Abstract
Nature exposure has presented positive relationships with satisfaction with life and negative relationships with loneliness. Nonetheless, studies on the underlying mechanisms that may mediate these relationships are still limited. Nature restorativeness may be one such mediating variable. This paper reports on the role of nature restorativeness in the relationship between exposure to nature and two mental health outcomes — satisfaction with life and loneliness. The sample included 235 women and 113 men with an average age of 35 years (SD = 16.2 years). They answered scales to measure (1) nature exposure, (2) restorative outcomes, (3) satisfaction with life and (4) loneliness. Nature exposure was positively associated with restorative outcomes and satisfaction with life, and negatively associated with loneliness. Hierarchical regressions and mediation analyses showed that nature restorativeness significantly fully mediated the link between nature exposure and (1) satisfaction with life and (2) loneliness. The findings suggested that restorative experiences in nature are likely to be relevant aspects of people’s experience of nature, and positive mental health outcomes are related to them.
According to the World Health Organization (2014), mental health is ‘a state of well-being in which [an] individual realizes his or her own potential, can cope with the normal stresses of life, can work productively and fruitfully, and is able to make a contribution to her or his community’. Therefore, mental health includes (i) the absence of mental illness and (ii) the presence of psychological well-being. In this paper we explore the relationships between nature exposure (NE) and two mental health outcomes: an indicator of positive mental health, satisfaction with life; and an indicator of negative mental health, loneliness. These outcomes reflect the dual continuum model of mental health (Iasiello & van Agteren, 2020). Additionally, the current study tests whether nature restorativeness mediates the relationships between NE, and satisfaction with life and loneliness.
Literature review
Nature exposure
Nature is a complex construct, being defined as ‘encompassing elements and phenomena of Earth’s lands, waters, and biodiversity, across spatial scales and degrees of human influence, from a potted plant or a small urban creek or park to expansive, “pristine” wilderness with its dynamics of fire, weather, geology, and other forces’ (Bratman et al., 2019, p. 2). NE concerns people’s contact with nature in both everyday life and outings or excursions (Kamitsis & Francis, 2013).
While NE can adversely impact mental health by means of the promotion of fear and anxiety, most research indicates positive associations (Cleary, 2017). NE provides substantial benefits to people (Andersen et al., 2021). Research has shown that NE is related to more positive body appreciation (Swami et al., 2022). Research has evidenced that NE is linked to enhanced mental health (Bratman et al., 2019; Fong et al., 2018). A meta-analysis found significant decreases in anger, anxiety and sadness, and improvements in energy after NE (Bowler et al., 2010). A review pointed out a negative association between NE and stress (Shuda et al., 2020).
A recognized instrument of NE is the four-item Nature Exposure Scale (NES, Kamitsis & Francis, 2013). The NES was developed to assess the level of NE in everyday life and activities and the level of exposure outside everyday environments. Prior studies have shown acceptable internal consistency (Kamitsis & Francis, 2013; Swami et al., 2016), one-dimensionality and criterion validity (Swami et al., 2016). A cross-cultural study with adults from 65 nations, including Portugal (Swami et al., 2024) suggested that the NES scores are unidimensional and evidenced full scalar invariance by gender and age groups.
Nature exposure and satisfaction with life
Satisfaction with life (SWL) refers to a cognitive evaluation of one’s subjective well-being (Diener et al., 1985) in which ‘individuals assess the quality of their lives on the basis of their own unique set of criteria’ (Pavot & Diener, 1993, p. 164). Prior studies have reported that SWL was positively linked to health, meaningful social relations and productivity, among others (Diener, 2013).
Bratman et al. (2021) reported NE as conducting to enhanced subjective well-being and positive affect. In this line there is research pointing out a positive link between NE and SWL. Individuals residing in urban locations with more green space reported lower mental distress and greater SWL than people residing in urban areas with less green space in the United Kingdom (White et al., 2013). One meta-analysis evidenced positive associations between green space and SWL (Houlden et al., 2018).
