Effects of street greenery and seasonality on health and wellbeing of cyclists: a Virtual Reality experiment

Effects of street greenery and seasonality on health and wellbeing of cyclists: a Virtual Reality experiment

Street vegetation can play an important multifunctional role in increasing green spaces for more just, healthy and sustainable cities. However, there is still limited knowledge on how greenery exposure and seasonal conditions affect health and wellbeing during everyday active travel. This study integrates Virtual Reality, biosensors and visual semantic segmentation to test and compare different green infrastructure scenarios: only built, low greenery, high greenery and a winter condition. Using a between-subjects design, 138 participants were randomised to cycle in one of the four conditions, while gathering data on various physiological (skin conductance, blink rate, pupil diameter) and psychological outcomes (self-reported pre and post measures of restoration, affective states, subjective vitality). The results provide important empirical evidence that street greenery can decrease negative emotions and provide momentary stress reduction even under rapid exposure during a cycling commute. The findings also demonstrate high reactivity to natural elements in virtual environments.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1038/s41598-026-64499-4.

Introduction

Current policies such as the UN’s Biodiversity Plan for Life on Earth and the EU 2024 Nature Restoration Law set ambitious goals for preserving, restoring and increasing green spaces for more just, healthy and sustainable cities1,2. As urban areas continue to densify, it is crucial to understand how small-scale green infrastructure such as street vegetation can provide a feasible solution for achieving these aims.

Street greenery interventions provide untapped opportunities to balance socio-ecological needs in a limited space. In addition to increasing overall green cover in cities, street greenery has been found to contribute to active mobility in several ways: by increasing cycling uptake, frequency and volume3–5, by promoting pleasantness and satisfaction6–9, by providing restoration and mental health benefits10,11, and by reducing adverse environmental exposures such as heat and noise, while increasing exposure to natural sounds and higher soundscape quality12,13. However, there is still limited knowledge on the causal mechanisms of health and wellbeing effects during travel, and how they are affected by different street greening designs14,15. This is caused primarily by the dominant residential-based and recreational perspectives in nature-health research16, the lack of consensus of what is considered street greenery, and the diverse methods used to study greenery exposure effects15,17,18. There is also limited knowledge of how seasonal variations in greenery exposure affect health and wellbeing during active mobility11,19.

To fill these gaps, this experimental study examines the immediate effects of different greenery and seasonal conditions on health and wellbeing of cyclists during everyday travel. We employed Virtual Reality (VR) as a novel and immersive method to study causal effects of environmental stimuli in laboratory settings. Increasingly used in the nature exposure and health domain, VR builds on a long tradition of nature simulation studies that originated in the 1970s to examine planning and management of forests, rural and recreational areas20,21. VR technologies facilitate control and customisation over designs and environmental conditions, reduce confounding effects, and save time and resources needed in real-world interventions such as allowing for vegetation to grow naturally or replicating assessments in multiple seasons. Virtual environments also provide greater immersion and sense of presence compared to traditional methods such as 2D photos or videos22.

The use of VR in various applications has been growing rapidly including nature exposure therapy and pain management23–25, psychological restoration26–30, green exercise31, nature connectedness32, environmental preferences33,34, and walking and cycling behaviour35–38. Recent literature reviews show that findings vary across VR designs and methodological approaches, but generally indicate psycho-physiological benefits of virtual nature, especially related to improved mood, attention and cognition, and perceived restorativeness and stress reduction35,39–41. However, mobility approaches in VR have remained limited, especially in relation to cycling36,39,41. In terms of environmental elements, there is significant evidence on the positive effects of greenery and vegetation, but previously less attention has been given to different environmental conditions or diversity of vegetation27,30,35.

Therefore, to contribute to the scarce literature on the effects of green infrastructure on active mobility, we developed and tested four future VR scenarios with different computer-generated street environments: (1) built (control, almost no vegetation, ambient traffic noise); (2) low level of greenery (low amount and diversity of vegetation, ambient traffic noise, few birdsongs); (3) high level of greenery (higher amount and diversity of vegetation, ambient traffic noise and more birdsongs); and (4) winter condition of environment (3). We compared the effects of these four exposure conditions and tested three main hypotheses that are based on previous evolutionary and theoretical perspectives of the biophilia hypothesis42, Stress Reduction Theory43, Attention Restoration Theory44, as well as the more recent greenery hypothesis45. These theories propose that humans have an inherent affinity for nature and that even brief or indirect exposure to natural elements can reduce stress and restore depleted cognitive resources. Therefore, we first hypothesise that street greenery promotes psycho-physiological restoration (i.e. reduces physiological stress and increases self-reported restoration) and improves mood (i.e. increases positive affect and reduces negative effect) and subjective vitality during cycling. Our second hypothesis is that these effects will be strongest for the VR environment with high level of greenery, which has higher auditory-visual diversity, compared to the low greenery and only built infrastructure environments. Since VR and active mobility studies focusing on seasonality are still scarce and findings have been mixed35,46, our final hypothesis is exploratory, postulating that there is a difference in the restoration and wellbeing effects between the summer and winter environments.

Since most previous empirical studies still lack comprehensive psycho-physiological integration47, here we test the theoretical rationale by combining three psychological (perceived restoration, affective states, and subjective vitality) and three physiological measures (skin conductance, blinking rate, and pupil diameter). While the psychological measures and skin conductance are well-established in the field47, the latter two are less studied eye-tracking biomarkers that can be used as indicators of emotion and attention restoration. Typically, eye blinking increases during moments when the probability of missing important visual information is low, indicating less directed attention and engagement48,49. Pupil diameter is used to measure pupil size which naturally adjusts to light conditions but can change independently of changes in luminance reflecting internal cognitive processes. Pupil dilation is also a marker of emotional stimuli, acute attention and cognitive load48,50. Overall, our aim is to advance psycho-physiological studies on mobility exposure by using emerging technologies and biophysical indicators.

Results

Overall differences between the VR cycling environments

After a stressful mental arithmetic task, 138 participants were blindly randomised to one of the four recovery conditions: built (N = 30), low greenery (N = 34), high greenery (N = 39) or winter (N = 35) (Fig. 1). Individual greenery exposure values were calculated based on the Green View Index (GVI) of participants’ recorded VR headset videos, which changed according to how much people looked around while cycling. Mean GVI values of the environments were statistically different from each other: Built µGVI = 0.001; Low greenery µGVI = 0.221; High greenery µGVI = 0.282; Winter µGVI = 0.064 (ANOVA F = 1152.54, p < 0.001, Tukey’s post-hoc p < 0.001). In relation to the outcome measures, participants exhibited the highest average skin conductance level (SCL) in the high greenery group both during the stress and recovery periods, and participants in the winter group had the lowest average SCL values, but mean values between groups did not differ significantly (Table 1; Fig. 2).

Table 1Mean and standard deviation (SD) of the physiological measures (mean/10 sec) during recovery period: Skin Conductance Level (µSiemens); Number of blinks; Pupil diameter (mm); and change in pre-post psychological measures: Restoration Outcome Scale (ROS), Positive and Negative Affect subscale (PANAS-POS and PANAS-NEG), Subjective Vitality Scale (SVS).

Environment SCL (µS)Mean ± SD N blinksMean ± SD Pupil dm (mm)Mean ± SD ROSMean ± SD PANAS -POSMean ± SD PANAS-NEGMean ± SD SVSMean ± SD

Built (N = 30) 11.4 (± 5.47) 3.16 (± 1.97) 4.37 (± 0.71) 1.53 (± 0.92) −0.026 (± 0.57) −0.086 (± 0.40) −0.052 (± 0.88)

Low greenery (N = 34) 11.8 (± 4.55) 4.95 (± 2.93) 4.67 (± 0.77) 1.57 (± 0.83) 0.165 (± 0.53) −0.188 (± 0.26) 0.228 (± 0.80)

High greenery (N = 39) 12.14 (± 4.15) 3.43 (± 1.83) 4.79 (± 1.11) 1.22 (± 0.78) 0.037 (± 0.58) −0.119 (± 0.29) 0.211 (± 0.79)

Winter (N = 35) 10.5 (± 3.95) 3.11 (± 1.75) 4.99 (± 0.87) 1.22 (± 0.94) −0.062 (± 0.52) −0.053 (± 0.43) 0.154 (± 0.87)

ANOVA (p) 0.512 0.002* 0.054* 0.183 0.353 0.444 0.539

When averaging over the entire recovery period, there were significant differences in the mean blink rates of participants in the four conditions (p = 0.002). Participants blinked most frequently in the low greenery environment (µblink rate = 4.95) and least in the winter environment (µ = 3.11). In addition, the mean pupil diameter was significantly different between environments (p = 0.054). As expected, mean pupil diameter was largest in the winter environment (µ = 4.99 mm), which indicates pupil dilation to adjust to darker light conditions. Pupil diameter was smallest (µ = 4.37 mm) in the built environment, in which the lack of vegetation may have caused higher brightness conditions.

In relation to changes in psychological measures, self-reported restoration (ROS values) significantly increased in all conditions in the post-questionnaire compared to pre-questionnaire (p < 0.001) (Fig. 3; Table 1 in Supplementary). Increase in positive affect (PANAS-POS) was reported for the low and slightly for the high greenery environments, although the changes were non-significant. However, there was a significant decrease in self-reported negative affect (PANAS-NEG) for participants in both greenery conditions: low (p = 0.003) and high (p = 0.04). There was also an increase in self-reported subjective vitality (SVS) for all conditions except the built, which showed slight decrease in mean scores, but the difference in pre-post measures was non-significant.

Main effects on psycho-physiological outcomes

First, we tested for overall effects of the four environments on the mean psychological (ROS, PANAS-POS, PANAS-NEG, SVS) and physiological (adjusted to baseline SCL, blink rates, pupil dm) outcomes with a regression model. The model included individual background variables (gender, age, stressful day, self-reported physical and mental health, nature connectedness, cycling frequency) as covariates to control for the potential effects on the outcomes (Table 1; for full model results, see Supplementary 2). When controlling for individual background variables, results from the regression models indicated very few significant variables that would explain the psychological outcomes. In general, no significant effect of the environment type was found, but there was a significant relationship between better self-reported mental health and increase in perceived restoration (p = 0.005) (Table 2).

