Why Is Nature Beneficial?The Role of Connectedness to Nature

Abstract

Three studies examine the effects of exposure to nature on positive affect and ability to reflect on a life problem. Participants spent 15 min walking in a natural setting (Studies 1, 2, & 3), an urban setting (Study 1), or watching videos of natural and urban settings (Studies 2 & 3). In all three studies, exposure to nature increased connectedness to nature, attentional capacity, positive emotions, and ability to reflect on a life problem; these effects are more dramatic for actual nature than for virtual nature. Mediational analyses indicate that the positive effects of exposure to nature are partially mediated by increases in connectedness to nature and are not mediated by increases in attentional capacity. The discussion focuses on the mechanisms that underlie the exposure to nature/well-being effects.

Full text

http://eab.sagepub.com
Environment and Behavior
DOI: 10.1177/0013916508319745
2008;
2009; 41; 607 originally published online Sep 5,Environment and Behavior and Kyffin Dolliver
F. Stephan Mayer, Cynthia McPherson Frantz, Emma Bruehlman-Senecal Why Is Nature Beneficial?: The Role of Connectedness to Nature
http://eab.sagepub.com/cgi/content/abstract/41/5/607
The online version of this article can be found at:
Published by:
http://www.sagepublications.com
On behalf of: Environmental Design Research Association
can be found at:Environment and Behavior Additional services and information for
http://eab.sagepub.com/cgi/alerts Email Alerts:
http://eab.sagepub.com/subscriptions Subscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://eab.sagepub.com/cgi/content/refs/41/5/607 Citations
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Why Is Nature Beneficial?
The Role of Connectedness
to Nature
F. Stephan Mayer
Cynthia McPherson Frantz
Oberlin College
Emma Bruehlman-Senecal
University of Illinois at Urbana–Champaign
Kyffin Dolliver
Oberlin College
Three studies examine the effects of exposure to nature on positive affect and abil-
ity to reflect on a life problem. Participants spent 15 min walking in a natural set-
ting (Studies 1, 2, & 3), an urban setting (Study 1), or watching videos of natural
and urban settings (Studies 2 & 3). In all three studies, exposure to nature
increased connectedness to nature, attentional capacity, positive emotions, and
ability to reflect on a life problem; these effects are more dramatic for actual nature
than for virtual nature. Mediational analyses indicate that the positive effects of
exposure to nature are partially mediated by increases in connectedness to nature
and are not mediated by increases in attentional capacity. The discussion focuses
on the mechanisms that underlie the exposure to nature/well-being effects.
Keywords: connectedness to nature; attentional capacity; positive affect;
ability to reflect
Environmentalists (e.g., Berry, 1997; Leopold, 1949; Orr, 1994) and nature
writers (e.g., Louv, 2005; Muir, 1894; Thoreau, 1854) have long main-
tained that humans derive physical and psychological benefits from spending
Environment and Behavior
Volume 41 Number 5
September 2009 607-643
© 2009 SAGE Publications
10.1177/0013916508319745
http://eab.sagepub.com
hosted at
http://online.sagepub.com
607
Authors’ Note: Parts of this article were presented at the 2005 annual meeting of the American
Psychological Society. The first two authors contributed to this article equally. The authors would
like to thank Emily Bielagus, Chelsea Doohan, Martha Friedman, Sam Kaplan, Evan Keeler-Wolf,
Zea Luce, Dalia Nazryan, Lisa Quay, Mindi Rock, Sam Sytsma, Allison Shauger, and Mallory
Yager for their help in conducting this research. Correspondence concerning this article should be
addressed either to F. Stephan Mayer (stephan.mayer@oberlin.edu) or to Cynthia McPherson
Frantz (cindy.frantz@oberlin.edu), Department of Psychology, Oberlin College, OH 44074.
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

time in the natural world. The past two decades of research in environmental
psychology have supported this contention. Using a variety of methodologies
and measures, researchers have shown that exposure to the natural world
decreases negative behaviors and states (e.g., aggression, anxiety, depression,
illness) and increases positive ones (e.g., affect, health, cognitive capacity).
The big picture is clear: Exposure to nature leads to many desirable outcomes
(see Health Council of the Netherlands and Dutch Council for Research on
Spatial Planning, 2004; van den Berg, 2005; Frumkin, 2001).
The details of this relationship, however, are not clear. For instance, how
does exposure to nature influence more complex cognitive-emotional
processes such as reflecting on a life problem? Is virtual nature as effective as
the real thing? Most important, through what mechanisms does nature have its
positive benefits (see report from the Health Council of the Netherlands, 2004;
Maas, Verheij, Groenewegen, de Vries, & Spreeuwenberg, 2006)?
This article aims to address these questions. The main goal of our research
is to investigate a potential mediator of the nature/well-being connection.
Specifically, the well-being effects we focus on are measures related to posi-
tive mood. Although exposure to nature has consistently been shown to affect
positive mood, to our knowledge there has not been a single experimental
study that has conducted a full mediational analysis addressing the question
of why nature produces beneficial effects. In particular, in the present work
we investigate whether increases in individuals’ experiential sense of connec-
tion to the natural world, as measured by our Connectedness to Nature Scale
(Mayer & Frantz, 2004), mediates nature’s effect on positive mood. In con-
ducting this research, we also had two secondary goals: (a) to determine
whether exposure to nature aids in the more complex socioemotional process
of reflecting on a life problem and (b) to examine the difference between real
and virtual nature for accruing benefits. With respect to these goals, the aims
of our work are more exploratory. To address these questions, in the three
studies we contrasted people who had spent time in an actual natural setting
to those who had been exposed to either an urban setting (Study 1), virtual
nature (Studies 2 and 3), or a virtual urban scene (Study 2).
Causal Mechanisms Underlying the
Beneficial Aspects of Nature
Understanding the causal mechanisms through which nature has its ben-
eficial effects is important not only for theory development but also for aid-
ing practitioners in more effective applications. To date, five potential
608 Environment and Behavior
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

mediators of nature’s benefits have been proposed (see report from the
Health Council of the Netherlands, 2004): recovery from stress and atten-
tion fatigue, encouragement to exercise, facilitating social contact, encour-
aging optimal development in children, and providing opportunities for
personal development and a sense of purpose. Although these five media-
tors have been identified, only recovery from stress and attention fatigue
have been extensively researched (see Kaplan & Kaplan, 1989, 2003;
Kaplan, 1992, 1993, 2001; Kaplan, 1995; Hartig, Böök, Garvil, Olsson, &
Gärling, 1996; Hartig, Evans, Jamner, Davis, & Gärling, 2003). In fact,
when reading the exposure to nature/well-being literature, there is often at
least the implicit sense that nature is beneficial because it is restoring, with-
out the acknowledgement that nature may be beneficial for many reasons.
If theory guides application, then it is important for practitioners to be
aware that the benefits of nature may extend beyond helping people to
recover from stress and attention fatigue. However, before theory guides
application, stringent tests of these potential mediators need to be con-
ducted. Specifically, to establish mediation, the independent variable (IV)
must significantly influence the dependent variable (DV), the IV must sig-
nificantly affect the potential mediator, the mediator must have a significant
relationship with the DV, and the relationship between the IV and DV
should be eliminated (full mediation) or weakened (partial mediation)
when the mediator is controlled for (Baron & Kenny, 1986; Preacher &
Hayes, 2004). To examine the last criterion, a statistical test, such as the
Sobel test, needs to be conducted to see whether the IV/DV correlation has
been significantly reduced when the mediator is controlled for.
Experiments examining the potential mediator of recovery from stress
and attention fatigue typically fulfill the requirements for Step 1. That is,
they investigate the impact that exposure to nature (the IV) has on overall
happiness and mood (the DVs). They typically also investigate the relation-
ship between exposure to nature (the IV) and a potential mediator, such as
directed attention (measured by errors on a proofreading task) or stress
reduction (measured by decreases in heart rate), fulfilling Step 2. However,
no published study we are aware of explicitly tests for the relationship
between the potential mediator to the DV (Step 3) or for the relationship
between the IV (exposure to nature) and DV (positive mood or happiness)
when controlling for the potential mediator (Step 4). It has been assumed
that the observed differences in the positive outcome between the nature
and urban environments must be due to their mediator of interest.
This area of research can move in new directions by conducting strin-
gent tests of what mediates the exposure to nature/well-being effects. Of the
Mayer et al. / A Mediational Analysis of Nature’s Benefits 609
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

five potential mediators that have been identified, our work highlights the
possibility that people gain purpose and meaning in life by feeling an expe-
riential sense of belonging to the natural world. In mainstream social psy-
chology, the need to belong to human groups, that is, to feel connected to
others and to feel like a valued member of a community, has been high-
lighted as a basic human need (see Baumeister & Leary, 1995; Fiske, 2004;
Myers, 2000). Extending this need to nature, the biophilia hypothesis (see
Wilson, 1984; Kellert & Wilson, 1993; Kellert, 1997) argues that people
have a biologically based need to affiliate with and feel connected to the
broader natural world. From an ecopsychological perspective, Roszak
(1995) also argued that this sense of belonging extends beyond our city lim-
its and includes a sense of belonging to the natural world. This argument
suggests that when people are in nature and meet this need to belong, they
will experience psychological benefits.
We have developed an operationalization of connectedness to the natural
world that builds on the work of Aldo Leopold (1949). Leopold argued that
in order for people to feel responsible for nature and to engage in eco-
friendly acts, they need to feel connected to nature as a plain and simple
member. Initially, our work was primarily interested in establishing the reli-
ability and validity of a scale and testing Leopold’s notion that increases in
connectedness predict eco-friendly behavior. For this purpose, we have cre-
ated the Connectedness to Nature Scale (CNS; Mayer & Frantz, 2004), a
14-item scale intended to tap into individuals’ sense of relatedness to nature.
The scale has been shown to have only one factor, to possess high internal
consistency (α=.84), test-retest reliability (r=.79), and has also been demon-
strated to correlate with biospheric values (r=.49; Schultz, 2000) and the New
Environmental Paradigm (NEP; Stern & Dietz, 1994; r=.35), a scale that
measures attitudes about environmental protection. Mayer and Frantz (2004)
also demonstrated that the scale significantly predicts eco-friendly behavior,
providing an important empirical test of Leopold’s argument that connected-
ness to nature does, in fact, promote proenvironmental actions.
In addition, six data sets (Frantz & Mayer, 2005, 2006; Mayer & Frantz,
2004, 2005) now demonstrate that the CNS significantly predicts partici-
pants’ degree of life satisfaction and overall happiness, in both community
and student samples. Furthermore, seven data sets (Frantz & Mayer, 2005,
2006, 2007; Mayer & Frantz, 2004) find CNS to be significantly related to
perspective-taking ability, an ability that has been shown to enable people
to be better able to resolve interpersonal problems (Arriaga & Rusbult,
1998) and moral dilemmas (Mason & Gibbs, 1993). Thus, connectedness
to nature is a measurable construct that is likely to have ramifications for
610 Environment and Behavior
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

