ResearchHealth and Biophilic DesignBiophilic Design Patterns and Their Evidence

Biophilic design patterns and the evidence

Which biophilic patterns apply outdoors, what experiments, reviews and population studies show about nature and well-being, and how far that supports design.

Field Method · AI-assisted research roundupPublished Sources checked

Biophilic design is an approach that draws on research about how people respond to nature. Terrapin Bright Green’s 14 patterns are one widely used framework for it. A landscape architect is asked to apply it outdoors, where the exposure is real and the design lever is the layout of planting, water, paths and edges. The research supports the general direction of that practice more firmly than it supports any single pattern or any quantity. We group the studies below by design so the weight of each is visible.

This is a reading of published research for design reasoning. It is not medical advice, and it does not tell a project team what health outcome a design will produce.

The pattern framework

Terrapin Bright Green’s report 14 Patterns of Biophilic Design groups its patterns in three categories. Nature in the Space covers seven patterns, from visual connection with nature to connection with natural systems. Natural Analogues covers biomorphic forms and patterns, material connection with nature, and complexity and order. Nature of the Space covers prospect, refuge, mystery, and risk or peril. Terrapin Bright Green, 14 Patterns

The authors say they reviewed more than 500 publications and describe the patterns as flexible guidance, not formulas. They also state that some cited studies are more rigorous than others. Each pattern carries a rating of zero to three asterisks. Three means the quantity and quality of peer-reviewed evidence is robust, and none means the research is minimal and the pattern rests partly on anecdotal information. The 2014 report gives these ratings in its summary table and pattern headings:

  • Three: visual connection with nature, prospect, refuge.
  • Two: non-visual connection with nature, non-rhythmic sensory stimuli, thermal and airflow variability, presence of water, dynamic and diffuse light, complexity and order, mystery.
  • One: biomorphic forms and patterns, risk or peril.
  • None: connection with natural systems, material connection with nature.

The report’s own text notes that health benefits of water depend on the water being perceived as clean, that there is limited scientific documentation of health effects for connection with natural systems and for natural materials, and that no metrics exist for how often or how long refuge should be used. Terrapin Bright Green, 14 Patterns

The report is a consultancy synthesis, useful as an index to the literature. The rating scale is the authors’ judgment, and we did not re-derive it.

An academic review reaches a similar split through a different framework. Gillis and Gatersleben reviewed the restorative-environments literature through the biophilic design categories of Kellert and Calabrese. They found much evidence for some attributes, such as the presence of natural elements, while empirical evidence for others, such as natural materials or processes, was lacking. They also note that most of the research they found depended on the visual sense. Their review is narrative and, they state, not exhaustive. Gillis and Gatersleben 2015

Which patterns map to landscape elements

This mapping is our interpretation, not a finding of the report. Most patterns in the first and third categories describe outdoor conditions directly:

  • Visual connection and connection with natural systems: views to planting, seasonal change, habitat and weather.
  • Non-visual and non-rhythmic stimuli: birdsong, wind in foliage, scent, texture underfoot.
  • Thermal and airflow variability: shade, sun and breeze across a site over a day.
  • Presence of water: ponds, rills and stormwater features, with the clean-water condition noted above.
  • Dynamic and diffuse light: filtered light under canopy.
  • Prospect, refuge, mystery: overlooks, sheltered seats, paths that reveal a view in stages.
  • Risk or peril: challenge elements such as stepping stones or heights, which the report says are deliberate and not suited to all users.

Heights, water and challenge elements fall under other requirements: play-area safety standards, accessibility law and standards, and, in healthcare settings, facility and infection-control requirements. These vary by jurisdiction, facility and edition. The project’s code authority and the client’s facility staff set those limits.

Biomorphic forms and material connection translate less directly, because outdoors the materials are already natural or are chosen as paving and structure.

Experimental evidence

Experiments assign people to conditions. They give the cleanest causal reading and the shortest exposures.