The findings for the physiological variables were more evident, especially for the eye-tracking data (Table 2). The environment type had no statistically significant overall effect on mean SCL levels, except for the high greenery condition, which had marginally significant positive effect (p = 0.059). Increasing age was associated with lower mean SCL, which reflects natural age-related decline in electrodermal activity51. However, contrary to expectations, higher nature connectedness was associated with higher SCL (p = 0.030). When compared to the built environment, there was a positive and significant association of both winter and high greenery environments with increase in pupil diameter, and a significant positive association between the low greenery condition and increased blink rate. Older age was associated with decreased pupil diameter and increase in number of blinks, while being a male was associated with reduction in blinking (Table 2).

Table 2The computed statistical significance (p-value) and the corresponding estimated effects (in parentheses) of environment type and individual background covariates on mean physiological and psychological (pre-post change) variables. Baseline factor levels: built environment, female gender, and 18–24 age group. ** p < 0.01, * p < 0.05.

Predictor SCL Blink rate Pupil diameter ROS PANAS-POS PANAS − NEG SVS

Intercept 0.810 (0.233) 0.013 (4.297) < 0.001 (4.958) 0.119 (−0.954) 0.814 (−0.101) 0.079 (−0.455) 0.612 (0.306)

Low greenery 0.151 (0.484) 0.007 (1.578) 0.079 (0.400) 0.083 (0.367) 0.090 (0.250) 0.447 (−0.068) 0.172 (0.285)

High greenery 0.059 (0.604) 0.974 (−0.018) 0.048 (0.436) 0.757 (0.063) 0.772 (0.041) 0.899 (−0.011) 0.260 (0.229)

Winter 0.859 (0.057) 0.640 (−0.265) 0.003 (0.669) 0.966 (0.009) 0.881 (−0.022) 0.411 (0.072) 0.569 (0.116)

Gender (Male) 0.976 (−0.007) 0.014(−0.961) 0.408 (-0.125) 0.661 (−0.063) 0.993 (0.001) 0.624 (0.030) 0.910 (−0.016)

Age (25–44) 0.046 (−0.621) 0.084 (0.919) 0.025 (−0.469) 0.517 (−0.125) 0.530 (0.085) 0.836 (0.017) 0.690 (−0.076)

Age (45–65) 0.005 (−1.147) 0.003 (2.087) 0.005 (−0.766) 0.705 (0.098) 0.294 (0.191) 0.075 (−0.196) 0.709 (0.096)

Stressful day 0.145 (0.089) 0.400 (−0.087) 0.506 (−0.027) 0.945 (0.003) 0.957 (0.001) 0.258 (−0.019) 0.186 (0.051)

Physical health 0.958 (−0.008) 0.170 (0.367) 0.921 (−0.010) 0.074 (0.171) 0.449 (−0.050) 0.293 (0.042) 0.081 (−0.165)

Mental health 0.256 (−0.192) 0.193 (−0.378) 1.000 (0.000) 0.005 (0.300) 0.842 (0.015) 0.070 (0.081) 0.199 (−0.133)

Nature connectedness 0.030 (0.226) 0.221 (−0.214) 0.548 (−0.041) 0.081 (0.117) 0.600 (0.024) 0.479 (0.020) 0.118 (0.103)

Cycling frequency 0.491 (0.073) 0.786 (−0.049) 0.399 (0.059) 0.998 (0.000) 0.882 (−0.007) 0.069 (−0.051) 0.375 (0.058)

Analysis of variance partitions (Supplementary 3. Variance partition plots for the regression models) of these regression models revealed that the proportion of the total variance explained by these models was modest, ranging from 3% (PANAS-POS change) to 17% (PANAS-NEG change). In practice, for instance, 5% of observed variance in pupil diameter was associated with environment type, while this was only 1% for the blink rate. Age explained the largest fraction of variance for mean blink rates and pupil diameter, SCL change and PANAS-NEG change, mental health for the SVS and ROS change (3% and 7% respectively), and environment type for the PANAS-POS change (~ 2%).

Temporal effects on physiological measures

We further assessed the temporal dynamics of physiological outcomes (baseline-adjusted SCL, blink rate and pupil dm) within the 8-minute cycling exposure using a Bayesian temporally-explicit regression model, with the individual GVI exposure values as a covariate. Figures 4 and 5 show the estimated marginal effects of the environment type (in time) and GVI under the three fitted models. We found clear effects of GVI on the physiological measurements. GVI had a significant positive effect on SCL and number of blinks, and a clearly negative effect on pupil diameter (Fig. 5). When looking at the temporal effects of the VR environments on the measured physiological responses of cyclists, we found that in the three summer conditions (built, low and high greenery) the SCL levels decreased after the induced stress with greatest reduction effect around 3–5 min of exposure, after which SCL in all three environments started to increase (Fig. 4). Notably, however, the decrease in SCL was estimated to be largest in the high greenery environment and lowest in the built environment, suggesting a different effect of the two environments during cycling. In addition, in the winter condition, the overall reduction in SCL in the early recovery phase could not be observed. Instead, the effect of winter environment on SCL level was consistently increasing in time, suggesting a cumulative arousal. Regarding the effect sizes, we note that the individual level variance in adjusted SCL varied mostly between 0.037 and 1.543 (5% and 95% empirical quantiles), so the estimated effects appear relevant in size.

When analysing the temporal effects of the VR environments on blink rates, no clear differences were found between the environments with a generally increasing positive effect towards the end of the cycling Recovery period. On the other hand, individual pupil diameter decreased with time in all vegetation environments. However, a different pattern was observed in the built environment, in which the pupil diameter was inferred to first increase in time and then decreased only towards the end of the recovery period. In all temporally-explicit models, high degree of autocorrelation between the consecutive measurements from the same individuals was found (0.99 for SCL, 0.93 for pupil diameter, 0.96 for blinking).

Discussion

Street greenery improves mood, but restoration effects were mixed

In this VR experimental study, we tested the effects of different street greening designs and seasonal conditions on the restoration, affective responses and subjective vitality of cyclists. Our findings partially support the pre-defined hypotheses. First, we hypothesised that street greenery increases psycho-physiological restoration and improves mood and subjective vitality. We found significant effects only for some measures. Cyclists’ exposure to both greenery environments (low and high greenery) significantly decreased negative affect (PANAS-NEG) compared to pre-experimental measures, which contributes to previous findings52,53. However, although participants reported increase in positive affect (PANAS-POS) and subjective vitality (SVS) after cycling in the two greenery environments, these changes were not statistically significant. This is not completely in accordance with previous studies53,54, which can be potentially attributed to the difference in between- or cross-over experimental designs. It should be also considered that while our experiment had four conditions, these studies have compared two very distinctive VR environments (control vs. nature; urban vs. nature), which may have contributed to the stronger difference in positive effect.

Further, in line with our first and second hypothesis, the high greenery environment, which had the highest level and diversity of natural auditory-visual elements, lowered individual SCL levels the most after the induced stress. This is in line with stress reduction theory43 and other outdoor55 and laboratory experiments56,57. This result also confirms previous findings that stress reduction is more effective within a combined auditory-visual stimuli (e.g. bird songs)29,54,56,58, although here we did not test for separate effects or interactions of these modalities. Recent studies have highlighted that soundscape attributes can have important positive non-linear effects on cycling behaviour, especially when there is audio-visual consistency between natural sounds and visual openness5,13,59. This is particularly relevant for future VR laboratory experiments, where researchers can control for sound sources, intensity and frequency of exposure56.

Furthermore, similar to previous experiments conducted while sitting55,56,60, the greatest stress reduction was reached around 4–5 min, showing momentary recovery potential of natural environments even under rapid exposure during a cycling commute. This finding implies that urban planning should ensure 1–2 km continuous vegetation sections in cycling corridors in order for stress reduction effects to occur. Since many cycling trips are short distance and for non-commute purposes, such recommendations can be expanded beyond commuting corridors to making neighbourhoods and school routes combine both greenery and accessibility61,62. In addition, previous research has shown that associations between environmental characteristics and pleasantness during commute and non-commute physical activity are highly contextual and spatially non-stationary6,63, which highlights the need for locally designed greenery interventions. We warrant for future research to further test the cumulative effects of continuous street vegetation exposure, as well as possible effects of repetitive exposure or distance decay in everyday travel.

The effects of street greenery on the eye-tracking measures were also confirmative to our hypotheses. Higher GVI increased blink rates and decreased pupil dilation in the Recovery period, which suggests reduction in directed attention and stimuli engagement. In addition, pupil dilation continuously decreased with time in all greenery conditions, but it took longer time for participants in the built environment group to reach such decrease only towards the end of the experiment. This is consistent with the soft fascination dimension of ART theory44, which proposes that natural environments can evoke effortless attention allowing recovery of directed attention and the opportunity for reflection.

However, self-reported restoration (ROS measures) increased after exposure to all VR cycling environments regardless of the condition, and similarly blink rates increased with time in all environments, suggesting accumulative effect on reduced directed attention. Also surprisingly, the built environment, which had no vegetation or birdsongs, had a perceived restorative effect and slightly reduced physiological arousal after the stress period. This suggests that cycling in VR as a form of exercise can relieve stress and contribute to cognitive restoration64. The lack of traffic, people or animals in our VR designs may have also contributed to this effect. Similarly, other studies have showed that virtual reality itself can have positive restorative impacts regardless of the condition. For example65, found a recovery effect for individuals with mild-to-moderate anxiety and depression in both urban and natural VR conditions, and the authors suggested that having a quiet urban environment without people can have a healing effect. In addition66 and53 showed that exposure to any VR condition reduced heart rate. Future research could include the role of sense of presence (i.e. the sense of being there) in VR and how it affects emotional responses and restoration mechanisms67. The strong sense of presence in VR environments may contribute to the feeling of “being away” which is another important aspect of Attention Restoration Theory44,68.

