There has been a growing research interest in mindfulness over the last decade, spurred by the manifold positive effects of mindfulness evidenced in diverse fields as psychological and physical health (Grossman et al., 2004), cognitive functioning (e.g., Malinowski, 2013), and affective states (e.g., Killingsworth and Gilbert, 2010). That mindfulness might also be positively related to ecological behavior has been suggested in various recent studies (e.g., Brown and Kasser, 2005; Amel et al., 2009; Barbaro and Pickett, 2015; Panno et al., 2017). Since the concepts of mindfulness and ecological behavior applied in these studies vary considerably, we start by defining both concepts.

Based on the works of Kabat-Zinn (1991) and Bishop et al. (2004) we define mindfulness as the awareness of momentary experiences with a non-identified, unbiased, open, and accepting attitude. The definition comprises the two recurring aspects of mindfulness described in the literature as awareness and acceptance (Rau and Williams, 2016). The first aspect is expressed in heightened awareness for internal and external stimuli and sensations and the capacity to act consciously without being distracted (acting with awareness, e.g., Baer et al., 2008). The second aspect refers to the quality of the attitude in which people experience this awareness, namely accepting and non-identified, expressed in the ability to impartially allow whatever sensation or thought arises in a given moment without identifying with it, evaluating it or reacting to it (see also Bishop et al., 2004). Authors who emphasize the root in Buddhist ethics, regard an inner stance of equanimity, openness, and friendliness as an integral aspect of mindfulness (Shapiro and Carlson, 2009; Grossman, 2015). Despite this multifaceted nature of the construct, various authors give factor-analytical evidence for the existence of a general mindfulness factor (Walach et al., 2006; Baer et al., 2008; Bergomi et al., 2014). The general characteristic of being mindful as a state accompanying and following meditation practice, is also considered an individual difference that can be cultivated via meditation practice over time (Kabat-Zinn, 2003; Rau and Williams, 2016).

Furthermore, we define ecological behavior in line with environmental psychology approaches as behaviors that protect/avoid harm to the environment and span all areas of life such as nutrition, mobility and transportation, energy and water consumption, waste avoidance, and consumerism (Kaiser, 1998; Gatersleben et al., 2002; Brown and Kasser, 2005; Steg and Vlek, 2009; Geiger et al., 2017). In terms of Stern's theory toward environmentally significant behavior (Stern, 2000) we focus on individual, private sphere conservation behaviors that were successfully related to dispositional mindfulness in former studies.

That mindfulness could be related to sustainable individual behavioral choices was first suggested in the field of consumer research, where mindfulness is believed to play the role of an antagonist to impulsive, automated acquisition habits that amount to unsustainable consumerism (Kottler, 1999; Rosenberg, 2004). In a recent systematic literature review, Fischer et al. (2017) identified three more potential ways in which mindfulness could exert a positive effect on ecological behaviors in general, namely the closure of the so-called attitude-behavior gap, re-orientation toward non-materialistic, simple lifestyles, and cultivation of pro-social, compassionate behaviors.

We will outline the latter two potential ways, relevant to our own approach. Ericson et al. (2014) summarize existing evidence on how increased mindfulness may lead people to re-think their values and increase subjective well-being independently from material consumption. In various studies mindfulness was related to increased subjective wellbeing and life satisfaction (Brown et al., 2009) and the endorsement of intrinsic, non-materialistic values and ecological conservation behaviors (e.g., Brown and Kasser, 2005).

Unveiling a further potential of mindfulness to increase ecological behavior, various studies have shown that mindfulness meditation increases pro-social behavior (Leiberg et al., 2011) and compassion for other people (Lim et al., 2015), both eventually increasing ecological behavior (Corral Verdugo et al., 2011; Pfattheicher et al., 2015; Geiger and Keller, 2017). More directly, Panno et al. (2017) showed that a positive relationship between trait mindfulness and ecological behavior was mediated through decreased social dominance orientation using a known group approach. This personality trait of questioning existing inequalities between social groups has been repeatedly shown to relate to environmentalism (Milfont et al., 2017). Barbaro and Pickett (2015) focused on biospheric instead of altruistic concerns and showed that the relationship of mindfulness on ecological behavior is mediated by their connectedness to nature (Mayer and Frantz, 2004). In an experimental study by Tang et al. (2017), biospheric and altruistic cognitive foci were compared to a selfish focus. Results indicated that mindful learning techniques are only instrumental for ecological behavior when the mind is set on the environment or other people, but could be detrimental if a selfish focus is predominant.