Nature exposure and loneliness
Loneliness is a growing mental health challenge. A perceived discrepancy between the desired and the actual level of social connections relatively to quality and/or quantity can be described as loneliness (Perlman & Peplau, 1981). Loneliness is linked to various negative physical problems, such as obesity and elevated blood pressure (Cacioppo et al., 2015), mental health outcomes, such as higher depressiveness, suicidal thoughts and aggressiveness (Bzdok & Dunbar, 2020; Neto, 2021), and high risk of mortality (Holt-Lunstad et al., 2015). Therefore, loneliness can be extremely detrimental from the standpoint of public health, and the World Health Organization (2023) has recently declared loneliness as a pressing global health threat.
A systematic review indicated that more green space exposure or experience was inversely associated with loneliness (Astell-Burt et al., 2022). Daily nature contact was a protective factor against feeling lonely for adults who worked mainly remotely during the pandemic (Klein et al., 2025).
Nature exposure and nature restorativeness
Nature restorativeness refers to a short-term state involving affective, physiological and attention restoration (S. Kaplan, 1995; Ulrich et al., 1991). These various elements of restoration are assessed in the Restoration Outcome Scale (ROS; Korpela & Ylén, 2009). There is research demonstrating that natural environments are more restorative than urban environments (Menardo et al., 2021). Two theories are usually used to justify the associations between NE and mental health and well-being: the Attention Restoration Theory (ART; R. Kaplan & Kaplan, 1989) and the Stress Reduction Theory (SRT; Ulrich, 1983).
The ART analyses the kinds of experiences that allow restoration from attentional fatigue. It relates to countering directed attentional fatigue. The theory suggests that certain environments can help restore attention by providing four key qualities: fascination, extent, being away and compatibility. The SRT posits that natural environments can facilitate recovery from stress. The theory suggests that exposure to nature can lead to restorative outcomes, including reduced physiological arousal, reduced psychological stress and negative emotions and enhanced positive emotions.
In accordance with S. Kaplan (1995), attention and stress are distinct but interacting benefits of restorative experience. ‘Both of these benefits have an important role in an individual’s life. Quite obviously anything that aids in the management of stress is essential. At the same time, however, directed attention also plays a significant role’ (S. Kaplan, 1995, p. 180). For both theories, NE provides positive influences on numerous aspects of well-being. Therefore, based on ART and SRT, nature exposure can help to restore mental health, enhancing positive-related emotions such as satisfaction with life and decreasing negative stress-related emotions such as loneliness.
There are other probable mechanisms relating natural environments with mental health, beyond those targeted by the ART and the SRT (Markevych et al., 2017). One potential mechanism is that natural environments may facilitate physical activity, which in turn benefits mental health (Galper et al., 2006). Another mechanism is that natural environments promote social cohesion, which in turn enhances health and well-being (Jennings & Bamkole, 2019). There is also research showing that connectedness to nature (Mayer & Franz, 2004) mediates the relationship between NE and the health benefits (Swami et al., 2016).
The present study
To summarize, the accessible studies have shown that NE is related to satisfaction with life and loneliness, but there are gaps relative to the mediating pathways. The current work intends to scrutinize the associations between NE and specific aspects of mental health, such as satisfaction with life and loneliness. Additionally, the current study tested the mediating role of nature restorativeness in the association between NE and satisfaction with life and loneliness.
In spite of the demonstration which suggests a positive association of NE with mental health, studies examining possible mechanisms underlying this relationship are limited (Baceviciene & Jankauskiene, 2022; Kamitsis & Francis, 2013; van den Berg et al., 2003). Kamitsis and Francis (2013) tested the mediating role of spirituality in the association between NE and psychological well-being. They found that nature exposure was positively related to psychological well-being and higher spirituality. Additionally, they found that spirituality emerged as a mediator of the relationships between NE and psychological well-being. These scholars suggested that future studies may test other possible mediating pathways. ‘For example, Attention Restoration Theory asserts that exposure to natural environments can help maintain or restore an individual’s attentional functioning. . . which may exert positive effects on psychological aspects of wellbeing’ (Kamitsis & Francis, 2013, p. 141). Following this suggestion, we tested the mediating role of nature restorativeness in the association between NE and two mental health outcomes: satisfaction with life and loneliness.