Finally, there was a difference in stress reduction in the summer compared to winter VR environments which supports our third hypothesis. Participants who cycled in the winter condition did not show reduction from the stressor and instead had a consistently increasing SCL response pattern, indicating a possible unique underlying mechanism. However, other results were not statistically significant, which corroborates the inconclusive findings in previous research. For example, results from a general population study in the Netherlands did not support the widespread belief that seasons influence mood to a great extent69. Previous research by46 also showed that exposure to virtual nature generally induce similar psycho-physiological responses regardless of season, except for stronger reduction in feelings of fatigue during autumn and winter. A limitation of this study is that we included only two seasons (summer and winter) to balance the resources and time needed for development, and reduce the complexity of the experimental design. Further research across all seasons is needed since studies assessing seasonal variation in benefits of greenery during cycling are still scarce.

Higher reactivity to natural elements in VR

Our findings showed that higher street greenery exposure (GVI) had a statistically significant positive effect on SCL values, which was a surprising result and contrary to the first hypothesis. In addition, higher nature connectedness of participants was also associated with increased SCL values. Despite previous evidence that biophilic design can reduce stress and anxiety60, contrasting results have been found by other laboratory studies indicating increase in skin conductance during exposure to simulated nature52 or natural (wooden) elements in an office setting70. Such arousal may reflect that the natural visual and auditory stimuli in virtual conditions may have an amplifying effect. In addition, in our study, none of the VR conditions were threatening or included any living beings such as people or animals, except the birdsongs that could be heard in the three greenery conditions. This may have caused participants, especially those that feel more connected to nature, to be more reactive to the vegetation elements and accompanying birdsongs as they are the only living elements in the otherwise rather static virtual world. Overall, our scenarios allowed testing greenery and seasonality effects in safe conditions, but this naturally brings challenges with generalising the findings into real-life urban travel environments.

Generally, SCL indicates only arousal of the sympathetic nervous system without showing valence i.e. how positive or negative an experience feels, which is the second key dimension in affective experience71. Although humans need to recover from both positive and negative arousal, moderate level of pleasant arousal could also increase alertness, the quality of sensory stimuli (more information made available per unit time) and increase short-term and long-term memory72,73. 52also showed that increased physiological arousal during virtual nature exposure was associated with the increase or maintenance of positive affective states. In this study, participants reported slight non-significant increase in positive affect scores after cycling in the greenery conditions, which suggests that higher arousal occurred at the same time as positive emotions were maintained. Although this requires further investigation, it generally highlights the important role of pleasantness on psycho-physiological restoration effects in green areas26,55 and during active mobility6,9.

Differences between the greenery environments were hard to detect in VR

Generally, differences in psycho-physiological responses to cycling especially in the high and low greenery conditions were not that obvious, and therefore, we could not confirm the second hypothesis. One explanation for lack of significant results could be that our computer-generated environments and greenery levels (low greenery µGVI = 0.221 and high greenery µGVI = 0.282) were simply not different enough to capture subtle or important elements that may shape human experience and perceptions. Future research can include self-reported ratings of visual greenery, complexity or level of naturalness to better account for perceptual differences between the experimental conditions. Similar lack of significant effects between greenery conditions were reported in previous studies. For example74, found that VR environments with high level of greenery (50%) reduced both physiological and affective stress for pregnant women compared to a non-green environment (0%), but the same effects were not found for the moderate level (12%). Also, interestingly and contrary to our hypothesis, the low greenery condition was associated with highest mean increase in self-reported restoration and highest blink rate, which suggests stronger effects in cognitive restoration. Past studies have also shown that a lower biodiversity VR environment can more effectively lower anxiety and stress compared to an urban and higher biodiversity VR environment58.

In our study, the cycling may also play an important role since previous research has demonstrated trade-offs between level of greenery and audio-visual complexity during active movement13,75. The authors suggested that landscapes that contain too much diversity in size, form, texture, and spatial characteristics may require additional directed attention and arousal, and therefore weaken restorative effects while on the move75. Also too dense vegetation may reduce the visual openness and perceived safety of cyclists76. Our study is one of the very few experiments on nature exposure while cycling in VR35and further research is needed to disentangle the effects of emotion, restoration and physical activity during active travel. Although we used a cycling baseline in a neutral VR environment to control for the effects of physical exercise, still we could not fully account for the confounding effects of movement. This is due to two important limitations in our experimental design: the lack of baseline eye-tracking data and repeated measures of psychological outcomes after each experimental phase. For example, blink rates are also shown to increase with increase in fatigue77. Therefore, it is advisable to collect an individual baseline for the eye-tracking data during cycling in a neutral VR condition to allow for distinguishing eye physiology as part of physical activity only compared to physical activity under a treatment condition. This can help create a moving baseline accounting for natural drift in the blink data. To reduce physical load, baseline measures should be kept short (up to 3 min) and brief rest periods can be introduced between experimental phases. Future studies can also include visual information processing aspects such contrast, spatial frequency, colour78, as well as visual enclosure, sky openness or depth of field5,13,76,79.

Finally, we acknowledge that our sample was biased towards younger and highly educated participants, which limits the generalisability of the findings to broader socio-demographic populations. In line with recommendations from the wider research community, future studies should prioritise targeting different socio-cultural groups in order to move beyond the dominant Western, Educated, Industrialized, Rich, and Democratic (WEIRD) sample in psychological and behaviour research80,81.

Methods

This study received an ethical approval statement by the Research Ethics Committee in the Humanities and Social and Behavioural Sciences at the University of Helsinki (48/2024), stating that all methods were carried out in accordance the relevant guidelines and regulations for research in the humanities and social and behavioural sciences issued by the Finnish Advisory Board on Research Integrity. Power analyses were conducted prior to the experiments to ensure sufficient statistical power using the “simr” package (v.1.0.9)82 in the R software. The power analyses were based on self-reported restoration scores and indicated that N = 140 represents the smallest sample size where the lower limit of the 95% confidence interval of the power curve exceeds 0.80. The study was pre-registered in OSF in February, 2025 (https://doi.org/10.17605/OSF.IO/XA4CD).

Participants

Finally, a total of 151 participants took part in VR experiments between February - May 2025. The participants were recruited as a convenience sample using university emails, posters and social media. The inclusion criteria were that study subjects are 18–65 years-old healthy individuals, while no prior VR or cycling experience was required. Mental and physical health and use of anticholinergic medications was self-reported in the survey. Participants were instructed to avoid alcohol and caffeinated drinks consumption at least two hours before the experiment.

After accounting for data loss due technical issues and excluding individuals for which the stress test did not work, the final sample included a total of 138 participants that were a mixture of university students and staff, as well as the general public. The sample included 58.7% female/38.4% male, mostly 25–44 years old (67.4%) individuals with generally high education level (79.8% with Bachelor degree or higher). The participants reported good mental (mean = 3.78) and physical (mean = 3.88) health, and not very stressful day before the experiments (mean = 3.97, 10-point scale). In total 71.9% of participants stated to have experienced Virtual Reality before. Participants also reported generally no cyber sickness symptoms during the experiment (mean = 1.32) based on the four-point Likert scale (1 = none; 4 = severe) of the Simulator Sickness Questionnaire83. Chi-squared tests indicated no statistically significant differences in background variables (age, gender, education, stressful day, physical health, mental health, cycling frequency and nature connectedness) between study subjects in the different groups, indicating successful randomisation.

Study design and experimental procedure

The experiment used a between-subjects design and each participant was blindly randomised to only one of four environmental VR conditions: one built and three greenery scenarios. We opted for a between-subjects design for two main reasons: to avoid carry-over effects of stressors and the VR environments within subjects, and to reduce the time of wearing the headset, which can cause discomfort or cybersickness due to prolonged presence in VR84.

The experiments were carried out in a laboratory setting at the university premises. The temperature, humidity and light conditions in the room were kept consistent. The experimental set up consisted of a bicycle with an attached Garmin Tacx bike trainer, a Varjo XR-3 MR headset, wireless earphones, and wireless BioNomadix transmitter bracelet measuring Electrodermal Activity (EDA) through skin conductance (see Fig. 6 for demonstration by a member of the research team). Skin Conductance Levels (SCL) was acquired continuously at a 2000 Hz rate using two dry electrodes on the middle and index fingers. A small fan for simulating breeze during cycling was used to increase the multisensory experience. Detailed description of the experimental set up, hardware and software components, codes and data collection protocols can be found in https://github.com/DigitalGeographyLab/GREENTRAVELVR.git and85. Informed Consent for publication of images were obtained (Figs. 6 and 7).

The experiment including several phases and it took approximately 1–1,5 h (Fig. 7). Prior to the experiment, participants were sent general instructions, overview of the research study and a Data Privacy Statement. Upon arrival, participants were welcomed and shown a short introduction video of the different phases and devices used. Then, they signed the informed consent form and filled in a pre-questionnaire on their background and current perceived restoration and affective state. Then, two baselines were taken: a three-minute baseline for measuring SCL during rest while sitting on a chair, and a three-minute cycling baseline in VR, during which the participants cycled in a neutral environment with no sound. The VR cycling baseline served two purposes: as a habitualisation period to help participants get used to the cycling experience and the VR effects, and to measure a skin conductance baseline level during physical movement (cycling). After the VR baseline, participants were stressed for five minutes using a mental arithmetic stress task by subtracting backwards 17 from 1000 and having five sec for each response. Participants’ responses were evaluated as either “correct”, “false” or “too late” (if later than 5 s) and incorrect responses led to participants starting from the beginning. The effect of stress, measured as the difference in the mean SCL levels between the stress period and sitting baseline, was significant (p = < 0.001) and positive, indicating that the mental arithmetic test induced stress successfully. There were no statistically significant differences between the groups in the effect of stress.

After the stress test, participants cycled in one of the four randomised VR recovery conditions. Participants could cycle only in a straight line but could turn their head and look around. They were instructed to cycle at a leisurely speed, and the speed was topped at 15 km/h to reduce the physiological effects of rigorous exercise. To simulate an everyday active behaviour, participants were asked to imagine that “it is an early morning and they are cycling to work, but on the way to work they need to pick up some food from the supermarket”. Their task was to reach the supermarket which was located at the end of the road. After the Recovery period (8 min), participants went back to sitting on a chair and filled a post-questionnaire including the same psychological scales, as well as their overall experience and cybersickness symptoms. All participants were given a 20-euro grocery store voucher or two movie tickets as compensation.

