1. Introduction
For centuries, people have used the natural environment to enhance their health and well-being [
1]. Empirical research has found that interaction with Nature is associated with better mental health and well-being [
2,
3,
4,
5,
6], positive emotions [
7,
8,
9,
10], and attention [
8,
11,
12], as well as reduced (physiological or perceived) stress [
4,
5,
6,
8,
13,
14,
15,
16,
17,
18]. Yet the natural environment is often “treated as uniform” [
19] (p. 48) as studies commonly compare broad urban and natural environment categories [
7,
8,
20] or analyze the amount of, or proximity to, Nature [
2,
3,
21,
22,
23]. There have been calls to go “beyond the green” to investigate the contribution different types and qualities of natural environments have on well-being [
24,
25,
26,
27,
28,
29,
30,
31,
32,
33,
34]. Specifically, Thompson Coon
et al. [
26] suggest “future studies might consider the impact of the perceived quality of the environment on mental and physical wellbeing outcomes.” (p. 1771).
“Quality” is often discussed in terms of the “aesthetics or attractiveness” of the natural environment [
35] (p. 27). Many of the indicators of environmental quality pertain to use, such as accessibility, maintenance, perceived safety, presence of amenities or absence of litter [
27,
29,
35,
36,
37]. Recently, a broader set of environmental quality indicators have begun to be acknowledged and researched, such as: biodiversity [
19,
38,
39,
40,
41,
42], naturalness [
36,
43], and perceived restorativeness [
33,
43].
Despite the predominate focus in the literature to test exemplars of natural and urban environments, recent research has begun to investigate the influence different types and qualities of natural environments have on health and well-being. The following will review the previous literature that has moved beyond the green to investigate the effect on well-being from different environment types and indicators of perceived environmental quality—specifically perceived naturalness, biodiversity and restorativeness.
1.1. Types of Natural Environments
Not all green spaces have an equal impact on well-being; some types of natural environments may have more of an effect on well-being than others. For example, de Vries
et al. [
2] found that the amount of agricultural green space in one’s neighbourhood was associated with greater physical and mental health, but the amount of urban green space, forest, or “nature areas” (p. 1722) in the neighbourhood had no effect [
2]. Similarly, living near to coastal environments have been shown to have an effect on positive mental health—over and above the effects of green space [
44,
45].
Use of specific types of natural environments for physical exercise has also been shown to have a differential effect on health and well-being. Exercise near a beach or river may have greater improvements in self-esteem and mood than exercising in urban green space, farmland and woodland environments [
46]. Walking alone in a maintained forest was associated with greater positive affect and less negative affect, compared to walking alone in an unmaintained forest [
47]. Marselle
et al. [
48] investigated the effect different environment types have on well-being and found that walking with others in farmland and green corridor environments were associated with less negative affect and perceived stress than walking with others in urban environments, whilst natural and semi-natural, urban green space, or coastal environments had no effect.
1.2. Perceived Naturalness
How natural an environment is perceived to be is an important predictor of well-being. People express greater positive affect and happiness in natural environments than in urban or indoor environments [
7]. Environments perceived as “more natural” (e.g., forest, woodland, or valley) have been associated with greater psychological well-being than “less natural” environments (e.g., parks, gardens, or farmland) [
9]. Perceived naturalness was a significant predictor of anxiety following a bout of green exercise; the more natural an environment is perceived, the larger the reductions in anxiety [
49]. However, van den Berg
et al. [
50] found perceived naturalness of an environment had no influence on restoration of emotional well-being following a scary movie.
1.3. Perceived Biodiversity
Biodiversity may be a useful environmental quality indicator for investigating the health and well-being impacts of natural environments [
39]. The level of objective biodiversity in the environment has been shown to have a positive influence on improved health [
28,
39,
51], psychological well-being [
40] and positive emotions [
41]. Our review here focuses on perceived biodiversity—an individual’s assessment of the species richness in an environment [
19,
39]. People have a general belief that the perceived biodiversity of flowers, birds, and trees in an urban park improves their well-being [
42]. In their
in-situ survey in riparian green space, Dallimer
et al. [
19] found psychological well-being was positively correlated with the number of bird, butterfly and plant species perceived in the environment. As investigations of biodiversity and well-being are a nascent research area, further research is needed to clarify the relationship between biodiversity and emotional well-being.