The conception of nature restorativeness as a mediator is grounded in three directions of reasoning. First, nature exposure was positively related to restorative experiences in nature (Baceviciene et al., 2021; Whitburn et al., 2019; White et al., 2010). Natural environments have elicited stronger restorative impacts than built environments (van den Berg et al., 2003). Second, higher levels of restorative experiences in nature have been found to be related to higher emotional well-being (Pasanen et al., 2018). Third, restorative experiences in nature have previously been researched as a mediator (Baceviciene & Jankauskiene, 2022; Korpela et al., 2014; van den Berg et al., 2003). For instance, Korpela et al. (2014) found evidence that restorative experiences in the most recent nature-based recreation visit mediated the relationship between the mean time spent on nature-based recreation and emotional well-being. Baceviciene and Jankauskiene (2022) found that nature restorativeness mediated the relationship between NE and enhanced quality of life. Both studies (Baceviciene & Jankauskiene, 2022; Korpela et al., 2014) used as dependent variables only pleasant affect indicators of well-being (emotional well-being and quality of life). We will use a positive and cognitive indicator of mental health (SWL) and a negative indicator (loneliness) of mental health.
Grounded on prior results, our expectation was that restorative experiences in nature would mediate the relationship between NE and increased SWL. In addition, we analysed the mediating role of restorative experiences in nature with the association between NE and loneliness. The increasing prevalence of loneliness (Astell-Burt et al., 2022) emphasizes the necessity to better comprehend the complexity of the mechanisms by which NE can decrease the feeling of loneliness. As far as we know, nature restorativeness has not yet been explored as the mediator between NE and satisfaction with life and loneliness. Therefore, the present study fills this gap, by exploring the mediating role of restorative experiences in nature with the relationships between NE and satisfaction with life and loneliness. The current work may provide novel insights to extant investigation and give empirical support that could be utilized to improve mental health.
In short, based on the aforementioned research, four hypotheses were formulated:
Hypothesis 1. Nature exposure and satisfaction with life will be significantly and positively associated.
Hypothesis 2. Nature exposure and loneliness will be significantly and negatively associated.
Hypothesis 3. Nature restorativeness will fully mediate the association between NE and satisfaction with life, with higher exposure to nature expected to enhance nature restorativeness and thereby increase satisfaction with life.
Hypothesis 4. Nature restorativeness will fully mediate the association between NE and loneliness, with higher NE expected to enhance nature restorativeness and thereby reduce loneliness.
Method
Participants
Our sample consisted of 348 participants (235 women and 113 men), aged between 18 and 84 years (M = 35.22, SD = 16.17). All respondents self-reported their nationality as Portuguese. As regards educational level, 5% had completed primary education, 30% had completed secondary education and 65% had attended higher education. Of the respondents, 58% were single, 39% were married or cohabiting with a partner and 3% were in other situations.
Instruments
Nature Exposure Scale (NES)
NES was designed by Kamitsis and Francis (2013). This scale has four statements (e.g., ‘How much do you notice the natural environments in your everyday life?’). Answer anchors differed, but all consisted of five-point scales. Greater values denote higher levels of NE. The NES displayed a one-dimension structure among English-speaking people (Swami et al., 2016), as well as in a cross-cultural study with 65 countries, including Portugal (Swami et al., 2024).
Restoration Outcome Scale (ROS)
ROS assesses restorative experiences after recent contact with nature (Korpela & Ylén, 2009). The measure consists of nine statements (e.g., ‘I calmed down’). Response options ranged from 1 (‘Not at all’) to 7 (‘Completely’). Greater scores denote more restorative experiences. The ROS displayed a one-dimension structure and adequate reliability (Baceviciene, et al., 2021; Korpela et al., 2014).
Satisfaction with life (SWLS)
SWLS includes five statements (e.g., ‘I am satisfied with my life’) (Diener et al., 1985). Ratings ranged from 1 (‘Strongly disagree’) to 7 (‘Strongly agree’). The SWLS displayed adequate reliability and construct validity in many countries, including in Portugal (Neto, 1993; Pavot & Diener, 1993).
The UCLA Loneliness Scale Short Version (ULS-6)
The short form of the Revised UCLA Loneliness Scale was utilized (Neto, 2014; Russell et al., 1980). The ULS-6 consists of six statements (e.g., ‘People are around me but not with me’). Ratings ranged from 1 (‘Never’) to 4 (‘Often’). A high score means greater loneliness.