Virtual environments

The Recovery period included cycling in a VR environment with a 1.5 km long straight street, which represents a typical distance for active mobility61,86. The VR environments were designed though an iterative process in a transdisciplinary team including VR technicians, urban planners and researchers with expertise in transportation, environmental psychology, health, ecology, and landscape architecture. The environments were generated in Unreal Game Engine (Epic Games) having built infrastructure and street dimensions with real-life proportions. To avoid place attachment and familiarity bias in responses, the environments were created to simulate the feeling of cycling in a city without being in a recognisable place. To increase ecological validity, vegetation assets and birdsongs were chosen to be typical for Nordic countries. High visual quality and multisensory experiences (auditory, wind through the fan) were used to increase sense of presence in the virtual world41. Since the experiments were conducted in Helsinki, local 3D assets (e.g. bus stops, traffic signs) were also used to increase contextual realism. In addition, no people or moving cars were present in the VR environments to provide feeling of safety and avoid confounding effects on physiological arousal.

We developed four VR environments representing different street greening designs and seasonal conditions: built, low level of greenery, high level of greenery and winter (Fig. 1, example videos also available in the GitHub dataset https://github.com/DigitalGeographyLab/GREENTRAVELVR.git). All environments had identical built infrastructure and distant ambient traffic noise, but vegetation and birdsongs were varied to respond to the diversity and level of naturalness. The built, low and high greenery environment also had the same light and seasonal conditions. The environments included the following elements:Built: control urban environment with very limited number of trees and vegetation; ambient traffic sound (continuous); summer sky and lighting conditions.

Low greenery: low level and diversity of greenery with monotonous tree-lined vegetation (trees and grass), limited number of bushes; in addition to the traffic sound, four times birdsong of Common Blackbird (Turdus merula) was played as spatial audio; summer sky and lighting conditions.

High greenery: higher quantity and more varied vegetation with structural diversity, more vegetation layers including bushes and seasonal pollinator flowers; green walls and other vertical vegetation; in addition to the traffic sound, eight times birdsong of Common Blackbird (Turdus merula), Chaffinch (Fringilla coelebs), or Great tit (Parus major) was played as spatial audio; summer sky and lighting conditions.

Winter: a multi-feature winter composite condition as a seasonal variation of the high greenery environment, same trees and bushes without leaves; snow piles randomly located on the ground; in addition to the traffic sound, one time a birdsong of House Sparrow (Passer domesticus) was played as spatial audio; winter overcast sky and low light condition.

Measures and data processing

Physiological measures

Skin conductance data was processed in AcqKnowledge programmme (BIOPAC Systems, Inc.), which was synchronised in real-time with the VR environments in Unreal Engine. We used tonic activity (skin conductance level) as outcome variable since it reflects progressive changes in galvanic skin response over time87. SCL data was first resampled to 200 Hz with a digital low pass filter at 1 hz. We then conducted visual investigation of the signal and correction of artefacts. Naturally, there is a high degree of variability across individuals in SCL responses. Therefore, we used an established approach of determining the relative changes of each participant average SCL during the experiment. SCLs were averaged every 10 s and adjusted to baseline measures for each individual according to the two types of physical state: sitting or cycling. Stress period SCL values (while sitting on the bike) were adjusted to the mean value of the last two minutes of the sitting baseline (ΔSCLstress = SCLstress− SCLsit_baseline). Recovery period (cycling in VR) was adjusted to the mean value of the last two minutes of the VR cycling baseline period (ΔSCLRecovery = SCLVR_Recovery− SCLVR_baseline). Adjusting recovery values to cycling in the neutral environment also allows for distinguishing the effect of the physical movement.

Pupil diameter and blink rates were collected with the built-in eye tracking features of the Varjo XR-3 headset. For the pupil diameter, we used the average diameter of both pupils. For blink rates, we used data loss on variables right_status and left_status in the eye-tracking system as a proxy for blinks88. Closures of the eye less than 30 ms and longer than 400 ms were ignored, while neighbouring blinks with gaps shorter than 100 ms were combined88–90. Lastly, the eye-tracking data was averaged for the 10 s intervals similarly to the SCL measures.

Psychological measures

Different psychological scales were used to measure participants’ perceived restoration, affective state and subjective vitality, which have been widely adopted in previous similar studies28,29,55,70. All scales were used before (pre-questionnaire) and after the experiment (post-questionnaire) to identify change in mental states.

The Restoration Outcome Scale (ROS)91 was used to measure self-reported restoration including six items (“I feel calm”, “I can concentrate well and feel alert”, “I have enthusiasm and energy for my everyday routines”, “I feel restored and relaxed”, “I can forget everyday worries”, “My thoughts are clear”). Self-reported emotional state was measured using the international Positive and Negative Affect Schedule short form (I-PANAS-SF) (Thompson, 2007), which is found to be a concise and reliable shorter version of PANAS92 suitable for various contexts. The scale includes five positive (Alert, Inspired, Determined, Attentive, Active) and five negative items (Upset, Hostile, Ashamed, Nervous, Afraid). To measure self-reported feeling of energy, the subjective vitality scale (SVS)93,94 was used including four of the original six items (“I feel alive and vital,” “I have energy and spirit,” “I look forward to each new day,” and “I don’t feel very energetic” as reversed item). The four items were selected to respond to local language nuances following similar approach to previous studies in the Finnish context28,95. ROS and SVS were rated on a seven-point Likert scale (1 = Not at all, 7 = Completely), and PANAS was rated on a five-point Likert Scale (1 = Very slightly or not at all, 5 = Extremely). All item scores were averaged and differences in pre and post responses were calculated for each participant.

Covariates

Several individual background variables were collected in the experiment including age, gender, education, self-reported overall physical and mental health (single-item WHO 5-point Likert scale; 1 = very bad, 5 = very good)96; how stressful was participants’ day before the experiment (10-point Likert scale, 0 = not stressful at all, 10 = extremely stressful), if they have used virtual reality before (yes/no), subjective connectedness with nature using the Inclusion of Nature in Self (INS) scale97, and participants cycling frequency (5-point Likert scale; 1 = less than once a month; 5 = 5–7 days/week).

Individual exposure to greenery

The Green View Index (GVI)98, the amount of visible greenery exposure of participants, was calculated by segmenting the participants headset video recordings at one second intervals. We used a pretrained ADE20K Dinat-L trained variant of the OneFormer Semantic segmentation model from the Hugging Face platform99–101. GVI was then calculated by assessing the share of the classes tree, grass, plant, flower from the total share of pixels and then, similar to the other measures, averaged for each 10 s interval.

Statistical analysis

First, we tested whether the environment type affects changes in psychological outcomes (ROS, PANAS-POS, PANAS-NEG, SVS) and the mean physiological outcomes (baseline-adjusted SCL, Pupil diam and N blinks). We fitted separate gaussian regression models to all these outcomes, using the environment type together with individual background variables (gender, age, stressful day, self-reported physical and mental health, nature connectedness and cycling frequency) as covariates. Here environment type, gender, and age groups were treated as categorical covariates, for which an effect was estimated for every factor level besides the baseline, while for the other covariates the effect was assumed to be linear.

We further tested the effect of the environment and the visible street greenery level (GVI) on the instantaneous physiological responses of each participant within the 8-min recovery period. We modelled the physiological response (below SCL) of participant i at time t in environment E(i) with an autoregressive regression model as follows:\documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$SCL_{{i,t}} \sim Normal\left( {\mu _{{i,t}} ,\sigma _{i} } \right)$$\end{document}\documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$\:{\mu\:}_{i,t}\:=\varphi\:{\mu\:}_{i,t-1}+\beta\:GV{I}_{i}\left(t\right)+{s}_{E\left(i\right)}\left(t\right)$$\end{document}

Here \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$\:GV{I}_{i}\left(t\right)$$\end{document} stand for the measured Green View Index of the participant i at time t (and 10 s before), \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$\:{s}_{E\left(i\right)}\left(t\right)$$\end{document} stands for the effect of the environment E(i) on the outcome as a function of time, and first term accounts for the autocorrelation in physiological measurements within individuals, where parameter ψ describes the degree of autocorrelation between the consecutive measurements from the same individual. The effects of the environments \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$\:{s}_{E\left(i\right)}\left(t\right)$$\end{document}, are assumed to be smooth non-linear functions in time, and are estimated from data. In addition, we assumed individual specific variation in SCL values, parameterized with \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$\:{\sigma\:}_{i}$$\end{document}, around their expected value. We assumed that this variation is constant throughout the experiment and unique for each individual, but assumed the \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$\:{\sigma\:}_{i}$$\end{document}‘s come from a shared population distribution.

Similar model structure was used to model pupil diameter and number of blinks during the Recovery period. For the SCL and pupil diameter measurements we assumed a Gaussian sampling distribution, while for the Number of blinks we assumed the outcomes to have a Poisson distribution. We did not model the individual-specific variances in the measurement outcomes for the number of blinks.

Thus, our target of inference are the following parameters ψ, β,\documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$\:{s}_{E\left(i\right)}\left(t\right)$$\end{document}, where the smooth effects of the environments are estimated for each of the four categories: built, low greenery, high greenery and winter. With \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$\:{\sigma\:}_{i}$$\end{document}‘s we adopted hierarchical modelling approach, in which we learn the population distribution of σ_i’s from data simultaneously with the individual \documentclass[12pt]{minimal}
\usepackage{amsmath}
\usepackage{wasysym}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{amsbsy}
\usepackage{mathrsfs}
\usepackage{upgreek}
\setlength{\oddsidemargin}{-69pt}
\begin{document}$$\:{\sigma\:}_{i}$$\end{document}‘s. We did Bayesian posterior inference on the model parameters, implementing the model fitting with MCMC sampling using brms-package (v.3) in R102, using default priors of the software for all the parameters. For each model we ran 4 MCMC chains, each with 5000 iterations, for which 2500 iterations were considered as warmup. This resulted in appropriate convergence diagnostics for all the models, e.g. Rhats being almost equal to one for all the parameters. Trace plots also indicated appropriate convergence of the chains.