So far, studies that empirically tested the supposed relationship between mindfulness and ecological behavior used different mediators, widely varying operationalizations of both concepts and report partially inconsistent correlations at the facet level of mindfulness (see Table 1). Most of these studies focused on self-transcendental values according to Schwartz's (1992) value model, such as pro-social, altruist, or biospheric concerns. In contrast, we suggest a mediation model based on self-enhancement, namely the improvement of personal health. Thus, we challenge the notion that an egoistic orientation is necessarily detrimental to environmental outcomes.

That mindfulness practice entails health and environmental co-benefits at the same time was suggested in a recent conceptual paper by Barrett et al. (2016). Unlike Barrett et al. who explicitly bank on positive side-effects of mindfulness-induced-behaviors (as e.g., vegetarian nutrition) on health and the environment, we propose a mediation model beyond simple co-benefits, in which increased mindfulness ignites increased health behavior, which in turn is positively related to general ecological behavior even without health benefits (see Figure 1). This supposition is backed up by a substantial body of empirical research for path a (from mindfulness to health behavior). For path b (from health behavior to ecological behavior) there are various conceptual considerations and recent evidence suggesting such a relation.

As the two studies only differed in terms of the scale used to measure mindfulness and incorporated covariates, they are reported together. The second study was included to add empirical evidence to the passionate debate on the true nature of mindfulness and its adequate assessment (e.g., Brown et al., 2011; Grossman, 2011). That mindfulness is not only a heightened capacity for paying attention, but additionally characterized by an ethical mind-set, is not fully accounted for in the KIMS scale used in study 1 (Baer et al., 2004), unlike the CHIME scale used in study 2 (Bergomi et al., 2014). This scale, validated in German, was published only after the design of study 1 in 2014, why we run a replication study with this improved measurement of our core construct.

We corroborated our mediation model, proposing an indirect effect of mindfulness on ecological behavior through personal health behavior in two independent studies.

In a first instance, we found a consistent direct effect of mindfulness on ecological behavior (c: β = 0.19^*/0.27^**) in both studies, replicating the findings by Barbaro and Pickett (2015) on a facet level (see Table 1). The awareness/observing facet correlated strongest with ecological behavior, which supports the notion that being aware of one's body and external surroundings are conducive to ecological behavior. The positive relationship of the facets insight and relativity of the CHIME measure explains the stronger total effect in study 2: a certain distance to subjective experience, which is common to both facets, might free up time and energy to engage in more self-transcendental pursuits as ecological behavior. On the other hand, contrary to studies using the MAAS scale (Brown and Kasser, 2005; Panno et al., 2017) acting with awareness was not related to ecological behavior. Acceptance/non-judging also showed a null-effect, pointing toward the possibility that some aspects of cultivating mindfulness might be detrimental for ecological behavior, e.g., through increased compliance or disinterest in changing the—albeit unpleasant—status quo.

Our main finding though concerns the indirect effect of the relationship between mindfulness and ecological behavior: it was substantially mediated by health behavior. In both studies we found an indirect effect of a comparable standardized effect size (ab_cs = 0.108/0.102), making up 55.67/37.08% of the total effect c respectively, speaking for the general importance of health behaviors in the mindfulness-ecological behavior relation. Dispositional mindfulness related to our broad measure of health behavior, ranging from eating and hygiene habits, to exercise and risk prevention in both studies (see path a Figures 1, 2). While in the first study all mindfulness facets of the KIMS related equally to health behavior (see Table 2), in the second study, the awareness/observing facet of the CHIME related strongest to health behavior (see Table 3), supporting the idea that being aware of one's own body and the surrounding world might be a driver behind increased health inclinations.

That personal health behavior in turn is positively related to general ecological behavior (see path b in Figures 1, 2) extends the findings by Kim (2017) to a general population and a wider range of behaviors. Moreover, this relation does not simply reflect co-benefits, since all such items were excluded (see Appendixes A, B in Supplementary Material). That health conscious people are more likely to also conserve the environment beyond direct personal health gains is noteworthy. It means, that the realization of the interdependency of personal health and ecological, “planetary” health is empirically reflected in people's behavior (Nisbet and Gick, 2008; Corral Verdugo et al., 2011; Corral Verdugo, 2012). Former research mainly investigated self-transcendental ecocentric or pro-social pathways to ecological behavior (e.g., through connectedness to nature, Barbaro and Pickett, 2015; intrinsic values, Brown and Kasser, 2005, reduced social dominance orientation, Panno et al., 2017). In contrast, our mediation model is based on a self-serving, egocentric interest to improve personal health. In this respect our findings corroborate former research indicating that egocentric self-enhancement values can also advocate conservation of nature for utilization motives (e.g., Milfont and Gouveia, 2006). It also complements recent publications about co-benefits of ecological behavior opening up new ways of promoting environmental action (Corral Verdugo, 2012; Bain et al., 2016). In our case, we provide empiric evidence of why mindfulness environmental educational approaches (Barrett et al., 2016; Stanszus et al., 2017) based on health and ecological co-benefits should be fruitful.