A walk in a natural setting. Bratman and colleagues randomly assigned 38 healthy adults to a 90-minute walk in a grassland and oak greenspace near Stanford (19 people) or along a busy urban thoroughfare in Palo Alto (19). Impervious surface covered about 10% of the area within 50 m of the nature path and about 76% within 50 m of the urban route. Blood perfusion in the subgenual prefrontal cortex, measured by MRI, fell after the nature walk (effect size d = 1.01) and did not change after the urban walk. Self-reported rumination fell in the nature group from a mean of 35.39 to 33.06 (d = 0.34), but the time-by-environment interaction was not significant at the usual threshold (P = 0.07). Seven participants were excluded from the perfusion analysis for movement. The walks also differed in elevation gain, 155 m against 4 m. Participants had no history of mental disorder, lived in urban areas and had a mean age of 26.6, and each took one walk. The authors read the brain result as pointing to a possible causal mechanism. They also say that, without the rumination self-reports for context, the same brain change could relate to processing sadness, guilt or similar states. Bratman et al. 2015

Real nature against a simulation. Browning and colleagues randomly assigned university students (82 analyzed, mean age 20) to one of three conditions, each six minutes of sitting plus six minutes of walking: in a forest, in a 360-degree video of the same forest in a virtual reality headset, or in an indoor control facing a blank wall. Group sizes were roughly 22, 30 and 30, because temperature limited outdoor sessions. Positive mood ended higher in both nature conditions than in the control, with regression coefficients of 0.77 for outdoors and 0.52 for virtual reality. Within groups, outdoor positive affect rose from 3.37 to 3.54, and in the control it fell from 3.00 to 2.42, so part of the contrast comes from a decline in the control group. Negative affect fell in all three groups. The study had no non-nature virtual reality condition, so novelty cannot be separated from nature. Browning et al. 2020

For design, these studies show that short exposures to outdoor nature can shift mood and some physiological measures in healthy young adults under controlled conditions. They do not tell a designer which element of the setting did it.

Reviews of short-term exposure

Reviews and meta-analyses pool studies like those above and show how consistent the picture is.

Mood, attention and physiology. Bowler and colleagues reviewed 25 studies from 24 articles that compared natural with synthetic environments: 13 crossover trials, 7 randomized trials and 5 observational studies. Pooled differences favored natural environments for anger (Hedges g = 0.46), fatigue (0.42) and sadness (0.36). Anxiety and tranquillity results were inconsistent. Blood pressure (g = 0.07) and cortisol (g = 0.03) showed little difference between environments. Attention showed a small positive pooled effect (g = 0.32) that lost significance once effect sizes were adjusted for pretest differences. The authors list self-selected, mostly student samples, expectation effects in self-reported mood, possible confounding in 13 studies, and very short exposures. Bowler et al. 2010

Attention. Ohly and colleagues reviewed 31 studies testing Attention Restoration Theory, which proposes that nature restores the capacity for directed attention. They ran 13 meta-analyses. Three outcome measures (Digit Span Forward, Digit Span Backward and Trail Making Test B) showed significant positive effects of natural environments. The other ten did not show marked benefit. The authors cite small numbers of studies, small samples and varied outcomes, and they conclude there is uncertainty about which aspects of attention are affected. Ohly et al. 2016

Structured activities in nature. Coventry and colleagues screened 14,321 records and included 50 studies of structured outdoor activities such as gardening and green exercise: 16 randomized trials, 18 controlled studies and 16 uncontrolled before-and-after studies. Risk of bias was low to moderate for the randomized trials and moderate to high for the others. Pooled randomized results favored the interventions for depressive mood (standardized mean difference −0.64, 95% confidence interval −1.05 to −0.23) and positive affect (0.95, 0.59 to 1.31), with high heterogeneity for depressive mood (I² = 85.7%). Blood pressure showed no clear effect, and evidence for physical health outcomes was thinner. The most effective programs ran 8 to 12 weeks, with doses of 20 to 90 minutes. Coventry et al. 2021

These are programs delivered to participants, such as gardening and guided activity. A designer can read them as a case for spaces that support such activities. They do not support a claim about a specific layout.

Observational evidence

Observational studies cover large populations and cannot separate nature from the things that go with it.