Supplementary Information

Below is the link to the electronic supplementary material.

References

1.The Convention on Biological Diversity (CBD). The Biodiversity Plan for Life on Earth. (2022).

2.European Union. Regulation Nature Restoration and Amending Regulation (EU) 2022/869. (2024).

3.Lu Y Yang Y Sun G Gou Z Associations between overhead-view and eye-level urban greenness and cycling behaviors Cities 2019 88 10 18 10.1016/j.cities.2019.01.003 Lu, Y., Yang, Y., Sun, G. & Gou, Z. Associations between overhead-view and eye-level urban greenness and cycling behaviors. Cities88, 10–18 (2019).

4.Wang R Lu Y Wu X Liu Y Yao Y Relationship between eye-level greenness and cycling frequency around metro stations in Shenzhen, China: A big data approach Sustainable Cities Soc. 2020 59 102201 10.1016/j.scs.2020.102201 Wang, R., Lu, Y., Wu, X., Liu, Y. & Yao, Y. Relationship between eye-level greenness and cycling frequency around metro stations in Shenzhen, China: A big data approach. Sustainable Cities Soc.59, 102201 (2020).

5.Gao M Fang C Unveiling the nonlinear effects of street-level audiovisual environments on urban cycling volume J. Transp. Health 2026 49 102313 10.1016/j.jth.2026.102313 Gao, M. & Fang, C. Unveiling the nonlinear effects of street-level audiovisual environments on urban cycling volume. J. Transp. Health. 49, 102313 (2026).

6.Korpilo S Green and blue spaces matter for active mobility: Results from mapping perceived environmental exposure in five European cities Urban Forestry Urban Green. 2025 113 129088 10.1016/j.ufug.2025.129088 Korpilo, S. et al. Green and blue spaces matter for active mobility: Results from mapping perceived environmental exposure in five European cities. Urban Forestry Urban Green.113, 129088 (2025).

7.Nawrath M Kowarik I Fischer LK The influence of green streets on cycling behavior in European cities Landsc. Urban Plann. 2019 190 103598 10.1016/j.landurbplan.2019.103598 Nawrath, M., Kowarik, I. & Fischer, L. K. The influence of green streets on cycling behavior in European cities. Landsc. Urban Plann.190, 103598 (2019).

8.Song J The effect of eye-level street greenness exposure on walking satisfaction: The mediating role of noise and PM2.5 Urban Forestry Urban Green. 2022 77 127752 10.1016/j.ufug.2022.127752 Song, J. et al. The effect of eye-level street greenness exposure on walking satisfaction: The mediating role of noise and PM2.5. Urban Forestry Urban Green.77, 127752 (2022).

9.Ta N Li H Chai Y Wu J The impact of green space exposure on satisfaction with active travel trips Transp. Res. Part. D: Transp. Environ. 2021 99 103022 10.1016/j.trd.2021.103022 Ta, N., Li, H., Chai, Y. & Wu, J. The impact of green space exposure on satisfaction with active travel trips. Transp. Res. Part. D: Transp. Environ.99, 103022 (2021).

10.Yu C Kwan MP Dynamic greenspace exposure, individual mental health status and momentary stress level: A study using multiple greenspace measurements Health Place 2024 86 103213 38447264 10.1016/j.healthplace.2024.103213 Yu, C. & Kwan, M. P. Dynamic greenspace exposure, individual mental health status and momentary stress level: A study using multiple greenspace measurements. Health Place. 86, 103213 (2024).10.1016/j.healthplace.2024.103213 38447264

11.Zhang L Zhou S Kwan MP Chen F Lin R Impacts of Individual Daily Greenspace Exposure on Health Based on Individual Activity Space and Structural Equation Modeling Int. J. Environ. Res. Public Health 2018 15 2323 30360440 PMC6210249 10.3390/ijerph15102323 Zhang, L., Zhou, S., Kwan, M. P., Chen, F. & Lin, R. Impacts of individual daily greenspace exposure on health based on individual activity space and structural equation modeling. Int. J. Environ. Res. Public Health. 15, 2323 (2018).10.3390/ijerph15102323 PMC6210249 30360440

12.Coleman AF Harper RW Eisenman TS Warner SH Wilkinson MA Street Tree Structure, Function, and Value: A Review of Scholarly Research (1997–2020) Forests 2022 13 1779 10.3390/f13111779 Coleman, A. F., Harper, R. W., Eisenman, T. S., Warner, S. H. & Wilkinson, M. A. Street tree structure, function, and value: a review of scholarly research (1997–2020). Forests13, 1779 (2022).

13.Gao M Fang C Decoding the impact of audiovisual street environment features on cycling volumes: Insights from street view imagery and machine learning Transp. Res. Part. A: Policy Pract. 2025 199 104586 Gao, M. & Fang, C. Decoding the impact of audiovisual street environment features on cycling volumes: Insights from street view imagery and machine learning. Transp. Res. Part. A: Policy Pract.199, 104586 (2025).

14.Poom A Willberg E Toivonen T Environmental exposure during travel: A research review and suggestions forward Health Place 2021 70 102584 34020232 10.1016/j.healthplace.2021.102584 Poom, A., Willberg, E. & Toivonen, T. Environmental exposure during travel: A research review and suggestions forward. Health Place. 70, 102584 (2021).10.1016/j.healthplace.2021.102584 34020232

15.Sarkar C Exploring associations between urban green, street design and walking: Results from the Greater London boroughs Landsc. Urban Plann. 2015 143 112 125 10.1016/j.landurbplan.2015.06.013 Sarkar, C. et al. Exploring associations between urban green, street design and walking: Results from the Greater London boroughs. Landsc. Urban Plann.143, 112–125 (2015).

16.Korpilo S Restoring nature, enhancing active mobility: The role of street greenery in the EU’s 2024 restoration law Ambio 2025 54 1611 1620 10.1007/s13280-025-02178-w 40208550 PMC12405100 Korpilo, S. et al. Restoring nature, enhancing active mobility: The role of street greenery in the EU’s 2024 restoration law. Ambio54, 1611–1620 (2025).10.1007/s13280-025-02178-w PMC12405100 40208550

17.Klein R Willberg E Korpilo S Toivonen T Temporal variation in travel greenery across 86 cities in Europe Urban Forestry Urban Green. 2024 102 128566 10.1016/j.ufug.2024.128566 Klein, R., Willberg, E., Korpilo, S. & Toivonen, T. Temporal variation in travel greenery across 86 cities in Europe. Urban Forestry Urban Green.102, 128566 (2024).

18.Torkko J Malekzadeh M Willberg E Toivonen T Same greenery, different Green View Index: Inconsistent greenery assessment in semantic segmentation and street view imagery Urban Forestry Urban Green. 2025 112 128969 10.1016/j.ufug.2025.128969 Torkko, J., Malekzadeh, M., Willberg, E. & Toivonen, T. Same greenery, different Green View Index: Inconsistent greenery assessment in semantic segmentation and street view imagery. Urban Forestry Urban Green.112, 128969 (2025).

19.Helbich M Spatiotemporal Contextual Uncertainties in Green Space Exposure Measures: Exploring a Time Series of the Normalized Difference Vegetation Indices IJERPH 2019 16 852 30857201 PMC6427170 10.3390/ijerph16050852 Helbich, M. Spatiotemporal contextual uncertainties in green space exposure measures: exploring a time series of the normalized difference vegetation indices. IJERPH16, 852 (2019).10.3390/ijerph16050852 PMC6427170 30857201

20.Stamps AE Use of Photographs to Simulate Environments: A Meta-Analysis Percept. Mot Skills 1990 71 907 913 10.2466/pms.1990.71.3.907 21162439 Stamps, A. E. Use of photographs to simulate environments: a meta-analysis. Percept. Mot Skills. 71, 907–913 (1990).10.2466/22.24.27.PMS.111.5.355-364 21162439

21.Zube EH Sell JL Taylor JG Landscape perception: Research, application and theory Landsc. Plann. 1982 9 1 33 10.1016/0304-3924(82)90009-0 Zube, E. H., Sell, J. L. & Taylor, J. G. Landscape perception: Research, application and theory. Landsc. Plann.9, 1–33 (1982).

22.Yeo NL What is the best way of delivering virtual nature for improving mood? An experimental comparison of high definition TV, 360° video, and computer generated virtual reality J. Environ. Psychol. 2020 72 101500 33390641 PMC7772948 10.1016/j.jenvp.2020.101500 Yeo, N. L. et al. What is the best way of delivering virtual nature for improving mood? An experimental comparison of high definition TV, 360° video, and computer generated virtual reality. J. Environ. Psychol.72, 101500 (2020).10.1016/j.jenvp.2020.101500 PMC7772948 33390641

23.Baghaei N Virtual Reality for Supporting the Treatment of Depression and Anxiety: Scoping Review JMIR Ment Health 2021 8 e29681 34554097 PMC8498902 10.2196/29681 Baghaei, N. et al. Virtual reality for supporting the treatment of depression and anxiety: scoping review. JMIR Ment Health. 8, e29681 (2021).10.2196/29681 PMC8498902 34554097

24.Cullen AJ Considerations and practical protocols for using virtual reality in psychological research and practice, as evidenced through exposure-based therapy Behav. Res. 2021 53 2725 2742 10.3758/s13428-021-01543-3 34080173 Cullen, A. J. et al. Considerations and practical protocols for using virtual reality in psychological research and practice, as evidenced through exposure-based therapy. Behav. Res.53, 2725–2742 (2021).10.3758/s13428-021-01543-3 34080173

25.White, M. P. et al. A prescription for “nature” – the potential of using virtual nature in therapeutics. NDT14, 3001–3013 (2018).10.2147/NDT.S179038 PMC6231506 30510422

26.Hedblom M Reduction of physiological stress by urban green space in a multisensory virtual experiment Sci. Rep. 2019 9 10113 10.1038/s41598-019-46099-7 31300656 PMC6625985 Hedblom, M. et al. Reduction of physiological stress by urban green space in a multisensory virtual experiment. Sci. Rep.9, 10113 (2019).10.1038/s41598-019-46099-7 PMC6625985 31300656

27.Huang Q Yang M Jane H Li S Bauer N Trees, grass, or concrete? The effects of different types of environments on stress reduction Landsc. Urban Plann. 2020 193 103654 10.1016/j.landurbplan.2019.103654 Huang, Q., Yang, M., Jane, H., Li, S. & Bauer, N. Trees, grass, or concrete? The effects of different types of environments on stress reduction. Landsc. Urban Plann.193, 103654 (2020).

