultural and regional pressures. These interacting forces allow realistic long‑term trends to emerge from small, everyday experiences.

The model was trained on four contrasting regions—Northumberland, Dorset, Birmingham and Leicester—and then tested on four others, including Liverpool and North Yorkshire, to check how well it generalises to regions it had never seen, giving confidence that its insights hold beyond the calibration areas. The model was also stress‑tested using sensitivity analyses. Much more sits behind the model, but this is the basic engine that reveals how two centuries of human–nature relationships can explain contemporary regional differences.

What can the model do?

Built from real‑world data, the model reliably mirrors both the long‑term national decline in nature connectedness and today’s regional variation, recreating two centuries of decline and reproducing contemporary differences across places. It not only fits the past, it fits the present and can help diagnoses the forces behind the trends over decades and differences we see today.

The model shows that regional conditions explain around two‑thirds of variation, with macro‑cultural trends (science, economy, secularisation) shaping the overall national trajectory. Urban areas are influenced more strongly by local socio‑environmental factors, while rural regions are more exposed to long‑term cultural pressures.

The Nature Connection Engine can also test interventions before they’re implemented, helping reveal why the same policy succeeds in one place but falters in another. The model found that children’s programmes raise nature connectedness across all regions and are the single most effective intervention, while neighbourhood greening strengthens the effect but rarely reverses decline on its own. However, combined action—children’s programmes plus nature‑based neighbourhoods—creates the conditions for sustained recovery, especially in urban settings where lower baseline connectedness allows for larger long‑term gains.

The model can be run into the future, and these scenarios show that under current trends, nature connectedness continues to fall, particularly in rural areas – something to think about. Encouragingly, model runs extended to 2100 show the potential for self‑sustaining improvement once supportive conditions are in place.

What next?

The model is complex, but reality is far more complicated and there can be further work to improve it, to make, for example, simulated Leicester more like real Leicester. So, I’m now seeking funding to:

• Incorporate more real urban and greenspace data to simulate neighbourhood‑level inequalities and opportunities.
• Extend the model with new mechanisms linking attention, experience, behaviour and feedback loops.
• Calibrate the future scenarios using results from the many individual projects running across the UK.
• Co‑design the outputs of the new version of the model to ensure relevance, usability and policy alignment.

The Vision

Policymakers lack tools that can reveal why interventions succeed in some places but not others, how cultural or economic trends influence outcomes, or how combinations of programmes could generate long‑term, self‑sustaining improvements in connectedness, wellbeing and pro‑nature behaviours.

The strength and stability of the initial model provide the foundations for a new policy tool— a Nature Connection Engine — a practical diagnostic and scenario‑testing approach that helps policymakers explore what works, where, why, and for whom. Under this view, nature connection becomes an engine for nature recovery and wider social benefits, rather than a narrowly individual or access‑based issue.

Such a model would provide a decision‑support dashboard that allows practitioners to test “what‑if” scenarios – e.g. rather than one successful community garden in Derby, what if there were 25? Together with the children’s programme that worked really well in Bradford? This will directly support urban planning, nature recovery, public health, wellbeing strategies and children’s nature engagement programmes.

That future is possible with a little more work, for now, the model shows how today’s regional differences emerged and why similar environments can lead to different levels of nature connectedness: individual relationships with nature are shaped by the interplay of culture, place and family. A key finding is that rural regions—despite greater access to nature—are vulnerable to long‑term cultural forces driving disconnection. The work also highlights that improving access to green space, while valuable, is a relatively shallow leverage point unless paired with interventions that shape attention and early‑life experience. In short, meaningful recovery requires changing not just the environments people live in, but the psychological and cultural conditions that sustain connection across generations.

Richardson, M., & Richardson, L. A. (2026). Reframing human–nature connectedness: A multi-scale agent-based model across time and region. Urban Forestry & Urban Greening, 120, 129399. https://doi.org/10.1016/j.ufug.2026.129399

Over the past year I’ve been focussing on the bigger picture of nature connection, what factors explain differences between nations and over 200 years? I looked at long-term trends in a paper in Earth last summer, then differences between nations in Ambio in November, followed by trends again earlier this week in Sustainability Science. The fourth paper in that series (just one more to come!) has just been published in Biological Conservation.

Although there’s global recognition that solving environmental crises requires a new relationship and reconnection with nature there are differences in that relationship across nations and cultures. This paper takes a very broad approach using the largest regional grouping, the Global North and Global South which form two economic worlds. Aside from economic differences, there are also differences between the worldviews of the Global North and Global South which may well contribute to differing human–nature relationships. Therefore, in this paper we are interested in two key research questions:

• Does nature connectedness differ between Global North and Global South?
• If so, what factors might explain those differences?