people’s experience of positive affect and sense of belonging. In the present
studies, we test to see whether exposure to nature increases individuals’
sense of feeling connected to nature and positive affect. Moreover, we test
to see whether connectedness to nature mediates the relationship between
the exposure to nature and positive affect.
Ability to Reflect as a Potential Benefit
of Exposure to Nature
Because exposure to nature influences mood and cognitive processing
in such positive ways, one might expect it to affect more complex socioe-
motional processes, such as the process of reflecting on a life problem.
This benefit of being able to think through and gain perspective on
personal problems, a benefit termed reflection (Herzog, Black, Fountaine,
& Knotts, 1997), has important implications for psychological health.
Although such an effect has been hypothesized (Kaplan & Kaplan, 1989),
a direct experimental test has not been conducted. Although research has
shown that people would prefer to be in a more natural setting to engage
in reflection (Herzog et al., 1997; Koole & Van den Berg, 2005, Study 1;
Korpela, 1992) or has examined the effects of nearby nature on life-
management issues (Kuo, 2001), no experiment has actually placed
people in a natural setting and measured whether they felt better able to
resolve a life issue.
Given that the CNS has been found to be significantly related to perspec-
tive-taking ability, in the present studies we are also interested in testing to
see whether exposure to nature increases individuals’ sense of feeling con-
nectedness to nature and ability to reflect. Moreover, we test to see whether
CNS scores mediate the relationship between the exposure to nature and the
ability to reflect. Although this aspect of the present studies is more
exploratory, we nevertheless argue that findings related to the ability to
reflect have important implications.
Study 1
The main purpose of Study 1 was to examine a potential mediator of the
exposure to nature/well-being effects. In particular, we were interested in
whether connectedness to nature mediates the relationship between exposure
to nature and the outcome variables of positive affect and ability to reflect.
Mayer et al. / A Mediational Analysis of Nature’s Benefits 611
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

In this study and the subsequent studies we also controlled for a potential
alternative explanation for the benefits of exposure to nature. Attention
Restoration Theory (ART; Kaplan & Kaplan, 1989) argues that the benefits
of nature are largely due to its ability to restore cognitive resources. The
capacity for focusing our attention on relevant stimuli is limited, and when
this mental resource is depleted, people then experience mental fatigue.
Kaplan and Kaplan (1989) argued that many natural settings possess features
that make it ideal for reducing mental fatigue and restoring attentional capac-
ity. In line with other work in this area (Hartig et al., 1996), we employed an
objective measure of cognitive performance as the mediating variable used to
test whether attentional capacity can account for the observed findings.
One line of research on ART (Berto, 2005; Hartig et al., 1996,
Experiment 1; Hartig et al., 2003; Laumann, Garling, & Stormark, 2003;
Staats & Hartig, 2004) has manipulated mental fatigue, to ensure that all
participants were depleted in their attentional capacity at the start of the
study. Because of limited resources, and because ART was not the primary
focus of our work, we did not induce mental fatigue in our participants. In
addition, it is important to note that the strongest support for ART actually
occurs under the relatively low mental fatigue conditions (Hartig et al.,
1996, Experiment 1; Hartig et al., 2003). For instance, after manipulating
mental fatigue in their first experiment and finding more support for ART
in the low mental fatigue condition, Hartig et al. (1996) decided not to
manipulate attention fatigue in their second experiment. Instead, it was
assumed that participants entered the study with some degree of mental
fatigue. Similarly, we also assume that our participants, who were predom-
inantly first year college students, entered the study with some degree of
mental fatigue, for they were not only dealing with the stresses associated
with transitioning to college but also with the stresses of coursework asso-
ciated with being at a highly selective and demanding college setting.
We also controlled for the potential alternative explanation that the positive
moods that people experience in nature are simply due to reduced private
and/or public self-awareness. Private self-awareness can be defined as a self-
focus in which the “individual evaluates (his/her) actions without reference to
others” and “reflects on private, autonomous, egocentric goals,” whereas
public self-awareness “pertains to aspects of behavior in which the needs,
desires, or reactions of others . . . are taken into account” and to a state where
people “consider the impact that an action may have on others’ impressions of
(them)selves” (Mor & Winquest, 2002). One of the central theoretical points
within objective self-awareness theory is that reduction of self-awareness, both
private and public, can be experienced as a positive state (see Duval &
612 Environment and Behavior
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Wicklund, 1972; Wicklund, 1975; Carver & Scheirer, 1981; Gibbons, 1990;
Duval, Silvia, & Lalwani, 2001). From this theoretical perspective, taking
people into a natural setting may very well lead these individuals to be less
self-focused. Our work (Frantz, Mayer, Norton, & Rock, 2005) also docu-
ments how simply being less privately self-aware results in higher CNS scores.
Consequently, obtaining and controlling for these measures seemed critical.
In this study and the subsequent studies, we also wished to control for
the impact that positive affect may have on the ability to reflect.
Fredrickson’s (1998; Fredrickson & Losada, 2005) broaden-and-build
theory of positive emotions had found that positive emotions facilitate the
broadening of people’s mindsets (Fredrickson & Joiner, 2002) and enhance
their psychological resources and coping strategies (Fredrickson, Brown,
Cohn, Conway, & Mikels, 2005). Consequently, in our analyses, we not
only examined whether connection to nature mediated the exposure to
nature/ability to reflect relationship but also whether CNS remained as a
significant predictor even when positive affect was controlled for.
Method
Participants
Seventy-six introductory psychology students (51 females, 22 males, 3
unidentified) participated in a two-celled study (nature vs. urban) in return
for partial course credit.
Materials
Attentional capacity measure. Participants completed a 10 min timed
memory loaded search task, adapted from Hartig, et al. (1996). This task was
designed to measure how the participants’ experimental environment influ-
enced their attentional capacities. Participants were given five target letters
to memorize and instructed to search lines of letters for these targets. They
were told to search each line of letters only once, and to cross off any target
letters found. Each line contained 59 letters, 0 to 8 of which were targets.
Lines were arranged in sets of six, with 18–27 target letters distributed
throughout each set. Four sets of lines were arranged on a sheet, and a total
of 95 targets were located throughout the entire task. The number of target
letters missed plus the number of nontargets incorrectly crossed off indi-
cated accuracy in the search.
Positive and negative affect. We used the positive and negative affect sched-
ule (PANAS; Watson, Clark, & Tellegen, 1988) to assess participants’ state
Mayer et al. / A Mediational Analysis of Nature’s Benefits 613
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

affect. This 20-item scale is divided into 2 separable 10-item positive and neg-
ative affect scales. Positive affect is characterized by feelings of enthusiasm,
engagement, and alertness, whereas negative affect is characterized by various
types of distress, including anger, contempt, disgust, guilt, fear, and nervous-
ness. Participants rated the degree to which they were presently experiencing
each mood state using a modified Likert-type scale (1 =very slightly or not at
all, 5 =extremely). The positive and negative affect scales were both reliable
(alpha of negative affect =.87, alpha of positive affect =.88).
Connectedness to nature. The Connectedness to Nature Scale (CNS)
measures participants’ sense of oneness with the natural world, sense of kin-
ship with animals and plants, and sense of equality between the self and
nature (Mayer & Frantz, 2004). Participants completed the trait version of
this scale as part of a prescreening procedure at the beginning of the semes-
ter. They completed the state version of this scale while in a natural or urban
environment. To create the state CNS, trait items were reworded to tap their
present feelings of connectedness to the natural world (one item could not
be reworded and was dropped, resulting in 13 items; see Appendix). This
scale correlates well with the trait CNS (r’s >.6) and has been used success-
fully in previous research to measure differences in connection to nature due
to experimental manipulation (Frantz et al., 2005). Both scales used a mod-
ified Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly
agree). A reliability analysis run on the 13-item state CNS scale proved this
scale to be reliable (α=.91). The trait scale was also reliable, α=.82.
Self-awareness measures. The Situational Self-Awareness Scale (SSAS)
was employed to measure various dimensions of self-awareness. This
scale consists of nine items, with three items assessing levels of private self-
awareness, three measuring public self-awareness, and three gauging aware-
ness of one’s immediate surroundings (Govern & Marsch, 2001). Participants
reported their level of agreement with a series of statements concerning
their present state, such as, (a) right now, I am conscious of my inner feelings
(private self-awareness), (b) right now, I am concerned about the way I
present myself (public self-awareness), and (c) right now, I am keenly aware
of everything in my environment (awareness of immediate surroundings).
This scale used a modified Likert-type scale ranging from 1 (strongly
disagree) to 5 (strongly agree). All three scales were reliable, private self-
awareness alpha =.81, public self-awareness alpha =.83, and environmental
awareness alpha =.81.
614 Environment and Behavior
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Procedure
All participants completed the trait CNS as part of a longer question-
naire administered to introductory psychology students during an in-class
mass-testing session. The rest of the data were collected on two consecu-
tive days in April that had identical, sunny weather. After giving informed
consent, participants were randomly assigned to board one of two buses.
Approximately 15-20 people were on each bus for each experimental ses-
sion. Two experimenters also boarded each bus. Both buses drove 20 min
to a nearby town. One bus went to an urban downtown area, whereas the
other went to a nature preserve. Toward the end of the bus ride, an experi-
menter asked participants to silently reflect on a loose end in their life that
needs tying. Participants were informed that this loose end could be any-
thing related to schoolwork, relationships, or any other subject, but it must
be an issue that is easily resolvable. The researcher cited finding time to
study for a test as a good example of something to reflect on because it is
easily resolvable. They described keeping your parents from getting
divorced as a bad example because it is too difficult an issue to resolve dur-
ing a 10-15 min time period. The researchers also requested that partici-
pants not talk with one another while walking to their destination, so they
could give their full attention to the landscape before them.
On arrival at their destination, participants were divided into smaller
groups of 8-10 and led by an experimenter on a 10 min walk. The participants
at the nature preserve walked to one of two locations that overlooked a bend
in a small river, approximately 20 feet wide. They were led to an open area
with woods immediately around them, and they had an open view of the bend
in the creek, with woods framing the far side of the creek. The participants in
the urban condition ended up in a concrete area near a building with an adja-
cent parking lot. Their location was at the back of the building, a relatively
quiet area, set back from a street that was approximately 20 yards away.
To further insure the independence of the participants’ experiences, after
the walk, participants were told to spread out from one another. They then
sat for 5 min, having been instructed to take in the scene. Following this,
they completed the questionnaires described above and rated their ability to
reflect by reacting to the statement, “I feel more prepared to ‘tie up my
loose end’ than I did before I began this study”. They responded to this
statement using a modified Likert-type scale, ranging from 1 (strongly dis-
agree) to 7 (strongly agree). Participants were also asked to indicate the
amount of time they spend outdoors on a typical day. Last, all participants
performed the memory loaded search task. Participants were fully
debriefed on the bus ride back to campus.
Mayer et al. / A Mediational Analysis of Nature’s Benefits 615
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Results
There were no preexisting differences between the two conditions on
trait connectedness to nature, F(1, 69) =.13, p=.72. The correlations
between the main dependent measures appear in Table 1. To test the effects
of condition on our DVs, we ran a series of 2 (condition: urban vs. natural) ×
2 (gender: male vs. female) analyses of variance.1
Outcome Variables
As in past research, spending time in nature led to a number of psycho-
logical benefits. Condition influenced positive affect, F(1, 69) =5.04, p<
.05, as measured by the PANAS. Those in the nature condition reported sig-
nificantly more positive emotions (M=2.55, SD =0.80) than those in the
urban condition (M=2.06, SD =0.70). The effects of gender and the gender ×
condition interaction on positive affect were not significant, p’s >.5.
Participants in the urban condition also did not experience greater negative
emotions than did the participants in the nature condition, F(1, 68) =.27,
p>.6. Both those in the nature and those in the urban condition reported
equal (and relatively low) levels of negative emotion (nature M=1.50,
SD =0.62; urban M=1.42, SD =0.47).
Those in the nature condition also reported a significantly greater ability
to reflect on their loose end (M=4.33, SD =1.55) than those in the urban
616 Environment and Behavior
Table 1
Pearson Product Moment Correlations
Between Dependent Variables in Study 1
12345678
1. Trait CNS —
2. State CNS .36*** —
3. Attentional capacity –.22* –.24** —
4. Negative affect .00 –.05 –.02 —
5. Positive affect .20* .54*** –.10 –.13 —
6. Ability to reflect –.10 .44*** –.24** –.03 .36*** —
7. Private self-awareness .13 .34*** –.11 .15 .11 .17 —
8. Public self-awareness –.13 –.40*** .37*** .12 –.32*** –.25** .00 —
9. Environmental .14 .55*** –.06 –.10 .38*** .27** .40*** –.08
self-awareness
Note: CNS =Connectedness to Nature Scale.
*p<.10. **p<.05. ***p<.01.
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