1.4. Perceived Restorativeness
Examining the perceived restorative quality of an environment is another way to “move beyond green” in analyses of Nature and health. Attention Restoration Theory (ART) posits that certain environments can facilitate restoration of one’s ability to direct attention or concentrate [
52,
53]. Theorized qualities of a restorative environment include: being away, fascination, coherence, and compatibility [
52,
53]. A restorative environment requires psychological and physical distance from tasks, thoughts, or environments which require directed attention (
being away). Fascinating stimuli are required to attract effortless, involuntary attention, which allows for the rest and restoration of directed attention (
fascination). Fascination can be sustained if the stimuli are organized in a coherent way and rich enough to foster the perception of being in a whole other world (
coherence). The theory acknowledges that a fit between the environmental setting and one’s purposes and inclinations is required for restoration; a compatible environment allows one to carry out his or her activities without struggle (
compatibility). Natural environments are theorized to be well endowed with these four restorative qualities [
52,
53].
Quantitative measures have been created to assess the perceived restorativeness of an environment, based on the four qualities described by ART. These measures have been positively correlated with greater emotional well-being in general [
54], and positive affect in particular [
55]. Specific examination of the restorative quality of “fascination” found it was correlated with greater positive affect, but non-significantly related to negative affect [
56].
1.5. Perceived Restorativeness as a Moderator
Environmental types and qualities do not occur in isolation; an environment can be experienced and assessed for its type, as well as its naturalness, biodiversity and restorative quality. For example, an urban green space could have a low level of perceived naturalness, moderate levels of perceived biodiversity and high perceived restorativeness. Similarly, a biodiverse environment can also be assessed for its naturalness, restorativeness, and environment type.
While there are multiple ways in which one could conceptualise the relationship between the environment and well-being, in this paper we specifically focus on how these environmental characteristics might interact with one another to influence emotional well-being. Interaction is also known as moderation [
57]. Moderators
qualify environment-behaviour relationships [
58]; they can answer
when the external environment will effect well-being—and when it will not [
57]. Moderation analyses of the relationship between environment and well-being have been called for by researchers [
59,
60]. Whilst previous studies have investigated gender [
56], social interaction [
61], activity type [
62], and the type of urbanity surrounding a natural area [
63] as moderators, few studies investigate an interaction between perceived environmental type and/or environmental qualities, and well-being.
In the current study, we investigate whether perceived restorativeness would moderate the relationship between perceived type, naturalness, or biodiversity of an environment, and emotional well-being. ART implies that perceived restorative quality may interact with a natural environment to influence restorative outcomes. In other words, it is possible that a natural environment with high perceived restorative quality may engender greater restoration than a natural environment with low restorative quality. Hartig
et al.’s [
55] analysis of perceived restorativeness and emotional wellbeing lends some support for this argument. Whilst not formally testing for moderation, study 1 found the relationship between positive affect and perceived restorativeness differed by the restorative quality of an outdoor environment. Natural and built outdoor environments
a priori expected to be high in restorative potential, had significant correlations between positive affect and perceived restorativeness, whereas those environments expected to be low in restorative potential demonstrated no significant relationship between positive affect and perceived restorativeness. Thus, environment and restorative quality give the appearance of having interacted, suggesting perceived restorativeness may effect when an outdoor environment influences emotional well-being and when it does not. In a study by Gonzalez
et al. [
64] the authors specifically investigated whether the restorative qualities of “being away” and “fascination” would moderate the effect of a therapeutic horticulture intervention on depression. Being away and fascination were measured as an average across multiple measurements, one of which included the respondent’s home. The authors found the overall level of “being away” moderated the change in depression of the therapeutic horticulture intervention, but ‘fascination’ did not. In other words, participating in the intervention was associated with greater decline in depression, among those who experienced a high level of “being away” in two environments (
i.e., home and the horticulture intervention setting). Due to the limited research, we believe there is scope to investigate whether perceived restorative quality interacts with perceived environmental type, naturalness or biodiversity to amplify well-being.
1.6. Walk Characteristics—Walk Duration and Intensity
Physical exercise itself can improve mood [
65,
66,
67]. As such, it is important to isolate its effect from the natural environment when studying green exercise [
16,
32,
68]. Duration and intensity of physical activity both have been shown to increase post-exercise positive affect [
69]. In this study, we measured duration and perceived intensity of the group walk to examine their independent influence on emotional well-being following an outdoor group walk.
1.7. Study Aims
The first aim of this study was to explore the health benefits of Nature beyond a “green” environment and investigate the effect of environment type and indicators of perceived environmental quality (i.e., naturalness, biodiversity, restorativeness) on emotional well-being following an outdoor group walk. Characteristics of the group walk—duration and intensity—were assessed to understand their independent relationship to emotional well-being. The second aim of this study was to investigate whether perceived restorative quality of an environment moderates the effect of perceived environment type or perceived environmental quality on emotional well-being. To our knowledge, the interaction of perceived restorative quality with perceived environment type, naturalness, and biodiversity are heretofore unknown and such moderation analyses are unique.