Clearly, this grouping hides great differences between and within nations, yet the North–South grouping provides a useful way to investigate macro dynamics. It captures historically rooted economic and cultural contrasts that shape human–nature relationships at scale. Understanding similarity and differences on such a scale is important when considering global policy.

Using data from over 60 countries we looked at how a range of objective socio-economic (e.g. urbanisation, biodiversity, economy, income and technology) and subjective cultural indicators (e.g. values related to technology, science versus faith and spirituality) explain any differences in nature connection between the Global North and South.

Perhaps unsurprisingly, on average, nature connectedness was higher in nations of the Global South relative to the Global North. Given the grouping is based upon economic development, this fits with the proposal that the dualistic outlook of the North and how the perceived right of exploitation of natural resources for economic progress strengthened a sense of separation between humans and nature.

However, there was no difference in nature connectedness when controlling for economic differences. It was the indicators of subjective values that explained more variance in nature connectedness than the economic indicators, pointing to the key role of worldviews and values in shaping group-level nature connectedness. Together, the objective indicators explained 39.0% of the variance in nature connectedness between the north and south. Whereas the subjective indicators explained 53.2% of the variance.

It is also noteworthy that, when controlling for the North–South difference, several objective indicators were surprisingly non-significant despite appearing as significant indicators in previous research. For example, as a block, economic development factors, which have been identified as a negative predictor of country-level nature connectedness in several studies, were not a significant predictor after accounting for the North–South dichotomy. The fact that such a simple categorical dichotomy based largely on economic development managed to outperform a collection of continuous measures of the very same suggests that economic factors may not fully explain the underlying differences that contribute to the nature-connectedness disparity.

Similarly, biodiversity, which has likewise been identified as a significant predictor of country-level nature connectedness, was a non-significant predictor of nature connectedness when we controlled for the North-South dichotomy. Again, this suggests that some other factor (e.g., cultural differences, in this case) may explain differences in the human–nature relationship between North–South. One important consideration is that for many Indigenous cultures, connection to nature is not primarily through biodiversity or individual species, but through a deep relational bond with land itself. This differs markedly from Western perspectives, which often frame land in terms of ownership or resource value.

So what does explain the North–South dichotomy?

Although the North–South dichotomy is largely based on economic development, this study suggests there’s a need to consider other dimensions on which the regions differ. Spirituality showed a significant positive relationship with relationship to nature connectedness. Attitudes towards science and faith also played a role. Specifically, a stronger belief that society is too dependent on science over faith was associated with higher nature connectedness. In the Global North, people are more open to depending on science over faith. In contrast, the openness to dependence on faith over science in the Global South is, instead, accompanied by a stronger relationship with nature.

Natural Disasters

It’s interesting to note that vulnerability to natural disasters had a significant, negative impact on nature connectedness, yet both are significantly lower in the Global North—whereas in the Global South, the human–nature relationship is stronger, despite increased exposure to events such as flooding, storms, drought and earthquakes. This inconsistency may, in fact, highlight the strength of worldviews, values and spirituality compared with environmental factors. That is, while this negative correlation implies that one would expect nature connectedness to be lower in the Global South (where natural disasters are more common), nature connectedness is, indeed, higher in the south. Considering that no other environmental variables predicted nature connectedness, the most plausible explanation is that cultural supports, such as relational and spiritual interpretations of nature, buffer against the negative effects of environmental harm.

The reason for the inconsistency notwithstanding, it is still worth recognizing the negative relationship between natural disasters and nature connectedness. With such events likely to increase with climate change—including in the North where the cultural supports are not as present—the negative relationship between exposure to natural disasters and nature connectedness could undermine efforts to renew the relationship with nature.

Conclusions

This research set out to explore how global differences in nature connectedness can inform a closer relationship with nature. The findings reveal a significant disparity in nature connectedness between the economically defined regions of the Global North and South, with the former generally exhibiting lower levels of connection. By exploring the global differences in nature connectedness, the results suggest that nature connectedness is not primarily a product of economic or environmental conditions, but is deeply embedded in cultural, spiritual, and value-driven contexts.

They point to the need for a multifaceted approach to reimagining human–nature relationships, one that addresses the adverse effects of disasters, draws on cultural and spiritual strengths, and better connects science and faith to foster a sustainable, interconnected worldview.