Weekly time in nature. White and colleagues analyzed a nationally representative English survey (19,806 adults in the adjusted analysis). Compared with no nature contact in the past week, people reporting 120 to 179 minutes had higher odds of good self-reported health (odds ratio 1.59, 95% confidence interval 1.31 to 1.92) and well-being (1.23, 1.08 to 1.40). Associations peaked at about 200 to 300 minutes. The design is cross-sectional with self-reported exposure and outcomes. The authors state that healthier or happier people may simply spend more time in nature. Time in nature explained about 1% of variance in unadjusted models. White et al. 2019

Childhood green space. Engemann and colleagues followed 943,027 people born in Denmark from 1985 to 2003, beginning at age 10, using satellite vegetation index values within 210 by 210 m around the home. Those in the lowest decile of green space had higher risk of psychiatric disorders than those in the highest, with an incidence rate ratio of 1.60 for any disorder in the capital center and 1.27 in rural areas, after adjusting for urbanization, socioeconomic factors and parental mental illness. The authors say the findings are correlational. The index does not capture vegetation quality, biodiversity or how people use the space, so it cannot say which features matter. Engemann et al. 2019

Health outcomes across studies. Twohig-Bennett and Jones pooled 143 studies, 103 observational and 40 interventional. Higher greenspace exposure was associated with a lower diastolic blood pressure (mean difference −1.97, I² = 82%), lower type II diabetes odds (0.72) and lower all-cause mortality odds (0.69, I² = 96%). They used eleven exposure definitions, from satellite indices to self-report. On the authors’ quality tools, 58% of the observational and 77% of the interventional studies scored at least 9 of 11, and the authors suggest the tools may not have been sensitive enough to separate stronger from weaker studies. Of the 40 interventional studies, 27 concerned forest bathing (shinrin-yoku). Several pooled results are heterogeneous. They state that mechanisms are largely unknown. Twohig-Bennett and Jones 2018

Large population studies support nature contact as a plausible contributor to better health. They do not test any design pattern, because their exposure measures are coarse counts of vegetation or minutes.

Biodiversity and complexity: mixed results

Terrapin gives connection with natural systems no asterisks, its lowest rating, and notes limited scientific documentation of its health effects. Two Sheffield studies, by overlapping author teams, show how unsettled the biodiversity question is.

Fuller and colleagues interviewed 312 users of 15 urban greenspaces larger than 1 ha. Plant species richness was positively associated with reflection and a sense of distinct identity, more strongly than area. Bird richness related to continuity with the past and attachment. Butterfly richness related to nothing, and tree cover was unrelated to any measure. The authors say causality is hard to establish because people choose parks and are not randomly distributed. Fuller et al. 2007

Dallimer and colleagues then surveyed 1,108 visitors across 34 riparian greenspaces in Sheffield. They found no consistent link between actual plant, butterfly and bird richness and well-being. Well-being tracked the richness that visitors perceived. Their explanation is that people identify biodiversity poorly. Dallimer et al. 2012

In our reading, the two results fit a design conclusion with a limit. Visible habitat variety may matter to people, while a species count measured by ecologists may not. Neither study was an experiment.

Evidence from built interventions

Hunter and colleagues synthesized 38 studies of urban green space interventions out of 6,997 records. Park-based and trail interventions that combined physical change with promotion showed strong evidence for use and activity (7 of 7 and 3 of 3 studies). Park or greenway changes without promotion were largely ineffective, with 7 of 12 studies showing no significant effect. Greening of vacant lots showed strong evidence in 4 of 4 studies for health, well-being and social outcomes. Most included studies were non-randomized, and the equity evidence was too limited for firm conclusions. Hunter et al. 2019

This is the nearest the literature comes to testing design decisions. It points to use and programming as active ingredients alongside physical form.

Limits the field names itself

Frumkin and colleagues’ research agenda lists the field’s gaps. Observational studies make up the preponderance of the literature, and most experiments measure short-term outcomes. Confounding and reverse causation are hard to exclude, and natural experiments are highly susceptible to bias. There is little agreement on how to define nature contact, and greenspace measures from different methods have generally shown poor to fair agreement. Dose is poorly captured, mechanisms are only partially understood, and most studies come from cool, temperate, high-income settings. Frumkin et al. 2017

What this means for design decisions

The following is our interpretation, not a finding of any source.