Demographics
Information on gender, age, education, marital status and nationality was obtained from the study respondents.
Procedure
Previously validated translations of the SWLS (Munoz Sastre et al., 2003; Neto, 1993) and the ULS-6 (Neto, 2014) were available. The NES and the ROS were translated into Portuguese using the back-translation technique (Brislin, 1986). The data were collected by one research assistant in diverse locations, including workplaces, shopping centres and public libraries in the Porto area (which has a temperate rainforest vegetation classification). The survey was self-administered using a standard paper-and-pencil format. Regarding ethical considerations, the work was achieved following the legal and ethical norms in the country and in accordance with the Declaration of Helsinki. Informed consent was obtained. Respondents were also informed that they could discontinue their participation at any time. Anonymity of responses was assured. No compensation was provided to respondents.
Data analysis
Preliminary analyses were accomplished to evaluate the psychometric features of the tools using Confirmatory Factor Analysis (CFA). The internal consistency of the scales was checked via Cronbach’s α and McDonald’s ω. Prior to these analyses, all scales were examined relatively to normality by assessing skewness and kurtosis. All normality indices were confirmed. To examine the associations between NE, nature restorativeness, satisfaction with life and loneliness, Pearson correlation analyses were run. Magnitudes between .1 and .3 were considered small, above .3 and below .5 were moderate, and equal or above .5 were strong (Cohen, 2013). Finally, to explore the mediating role of nature restorativeness in the association between NE and satisfaction with life and loneliness, path analyses were performed to evaluate the supposed associations within the hypothesized models. Analyses were carried out via IBM SPSS and AMOS (version 29.0). The significance level was .05.
Results
Psychometric properties
Prior to testing our hypotheses, we evaluated the constructs’ psychometric properties and their internal consistencies.
Nature exposure
The Nature Exposure Scale (NES) was incorporated as an independent variable. The construct validity of the NES was examined by CFA. CFA was scrutinized with a one-dimensional structure grounded on previous research (Swami et al., 2024). The one-dimensional structure of the NES revealed acceptable fit to the data, disregarding χ2 and RMSEA, given the small degrees of freedom of this model, (χ2 = 8.24, df = 1, χ2/df = 8.24, CFI = .98, GFI = .99, SRMR = .03, RMSEA = .14, 90% CI [.07, .24]) (Hu & Bentler, 1999). To accomplish the model, a correlation between two statements’ errors (statements 1 and 4) was specified. Estimates (λ) of the statements were significant at p < .001, ranging from .52 (statement 3) to .81 (statement 4), with a mean of .71. The NES presented an α of .76 and an ω of .74. These findings seem to indicate that the NES is a valid and reliable measure of nature exposure in the Portuguese cultural context.
Restoration Outcome
The Restoration Outcome Scale (ROS) was used as a mediating variable. CFA was scrutinized with one-dimensional structure grounded on previous research (Korpela et al., 2014). The one-dimensional structure of the ROS revealed good fit to the data (χ2 = 85.93, df = 22, χ2/df = 3.91, CFI = .98, GFI = .95, SRMR = .03, RMSEA = .08, 90% CI [.06, .10]) (Hu & Bentler, 1999). However, to accomplish the model, a correlation between several errors in pairs of statements (statements 1 and 2; 1 and 3; 2 and 3; 5 and 6; 8 and 9) was established. Estimates (λ) of the statements were significant at p < .001, ranging from .75 (statement 5) to .91 (statement 7), with a mean of .82. The ROS presented an α of .95 and an ω of .95.
Satisfaction with life
The Satisfaction with Life Scale (SWLS) was utilized as a dependent variable. CFA was scrutinized with a one-dimensional structure grounded on previous research (Neto, 1993). The one-dimensional structure of the SWLS revealed good fit to the data (χ2 = 15.64, df = 5, χ2/df = 3.13, CFI = .99, GFI = .98, SRMR = .03, RMSEA = .07, 90% CI [.03, .11]) (Hu & Bentler, 1999). Estimates (λ) of the statements were significant at p < .001, ranging from .60 (statement 5) to .88 (statement 3), with a mean of .73. The SWLS presented an α of .85 and an ω of .84.