28.Kari T Ojala A Kurkilahti M Tyrväinen L Comparison between three different delivery technologies of virtual nature on psychological state related to general stress recovery: An experimental study J. Environ. Psychol. 2024 100 102452 10.1016/j.jenvp.2024.102452 Kari, T., Ojala, A., Kurkilahti, M. & Tyrväinen, L. Comparison between three different delivery technologies of virtual nature on psychological state related to general stress recovery: An experimental study. J. Environ. Psychol.100, 102452 (2024).

29.Ojala A Short virtual nature breaks in the office environment can restore stress: An experimental study J. Environ. Psychol. 2022 84 101909 10.1016/j.jenvp.2022.101909 Ojala, A. et al. Short virtual nature breaks in the office environment can restore stress: An experimental study. J. Environ. Psychol.84, 101909 (2022).

30.Yen HY Huang HY Virtual cycling in urban green and blue vs gray spaces: The efficacy of mental health and physical activity Public. Health 2025 243 105723 40250130 10.1016/j.puhe.2025.105723 Yen, H. Y. & Huang, H. Y. Virtual cycling in urban green and blue vs gray spaces: The efficacy of mental health and physical activity. Public. Health. 243, 105723 (2025).10.1016/j.puhe.2025.105723 40250130

31.Mavrantza AM Bigliassi M Calogiuri G Psychophysiological mechanisms underlying the effects of outdoor green and virtual green exercise during self-paced walking Int. J. Psychophysiol. 2023 184 39 50 36572348 10.1016/j.ijpsycho.2022.12.006 Mavrantza, A. M., Bigliassi, M. & Calogiuri, G. Psychophysiological mechanisms underlying the effects of outdoor green and virtual green exercise during self-paced walking. Int. J. Psychophysiol.184, 39–50 (2023).10.1016/j.ijpsycho.2022.12.006 36572348

32.Brambilla E Effects of immersive virtual nature on nature connectedness: A systematic review and meta-analysis Digit. HEALTH 2024 10 20552076241234639 38533309 PMC10964443 10.1177/20552076241234639 Brambilla, E. et al. Effects of immersive virtual nature on nature connectedness: A systematic review and meta-analysis. Digit. Health. 10, 20552076241234639 (2024).10.1177/20552076241234639 PMC10964443 38533309

33.Birenboim A The utilization of immersive virtual environments for the investigation of environmental preferences Landsc. Urban Plann. 2019 189 129 138 10.1016/j.landurbplan.2019.04.011 Birenboim, A. et al. The utilization of immersive virtual environments for the investigation of environmental preferences. Landsc. Urban Plann.189, 129–138 (2019).

34.Yuan S A virtual reality investigation of factors influencing landscape preferences: Natural elements, emotions, and media creation Landsc. Urban Plann. 2023 230 104616 10.1016/j.landurbplan.2022.104616 Yuan, S. et al. A virtual reality investigation of factors influencing landscape preferences: Natural elements, emotions, and media creation. Landsc. Urban Plann.230, 104616 (2023).

35.Ghanbari M Dijst M McCall R Perchoux C The use of Virtual Reality (VR) to assess the impact of geographical environments on walking and cycling: a systematic literature review Int. J. Health Geogr. 2024 23 15 10.1186/s12942-024-00375-6 38851727 PMC11162039 Ghanbari, M., Dijst, M., McCall, R. & Perchoux, C. The use of Virtual Reality (VR) to assess the impact of geographical environments on walking and cycling: a systematic literature review. Int. J. Health Geogr.23, 15 (2024).10.1186/s12942-024-00375-6 PMC11162039 38851727

36.Lieze M Using virtual reality to investigate physical environmental factors related to cycling in older adults: A comparison between two methodologies J. Transp. Health 2020 19 100921 10.1016/j.jth.2020.100921 Lieze, M. et al. Using virtual reality to investigate physical environmental factors related to cycling in older adults: A comparison between two methodologies. J. Transp. Health. 19, 100921 (2020).

37.Nakamura K Experimental analysis of walkability evaluation using virtual reality application Environ. Plann. B: Urban Analytics City Sci. 2021 48 2481 2496 Nakamura, K. Experimental analysis of walkability evaluation using virtual reality application. Environ. Plann. B: Urban Analytics City Sci.48, 2481–2496 (2021).

38.Nazemi M Rababah B Ramos D Zhao T Farooq B Decoding pedestrian stress on urban streets using electrodermal activity monitoring in virtual immersive reality Transp. Res. Part. C: Emerg. Technol. 2025 171 104952 10.1016/j.trc.2024.104952 Nazemi, M., Rababah, B., Ramos, D., Zhao, T. & Farooq, B. Decoding pedestrian stress on urban streets using electrodermal activity monitoring in virtual immersive reality. Transp. Res. Part. C: Emerg. Technol.171, 104952 (2025).

39.Browning MHEM Saeidi-Rizi F McAnirlin O Yoon H Pei Y The Role of Methodological Choices in the Effects of Experimental Exposure to Simulated Natural Landscapes on Human Health and Cognitive Performance: A Systematic Review Environ. Behav. 2021 53 687 731 10.1177/0013916520906481 Browning, M. H. E. M., Saeidi-Rizi, F., McAnirlin, O., Yoon, H. & Pei, Y. The role of methodological choices in the effects of experimental exposure to simulated natural landscapes on human health and cognitive performance: a systematic review. Environ. Behav.53, 687–731 (2021).

40.Diniz Bernardo P Bains A Westwood S Mograbi DC Mood Induction Using Virtual Reality: a Systematic Review of Recent Findings J. technol. behav. sci. 2021 6 3 24 10.1007/s41347-020-00152-9 Diniz Bernardo, P., Bains, A., Westwood, S. & Mograbi, D. C. Mood induction using virtual reality: a systematic review of recent findings. J. Technol. Behav. Sci.6, 3–24 (2021).

41.Spano G Virtual nature, psychological and psychophysiological outcomes: A systematic review J. Environ. Psychol. 2023 89 102044 10.1016/j.jenvp.2023.102044 Spano, G. et al. Virtual nature, psychological and psychophysiological outcomes: A systematic review. J. Environ. Psychol.89, 102044 (2023).

42.Wilson, E. O. Biophilia (Harvard University Press, 1984).

43.Ulrich RS Stress recovery during exposure to natural and urban environments J. Environ. Psychol. 1991 11 201 230 10.1016/S0272-4944(05)80184-7 Ulrich, R. S. et al. Stress recovery during exposure to natural and urban environments. J. Environ. Psychol.11, 201–230 (1991).

44.Kaplan, R. & Kaplan, S. The Experience of Nature: A Psychological Perspective (Cambridge University Press, 1989).

45.Fukano Y Soga M Greenery hypothesis: An evolutionary explanation for why presence/absence of green affects humans People Nat. 2024 6 422 434 10.1002/pan3.10619 Fukano, Y. & Soga, M. Greenery hypothesis: An evolutionary explanation for why presence/absence of green affects humans. People Nat.6, 422–434 (2024).

46.Litleskare S Calogiuri G Seasonal Variations in the Effectiveness of Immersive Virtual Nature HERD 2023 16 219 232 36154715 10.1177/19375867221127420 PMC9755693 Litleskare, S. & Calogiuri, G. Seasonal variations in the effectiveness of immersive virtual nature. HERD16, 219–232 (2023).10.1177/19375867221127420 PMC9755693 36154715

47.Du X Yuan S Lam TLH Browning MHEM Tan Z Concordance between psychological and physiological measures of nature contact’s mental benefits: A systematic review and exploratory synthesis J. Environ. Psychol. 2026 109 102863 10.1016/j.jenvp.2025.102863 Du, X., Yuan, S., Lam, T. L. H., Browning, M. H. E. M. & Tan, Z. Concordance between psychological and physiological measures of nature contact’s mental benefits: A systematic review and exploratory synthesis. J. Environ. Psychol.109, 102863 (2026).

48.Moreno-Arjonilla J López-Ruiz A Jiménez-Pérez JR Callejas-Aguilera JE Jurado JM Eye-tracking on virtual reality: a survey Virtual Real. 2024 28 38 10.1007/s10055-023-00903-y Moreno-Arjonilla, J., López-Ruiz, A., Jiménez-Pérez, J. R., Callejas-Aguilera, J. E. & Jurado, J. M. Eye-tracking on virtual reality: a survey. Virtual Real.28, 38 (2024).

49.Ranti C Jones W Klin A Shultz S Blink Rate Patterns Provide a Reliable Measure of Individual Engagement with Scene Content Sci. Rep. 2020 10 8267 10.1038/s41598-020-64999-x 32427957 PMC7237680 Ranti, C., Jones, W., Klin, A. & Shultz, S. Blink rate patterns provide a reliable measure of individual engagement with scene content. Sci. Rep.10, 8267 (2020).10.1038/s41598-020-64999-x PMC7237680 32427957

50.Slovák M Emotional arousal in patients with functional movement disorders: A pupillometry study J. Psychosom. Res. 2022 162 111043 36166959 10.1016/j.jpsychores.2022.111043 Slovák, M. et al. Emotional arousal in patients with functional movement disorders: A pupillometry study. J. Psychosom. Res.162, 111043 (2022).10.1016/j.jpsychores.2022.111043 36166959

51.Boucsein, W. Electrodermal Activity (Springer, 2012).