condition (M=3.00, SD =1.84). The effects of gender and the gender ×
condition interaction on the ability to reflect were not significant, p’s >.13.
We next tested to see whether condition influenced our two potential
mediators, CNS and attentional capacity. We ran a 2 (condition) ×2 (gender)
analysis of covariance (ANCOVA) with trait CNS as a covariate and state
CNS as the DV. Thus, all results show differences in connectedness to nature
scores controlling for initial level of connection. Trait CNS scores signifi-
cantly predicted state CNS scores, F(1, 67) =16.27, r=.36, p<.01. Those
initially high on the CNS tended to remain so. Condition also influenced
CNS state scores, F(1, 67) =11.63, p<.001. Those who spent time in nature
had significantly higher CNS scores (M=4.69, SD =1.12) than those in the
urban setting (M=3.73, SD =1.06). The effects of gender and the gender ×
condition interaction on state CNS were not significant, p’s >.15.
Condition also significantly influenced attentional capacity, as measured
by the number of errors per line completed (higher numbers reflect more
errors and thus less attentional capacity), F(1, 67) =8.49, p<.01. Those in
the nature condition made significantly fewer errors (M=1.18, SD =0.47)
than those in the urban condition (M=1.60, SD =0.74). The effects of
gender and the gender ×condition interaction on number of errors were not
significant, p’s >.3.
Finally, we tested to see whether exposure to nature influenced self-
awareness. There were no significant differences due to gender or condition
on environmental awareness (F’s <1) or on private self-awareness (F’s <
1.2). However, public self-awareness was significantly higher in the urban
condition (M=11.44, SD =3.8) than in the nature condition (M=8.22,
SD =3.79), F(1, 68) =12.80, p<.001.
Mediational Analyses
The major goal of this study was to determine what mediates the positive
effects of exposure to nature. We used Baron and Kenny’s (1986) procedure
for establishing whether mediation has occurred. To show mediation, the IV
must significantly influence the potential mediator, the mediator must have
a significant relationship with the DV, and the relationship between the IV
and DV should be eliminated (full mediation) or weakened (partial media-
tion) when the mediator is controlled for. We used a series of regression
equations and the Sobel test (Preacher & Hayes, 2004) to test for mediation.
Our primary interest was in connectedness to nature as a potential medi-
ator (see Table 2). As established above, condition was found to signifi-
cantly affect positive mood. Condition also significantly influenced state
CNS scores. Moreover, state CNS scores were found to be significantly
Mayer et al. / A Mediational Analysis of Nature’s Benefits 617
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

related to positive affect. When condition and state CNS scores were both
used to predict positive affect, the relationship between condition and pos-
itive affect was significantly weakened. These analyses provide support for
the role of connection to nature as at least a partial mediator of condition’s
effect on the positive affect scores measured here.
To eliminate the alternative explanation that ART could account for our
findings, we repeated the same analyses with attentional capacity as the
potential mediator (see Table 2). First, as established above, condition
influenced the DV of interest. In addition, the IV (condition) was observed
to significantly influence the potential mediator (attentional capacity).
618 Environment and Behavior
Table 2
Test of Attentional Capacity and Connection to Nature
as Mediators of Condition Effects in Study 1
Positive Affect
Mediator Connectedness to Nature Attentional Capacity
Equation btSobel2bt Sobel
Condition1predicting mediator 1.01 4.30**** –.43 3.01**
Predictors of positive affect
1. Mediator 0.343 5.45** –.12 0.81
2. Condition 0.48 2.83** .48 2.83**
3. Mediator Condition 0.32 4.49**** 3.11** .01 0.06 .07
0.16 0.93 .48 2.56**
Ability to Reflect
Mediator Connectedness to Nature Attentional Capacity
Equation btSobel bt Sobel
Condition1predicting mediator 1.01 4.30**** –.43 3.01**
Predictors of ability to reflect
1. Mediator 0.67 4.21**** –.68 2.06**
2. Condition 1.39 3.47**** 1.39 3.47****
3. Mediator condition 0.52 2.94** 2.43** –0.34 1.03 .97
0.82 1.92* 1.30 3.02**
Note: 1. When predicting state CNS, trait CNS was controlled for. 2. The Sobel test indicates
whether including the mediator (Equation 3) significantly reduces the condition regression
weight (Equation 2 vs. Equation 3).
*p<.10. **p<.05. ****p<.001.
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

However, attentional capacity did not predict positive affect, arguing
against its role as a mediator. Thus, the criteria for mediation were not met
by attentional capacity.
We also tested to see if the increase in positive affect could be explained
by a reduction in private and/or public self-awareness. To test this, public and
private self-awareness were added to the regression equation predicting pos-
itive affect, along with condition, trait, and state CNS scores. Neither public
nor private self-awareness was a significant predictor of positive affect (b’s <
.02, t’s <.8); state CNS remained significant, b=0.33, t(65) =3.64, p<.001.
A secondary goal of this study was to determine what mediates the expo-
sure to nature/ability to reflect relationship. We repeated the same analyses
that we performed on the exposure to nature/positive mood relationship
(see Table 2). Once again, condition was observed to be significantly
related to participants’ ability to reflect. As established above, condition
also significantly influenced state CNS scores. Consistent with the mood
findings, state CNS scores were also found to significantly predict the abil-
ity to reflect. When condition and state CNS scores were used to predict the
ability to reflect, the relationship between condition and ability to reflect
was significantly weakened. Once again, these analyses provide support for
the role of connection to nature as at least a partial mediator of condition’s
effect on the positive outcomes measured here.
We then tested attentional capacity as a potential mediator. As estab-
lished above, the IV (condition) did significantly influence the potential
mediator (attentional capacity). In this instance, attentional capacity was
also found to predict the ability to reflect. However, when both attentional
capacity and condition were included in the same regression equation, the
effect of condition was not weakened. Thus, once again the criteria for
mediation were not met by attentional capacity.
Finally, because state CNS and positive affect were so highly correlated
with each other (r=.54, p<.001) we tested to see whether the effect of con-
nection to nature contributed to the ability to reflect independently of posi-
tive affect.2Both positive affect and state CNS were included in a regression
equation predicting the ability to reflect. Controlling for state CNS, positive
affect was not a significant predictor of the ability to reflect, b=0.39, t=
1.30, ns. State CNS scores did significantly predict the ability to reflect, b=
0.53, t=2.77, p<.01, above and beyond the effects of positive affect.
Discussion
Study 1 provides strong support for state CNS as a mediator of nature’s
effects on well-being. As such, this study constitutes the first time that a full
Mayer et al. / A Mediational Analysis of Nature’s Benefits 619
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