  1. Direction. Time in planted, naturalistic outdoor settings has fairly consistent, modest support for better self-reported mood, mostly in short exposures with healthy adults. Support for physiological stress measures is weaker. The studies tested visits to existing nature, not the provision of new space.
  2. Specific patterns. The sources do not isolate the effect of one pattern outdoors. A pattern list is a design heuristic resting on a general evidence base.
  3. Effect sizes. Where effects are reported, they range from small (rumination, d = 0.34, interaction not significant) to large (perfusion, d = 1.01, in a sample of 38; Coventry’s positive affect, 0.95). Coventry’s depressive mood result is medium by the usual convention (−0.64) and highly heterogeneous. Blood pressure and cortisol results are weak in Bowler’s short experiments, while Twohig-Bennett and Jones, pooling observational studies with intervention studies that are mostly forest bathing, find associations. The two bodies of evidence measure different things.
  4. Claims in project documents. Wording such as “supports mood and attention restoration” overstates the attention evidence, given Ohly’s results. No source we read supports a specific health outcome from a specific design.
  5. Use and maintenance. Hunter’s synthesis suggests that spaces people are encouraged to use do better than spaces that only change physically. Programming and upkeep belong in the same conversation as pattern selection.
  6. Settings and populations. None of the studies we read tested hospital patients, healthcare staff, or children at school. The experiments used healthy adults and students, and Engemann measured green space around the home. For a healthcare garden or school yard, the evidence is a general case for planted outdoor space, not evidence for that population.
  7. Wording a team can defend. Descriptive wording stays within the evidence, for example: “Published experiments and reviews associate short visits to natural settings with better self-reported mood in healthy adults. The evidence for specific design features is limited.”

What we could not settle

  • Which patterns hold up when tested on real sites against alternative designs. We found no such comparison.
  • What Terrapin’s ratings rest on, pattern by pattern. We did not trace the studies behind each rating.
  • Kahn and colleagues (2008) on real and simulated window views. Terrapin reports a heart rate figure from it, which we could not confirm in the original paper.
  • Whether the effects persist or accumulate over months of use. Most experiments are single exposures.
  • Effects on patients, healthcare staff, school children, older adults, people with disabilities, and populations outside temperate high-income countries.
  • Studies not read in full here (Ulrich 1984 on hospital window views, Barton and Pretty 2010 on dose of green exercise) that a later revision should read and weigh.
  • Results for dry-climate and Mediterranean-climate landscapes, which matter for California and San Diego County.

Using this in Field Method

Photo tools puts site photographs on a map. Its key view package makes a key view map with a numbered pin for each photograph and a view wedge where the photograph records a direction. It does not measure well-being or rate a view. A key view map can document the existing views and overlooks that the visual connection and prospect patterns refer to, for discussion in a design review.

Sources

  1. Terrapin Bright Green — 14 Patterns of Biophilic Design: Improving Health & Well-Being in the Built Environment
  2. Gillis and Gatersleben (2015) — A Review of Psychological Literature on the Health and Wellbeing Benefits of Biophilic Design
  3. Bratman et al. (2015) — Nature experience reduces rumination and subgenual prefrontal cortex activation
  4. Browning et al. (2020) — Can simulated nature support mental health? Comparing short, single-doses of 360-degree nature videos in virtual reality with the outdoors
  5. Bowler et al. (2010) — A systematic review of evidence for the added benefits to health of exposure to natural environments
  6. Ohly et al. (2016) — Attention Restoration Theory: A systematic review of the attention restoration potential of exposure to natural environments
  7. Coventry et al. (2021) — Nature-based outdoor activities for mental and physical health: Systematic review and meta-analysis
  8. White et al. (2019) — Spending at least 120 minutes a week in nature is associated with good health and wellbeing
  9. Engemann et al. (2019) — Residential green space in childhood is associated with lower risk of psychiatric disorders from adolescence into adulthood
  10. Twohig-Bennett and Jones (2018) — The health benefits of the great outdoors: a systematic review and meta-analysis of greenspace exposure and health outcomes
  11. Fuller et al. (2007) — Psychological benefits of greenspace increase with biodiversity
  12. Dallimer et al. (2012) — Biodiversity and the feel-good factor: understanding associations between self-reported human well-being and species richness
  13. Hunter et al. (2019) — Environmental, health, wellbeing, social and equity effects of urban green space interventions: a meta-narrative evidence synthesis
  14. Frumkin et al. (2017) — Nature Contact and Human Health: A Research Agenda

These tools support site review and design studies. Read each tool’s stated capabilities and limits before use.

Explore Photo tools
Back to Research