52.Browning MHEM Mimnaugh KJ Van Riper CJ Laurent HK LaValle SM Can Simulated Nature Support Mental Health? Comparing Short, Single-Doses of 360-Degree Nature Videos in Virtual Reality With the Outdoors Front. Psychol. 2020 10 2667 32010003 PMC6974516 10.3389/fpsyg.2019.02667 Browning, M. H. E. M., Mimnaugh, K. J., Van Riper, C. J., Laurent, H. K. & LaValle, S. M. Can Simulated Nature Support Mental Health? Comparing Short, Single-Doses of 360-Degree Nature Videos in Virtual Reality With the Outdoors. Front. Psychol.10, 2667 (2020).10.3389/fpsyg.2019.02667 PMC6974516 32010003

53.Yu CP Lee HY Luo XY The effect of virtual reality forest and urban environments on physiological and psychological responses Urban Forestry Urban Green. 2018 35 106 114 10.1016/j.ufug.2018.08.013 Yu, C. P., Lee, H. Y. & Luo, X. Y. The effect of virtual reality forest and urban environments on physiological and psychological responses. Urban Forestry Urban Green.35, 106–114 (2018).

54.Mostajeran F Fischer M Steinicke F Kühn S Effects of exposure to immersive computer-generated virtual nature and control environments on affect and cognition Sci. Rep. 2023 13 220 10.1038/s41598-022-26750-6 36604527 PMC9815073 Mostajeran, F., Fischer, M., Steinicke, F. & Kühn, S. Effects of exposure to immersive computer-generated virtual nature and control environments on affect and cognition. Sci. Rep.13, 220 (2023).10.1038/s41598-022-26750-6 PMC9815073 36604527

55.Korpilo S Landscape and soundscape quality promote stress recovery in nearby urban nature: A multisensory field experiment Urban Forestry Urban Green. 2024 95 128286 10.1016/j.ufug.2024.128286 Korpilo, S. et al. Landscape and soundscape quality promote stress recovery in nearby urban nature: A multisensory field experiment. Urban Forestry Urban Green.95, 128286 (2024).

56.Hedblom M Reduction of physiological stress by urban green space in a multisensory virtual experiment Sci. Rep. 2019 9 10113 10.1038/s41598-019-46099-7 31300656 PMC6625985 Hedblom, M. et al. Reduction of physiological stress by urban green space in a multisensory virtual experiment. Sci. Rep.9, 10113 (2019).10.1038/s41598-019-46099-7 PMC6625985 31300656

57.Anderson AP Relaxation with Immersive Natural Scenes Presented Using Virtual Reality aerosp. med. hum. perform. 2017 88 520 526 28539139 10.3357/AMHP.4747.2017 Anderson, A. P. et al. Relaxation with Immersive Natural Scenes Presented Using Virtual Reality. Aerosp. Med. Hum. Perform.88, 520–526 (2017).10.3357/AMHP.4747.2017 28539139

58.Schebella, M. F., Weber, D., Schultz, L. & Weinstein, P. The nature of reality: human stress recovery during exposure to biodiverse, multisensory virtual environments. IJERPH17, 56 (2019).10.3390/ijerph17010056 PMC6982328 31861700

59.Zhang K Synergistic role of audio-visual perceptions in promoting bikeshare for active travel Transp. Res. Part. D: Transp. Environ. 2025 145 104806 10.1016/j.trd.2025.104806 Zhang, K. et al. Synergistic role of audio-visual perceptions in promoting bikeshare for active travel. Transp. Res. Part. D: Transp. Environ.145, 104806 (2025).

60.Yin J Effects of biophilic indoor environment on stress and anxiety recovery: A between-subjects experiment in virtual reality Environ. Int. 2020 136 105427 31881421 10.1016/j.envint.2019.105427 Yin, J. et al. Effects of biophilic indoor environment on stress and anxiety recovery: A between-subjects experiment in virtual reality. Environ. Int.136, 105427 (2020).10.1016/j.envint.2019.105427 31881421

61.Goel R Cycling behaviour in 17 countries across 6 continents: levels of cycling, who cycles, for what purpose, and how far? Transp. Reviews 2022 42 58 81 10.1080/01441647.2021.1915898 Goel, R. et al. Cycling behaviour in 17 countries across 6 continents: levels of cycling, who cycles, for what purpose, and how far? Transp. Reviews. 42, 58–81 (2022).

62.Willberg E Fink C Klein R Heinonen R Toivonen T Green or short: choose one’ - A comparison of walking accessibility and greenery in 43 European cities Comput. Environ. Urban Syst. 2024 113 102168 10.1016/j.compenvurbsys.2024.102168 Willberg, E., Fink, C., Klein, R., Heinonen, R. & Toivonen, T. Green or short: choose one’ - A comparison of walking accessibility and greenery in 43 European cities. Comput. Environ. Urban Syst.113, 102168 (2024).

63.Huang D Kyttä M Kajosaari A Xie X Zhang J One approach does not fit all settings: Exploring the effects of natural and built environments on running pleasantness across places Build. Environ. 2023 245 110961 10.1016/j.buildenv.2023.110961 Huang, D., Kyttä, M., Kajosaari, A., Xie, X. & Zhang, J. One approach does not fit all settings: Exploring the effects of natural and built environments on running pleasantness across places. Build. Environ.245, 110961 (2023).

64.Calogiuri, G., Litleskare, S. & Fröhlich, F. Physical activity and virtual nature: perspectives on the health and behavioral benefits of virtual green exercise. In Nature and Health: Physical Activity in Nature (Routledge, 2021).

65.Li H Effect of a Virtual Reality-Based Restorative Environment on the Emotional and Cognitive Recovery of Individuals with Mild-to-Moderate Anxiety and Depression IJERPH 2021 18 9053 34501643 PMC8430968 10.3390/ijerph18179053 Li, H. et al. Effect of a virtual reality-based restorative environment on the emotional and cognitive recovery of individuals with mild-to-moderate anxiety and depression. IJERPH18, 9053 (2021).10.3390/ijerph18179053 PMC8430968 34501643

66.Mostajeran F Krzikawski J Steinicke F Kühn S Effects of exposure to immersive videos and photo slideshows of forest and urban environments Sci. Rep. 2021 11 3994 10.1038/s41598-021-83277-y 33597577 PMC7889877 Mostajeran, F., Krzikawski, J., Steinicke, F. & Kühn, S. Effects of exposure to immersive videos and photo slideshows of forest and urban environments. Sci. Rep.11, 3994 (2021).10.1038/s41598-021-83277-y PMC7889877 33597577

67.Felnhofer A Is virtual reality emotionally arousing? Investigating five emotion inducing virtual park scenarios Int. J. Hum. Comput. Stud. 2015 82 48 56 10.1016/j.ijhcs.2015.05.004 Felnhofer, A. et al. Is virtual reality emotionally arousing? Investigating five emotion inducing virtual park scenarios. Int. J. Hum. Comput. Stud.82, 48–56 (2015).

68.Kaplan S The restorative benefits of nature: Toward an integrative framework J. Environ. Psychol. 1995 15 169 182 10.1016/0272-4944(95)90001-2 Kaplan, S. The restorative benefits of nature: Toward an integrative framework. J. Environ. Psychol.15, 169–182 (1995).

69.Winthorst WH Bos EH Roest AM De Jonge P Seasonality of mood and affect in a large general population sample PLoS ONE 2020 15 e0239033 32925966 PMC7489524 10.1371/journal.pone.0239033 Winthorst, W. H., Bos, E. H., Roest, A. M. & De Jonge, P. Seasonality of mood and affect in a large general population sample. PLoS ONE. 15, e0239033 (2020).10.1371/journal.pone.0239033 PMC7489524 32925966

70.Ojala A Psychological and physiological effects of a wooden office room on human well-being: Results from a randomized controlled trial J. Environ. Psychol. 2023 89 102059 10.1016/j.jenvp.2023.102059 Ojala, A. et al. Psychological and physiological effects of a wooden office room on human well-being: Results from a randomized controlled trial. J. Environ. Psychol.89, 102059 (2023).

71.Russell JA A circumplex model of affect J. Personal. Soc. Psychol. 1980 39 1161 1178 10.1037/h0077714 Russell, J. A. A circumplex model of affect. J. Personal. Soc. Psychol.39, 1161–1178 (1980).

72.Maljkovic V Martini P Short-term memory for scenes with affective content J. Vis. 2005 5 6 10.1007/BF03182595 PMC1307514 15929647 Maljkovic, V. & Martini, P. Short-term memory for scenes with affective content. J. Vis.5, 6 (2005).10.1167/5.3.6 PMC1307514 15929647

73.Quevedo J Differential effects of emotional arousal in short- and long-term memory in healthy adults Neurobiol. Learn. Mem. 2003 79 132 135 12591221 10.1016/S1074-7427(02)00034-5 Quevedo, J. et al. Differential effects of emotional arousal in short- and long-term memory in healthy adults. Neurobiol. Learn. Mem.79, 132–135 (2003).10.1016/s1074-7427(02)00034-5 12591221

74.Sun Y Physiological and affective responses to green space virtual reality among pregnant women Environ. Res. 2023 216 114499 36208780 10.1016/j.envres.2022.114499 Sun, Y. et al. Physiological and affective responses to green space virtual reality among pregnant women. Environ. Res.216, 114499 (2023).10.1016/j.envres.2022.114499 36208780

75.Jiang B He J Chen J Larsen L Wang H Perceived Green at Speed: A Simulated Driving Experiment Raises New Questions for Attention Restoration Theory and Stress Reduction Theory Environ. Behav. 2021 53 296 335 10.1177/0013916520947111 Jiang, B., He, J., Chen, J., Larsen, L. & Wang, H. Perceived green at speed: a simulated driving experiment raises new questions for attention restoration theory and stress reduction theory. Environ. Behav.53, 296–335 (2021).

76.Song Q Huang Y Li W Chen F Qiu W Unraveling the effects of micro-level street environment on dockless bikeshare in Ithaca Transp. Res. Part. D: Transp. Environ. 2024 132 104256 10.1016/j.trd.2024.104256 Song, Q., Huang, Y., Li, W., Chen, F. & Qiu, W. Unraveling the effects of micro-level street environment on dockless bikeshare in Ithaca. Transp. Res. Part. D: Transp. Environ.132, 104256 (2024).