mediational analysis has been performed on the exposure to nature/positive
mood relationship. In addition, although exposure to nature was shown to sig-
nificantly affect attentional capacity, attentional capacity did not predict pos-
itive affect. That attentional capacity was demonstrated not to mediate this
relationship also highlights the importance of full mediational analyses for
controlling for alternative explanations. Thus, this study also pushes this
research in new directions, for it illustrates how another category of potential
mediators besides stress and attentional restoration can affect well-being.
Study 1 is also important because in this literature few studies have
taken people into actual nature or urban environments to test their hypothe-
ses. Given the logistics involved, especially in more northern climates, it is
understandable why these types of studies are rare. Nevertheless, experi-
ments of this type are important not just to investigate causal mechanisms
but also to increase the external validity of the work.
The self-awareness findings are also of theoretical interest. The fact that
private self-awareness did not significantly decrease in nature versus the
urban setting clearly argues against the explanation that increases in posi-
tive emotions in nature were due to a reduction in private self-awareness. In
addition, controlling for public and private self-awareness did not reduce
the significant effect of connectedness to nature on positive affect. Self-
awareness is thus not a viable mediator for nature’s beneficial effects.
Study 1 also provides support for the benefits of nature on increasing the
ability to reflect. Although we view this aspect of the study as being more
exploratory, Study 1 does provide the first experimental support for
the hypothesis that actually being in nature helps people to reflect on a
life issue. Moreover, it is important to note that this effect emerged even
though our participants had no special connection to the natural area, and
spent a relatively short time there. Thus, the nature/reflection benefit effect
is not limited to some place of special meaning to a person but can even
occur in a natural setting that is unfamiliar to a person.
In addition, it is important to note that the state CNS/ability to reflect rela-
tionship remained significant even when controlling for positive affect, which
strongly argues that this relationship cannot be accounted for by Fredrickson’s
(1998) broaden-and-build theory of positive emotions. Moreover, although
attentional capacity did predict individuals’ ability to reflect, it was clearly
shown not to mediate the effect between condition and ability to reflect. Thus,
attention restoration also is not a plausible explanation for our effects.
Given that participants were actually in nature, however, may raise the
possibility that demand characteristics could account for the findings. That is,
perhaps participants in the nature condition expected that the experimenter
620 Environment and Behavior
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

intended for them to experience more positive emotions and be better able
to reflect on a life issue. Although on the face of it demand does seem plau-
sible, a closer look brings this alternative explanation into question.
Participants in the nature condition felt significantly less publicly self-aware;
reduced public self-awareness should make the participants in the nature con-
dition care less about wanting to please the experimenter and confirm the
expected hypotheses.
Study 2
Besides addressing the research goals identified in Study 1, Study 2 also
addresses the question of whether real nature provides substantially more
psychological benefits than virtual nature. As noted above, most research
on the restorative aspects of nature has been done using virtual nature:
nature viewed through a window or on a screen. This research consistently
suggests that benefits can be accrued from indirect experiences of nature.
However, are they equivalent to direct experiences? Although it may seem
self-evident that merely viewing nature could not provide the same benefits
as being immersed in nature, Levi and Kocher (1999) had called this
assumption into question. They argued that natural simulations can serve as
psychologically satisfying substitutes for the real thing. Levi and Kocher
supported this contention by citing previous studies that demonstrated high
correlations between on-site and photographic/audiovisual ratings of land-
scape preferences and scenic quality (beauty) judgments (Anderson,
Mulligan, Goodman, & Regen, 1983; Daniel, 1990; Shuttleworth, 1980;
Zube, Pitt, & Anderson, 1975).
Based on these findings, it is possible that virtual nature can substitute for
actual nature. However, that is not the only interpretation. It may be that land-
scape preference and scenic beauty assessments are based heavily on visual
input from the environment, whereas psychological benefits of nature derive
from other senses as well. Natural environments tend to engage a broader range
of senses than do photographs and videos. Although videos are composed of
visual and auditory components, and photographs only contain a visual dimen-
sion, real environments are much more complex, involving the sense of sight,
sound, touch, and smell. The eye is also capable of taking in a larger field of
vision at any one moment than many cameras (Shuttleworth, 1980).
Thus, the issue raised by Levi and Kocher remains unanswered: Is vir-
tual nature equivalent to the real thing? Study 2 addresses this question by
contrasting individuals who have had an actual experience in nature with
Mayer et al. / A Mediational Analysis of Nature’s Benefits 621
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

those having a virtual experience. Specifically, this study is designed to
address whether an individual can obtain the same psychological benefits
from viewing audiovisual images of nature as they can from walking in a
natural area. In addition, similar to Study 1, Study 2 addresses our research
goals concerning reflection and the mediators of the nature/well-being ben-
efits. These goals are addressed by examining whether an actual experience
in nature has a greater impact on reflection than a virtual experience and by
examining whether connectedness to nature mediates the relationship
between exposure to nature and the psychological benefits of positive affect
and reflection.
Method
Participants
Participants consisted of 92 male and female undergraduates (28 male,
61 female, and 3 unidentified) enrolled in introductory psychology at
Oberlin College. They received course credit for participating in this three-
celled study (actual nature vs. virtual nature vs. virtual urban).
Materials
Videotapes. The experimental procedure made use of three videotapes
filmed by the research team using a handheld digital camera. Two of these
videos were nature videos taken in the local arboretum, which replicated the
path and duration of the outdoor group’s walk. The nature videos were iden-
tical, except one was filmed on a sunny day and the other on a cloudy day.
The nature video shown during a given experimental session was yoked to the
weather conditions outdoors. In addition, there was one urban video contain-
ing footage of a busy metropolitan street on a partially cloudy day. The urban
video was approximately equivalent in length to the nature videos.
Additional measures. Participants completed the same memory loaded
search task used in the Study 1. They also completed the PANAS (alpha of
negative affect =.83, alpha of positive affect =.89, the trait CNS (α=.84),
and the state CNS (α=.82). In addition, they completed measures of envi-
ronmental awareness (α=.79), private self-awareness (α=.79), and public
self-awareness (α=.89).
Procedure
Once again, all participants completed the trait CNS as part of a longer
questionnaire administered to introductory psychology during an in-class
622 Environment and Behavior
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

mass-testing session. All subsequent data collection took place between mid-
October through late November. Each experimental session was randomly
preassigned to either the nature condition or the video conditions; all condi-
tions were run on a given day, within 15–30 min of each other, to control for
weather effects. However, the outdoor condition was not run in the rain.
All participants met in a building on campus that was equidistant to the
psychology department and the arboretum used in the outdoor condition.
Participants gave informed consent before departing for their respective
destinations. The outdoor group walked to the arboretum accompanied
by a researcher. Those in the video conditions were lead by a second
researcher to the psychology building.
The researchers explained the purpose of the experiment in transit to their
destinations. All participants were informed that they were taking part in a
study designed to assess the effects of walking on environmental percep-
tions. The researchers requested that participants should not talk with one
another while at the arboretum or while watching the videos, so they could
give their full attention to the landscape/video before them. Also in transit,
the researchers asked participants to silently reflect on a loose end in their
life that needs tying, using the same instructions from Study 1.
On arrival at the psychology building, those in the video conditions were
randomly assigned to either a nature or an urban video group. The
researcher explained that they would be viewing identical videos but that
this division allowed them to administer the questionnaires in a more time-
efficient manner. Participants then watched a 10 min video of a walk in
their assigned environment. Participants in the actual nature condition
walked for 10 min through a wooded area. Afterwards, all participants com-
pleted the PANAS, the self-awareness measures, and the CNS. After finish-
ing these questionnaires, participants rated their ability to reflect by
reacting to the statement, “I feel more prepared to ‘tie up my loose end’
than I did before I began this study.” Last, all participants performed the
memory loaded search task.
After participants completed the questionnaire and concentration task, they
were debriefed by the on-site researcher and thanked for their participation.
Results
There were no preexisting differences between the three conditions on
trait connectedness to nature, F(2, 75) =.73, p=.48. The correlations
between the main dependent measures appear in Table 3. To test the effects
of condition on our DVs, we ran a series of 3 (condition: nature vs. virtual
Mayer et al. / A Mediational Analysis of Nature’s Benefits 623
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