Loneliness
The brief six-item scale of loneliness (ULS-6) was used as dependent variable. CFA was scrutinized with a one-dimensional structure grounded on previous research (Neto, 2014). The one-dimensional structure of the ULS-6 revealed good fit to the data (χ2 = 34.04, df = 9, χ2/df = 3.78, CFI = .97, GFI = .97, SRMR = .04, RMSEA = .08, 90% CI [.05, .11]) (Hu & Bentler, 1999). Estimates (λ) of the statements were significant at p < .001, ranging from .51 (statement 2) to .90 (statement 4), with a mean of .68. The ULS-6 presented an α of .82 and an ω of .83.
Testing hypotheses
Table 1 displays descriptive statistics for the scales utilized. Notably, the difference between the mean value of the NES (M = 3.64; SD = .79) and the neutral midpoint of the NES (3) was significant, [t(347) = 15.17, p < .001, [95% CI = (.56, .73)]. Therefore, this value suggests that participants reported a positive level of nature exposure.
Descriptive statistics, reliability indices and inter-correlations for the study variables.
Note: M = Mean; SD = Standard deviation; Sk = Skewness; Ku = Kurtosis; α = Cronbach”s alpha; ω = McDonald”s omega
Values in square brackets indicate the 95% confidence interval for each correlation. *p < .05; **p < .01; ***p < .001.
Testing of associations
Correlation coefficients were computed to evaluate bivariate relationships among the measures of interest: NE, nature restorativeness, satisfaction with life and loneliness. Table 1 presents these findings. All associations were statistically significant. Importantly, NE was significantly and positively correlated with satisfaction with life (r = .31, p < .001). According to Cohen’s guidelines, this correlation was moderate. This finding supports our first hypothesis. In addition, NE was significantly and negatively correlated with loneliness (r = −.13, p < .05). According to Cohen’s guidelines, this correlation was small. This finding supports the second hypothesis.
Mediation analysis
Since we found statistically significant associations between all measures, mediation analyses were implemented. Two mediation models were explored. For the first model, nature restorativeness was explored as the mediator of the association between NE and satisfaction with life. For the second model, nature restorativeness was explored as the mediator of the association between NE and loneliness.
The findings of regression analysis examining the potential mediating role of nature restorativeness in the association between NE and SWL are shown in Figure 1. The findings indicated that NE was indirectly linked to SWL via its relationship with nature restorativeness. Nature exposure significantly predicted nature restorativeness, β = .59, p < .001, and nature restorativeness significantly predicted SWL, β = .37, p < .001. These findings are consistent with the mediational hypothesis. Nature exposure did not significantly predict satisfaction with life after controlling for the mediator, nature restorativeness, β = .09, p = .13, in consonance with total mediation. About 19% of the variance in SWL was accounted for by the predictor measures. The indirect impact was calculated based on a percentile bootstrap estimation approach with 1,000 samples, conducted with IBM SPSS Amos. These findings indicated that the indirect coefficient was significant, β = .22, 95% CI = .15, .30, p < .001, and thus hypothesis 3 was supported.
The pattern of the hypothesized mediation analysis.
The findings of the regression analysis examining the potential mediating role of nature restorativeness in the association between NE and loneliness are shown in Figure 2. The findings indicated that nature exposure was indirectly linked to loneliness via its association with nature restorativeness. Nature exposure significantly predicted nature restorativeness, β = .59, p < .001, and nature restorativeness significantly predicted loneliness, β = −.29, p < .001. These findings are consistent with the mediational hypothesis. Nature exposure did not significantly predict loneliness after controlling for the mediator, nature restorativeness, β = .04, p = .55, in consonance with total mediation. About 8% of the variance in loneliness was accounted for by the predictor measures. The findings revealed that the indirect coefficient was significant, β = −.17, 95% CI = −.26, −.10, p < .001, and thus hypothesis 4 was supported.
The pattern of the hypothesized mediation analysis.
Discussion
The current research aimed to determine the extent to which NE is linked to two indicators of mental health, satisfaction with life and loneliness, and the role of restorative experiences in nature in these relationships. Our first hypothesis was supported since higher nature exposure was significantly (but moderately) associated with greater satisfaction with life. This finding is in accordance with prior work demonstrating that greater subjective well-being and positive affect were related to exposure to natural environments (Bratman et al., 2021). For instance, this may be important for workers given that well-being as evaluated by satisfaction with life contributed to increased work productivity (Judge et al., 2001).