A strength of the current study is that all paths were tested with validated scales for the respective constructs, comprising a wide variety of subfacets each. The two behavioral scales encompass a great range of relevant behaviors instead of focusing on single behaviors (e.g., smoking or recycling), as often the case in health or environmental studies. That the results have been conceptually replicated with two different mindfulness scales underlines the robustness and general character of the findings.

Neither familiarity with mindfulness, nor age or gender did explain any proportion of behavioral variance. In the case of age, our young and unrepresentative sample might have prevented a weak relation found in other studies (for an overview of age effects in ecological behavior, see Wiernik et al., 2013; Otto and Kaiser, 2014). Regarding previous exposure to the mindfulness concept, a more detailed measure of actual meditation experience in terms of regularity and length of practice would be desirable in future studies. Assessing previous practice might have yielded more insightful results into the role of actual meditation practice in cultivating sustainable lifestyles than the short measure we used. In the second study, income was a negative predictor of ecological behavior, showing that higher income is usually detrimental to ecological behavior, corroborating findings of representative studies (e.g., Kleinhückelkotten et al., 2016). This shows, that ecological behavior could also be partially motivated by saving money, another self-centered motive.

A caveat all cross-sectional studies face, lays in the inconclusiveness with regards to the assumed causal direction. The mediation analysis of cross-sectional data does not per se warrant the interpretation of an intrapersonal, temporal process (Winer et al., 2016), in our case from mindfulness via health behavior to ecological behavior. Alternative models, e.g., where health problems make people turn simultaneously to mindfulness practices and ecological behavior is theoretically likewise conceivable. For further possible interpretations of mediation models, see Roe (2012). We base the interpretation of our data as an intrapersonal process on former intervention studies, which showed (for path a) that mindfulness practice fosters health behaviors such as balanced nutrition or increased exercise (e.g., Dutton, 2008; Salmoirago-Blotcher et al., 2013). This causal chain has yet to be established for path b in future research with, for example, intervention studies aiming at health behaviors and measuring ecological behaviors as a dependent variable.

A possible alternative to a causal intrapersonal process includes a common third factor, e.g. an inclination for a nature-related, spiritual lifestyle, which could be more prevalent in our voluntary samples with a propensity for the topic. As connectedness to nature was shown to mediate the relation between mindfulness and ecological behavior in a cross-sectional study similar to ours (Barbaro and Pickett, 2015), their effect could also be interpreted as a third-factor confound instead of a causal process. Appreciation of nature could be another such factor: it is partially fuelled by a health motive, i.e., people seeking restoration effects in nature settings (Maller et al., 2006; Carrus et al., 2017; see also Mercado-Doménech et al., 2017 for a recent discussion on motivation and evaluation in the perception of environmental stimuli) and has been proven to spur ecological behavior (Hartig et al., 2001; Otto et al., 2014). How these phenomena relate to each other and to mindfulness practice, health behavior and ecological behavior, will have to be explored in future studies. Furthermore, a recent study with children showed the positive results of exposure to nature for pro-social outcomes (Carrus et al., 2015) and environmental behavior (Otto and Pensini, 2017). Thus, another promising avenue for future research would be an intervention study combining nature-exposure and mindfulness practices, measuring the outcomes on health and environmental behavior.

The presented evidence on the relation between mindfulness and ecological behavior favors a mediation model via personal health behavior and partially supports former research on a direct, modest link between the two concepts. Our findings suggest that the increased ecological behavior of mindful people is an indirect consequence of increased personal health behavior. The overall positive effect of mindfulness on ecological behavior seems to warrant considerations to include mindfulness trainings and practices into the intervention canon of environmental education. A combination of mindfulness and nature experiences might be an effective way to enhance the positive health effects while at the same time promoting ecological behavior. Generally, the results support a new narrative of the compatibility of self-serving, personal, and planetary health benefits which might be a more fruitful approach to promote ecological behavior than mainly cognitive, moral, or normative appeals.

SG: Analyzed data, did literature research, concept of the paper, wrote the manuscript, expertise on mindfulness and ecological behavior. SO: Collected and Rasch-modeled the data, expertise on health and ecological behavior. US: Overall conceptualization on mindfulness and ecological behavior.