nature vs. virtual urban) ×2 (gender: male vs. female) analyses of vari-
ance.3Planned comparisons between virtual nature and virtual urban and
between virtual nature and nature were run on all variables.
Outcome Variables
Consistent with Study 1, condition significantly influenced positive
affect, F(2, 83) =21.56, p<.001. Those in the nature condition reported
significantly more positive emotions than those in the other two conditions
(see Table 4 for means and standard deviations of this and subsequent vari-
ables, by condition). The effects of gender and the gender ×condition inter-
action on positive affect were not significant, p’s >.6. Unlike Study 1,
condition had a marginal main effect on negative affect, F(2, 83) =2.52,
p=.09. The virtual urban condition reported significantly more negative
affect than the other two conditions. The effects of gender and the gender ×
condition interaction on negative affect were not significant, p’s >.15.
Thus, similar to Study 1, exposure to nature once again was shown to lead
to more positive emotions, although in this instance exposure to urban
scenes led to a marginally significant increase in more negative emotions.
Replicating Study 1, condition also affected participants’ ability to
reflect, F(2, 83) =6.61, p<.01. The urban video condition was signifi-
cantly less able to reflect than the other two conditions; the actual nature
condition was not significantly better able to reflect than the video nature
condition. There was also a main effect of gender, F(1, 83) =4.42, p<.05.
624 Environment and Behavior
Table 3
Pearson Product Moment Correlations Between
Dependent Variables in Study 2
12345678
1. Trait CNS —
2. State CNS .36*** —
3. Attentional capacity –.08 –.24* —
4. Negative affect .10 –.26** .01 —
5. Positive affect .09 .46*** –.20 –.21** —
6. Ability to reflect –.03 .36*** –.12 –.32*** .37*** —
7. Private self-awareness .14 .39*** –.38*** .00 .44*** .46*** —
8. Public self-awareness –.03 –.20* .13 .25** –.21** –.09 .09 —
9. Environmental .05 .39*** –.26** –.23** .39*** .35*** .36*** –.11
self-awareness
*p<.10. **p<.05. ***p<.01.
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Women reported a greater ability to reflect (M=3.75, SD =0.94) than men
(M=3.36, SD =0.95). The gender ×condition interaction was not signifi-
cant, p>.25.
We also examine whether condition influenced our two potential medi-
ators, connectedness to nature and attentional capacity. We once again
included trait CNS as a covariate; thus, all results show changes in connect-
edness to nature scores. Trait CNS scores significantly predicted state CNS
scores, F(1, 74) =9.35, r=.36, p<.01. Those initially high on the CNS
tended to remain so. Condition significantly affected state CNS scores, F(2,
74) =8.66, p<.001. All three groups differed significantly from each other,
with those in the nature conditions feeling most connected and those in the
urban video condition feeling least connected. There was a marginal main
effect of gender, F(1, 74) =3.42, p=.07. Women were marginally higher
on the CNS (M=3.40, SD =0.70) than males (M=3.11, SD =0.63). The
gender ×condition interaction was not significant, p>.6.
On examining the scores on the cognitive task, we discovered that one
participant in the virtual nature condition had an accuracy score 6 SD’s
below the mean; this participant was not included in the following analy-
ses. Condition did not have a significant effect on attentional capacity, as
measured by the number of errors per line completed, F(1, 56) =1.31,
p=.28. However, the means were in the predicted direction (see Table 4).
Mayer et al. / A Mediational Analysis of Nature’s Benefits 625
Table 4
Study 2: Means and Standard Deviations of Mediating
and Dependent Variables, by Condition
Nature Virtual Nature Virtual Urban
DV MSD M SD M SD
CNS 3.60a.64 3.28b.54 2.80c.68
Errors on task 1.00a.47 1.15a.59 1.41a.49
Positive affect 3.15a.72 2.33b.56 2.00b.48
Negative affect 1.35a.50 1.39a.38 1.66b.59
Ability to reflect 3.88a.78 3.69a.84 3.09b1.19
Private self-awareness 3.91a.77 3.60b.64 3.50b1.00
Public self-awareness 2.19a.96 2.65b.92 3.08b1.03
Environmental self-awareness 3.74a.63 3.10b.68 3.21b.82
Note: For each DV, means with different superscripts are significantly different at the p=.05
level.
DV =dependent variable.
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Finally, we tested to see whether condition affected participants’ levels
of self-awareness. As in Study 1, there was a significant effect of condition
on public self-awareness, F(2, 83) =4.57, p<.05. The nature condition
(M=2.2, SD =0.96) was less publicly self-aware than the two virtual
groups (M=3.07, SD =1.03). There was also a marginal condition ×
gender interaction, F(2, 83) =2.63, p=.08. Women in the urban video con-
dition reported feeling particularly publicly self-aware, M=3.36, SD =
0.98. The main effect of gender was not significant.
In addition, there was a main effect of condition on environmental
awareness, F(2, 83) =6.22, p<.01. Participants in the nature condition
experienced more awareness of the environment (M=3.74, SD =0.63) than
did the two virtual groups (M=3.21, SD =0.82), t(62) =2.91, p<.01. No
other effects were significant, F’s <1.4. There was also a marginal effect of
condition on private self-awareness, F(2, 83) =2.58, p=.08. Those in
nature were significantly more self-aware (M=3.91, SD =0.76) than those
in the virtual conditions (M=3.55, SD =0.82), t(90) =2.17, p<.05.
Finally, women (M=3.89, SD =0.66) reported feeling more privately self-
aware then men (M=3.36, SD =0.98), F(1, 83) =9.29, p<.01.
Mediational Analyses
Using the procedures employed in the previous study, we tested to see
whether connectedness to nature and attention restoration explained the
positive outcomes due to condition (see Table 5). To maximize power,
the two most contrasting conditions (nature vs. virtual urban) were used in
the analyses. Condition did significantly influence the DVs of interest
(positive affect and ability to reflect). As established above, condition sig-
nificantly influenced state CNS scores. State CNS scores also significantly
predicted positive affect and ability to reflect. When condition and state
CNS scores were both used to predict positive affect, the relationship
between condition and positive affect was significantly weakened. When
condition and state CNS scores were used to predict the ability to reflect,
the relationship between condition and ability reflect was significantly
weakened. These analyses provide support for the role of connection to
nature as at least a partial mediator of condition’s effect on the positive
outcomes measured here.
We also tested to see if the increase in positive affect could be explained
by a reduction in private and/or public self-awareness. To test this, public and
private self-awareness were added to the regression equation predicting pos-
itive affect, along with condition, trait, and state CNS scores. Neither public
626 Environment and Behavior
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

nor private self-awareness was a significant predictor of positive affect (b’s <
.12, t’s <1.1); state CNS remained significant, b=0.33, t(51) =2.36, p<.05.
We also conducted mediational analyses with attentional capacity as the
potential mediator. Condition did significantly predict attentional capacity.
However, attentional capacity only marginally predicted positive affect (r=
.23, p=.07) and the ability to reflect (r=.22, p=.08). When both atten-
tional capacity and condition were included in the same regression equa-
tion, attentional capacity failed to remain even marginally significant.
Condition remained significant. Thus, the conditions for mediation were
not met by attentional capacity.
Mayer et al. / A Mediational Analysis of Nature’s Benefits 627
Table 5
Test of Attentional Capacity and Connection to Nature
as Mediators of Condition Effects in Study 2
Positive Affect
Mediator Connectedness to Nature Attentional Capacity
Equation btSobel2btSobel
Condition1predicting mediator .36 3.99**** –.42 2.50**
Predictors of positive affect
1. Mediator .62 5.20**** –.17 1.46
2. Condition .59 6.87**** .59 6.87****
3. Mediator condition .38 3.51**** 2.64** .02 0.17 –.21
.46 5.39**** .58 5.12****
Ability to Reflect
Mediator Connectedness to Nature Attentional Capacity
Equation btSobel btSobel
Condition1predicting mediator .36 3.99**** –.42 2.50**
Predictors of ability to reflect
1. Mediator .61 4.13**** –.12 1.0
2. Condition .40 3.21** .40 3.21**
3. Mediator condition .49 3.07** 2.44** –.03 0.25 .24
.24 1.86* .28 1.91*
Note: 1. When predicting state CNS, trait CNS was controlled for. 2. The Sobel test indicates
whether including the mediator (Equation 3) significantly reduces the condition regression
weight (Equation 2 vs. Equation 3).
*p<.10. **p<.05. ****p<.001.
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Finally, because state CNS and positive affect were so highly correlated
with each other (r=.46, p<.001) we tested to see whether the effect of
connection to nature contributed to the ability to reflect independently
of positive affect. Both positive affect and state CNS were included in a
regression equation predicting the ability to reflect. Controlling for state
CNS, positive affect marginally predicted the ability to reflect, b=0.30,
t=1.93, p<.06. State connectedness to nature significantly predicted the
ability to reflect, b=0.44, t=2.54, p<.05, above and beyond the effects of
positive affect.
Discussion
Study 2 largely replicated the effects of Study 1: Exposure to nature led
to increases in connectedness to nature, positive affect, ability to reflect,
and a decrease in public self-awareness. The results also provide strong
support for state CNS as a mediator of the exposure to nature/well-being
effects; these effects could not be accounted for by attentional capacity
or changes in private or public self-awareness. In addition, the state
CNS/ability to reflect relationship remained even when controlling posi-
tive affect.
Unlike Study 1, Study 2 demonstrated that in contrast to those in virtual
environments, participants in the nature condition experienced both higher
levels of private self-awareness and awareness of their immediate environ-
ment. Although demand characteristics could potentially explain the
increases in self-reported positive affect and ability to reflect, a demand
explanation is not consistent with this pattern of results on the self-
awareness measures. In addition, given the added findings from the virtual
nature condition, a demand explanation would assume that somehow
people in the actual nature condition knew to feel very positive, in the
virtual nature condition only somewhat positive, and in the virtual urban to
feel the worse off. This seems implausible. Moreover, lessened public self-
awareness in the nature condition should decrease the likelihood that these
individuals would engage in behavior to please the experimenter.
The unique findings from Study 2 primarily concern whether virtual
nature produces psychological effects similar to actual nature. Although the
group means of all variables showed a trend for the virtual nature condition
to have less impact than the actual nature condition, the two groups were
significantly different only in their effect on connectedness to nature and on
positive affect.
628 Environment and Behavior
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Study 3
In Study 2, it was observed that although participants in the real nature
condition felt more positive than did the participants in either the virtual
nature or virtual urban conditions, this effect did not extend to the ability
to reflect measure. Specifically, although participants in the real nature
condition were observed to have higher reflection scores than participants
in the virtual urban condition, the real nature/virtual nature contrast was
not significant. Several factors could account for this finding. On the one
hand, it may be the case that real nature and virtual nature do not differ
from one another in terms of affecting participants’ ability to reflect. On
the other hand, it may be that in this more subtle contrast, we simply did
not have enough statistical power to ascertain whether a difference was
present. To address this question, in this study we only created a two
group comparison (i.e., between real nature vs. virtual nature). Otherwise,
Study 3 was identical to Study 2 in terms of its goals and procedures,
although different nature videos were employed in this study than in
Study 2.
Method
Participants
Sixty-four introductory psychology students (33 females, 29 males, 2
unidentified) participated in a two-cell study (actual nature vs. virtual
nature) in return for partial course credit.
Materials
Videotape. In this study, a professional-looking video was shot by a cin-
ema studies student with extensive filming experience just a few days before
participants were run (to match the amount of foliage as closely as possible).
The video replicated the path and duration of the outdoor group’s walk.
Additional measures. As in the previous studies, participants completed
the memory loaded search task. They also completed the PANAS (alpha of
negative affect =.85, alpha of positive affect =.93), the state CNS (α=.91),
trait CNS (α=.82), and the three self-awareness measures (environmental
awareness alpha =.66, private self-awareness alpha =.76, and public self-
awareness alpha =.83).
Mayer et al. / A Mediational Analysis of Nature’s Benefits 629
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Procedure
All participants completed a CNS premeasure as part of a longer ques-
tionnaire administered to introductory psychology during an in-class mass-
testing session.
All subsequent data collection took place between mid-April and mid-
May. Each experimental session was randomly preassigned to one of the two
conditions; both conditions were run on a given day, within 15–30 min of
each other (to control for weather effects.) All participants met in a building
on campus that was equidistant to the psychology department and a local
arboretum used in the outdoor condition. Participants gave informed consent
before departing for their respective destinations. The outdoor group walked
to the arboretum accompanied by a researcher. Those in the video condition
were led by a second researcher to the psychology building.
The researchers explained the purpose of the experiment in transit to
their destinations. All participants were informed that they were taking part
in a study designed to assess the effects of walking on environmental per-
ceptions. The researchers requested that participants should not talk with
one another while at the arboretum or while watching the video, so they
could give their full attention to the landscape/video before them. As in the
previous study, the researchers asked participants to silently reflect on a
loose end in their life that needs tying.
Participants then walked for 10 min through the arboretum or watched a
10-min video of the same walk through the arboretum. Afterwards, all par-
ticipants completed the measures described above. Those in the arboretum
condition entered a nearby building and completed the questionnaires in a
lounge area. Those in the video condition remained in their seats. After fin-
ishing these questionnaires, participants rated their ability to reflect by
responding to the same question used in the previous studies. Last, all par-
ticipants performed the memory loaded search task.
After participants completed the questionnaire and concentration task,
they were debriefed by the on-site researcher and thanked for their partici-
pation.
Results
There were no preexisting differences between the two conditions on
trait connectedness to nature, F(1, 58) =1.61, p=.21. The correlations
between the main dependent measures appear in Table 6. To test the effects
of condition on our DVs, we ran a series of 2 (condition: virtual nature vs.
actual nature) ×2 (gender: male vs. female) analyses of variance.4
630 Environment and Behavior
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Outcome Variables
As in the previous studies, condition influenced positive affect, F(1, 57) =
12.67, p<.001. Those in the nature condition reported significantly more
positive emotions (M=5.28, SD =1.74) than those in the virtual nature
condition (M=3.83, SD =1.33). The effects of gender and the gender ×
condition interaction on positive affect were not significant, p’s >.4. In
addition, similar to Study 1, the effect of condition on negative affect was
not significant, F(1, 57) =.30, p=.59, nor were other effects, p’s >.2.
Of particular interest, by increasing the statistical power in the present
study it was now observed that condition significantly affected the ability
to reflect, F(1, 58) =12.24, p<.001. Those in the nature condition reported
a significantly greater ability to reflect on their loose end (M=4.52, SD =
1.66) than those in the virtual nature condition (M=3.10, SD =1.68). The
effects of gender and the gender ×condition interaction on the ability to
reflect were not significant, p’s >.10.
We once again ran a 2 (condition) ×2 (gender) ANCOVA with trait CNS
as a covariate and state CNS as the DV. Thus, as in the previous studies, all
results show group differences in connectedness to nature scores that
control for initial level of connection. Trait CNS scores significantly pre-
dicted state CNS scores, F(1, 57) =10.59, r=.47, p<.01. Those initially
high on the CNS tended to remain so. Condition also influenced state CNS
scores, F(1, 57) =5.79, p<.05. Those who spent time in actual nature had
significantly higher state CNS scores (M=3.00, SD =0.70) than those in
Mayer et al. / A Mediational Analysis of Nature’s Benefits 631
Table 6
Pearson Product Moment Correlations Between
Dependent Variables in Study 3
12345678
1. Trait CNS —
2. State CNS .47*** —
3. Attentional capacity .04 .05 —
4. Negative affect –.18 –.07 .10 —
5. Positive affect .08 .41*** –.13 .01 —
6. Ability to reflect .04 .40*** .08 .05 .44*** —
7. Private self-awareness .14 .26** –.08 –.24* .33*** .31** —
8. Public self-awareness –.002 –.15 .14 –.25* –.09 –.20 .05 —
9. Environmental .03 .41*** –.12 –.03 .49*** .23* .17 .21
self-awareness
*p<.10. **p<.05. ***p<.01.
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