Our second hypothesis was also supported since higher nature exposure was significantly (but weakly) associated with lower perceived loneliness. This finding aligns with the idea that natural environments promote social cohesion, since they may provide the opportunity for social interactions (Jennings & Bamkole, 2019). It is in accordance with prior investigation demonstrating that lower stress and negative affect and a decline in risk of mental health disorders were associated with exposure to natural environments (Bratman et al., 2021).
We tested nature restorativeness as a potential explanatory mechanism for the association between NE and two indicators of mental health, satisfaction with life and loneliness. Regarding the expected models of mediation, findings supported that nature restorativeness significantly fully mediated the link between NE and (1) SWL and (2) loneliness. Current results expand previous research and are in line with the prior research demonstrating that nature restorativeness represented a mediator between nature-based recreation and emotional well-being (Korpela et al., 2014) and between NE and quality of life (Baceviciene & Jankauskiene, 2022).
The present research revealed that NE is indirectly related to SWL and loneliness, via their relationship to nature restorativeness. This suggests that NE provides an opportunity for restorative experiences, which then relate to SWL and loneliness. Nature restorativeness may therefore be a relevant mechanism underlying the salutogenic effects of nature.
Current results may be explained by SRT (Ulrich, 1983) and ART (R. Kaplan & Kaplan, 1989). The human–nature relationship is embedded within restorative experiences. Higher immersion in nature facilitates emotional restoration. In turn, improvements in emotional restoration may promote more positive mental health. Urbanized environments are greatly appearance-oriented, while being in natural environments may attenuate stress and enhance positive emotions. As such, these findings suggest that restorative experiences in nature may constitute a pathway through which experiences with natural environments are associated with positive mental health. Our findings align with an incorporation of restorative experiences theories and emphasize a transactional relation of the environment with the individual.
Before concluding, we will consider some limitations of the current work. First, the cross-sectional nature of this study precludes causal inferences. For instance, while nature exposure may lead towards both higher SWL and lower loneliness, nature exposure may also result from higher SWL and lower loneliness. Further research applying a longitudinal design is needed to test the direction of the relationships. A second concern of the present study involves the measurement of NE. The NES may be conceptually problematic given that it evaluates various kinds of NE. The RMSEA value of the NES might suggest a poor fit. However, similar values were found in a cross-cultural study, and this can be related to the small degrees of freedom (Swami et al., 2024). Further research should reexamine this aspect. Third, in terms of sampling, the recruitment strategy was opportunistic, and as such, the sample is not representative of the Portuguese population. Therefore, the findings cannot be generalized to the Portuguese population without further research. Fourth, we have not controlled sociodemographic variables. Possible moderating effects of gender, age and educational level might be tested in future research. Finally, this work used self-report assessments. Further work should explore whether the noted relationships are replicated when utilizing behavioural assessments.
In spite of these limitations, the current results have significant implications for practice. Exposure to natural environments likely impacts mental health positively, which might translate into higher satisfaction with life and lower loneliness. The findings confirm the promotion of engagement with natural environments as an avenue for better mental health. Since NE relates to several outcomes with regard to mental health, for example, managers may wish to implement natural environments into the work context. For instance, work organizations might design break rooms to mimic natural environments (Craig et al., 2015). Past research also shows that plants in the work context are beneficial for performance (Raanaas et al., 2011). Educational institutes, for example, can implement cultural and academic activities that provide students opportunities to increase their desire to explore NE. Current results support the practical use of restorative qualities to design environments (R. Kaplan et al., 1998). They can interest urban planners to implement healthy environments, for example, trying to design environments with a sense of distance or fascination. Mental health agencies can help by developing and carrying out programmes that imbue the desire to be exposed to natural environments among people affected by poor satisfaction with life and high loneliness. In particular, our findings point out the potential of NE to be part of public health policy strategy to reduce loneliness and so promote optimal function. Such policies represent relevant contributions to developing sound health and well-being as well as sustainable communities (United Nations, 2015).