77.Stern JA Boyer D Schroeder D Blink Rate: A Possible Measure of Fatigue Hum. Factors 1994 36 285 297 8070793 10.1177/001872089403600209 Stern, J. A., Boyer, D. & Schroeder, D. Blink Rate: A Possible Measure of Fatigue. Hum. Factors. 36, 285–297 (1994).10.1177/001872089403600209 8070793

78.Valtchanov D Ellard CG Cognitive and affective responses to natural scenes: Effects of low level visual properties on preference, cognitive load and eye-movements J. Environ. Psychol. 2015 43 184 195 10.1016/j.jenvp.2015.07.001 Valtchanov, D. & Ellard, C. G. Cognitive and affective responses to natural scenes: Effects of low level visual properties on preference, cognitive load and eye-movements. J. Environ. Psychol.43, 184–195 (2015).

79.Zhang X Beyond just green: Explaining and predicting restorative potential of urban landscapes using panorama-based metrics Landsc. Urban Plann. 2024 247 105044 10.1016/j.landurbplan.2024.105044 Zhang, X. et al. Beyond just green: Explaining and predicting restorative potential of urban landscapes using panorama-based metrics. Landsc. Urban Plann.247, 105044 (2024).

80.Thalmayer AG Toscanelli C Arnett JJ The neglected 95% revisited: Is American psychology becoming less American? Am. Psychol. 2021 76 116 129 32271027 10.1037/amp0000622 Thalmayer, A. G., Toscanelli, C. & Arnett, J. J. The neglected 95% revisited: Is American psychology becoming less American? Am. Psychol.76, 116–129 (2021).10.1037/amp0000622 32271027

81.Henrich J Heine SJ Norenzayan A The weirdest people in the world? Behav. Brain Sci. 2010 33 61 83 20550733 10.1017/S0140525X0999152X Henrich, J., Heine, S. J. & Norenzayan, A. The weirdest people in the world? Behav. Brain Sci.33, 61–83 (2010).10.1017/S0140525X0999152X 20550733

82.Green P MacLeod CJ SIMR: an R package for power analysis of generalized linear mixed models by simulation Methods Ecol. Evol. 2016 7 493 498 10.1111/2041-210X.12504 Green, P. & MacLeod, C. J. SIMR: an R package for power analysis of generalized linear mixed models by simulation. Methods Ecol. Evol.7, 493–498 (2016). https://cran.r-project.org/web/packages/simr/

83.Kennedy RS Lane NE Berbaum KS Lilienthal MG Simulator Sickness Questionnaire: An Enhanced Method for Quantifying Simulator Sickness Int. J. Aviat. Psychol. 1993 3 203 220 10.1207/s15327108ijap0303_3 Kennedy, R. S., Lane, N. E., Berbaum, K. S. & Lilienthal, M. G. Simulator Sickness Questionnaire: An Enhanced Method for Quantifying Simulator Sickness. Int. J. Aviat. Psychol.3, 203–220 (1993).

84.Cossio S Cybersickness and discomfort from head-mounted displays delivering fully immersive virtual reality: A systematic review Nurse Educ. Pract. 2025 85 104376 40267853 10.1016/j.nepr.2025.104376 Cossio, S. et al. Cybersickness and discomfort from head-mounted displays delivering fully immersive virtual reality: A systematic review. Nurse Educ. Pract.85, 104376 (2025).10.1016/j.nepr.2025.104376 40267853

85.Ravindran, O. et al. GREENTRAVEL VR: Technical Report University of Helsinki, Helsinki, (2025).

86.Silva C Büttner B Seisenberger S Rauli A Proximity-centred accessibility—A conceptual debate involving experts and planning practitioners J. Urban Mobil. 2023 4 100060 10.1016/j.urbmob.2023.100060 Silva, C., Büttner, B., Seisenberger, S. & Rauli, A. Proximity-centred accessibility—A conceptual debate involving experts and planning practitioners. J. Urban Mobil.4, 100060 (2023).

87.Boucsein W Fowles D Grimnes S Ben-Shakhar, G. Publication recommendations for electrodermal measurements Psychophysiology 2012 49 1017 1034 22680988 10.1111/j.1469-8986.2012.01384.x Boucsein, W., Fowles, D. & Grimnes, S. Ben-Shakhar, G. Publication recommendations for electrodermal measurements. Psychophysiology49, 1017–1034 (2012).10.1111/j.1469-8986.2012.01384.x 22680988

88.Nyström M Andersson R Niehorster DC Hessels RS Hooge I T. C. What is a blink? Classifying and characterizing blinks in eye openness signals Behav. Res. 2024 56 3280 3299 10.3758/s13428-023-02333-9 PMC11133197 38424292 Nyström, M., Andersson, R., Niehorster, D. C., Hessels, R. S. & Hooge, I. T. C. What is a blink? Classifying and characterizing blinks in eye openness signals. Behav. Res.56, 3280–3299 (2024).10.3758/s13428-023-02333-9 PMC11133197 38424292

89.Demiral ŞB Striatal D1 and D2 receptor availability are selectively associated with eye-blink rates after methylphenidate treatment Commun. Biol. 2022 5 1015 10.1038/s42003-022-03979-5 36163254 PMC9513088 Demiral, Ş. B. et al. Striatal D1 and D2 receptor availability are selectively associated with eye-blink rates after methylphenidate treatment. Commun. Biol.5, 1015 (2022).10.1038/s42003-022-03979-5 PMC9513088 36163254

90.Hoogerbrugge AJ Seeing the Forrest through the trees: Oculomotor metrics are linked to heart rate PLoS ONE 2022 17 e0272349 35917377 PMC9345484 10.1371/journal.pone.0272349 Hoogerbrugge, A. J. et al. Seeing the Forrest through the trees: Oculomotor metrics are linked to heart rate. PLoS ONE. 17, e0272349 (2022).10.1371/journal.pone.0272349 PMC9345484 35917377

91.Korpela KM Ylén M Tyrväinen L Silvennoinen H Determinants of restorative experiences in everyday favorite places Health Place 2008 14 636 652 18037332 10.1016/j.healthplace.2007.10.008 Korpela, K. M., Ylén, M., Tyrväinen, L. & Silvennoinen, H. Determinants of restorative experiences in everyday favorite places. Health Place. 14, 636–652 (2008).10.1016/j.healthplace.2007.10.008 18037332

92.Watson D Clark LA Tellegen A Development and validation of brief measures of positive and negative affect: The PANAS scales J. Personal. Soc. Psychol. 1988 54 1063 1070 10.1037/0022-3514.54.6.1063 3397865 Watson, D., Clark, L. A. & Tellegen, A. Development and validation of brief measures of positive and negative affect: The PANAS scales. J. Personal. Soc. Psychol.54, 1063–1070 (1988).10.1037//0022-3514.54.6.1063 3397865

93.Bostic TJ McGartland Rubio D Hood M A Validation of the Subjective Vitality Scale Using Structural Equation Modeling Soc. Indic. Res. 2000 52 313 324 10.1023/A:1007136110218 Bostic, T. J., McGartland Rubio, D. & Hood, M. A Validation of the Subjective Vitality Scale Using Structural Equation Modeling. Soc. Indic. Res.52, 313–324 (2000).

94.Ryan RM Frederick C On Energy, Personality, and Health: Subjective Vitality as a Dynamic Reflection of Well-Being J. Pers. 1997 65 529 565 9327588 10.1111/j.1467-6494.1997.tb00326.x Ryan, R. M. & Frederick, C. On Energy, Personality, and Health: Subjective Vitality as a Dynamic Reflection of Well-Being. J. Pers.65, 529–565 (1997).10.1111/j.1467-6494.1997.tb00326.x 9327588

95.Ojala A Korpela K Tyrväinen L Tiittanen P Lanki T Restorative effects of urban green environments and the role of urban-nature orientedness and noise sensitivity: A field experiment Health Place 2019 55 59 70 30502229 10.1016/j.healthplace.2018.11.004 Ojala, A., Korpela, K., Tyrväinen, L., Tiittanen, P. & Lanki, T. Restorative effects of urban green environments and the role of urban-nature orientedness and noise sensitivity: A field experiment. Health Place. 55, 59–70 (2019).10.1016/j.healthplace.2018.11.004 30502229

96.Jürges H Avendano M Mackenbach JP Are different measures of self-rated health comparable? An assessment in five European countries Eur. J. Epidemiol. 2008 23 773 781 10.1007/s10654-008-9287-6 18814040 Jürges, H., Avendano, M. & Mackenbach, J. P. Are different measures of self-rated health comparable? An assessment in five European countries. Eur. J. Epidemiol.23, 773–781 (2008).10.1007/s10654-008-9287-6 18814040

97.Schultz PW Schmuck P Schultz WP Inclusion with Nature: The Psychology Of Human-Nature Relations Psychology of Sustainable Development 2002 Boston, MASpringer US61 78 10.1007/978-1-4615-0995-0_4 Schultz, P. W. Inclusion with Nature: The psychology of human-nature relations. In Psychology of Sustainable Development (eds (eds Schmuck, P. & Schultz, W. P.) 61–78 (Springer, 2002).

98.Yang J Zhao L Mcbride J Gong P Can you see green? Assessing the visibility of urban forests in cities Landsc. Urban Plann. 2009 91 97 104 10.1016/j.landurbplan.2008.12.004 Yang, J., Zhao, L., Mcbride, J. & Gong, P. Can you see green? Assessing the visibility of urban forests in cities. Landsc. Urban Plann.91, 97–104 (2009).

99.Hassani, A. & Shi, H. Dilated neighborhood attention transformer. 10.48550/ARXIV.2209.15001 (2022).

100.Jain, J. et al. OneFormer: one transformer to rule universal image segmentation. 10.48550/ARXIV.2211.06220 (2022).

101.Zhou, B. et al. Scene parsing through ADE20K dataset. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 5122–5130 (IEEE, 2017).

102.Bürkner, P. C. brms: An R package for Bayesian multilevel models using Stan. J. Stat. Soft. 8010.18637/jss.v080.i01 (2017).