the virtual nature condition (M=2.59, SD =0.44). The effects of gender
and the gender ×condition interaction on state CNS were not significant,
p’s >.4.
Condition did not have a significant effect on attentional capacity, as
measured by the number of errors per line completed (higher numbers
reflect more errors, and thus less attentional capacity), F(1, 54) =2.38, p=
.13. However, the means were in the predicted direction. The nature condi-
tion committed an average of 1.07 errors/line (SD =0.54), whereas the vir-
tual condition committed an average of 1.39 errors/line (SD =0.87).
Finally, we tested to see whether condition affected participants’ levels
of self-awareness. Unlike the previous studies, there was no effect of con-
dition on public self-awareness, F’s <.5. There was a main effect of condi-
tion on environmental awareness, F(1, 60) =6.83, p<.01. Participants in
the nature condition experienced more awareness of the environment (M=
3.45, SD =0.64) than the virtual nature condition (M=3.06, SD =0.59).
No other effects were significant, F’s <1. There was also a significant
effect of condition on private self-awareness, F(1, 60) =4.07, p<.05. Once
again, those in nature were significantly more self-aware (M=3.87, SD =
0.78) than those in the virtual conditions (M=3.49, SD =0.77). No other
effects were significant, F’s <1.
Mediational Analyses
We followed the same procedures used in the previous studies for test-
ing whether connectedness to nature scores mediated the effects of condi-
tion on positive outcomes (see Table 7). Connectedness to nature received
some support as a mediator. As established above, condition significantly
influenced the DVs of interest (positive affect and ability to reflect). In
addition, condition was observed to significantly influence the state CNS
scores. State CNS scores also significantly predicted positive affect and
ability to reflect. When condition and state CNS scores were both used to
predict positive affect and ability to reflect, the condition effect was weak-
ened for both outcome variables. However, the Sobel test was only margin-
ally significant in both analyses. Thus, although consistent with the
findings from the previous studies, there is less compelling evidence for
state CNS as a mediator in this study. However, given the subtlety of the
comparison (real nature vs. virtual nature), it is not surprising that these
effects are weaker than the previously reported effects.
We also tested to see if the increase in positive affect could be explained
by a reduction in private and/or public self-awareness. To test this, public
632 Environment and Behavior
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

and private self-awareness were added to the regression equation predicting
a positive affect, along with condition, trait, and state CNS scores. Public
self-awareness did not predict positive affect, b=–0.16, t(56) =.49, ns.
Private self-awareness marginally predicted positive affect (b=0.76,
t(56) =1.77, p=.08). However, state CNS remained significant, b=4.84,
t(56) =2.60, p<.05.
Attentional capacity was only observed to be marginally influenced by
exposure to nature (see Table 7). Primarily, however, given that attentional
capacity did not predict positive affect or ability to reflect, it cannot be con-
sidered as a possible mediator of the condition/nature benefit effects.
Mayer et al. / A Mediational Analysis of Nature’s Benefits 633
Table 7
Test of Attentional Capacity and Connection to Nature
as Mediators of Condition Effects in Study 3
Positive Affect
Mediator Connectedness to Nature Attentional Capacity
Equation btSobel2bt Sobel
Condition1 predicting mediator 0.33 2.35** –0.33 1.81*
Predictors of Positive Affect
1. Mediator 0.56 3.50**** –0.16 0.99
2. Condition 0.65 3.18** 0.65 3.18**
3. Mediator Condition 0.44 2.69** 1.77* –0.07 0.42 0.43
0.47 2.29** 0.58 2.61
Ability to Reflect
Mediator Connectedness to Nature Attentional Capacity
Equation btSobel bt Sobel
Condition1 predicting mediator 0.33 2.35** –0.33 1.81*
Predictors of ability to reflect
1. Mediator 1.12 3.42**** 0.21 0.64
2. Condition 1.43 3.44**** 1.43 3.44****
3. Mediator Condition 0.84 2.51** 1.75* 0.46 1.55 –1.18
1.07 2.54** 1.56 3.69****
Note: 1. When predicting state CNS, trait CNS was controlled for. 2. The Sobel test indicates
whether including the mediator (Equation 3) significantly reduces the condition regression
weight (Equation 2 vs. Equation 3).
*p<.10. **p<.05. ****p<.001.
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Finally, because state CNS and positive affect were so highly correlated
with each other (r=.41, p<.001) we tested to see whether the effect of con-
nection to nature contributes to the ability to reflect independently of posi-
tive affect. Both positive affect and state CNS were included in a regression
equation predicting the ability to reflect. Controlling for state CNS, positive
affect significantly predicted the ability to reflect, b=0.69, t=2.72, p<.01.
However, state CNS also significantly predicted the ability to reflect,
b=0.70, t=2.03, p<.05, above and beyond the effects of positive affect.
Discussion
Besides addressing the first two goals of our research, Study 3 was also
designed to follow up on Study 2’s ambiguous results on whether exposure
to real nature provides substantially more psychological benefits than expo-
sure to virtual nature. The answer from this study is clearly yes: people in
the actual nature condition did reap greater psychological benefits than did
the individuals in the virtual nature condition. In addition, as in Study 1, we
found that people actually in nature were better able to reflect than people
not actually in nature.
Although public self-awareness did not vary by condition in this study,
participants in the nature condition did feel significantly more privately
self-aware, and they were also significantly more aware of their immediate
environment than were the participants in the virtual-nature condition. The
increase in private self-awareness in the real-nature condition argues against
a demand alternative explanation, because heightened self-awareness is
associated with individuals acting in a manner more reflective of their
self-and with less concern for meeting the expectations of others.
The tests of mediation in this study were decidedly more equivocal than in
the previous studies. Connectedness to nature was found to only marginally
reduce the relationship between exposure to nature and the positive outcomes.
However, this is not surprising given that the contrast between conditions in
Study 3 (virtual vs. actual nature) is much less dramatic than the contrast in
Study 1 (actual nature vs. actual urban) and Study 2 (actual nature vs. virtual
urban). The test of attentional capacity was unequivocal: It did not mediate the
benefits of exposure to nature. Furthermore, as in Studies 1 and 2, the state
CNS/ability to reflect relationship remained significant even when controlling
for positive affect. Once again, these findings strongly argue that the impact
that CNS has on ability to reflect cannot be accounted for by Fredrickson’s
(1998) broaden-and-build theory of positive emotions or by ART.
634 Environment and Behavior
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