4. Discussion
This study explored the health benefits of Nature beyond a “green” environment to investigate the effects environment type and indicators of perceived environmental quality—naturalness, biodiversity and restorativeness—had on short-term emotional well-being following an outdoor group walk. Characteristics of the walk (i.e., walk duration, intensity) were assessed to understand their independent relationship to emotional well-being. We also investigated whether perceived restorative quality moderated the effect of perceived environment type or perceived environmental quality on emotional well-being.
We found that perceived restorative quality of the environment was a significant predictor of emotional well-being following a group walk, associated with an increase in positive affect and happiness as well as a reduction in negative affect. The identified relationship between perceived restorativeness and positive affect mirrors findings in research by Hartig
et al. [
7,
55] and Sato [
56]. To our knowledge, the influence of perceived restorativeness on negative affect is a unique finding. Further research is required to fully understand the effect of perceived restorativeness on negative affect. The type of environment and the other two indicators of perceived environmental quality—naturalness and biodiversity—were nonsignificant predictors of emotional well-being when combined in the same model with perceived restorative quality. These results suggest that restorative quality of an environment may be an important element for enhancing emotional well-being.
Moreover, our analyses also showed that perceived restorative quality moderated the association between perceived naturalness and post-walk positive affect. In other words, perceived restorativeness and perceived naturalness interacted to enhance positive affect following an outdoor group walk. This finding suggests these two indicators may work together to amplify the experience of positive emotions in Nature. The significant interaction of perceived naturalness and restorative quality is supportive of ART, which considers natural environments to have plentiful restorative qualities [
52,
53]. Previous research has found that people differ in their assessments of an environment’s restorative quality based on its level of naturalness, in that more natural environments are rated as higher in perceived restorative quality than less natural environment [
13,
43,
55,
101,
102,
103,
104]. To date no research has specifically investigated the interaction of the two on affective outcomes. Perhaps the closest study to ours was Gonzalez et al [
64] who found the restorative quality ‘being away’ moderated the effect of a therapeutic horticulture intervention on depression [
64]. It is important to note that Gonzalez
et al. [
64] captured the restorative experience of two environments (
i.e., home and the horticulture intervention setting) in their measure of being away. Thus to our knowledge, the interaction between perceived restorativeness and perceived naturalness on emotional well-being found here is novel.
Taken together, these two results emphasize the importance of considering the transactional relationship between person and environment in Nature-health research. Perceived restorative quality is about a person’s experience of an environment as restorative, which is related to but separate from the other two indicators of perceived environmental quality examined in this study: naturalness and biodiversity. Our results suggest the short-term emotional well-being benefits of Nature may be a consequence of an individual’s experience of the physical environment as restorative, rather than the environment itself. These findings suggest a move away from a deterministic approach to environmental design or Nature and health research, in which the assumption that including a particular feature (e.g., water; variety of shrubs) into an environment or living in proximity to a certain environment type (e.g., coast) will result in greater emotional well-being.
In terms of walk characteristics, perceived walk intensity was a significant predictor of greater positive affect and happiness following an outdoor group walk. The identified relationship between intensity and positive affect is consistent with previous research that found immediate gains in positive affect following objectively measured exercise intensity [
69]. We found no significant predictive relationship of perceived intensity on post-walk negative affect. Duration of the group walks was not a significant predictor of emotional well-being. This finding differs from that of Ekkekakis [
69] who identified a significant positive relationship between duration of physical exercise and post-exercise positive affect. Thus it may be that the perceived physical intensity of the walk contributes to greater positive emotional well-being following a group walk and that the duration of walking may not matter after accounting for intensity. The significance of physical intensity on post-walk positive emotional well-being highlights physical exercise as a possible confounder in green exercise studies. Moreover, the result calls attention to the need to isolate the effects of physical activity—such as walk intensity—from the effects of perceived environmental quality when investigating the health benefits of Nature [
16,
32,
68]. This is especially important to tease out since providing a space for or enhancing the effects of physical activity has been suggested as one of the mechanisms through which Nature can affect human health [
32].