Richardson, M., Rocha, N. M. F. D., & Lengieza, M. L. (2026). How can global differences in nature connection inform new relationships with nature? Biological Conservation, 316, 111768. https://doi.org/10.1016/j.biocon.2026.111768

Earth image By Reto Stöckli (land surface, shallow water, clouds)Robert Simmon (enhancements: ocean color, compositing, 3D globes, animation)Data and technical support: MODIS Land Group; MODIS Science Data Support Team; MODIS Atmosphere Group; MODIS Ocean GroupAdditional data: USGS EROS Data Center (topography); USGS Terrestrial Remote Sensing Flagstaff Field Center (Antarctica); Defense Meteorological Satellite Program (city lights). – https://visibleearth.nasa.gov/view.php?id=57723, Public Domain, https://commons.wikimedia.org/w/index.php?curid=306260

Understanding how our relationship with nature has changed over time helps identify the forces that have shaped the long‑term decline, and where recovery might come from. Using nature‑related word use in books as a proxy for the human–nature relationship over the past two centuries, my latest research paper, published in Sustainability Science, explores that long-term decline using a simple machine learning approach.

This new study explores whether wider socio‑economic and cultural factors, represented through their own word‑use trends, can predict the long‑term changes in the human–nature relationship. In short: can non‑nature words explain the fall in nature words?

Tracing the human–nature relationship through language

The human–nature relationship proxy comes from the frequency of use of 28 experiential nature words such as river, blossom, mosses and meadow. The set spans sensory modalities (e.g., bark, stream), experiential places (e.g. mountain, meadow, coast), living elements (e.g. birds) and archetypal aspects (e.g. river). The words evoke lived and sensory experiences in nature and tap into deep symbolic or cultural meanings of the type related to closer relationships with nature.

These words reflect what people noticed, valued, and wrote about. And when their use is plotted over time, a clear decline of around 60% is revealed. Particularly from 1850, a time when industrialisation and urbanisation grew rapidly.

The new model pairs this nature‑word trend with seven macro‑factors, each represented by their own cluster of words:

• Natural disasters
• Science and technology
• Religiosity
• Industrialisation
• Humanism
• Urbanisation
• Economic development

These wider clusters capture structural and cultural shifts known to influence people’s relationship with nature.

Word use trends since 1800

Testing whether macro‑factors predict nature‑related word use

So through varying the weight attached to each macro-factor trend, can the seven trends shown in the chart above be combined to form the nature words trend?

The study tested over 48 million combinations of factor weights using a machine learning approach before refining the best solution using an approach called gradient descent. The resulting model fits the observed HNR trend closely, see chart below, explaining around 98.6% of the variance. Trends in non‑nature words can explain the fall in nature words!

Nature word trend versus trend constructed from non-nature words

Looking at the weights attached to the non-nature words that are needed to explain the use of nature words, three points stand out:

1. Urbanisation and economic development are the dominant influences.

This aligns with the shift to urban living and increased consumerism.

2. Industrialisation mattered earlier but less so now.

Its influence peaked in the early 20th century, reflecting its key role in the initial decline.

3. Humanism shows a steady and growing influence.

Although assigned a lower weight, its consistent rise, particularly since 1990, suggests an increasing cultural focus on human‑centred values, potentially at the expense of meaning derived from the wider natural world.

Science, natural disasters and religiosity show comparatively modest contributions in the model, despite large changes in their own trajectories.

Overall, the weightings align well with historical intuition: large‑scale structural shifts, especially urbanisation and the economy, map closely onto the long-term decline in the cultural presence of nature.

Looking forward: on‑trend futures and possible recovery

The model’s fit over time allowed two simple scenarios to 2050 to be tested.

On‑trend scenario showed that if current patterns continue, the recent uptick in nature‑word use is unlikely to continue. Increasing urbanisation and continued rise in human‑centred values may push the HNR downwards again.

The recovery scenario suggested that if urbanisation is softened through approaches such as biophilic city design, and if humanistic values broaden towards an “all beings” viewpoint, as seen in the rights of nature and One Health movements, the model suggests a recovery to levels last seen around the mid‑20th century.

Where is the human-nature relationship heading?

These are not forecasts, but illustrations of how socio‑cultural shifts could influence the future HNR.

Three independent approaches, one pattern

This work now sits alongside earlier approaches:

• the agent‑based model simulating the decline in HNR from 1800 to 2020, and
• the raw nature‑word trend itself.

All three show a remarkably similar pattern: a strong decline beginning around 1850, a long trough, and a modest recent rise. Convergence across methods increases confidence in the underlying trend.

Conclusions

The results show that the frequency of use of non‑nature words can indeed explain nature‑related word use. Language about industry, cities, economics and human‑centred values explains the long‑term weakening of the human–nature relationship. Of these, urbanisation and economic development have the strongest influence. The findings reinforce the importance of structural and cultural drivers, not just individual behaviour, in shaping connection with nature. They also highlight pathways for recovery: greener urban living, and worldviews that place humans within, rather than apart from, the wider community of life.

Richardson, M. (2026). The human–nature relationship across two centuries: Macro factor insights from a machine learning model. Sustainability Science. https://doi.org/10.1007/s11625-026-01807-x