General Discussion
The present set of studies attempted to address three questions: (a) to
investigate connectedness to nature as a potential mediator of the nature/
well-being (i.e., positive mood) effect, (b) to determine whether exposure to
nature aids in the more complex socioemotional process of reflecting on a
life problem, and (c) to determine the difference between real and virtual
nature for accruing these benefits. Overall, the results were consistent across
three independent samples, despite differences in season and in both the
actual and virtual nature experiences. The consistency of findings across these
variations of time, place, method, and participants argues for the internal
validity of these effects. The fact that we actually took our participants into
the field to test our hypotheses argues for the external validity of the work.
In particular, these studies highlight another potential mediator other than
stress reduction and attention restoration that contributes to the exposure to
nature/well-being effects. In addition, for the first time these studies also
examine a potential mediator of the exposure to nature/well-being effect
through a full mediational analysis. These effects held even when control-
ling for alternative explanations (the broaden and build theory of positive
emotions, self-awareness theory, ART). Further, the pattern of findings is
inconsistent with a demand explanation for nature’s positive benefits.
In other words, researchers need to consider more than hectic lifestyles
and the associated stress when thinking of the exposure to nature/
well-being effects. Humans have lived the vast majority of their lives
embedded in nature, belonging to the natural world in very real ways. In
geological time, it is only a tick of the clock that we have spent in highly
urban settings, working in concrete buildings, driving in climate controlled
cars, and living in relatively densely populated areas, shut off from nature.
As Pretty (2002) estimated, for 350,000 generations humans have lived
close to the land as hunter-gatherers; a sense of belonging, place, and feeling
embedded within the broader natural world characterized these cultures. In
some ways, then, it would be surprising if the modern life of being divorced
from nature did not have some negative consequences associated with it
and that being in nature had positive benefits.
When practitioners think of how to create settings to help clients feel
better, they may want to think of more than simply how nature can restore
depleted attentional capacity and reduce stress. They may also want to think
of how people need to feel a sense of belonging to something larger than
themselves and that this need may be fulfilled through a sense of belonging
or connectedness to the natural world.
Mayer et al. / A Mediational Analysis of Nature’s Benefits 635
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Although we have noted the differences between ART and our approach,
the two approaches in all likelihood share some very strong relationships.
For instance, the factors that lead to attention restoration (being away,
extent, fascination, and compatibility) probably also increase a person’s
experiential sense of feeling connected to nature. This poses both problems
and opportunities for both approaches. Particularly, it will be important for
researchers not to assume that simply because they vary these factors that
they are necessarily affecting attention restoration, or that they are only
affecting attention restoration. Intuitively, it makes sense that exposure to
nature helps restore energy drained by our hectic modern lifestyle. Yet our
work highlights that, in fact, connectedness to nature may be playing a
more important role in some of the well-being effects.
We also want to highlight that we are not arguing that attentional capac-
ity cannot mediate the exposure to nature/well-being relation or that con-
nectedness to nature is the better mediator. As stated in the introduction,
five potential mediators have been proposed, and the purpose of our work
has been to highlight a mediator other than attentional capacity. To more
fully contrast connectedness to nature with attentional capacity, we would
have had to manipulate attentional fatigue. Because of limited resources
and because ART was not the primary focus of our work, we did not manip-
ulate mental fatigue in our participants but simply tried to control for atten-
tional capacity as an alternative explanation for our findings.
It could be argued that attentional capacity was not found to be a medi-
ator in the present studies due to the fact that perhaps we used an insensi-
tive measure of attentional capacity. Future research should indeed use
different measures of attentional capacity (e.g., the Necker Cube Pattern
Control Test, as used by Hartig et al., 2003). It should be noted, however,
that in our clearest and strongest contrast of nature with urban settings
(Study 1), the condition manipulation had a very strong effect on attentional
capacity but was still not found to be a mediator. Thus although prelimi-
nary, these results do suggest that restored attentional capacity is not a
requirement for nature to have beneficial effects.
It is also interesting to contrast our work with the work of Kuo (2001;
Kuo & Sullivan, 2001). In her quasi-experimental work, Kuo has found that
attentional capacity mediates the relationship between exposure to nature
and what we construe as more negative aspects of thought, mood, and
behavior (i.e., the ability to cope with major life problems and aggression).
Given that attentional fatigue is typically associated with increases in irri-
tability, we wonder whether attentional capacity may play a greater role in
relieving people from negative affective states whereas connectedness to
636 Environment and Behavior
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

nature may play a greater role in enhancing feelings of positive affect.
Future research might explore these possibilities.
This work also constitutes the first time that a true experimental test and
a full mediational analysis of the reflection hypothesis has been conducted
with people actually present in a natural environment. In each of the three
studies, we found that our participants in the actual nature condition reported
being better able to tie up a loose end than were the participants in either the
actual urban, virtual nature, or virtual urban conditions. This speaks to the
importance that actually being in nature has for the more complex socioe-
motional process of dealing with a life problem. Moreover, although previ-
ous researchers (e.g., Korpela, 1992; Korpela, Hartig, Kaiser, & Fuhrer,
2001) have suggested that reflection may occur primarily in places that hold
special significance to individuals, our studies suggest that the special sig-
nificance of a place is not a necessary condition for reflection to occur. Our
participants reaped the benefit of reflection even though they had never been
to this place before. Thus, although a place that holds a special significance
to a person may create more powerful effects, practitioners who wish to help
someone reflect on an issue need not limit their thinking to only such places.
One weakness of this research is that the measure of ability to reflect
was a single item self-report measure. As such, it is open to the criticism of
being unreliable. However, given that in three studies we observed similar
results with this measure, the issue of reliability seems to be minimized to
some extent. Although future research should employ multi-item scales or
perhaps behavioral measures of ability to reflect, we nonetheless think that
the present studies provide researchers with a good starting point to jump
off from in their future inquiries.
The fact that exposure to nature affects one’s ability to reflect has impor-
tant implications. When confronted with hassles and loose ends, our work
argues for the benefits of getting outside and walking in nature. However,
our results may not generalize to more major life problems. Our participants
were instructed to not think of a major issue, such as resolving the conflict
between their parents that might lead to their divorce, because 10 min in
nature was not thought to be enough time to reflect on this problem. Future
research should explore the severity of the issues that can be resolved in
nature and whether length of exposure to nature is linked with the magnitude
of the problem that a person can gain insight into. At this point, we only
know that brief exposures can help with relatively minor issues.
In addition, in two of our three studies, positive affect predicted the abil-
ity to reflect, above and beyond the variance explained by connectedness to
nature. These findings point to another possible mechanism to consider in
Mayer et al. / A Mediational Analysis of Nature’s Benefits 637
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

future research when addressing why exposure to nature leads to enhanced
reflection. Fredrickson’s (1998) broaden-and-build theory of positive emo-
tions may prove to be helpful in guiding this research.
Our third goal, to determine the difference between real and virtual
nature for accruing these benefits, also resulted in interesting findings.
Generally, exposure to real nature was associated with greater psychologi-
cal benefits than with virtual nature. Thus, taking a walk in the woods is not
equivalent to watching a nature film if one is interested in accruing the
greatest psychological benefit for either oneself or another. Nevertheless,
not everyone has this opportunity. Shut-ins, people with busy lives, and oth-
ers living in certain urban areas may not have the luxury of having an
arboretum nearby. Consistent with previous research, our research also
speaks of the positive aspects of exposure to nature in the media.
It is also possible that more elaborate simulation technology could
approximate actual nature fairly closely. For instance, although we used
videos, there are more sophisticated virtual headsets that could make the
visual experience even more real. In addition, other sensory inputs could be
employed to even further enhance the realism of the presentation and, pos-
sibly, the psychological benefits that would result. Staff at hospitals, nurs-
ing homes, hospice workers, and so on could potentially provide benefit to
their clients through such technology.
The current findings also point out several other potential directions for
future research. In addition to exploring the complimentary roles of atten-
tion restoration and connection to nature, future research should look for
other factors that explain nature’s beneficial effects. Self-transcendence,
awe, and flow are all states of mind that are often associated with the time
spent in nature and, as such, are likely candidates.
Last, when environmentalists exhort modern humans to change their
ways, they often rely on fear communications: Unless we change our
lifestyles, societal collapse is right around the corner. Although fear com-
munications can produce attitude change (Leventhal, Watts, & Pagano,
1967), they can also lead to denial and accusations that the communicator
is an alarmist. A more positive approach would be to highlight the benefits
associated with being connected to nature. Besides the psychological ben-
efits demonstrated in this article, our previous work also indicates that
people who feel more connected to nature are more likely to engage in eco-
friendly acts (Mayer & Frantz, 2004). Perhaps, offering people a different
psychological orientation toward the world—that of being a plain and
simple member of a natural community—to achieve health benefits might
prove to be a message that people will be more willing to hear and act on.
638 Environment and Behavior
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Appendix
Connectedness to Nature Scale (State)
Please answer each of these questions in terms of the way you feel at the present
moment. There are no right or wrong answers. Using the following scale, in the
space provided next to each question simply state as honestly and candidly as you
can what you are presently experiencing.
123 4 567
Strongly Disagree Neutral Strongly Agree
____ 1. Right now I’m feeling a sense of oneness with the natural world around me.
____ 2. At the moment, I’m feeling that the natural world is a community to which
I belong.
____ 3. I presently recognize and appreciate the intelligence of other living organisms.
____ 4. At the present moment, I don’t feel connected to nature.
____ 5. At the moment, I can imagine myself as part of the larger cyclical process
of living.
____ 6. At this moment, I’m feeling a kinship with animals and plants.
____ 7. Right now, I feel as though I belong to the earth just as much as it belongs
to me.
____ 8. Right now, I am feeling deeply aware of how my actions affect the natural
world.
____ 9. Presently, I feel like I am part of the web of life.
____10. Right now, I feel that all inhabitants of earth, human and nonhuman, share
a common life force.
____11. At the moment, I am feeling embedded within the broader natural world,
like a tree in a forest.
____12. When I think of humans’ place on earth right now, I consider them to be
the most valuable species in nature.
____13. At this moment, I am feeling like I am only a part of the natural world
around me, and that I am no more important than the grass on the ground
or the birds in the trees.
Notes
1. Because it is possible that participants’ trait level of connection to nature could influ-
ence their reaction to the condition, trait CNS was initially included as a covariate in all analy-
ses, and the interaction between trait CNS and condition was tested. The interaction term was
not significant for any of the DVs.
Mayer et al. / A Mediational Analysis of Nature’s Benefits 639
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