Analyses also revealed perceived bird biodiversity was a significant predictor of post-walk negative affect. In contrast to previous research [
19,
41,
42], negative affect increased as the perceived biodiversity of birds increased, specifically from 0–4 to 5–14 species types. The relationship could perhaps be explained by the type of bird species and its acoustic properties (pitch, intensity, roughness) as well as whether the increased number of birds was compatible with group walkers’ use of the environment [
105]. For example, it is possible that participants made their assessment of perceived bird biodiversity based on bird song and calls, as this can be an one way in which to identify them [
105]. In general, listening to birdsong improves mood [
106]. However, not all bird sounds are perceived as positive, as the songs and calls of certain bird species common in English urban green spaces (e.g., magpies, crows, owls) are associated with negative emotions [
105]. Perhaps the increased number of birds perceived by participants was one of these bird species, or that the additional bird species were incompatible to the participant’s use of the environment whilst on a group walk.
In this study, environment type, perceived biodiversity of butterflies, and plants and trees were not significant predictors of the change in emotional well-being following a group walk. Previous research [
48] also found a nonsignificant effect of environment type for a group walk on positive affect. However, the authors did find a reduction in negative affect associated with group walks in farmland and green corridor environments. Methodological differences in the measurement of the emotional well-being between that study and ours make direct comparisons difficult; the previous study assessed longer-term emotional well-being associated with outdoor group walks in certain environments, whilst here we assessed short-term emotional well-being,
i.e., immediately following the walk. Three reasons for the nonsignificant effects of butterflies and plants and trees are discussed below.
6. Limitations and Future Directions
This study has a number of limitations. First, the data may reflect a seasonal effect, as the study took place over the changing seasons from late summer through autumn. As changes in temperature and weather may influence response to perceived restorativeness and emotional well-being measures, future research may consider data collection during a single season. Second, it is beneficial to conduct a power analysis and estimate the required sample size prior to the data collection to ensure detection of moderation effects [
107]. However, given the study design—in which the number of returned questionnaires was dependent on the number of group walks taken by participants during the study— an initial power analysis could not be performed. Third, low response frequencies for certain biodiversity categories might have increased the sensitivity of our models to external interactions; potential conclusions and inference will need to take this into account. Fourth, the data collection protocol could mean that participants did not necessarily complete the questionnaire immediately before and after the group walk, which could affect internal validity. Future smaller scale studies could place the experimenter with participants to ensure adherence to the data collection protocol. Finally, walking in a group may result in less interaction with the environment [
108] and less perceived restorativeness [
61]. As such, future studies may wish to replicate this study with solo walkers.
To our knowledge, the identified relationship between perceived biodiversity and emotional well-being following a group walk is novel. As research in perceived and objective biodiversity and subjective well-being is a nascent research area, we suggest our findings here be considered with caution until a greater evidence base is developed. We give three reasons for caution. First, over the course of the data collection (22 August to 14 November 2011), the number of actual species of birds and butterflies present in the walk setting likely diminished, and cues of different types of plants and trees may have become reduced as well. Second, the measures of perceived biodiversity used in this study were created specifically for investigating species richness in urban green spaces [
40], but were applied here to assess perceived biodiversity in seven different environment types. As such, use of setting-specific perceived biodiversity response categories to assess perceived biodiversity in other environment types may be inappropriate. Finally, the perceived biodiversity measures were originally designed such that responses from participants could be compared to objective ecological survey data on species richness [
40]. Consequently, these measures ask participants to make a numerical assessment, on a categorical response option scale, of the number of birds, butterflies and plants/trees in an environment. If researchers are not seeking to align measures of objective and subjective biodiversity, then future studies may want to use a more subjective scale of perceived biodiversity, like the Biodiversity Experience Index [
41].
In this paper we specifically focused on how perceived restorativeness might interact with perceived environment type, naturalness and biodiversity to influence well-being. An alternative examination is to investigate perceived restorativeness as a mediator. Indeed, suggestive evidence of mediation appears in the reported bivariate correlations; measures of perceived naturalness and biodiversity did not correlate with post-walk emotional well-being but did significantly correlate with perceived restorativeness, which was significantly correlated with the post-walk emotional well-being. Thus, perceived naturalness and biodiversity may
indirectly influence post-walk emotional well-being
via perceived restorativeness. Future studies could usefully investigate a mediation model in which perceived restorativeness mediates the relationship between environment type, naturalness and biodiversity on well-being [
109]. We are currently reanalysing our data to explore whether such a multilevel mediation model exists.
Environmental experiences are multi-dimensional as environmental types and qualities co-occur. As such, future studies on Nature and health could usefully investigate the interaction of environment type and/or indicators of perceived environmental quality on health and well-being outcomes. Further research is required to determine whether perceived restorative quality moderates the Nature-health relationship.