2. In this study and in Studies 2 and 3, we also tested to see if positive affect mediated the
relationship between CNS and ability to reflect. It did not meet the criteria of mediation in any
of the three studies.
3. As in Study 1, trait CNS was initially included as a covariate in all analyses, and the
interaction between trait CNS and condition was tested. The interaction term was not signifi-
cant for any of the DVs.
4. Once again, trait CNS was initially included as a covariate in all analyses, and the inter-
action between trait CNS and condition was tested. The interaction between trait CNS and con-
dition was significant only for state CNS, F(1, 56) =4.17, p<.05. Those in the actual nature
condition showed a stronger relationship between state and trait CNS (r=.57) than those in the
virtual nature condition (r=.23). This interaction did not qualify the main effect, however.
References
Anderson, L. M., Mulligan, B. E., Goodman, L. S., & Regan, H. Z. (1983). Effects of sounds
on preferences for outdoor settings. Environment and Behavior, 15, 539-566.
Arriaga, X. B., & Rusbult, C. E. (1998). Standing in my partner’s shoes: Partner perspective
taking and reactions to accommodative dilemmas. Personality and Social Psychology
Bulletin, 24, 927-948.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51, 1173-1182.
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attach-
ments as a fundamental human motivation. Psychological Bulletin, 3, 497-529.
Berry, W. (1997). The unsettling of America: Culture and agriculture. San Francisco: Sierra
Club Book.
Berto, R. (2005). Exposure to restorative environments helps restore attentional capacity.
Journal of Environmental Psychology, 25, 249-259.
Carver, C. S., & Scheirer, M. F. (1981). Attention and self-regulation. New York: Springer.
Daniel, T. (1990). Measuring the quality of the natural environment: A psychopysiological
approach. American Psychologist, 45, 633-637.
Duval, T. S., & Wicklund, R. A. (1972). A theory of objective self-awareness. New York:
Academic Press.
Duval, T. S., Silvia, P. J., & Lalwani, N. (2001). Self-awareness and causal attribution: A dual
systems theory. Boston: Kluwer Academic.
Fiske, S. T. (2004). Social beings: Core motives in social psychology. New York: Wiley.
Frantz, C. M., & Mayer, F. S. (2005). [Introductory psychology mass testing data].
Unpublished raw data.
Frantz, C. M., & Mayer, F. S. (2006). [Introductory psychology mass testing data].
Unpublished raw data.
Frantz, C. M., & Mayer, F. S. (2007). [Introductory psychology mass testing data].
Unpublished raw data.
Frantz, C. M., Mayer, F. S., Norton, C., & Rock, M. (2005). There is no “I” in nature: The
influence of self-awareness on connectedness to nature. Journal of Environmental
Psychology, 25, 425-436.
640 Environment and Behavior
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Fredrickson, B. L. (1998). What good are positive emotions? Review of General Psychology,
2, 300-319.
Fredrickson, B. L., Brown, S., Cohn, M. A., Conway, A., & Mikels, J. (2005, January). Positive
emotions build personal resources and predict future subjective well-being. Paper pre-
sented at the Sixth Annual Meeting of the Society for Personality and Social Psychology.
New Orleans, LA.
Fredrickson, B. L., & Joiner, T. (2002). Positive emotions trigger upward spiral toward emo-
tional well-being. Psychological Science, 13, 172-175.
Fredrickson, B. L., & Losada, M. F. (2005). Positive affect and the complex dynamics of
human flourishing. American Psychologist, 60, 678-686.
Frumkin, H. (2001). Beyond toxicity: Human health and the natural environment. American
Journal of Preventive Medicine, 3, 234-240.
Gibbons, F. X. (1990). Self-attention and behavior: A review and theoretical update. Advances
in Experimental Social Psychology, 23, 249-303.
Govern J. M., & Marsch, L. A. (2001). Development and validation of the Situational Self-
Awareness Scale. Consciousness and Cognition: An International Journal, 10, 366-378.
Hartig, T., Böök, A., Garvill, J., Olsson, T., & Gärling, T. (1996). Environmental influences on
psychological restoration. Scandinavian Journal of Psychology, 37, 378-393.
Hartig, T., Evans, G. W., Jamner, L. D., Davis, D. S., & Gärling, T. (2003). Tracking restora-
tion in natural and urban field settings. Journal of Environmental Psychology, 23, 109-123.
Health Council of the Netherlands and Dutch Advisory Council for Research on Spatial
Planning, Nature, and the Environment. (2004). Nature and health: The influence of nature
on social, psychological and physical well-being (publication no. 2004/09E; RMNO publi-
cation no. A02ae.). The Hague, Netherlands: Health Council of the Netherlands and RMNO.
Herzog, T. R., Black, M. A., Fountaine, K. A., & Knotts, D. J. (1997). Reflection and atten-
tion recovery as distinctive benefits of restorative environments. Journal of Environmental
Psychology, 17, 165-170.
Kaplan, R. (1992). The psychological benefits of nearby nature. In D. Relf (Ed.). The role of
horticulture in human well-being and social development: A national symposium (pp. 125-133),
Portland, OR: Timber Press.
Kaplan, R. (1993). The role of nature in the context of the workplace. Landscape urban plan-
ning, 26, 193-201.
Kaplan, R. (2001). The nature of the view from home: Psychological benefits. Environmental
and Behavior, 33, 507-542.
Kaplan, R., & Kaplan, S. (1989). The experience of nature: A psychological perspective.
Cambridge, UK: Cambridge University Press.
Kaplan, S. (1995). The restorative benefits of nature: Towards an integrative framework.
Journal of Environmental Psychology, 15, 169-182.
Kaplan, S., & Kaplan, R. (2003). Health, supportive environments, and the reasonable person
model. American Journal of Public Health, 9, 1484-1489.
Kellert, S. R. (1997). Kinship to mastery: Biophilia in human evolution and development.
Washington, DC:Island Press.
Kellert, S. R., & Wilson, E. O. (Eds.). (1993). The biophilia hypothesis. Washington, DC:
Island Press.
Koole, S. L., & Van den Berg, A. E. (2005). Lost in the wilderness: Terror management, action ori-
entation, and nature evaluation. Journal of Personality and Social Psychology, 88, 1014-1028.
Korpela, K. M. (1992). Adolescents’ favourite places and environmental self-regulation.
Journal of Environmental Psychology, 12, 249-258.
Mayer et al. / A Mediational Analysis of Nature’s Benefits 641
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Korpela, K. M., Hartig, T., Kaiser, F. G., & Fuhrer, U. (2001). Restorative experience and self-
regulation in favorite places. Environment and Behavior, 4, 572-589.
Kuo, F. E. (2001). Coping with poverty: Impacts of environment and attention in the inner city.
Environment and Behavior, 1, 5-34.
Kuo, F. E., & Sullivan, W. C. (2001). Aggression and violence in the inner city: Effects of envi-
ronment via mental fatigue. Environment and Behavior, 4, 543-571.
Laumann, K., Garling, T., & Stormark, K. M. (2003). Selective attention and heart rate responses
to natural and urban environments. Journal of Environmental Psychology, 2, 125-134.
Leopold, A. (1949). A sand county almanac: With essays on conservation from Round River.
New York: Ballantine.
Leventhal, H., Watts, J. C., & Pagano, F. (1967). Effects of fear and instructions on how to
cope with danger. Journal of Personality and Social Psychology, 6, 313-321.
Levi, D., & Kocher, S. (1999). Virtual nature: The future effects of information technology on
our relationship to nature. Environment and Behavior, 34, 203-226.
Louv, R. (2005). Last child in the woods: Saving our children from nature-deficit disorder.
Chapel Hill, NC: Algonguin Books of Chapel Hill.
Maas, J., Verheij, R. A., Groenewegen, P. P., de Vries, S., & Spreeuwenberg, P. (2006). Green
space, urbanity, and health: How strong is the relation? Journal of Epidemiology and
Community Health, 60, 587-592.
Mason, M. G., & Gibbs, J. C. (1993). Social perspective taking and moral judgment among
college students. Journal of Adolescent Research, 8, 109-123.
Mayer, F. S., & Frantz, C. M. (2004). The Connectedness to Nature Scale: A measure of individ-
uals’ feeling in community with nature. Journal of Environmental Psychology, 24, 504-515.
Mayer, F. S., & Frantz, C. M. (2005). [Connectedness to nature, life satisfaction, and recycling
behavior in a community sample]. Unpublished raw data.
Mor, N., & Winquest, J. (2002). Self-focused attention and negative affect: A meta-analysis.
Psychological Bulletin, 4, 638-662.
Muir, J. (1894). The mountains of California. New York: The Century.
Myers, D. G. (2000). The funds, friends, and faith of happy people. American Psychologist,
55, 56-67.
Orr, D. W. (1994). Earth in mind: On education, environment, and the human prospect.
Washington, DC: Island Press.
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect
effects in simple mediation models. Behavior Research Methods, Instruments &
Computers, 36, 717-731.
Pretty, J. (2002). Agri-culture: Reconnecting people, land, and nature. London: Earthscan.
Roszak, T. (1995). Where psyche meets Gaia. In T. Roszak, M. E. Gomes, & A. D. Kanner (Eds.),
Ecopsychology: Restoring the earth, healing the mind. San Francisco: Sierra Club Books.
Schultz, P. W. (2000). Empathizing with nature: The effects of perspective taking on concern
for environmental issues. Journal of Social Issues, 56, 391-406.
Shuttleworth, S. (1980). The use of photographs as an environment presentation medium in
landscape studies. Journal of Environmental Management, 11, 61-76.
Staats, H., & Hartig, T. (2003). Alone or with a friend: A social context for psychological restora-
tion and environmental preferences. Journal of Environmental Psychology, 24, 199-211.
Stern, P. C., & Dietz, T. (1994). The value basis of environmental concern. Journal of Social
Issues, 50, 65-84.
Thoreau, H. D. (1854). Walden: Or, life in the woods. Boston: Ticknor and Fields.
642 Environment and Behavior
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from

Van den Berg, A. E. (2005). Health impacts of healing environments: A review of evidence for
benefits of nature, daylight, fresh air, and quiet in healthcare settings. Groningen,
Netherlands: Foundation 200 years University Hospital Groningen.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures
of positive and negative affect: The PANAS scale. Journal of Personality and Individual
Differences, 20, 545-550.
Wicklund, R. A. (1975). Objective self-awareness. Advances in Experimental Social
Psychology, 8, 233-273.
Wilson, E. O. (1984). Biophilia: The human bond with other species. Cambridge, MA:
Harvard University Press.
Zube, E. H., Pitt, D., & Anderson, T. W. (1975). Perception and prediction of scenic resource
values of the Northeast. In E. H. Zube, R. O. Brush, & J. G. Fabos (Eds.), Landscape
assessment: Values, perceptions and resources (pp. 151-167). Stroudsburg, PA: Dowden,
Hutchinson and Ross.
F. Stephan Mayer is a professor in the psychology department at Oberlin College. His research
interests focus on issues related to connectedness to nature and environmental sustainability.
Cynthia McPherson Frantz is an associate professor in the psychology department at
Oberlin College. She conducts research on humans’ psychological relationship to the natural
world and on the benefits of exposure to nature.
Emma Bruehlman-Senecal is a graduate student in the landscape and human health labora-
tory at the University of Illinois at Urbana–Champaign. Her current research focuses on the
effects of natural environments on psychological and social functioning.
Kyffin Dolliver is a graduate of Oberlin College where he majored in psychology.
Mayer et al. / A Mediational Analysis of Nature’s Benefits 643
at OBERLIN COLLEGE LIBRARY on November 23, 2009 http://eab.sagepub.comDownloaded from