Environment and Behavior Perceptual Evaluation of © 2014 SAGE Publications Natural...

Environment and Behavior 1 –23

© 2014 SAGE Publications Reprints and permissions:

sagepub.com/journalsPermissions.navDOI: 10.1177/0013916513520604

eab.sagepub.com

Article

Perceptual Evaluation of Natural Landscapes: The Role of the Individual Connection to Nature

I-Chun Tang1, William C. Sullivan2, and Chun-Yen Chang1

AbstractMany people do not seem to have interesting perceptual experiences while in contact with nature. To identify potential antecedents of positive perceptual experiences in natural settings, we investigated the role of a personal connection to nature on perceptual evaluation of preference-related environmental information in varying natural settings. The participants (N = 77) rated three different types of images of rural forest landscapes in terms of perceived environmental information, including sense of safety, coherence, complexity, legibility, mystery, attentional restorativeness, familiarity, and preference. They also reported their personal connection to nature. The results showed that deeper personal connections to nature are associated with greater perceptual evaluations of sense of safety, legibility, mystery, and attentional restorativeness after accounting for landscape type and familiarity. A personal connection to nature is likely to enhance a person’s perceptual experiences of natural landscapes.

Keywordsconnection to nature, landscape perception, preference, attentional restorativeness, sense of safety

1National Taiwan University, Taipei, Taiwan2University of Illinois at Urbana–Champaign, USA

Corresponding Author:Chun-Yen Chang, Department of Horticulture and Landscape Architecture, National Taiwan University, No. 138, Sec. 4 Keelong Road, Taipei 10673, Taiwan. Email: [email protected]

520604 EABXXX10.1177/0013916513520604Environment and BehaviorTang et al.research-article2014

at NATIONAL TAIWAN UNIV LIB on February 28, 2014eab.sagepub.comDownloaded from

mailto:[email protected]

http://eab.sagepub.com/


2 Environment and Behavior

Despite credible evidence demonstrating the health-promoting qualities of natural environments (e.g., Berman, Jonides, & Kaplan, 2008; De Vries, Verheij, Groenewegen, & Spreeuwenberg, 2003; van den Berg, Maas, Verheij, & Groenewegen, 2010), many people do not seem to perceive suffi-cient restorative and preference-related qualities while in contact with nature. Research has shown that certain individuals perceive lesser restorative quali-ties in natural settings than others (Hartig, Kaiser, & Bowler, 2001; Hartig, Kaiser, & Strumse, 2007). For these types of people, the uninspiring percep-tual experiences that they have had in nature may diminish their preference for natural environments (Purcell, Peron, & Berto, 2001) and consequently, decrease their willingness to visit this type of setting. Besides, research find-ings also indicate that at least some people tend to underestimate nature’s hedonic benefit and choose to avoid nearby nature (Nisbet & Zelenski, 2011). Although the reason for underestimation might be complicated, their evalua-tions of natural environments may partly be caused by lacking positive per-ceptual experiences in the past. These findings suggest that certain individuals may not be able to have interesting perceptual experiences while in contact with nature. To find a starting point to enrich perceptual experiences of natu-ral environments, we worked to identify a potential antecedent of positive perceptual experiences in natural settings.

In this article, we propose a subjective sense of connection to nature as a potential antecedent of positive perceptual experiences in natural settings. The concept of connection to nature describes a deep appreciation of, and affiliation with, natural environments; and individuals with higher levels of connection to nature are suspected to view natural landscapes as more attractive and fascinat-ing. When introducing ways to enhance restorative effects of nature, S. Kaplan (2001) emphasized the resonance between human and natural landscape and stated that the restorative effect would be greater when the mental activity is resonant with what is in the environment. The resonance of the human mind and the environment brings fascination into the restoration process. Similarly, a sense of connection to nature should encourage individuals to pay more atten-tion to natural landscapes and to resonate with the environment spontaneously. Therefore, individuals with higher levels of connection to nature should be able to find fascinating information in the encountered natural landscapes. We expect that individuals with higher levels of connection to nature will perceive much more restorative and preference-related qualities from natural settings than their counterparts who have weaker connections to nature.

Connection to NatureThe concept of connection to nature has been developed to represent a per-son’s psychological relationship with the natural world. It encompasses a




Tang et al. 3

sense of belonging to nature (Mayer & Frantz, 2004) or a sense of community with nature (Dutcher, Finley, Luloff, & Johnson, 2007). A growing body of research has shown that a connection to nature is strongly associated with environmental concern and responsible behavior toward the environment (e.g., Davis, Green, & Reed, 2009; Kals, Schumacher, & Montada, 1999).

Multi-Dimensional ApproachIn the current study, we adopted the multi-dimensional concept developed by Nisbet, Zelenski, and Murphy (2009) and defined connection to nature as including not only love and care for nature but also an appreciation for, and understanding of, our interconnectedness with all other living things on the earth. There are two reasons for adopting this concept and the corresponding measure, the nature relatedness (NR) scale. First, this concept captures three important dimensions of the human–nature relationship: an individual’s iden-tification with nature, perspectives of human impact on the environment, and enjoyment of nature. Hence, individuals with higher NR scores appear to have a greater sense of being part of the natural world, both physically and psychologically. A recent comparison of concepts and measures related to the connection to nature also suggested the adequacy of adopting a multi-dimen-sional approach to the human–nature relationship (Tam, 2013). Second, col-lectively, it describes an individual’s deep appreciation of and affiliation with natural environments and represents a possible motivation to resonate with the natural environment and to seek positive perceptual experiences in natu-ral settings.

In Relation to FamiliarityTo fully demonstrate the role of connection to nature on perceptual evalua-tion of natural landscape, we need to consider the influence of familiarity. Research indicates that to a certain extent, familiarity plays a significant role in landscape preference judgments (R. Kaplan & Kaplan, 1989; Purcell et al., 2001). Because some research suggests that a sense of connection to nature is formed through knowledge of, and frequent contact with, natural environ-ments (e.g., Chawla, 1999; Mayer & Frantz, 2004; Nisbet et al., 2009), a person with a deeper connection to nature may simultaneously be more familiar with natural environments. Thus, it is necessary to distinguish the effects of the two variables.

How is one’s connection to nature distinct from or related to familiarity? According to the definition, a sense of connection to nature encompasses not only an acquaintance with nature but also an appreciation of, and affiliation





with, the natural world. This identity of “being fond of nature” should distin-guish the effect of connection to nature from mere familiarity with nature, and therefore, has more influence on one’s attention to and interpretation of perceived natural information. Although familiarity is likely to have signifi-cant predictive power for restorative and preference-related perceptual expe-riences in natural environments (R. Kaplan & Kaplan, 1989; Purcell et al., 2001), connection to nature should exceed this power and should have addi-tional effects on landscape perception in natural settings.

Preference-Related Informational VariablesTo understand differences in the perceptions of natural environments among individuals, we include several preference-related informational variables as indicators of positive perceptual experiences in natural settings. In the fol-lowing paragraphs, we will discuss these informational variables and their links to individuals’ connection to nature.

Sense of SafetyNatural environments may contain a number of potential dangers, such as dangerous animals, unseen obstacles, or offenders in hiding, and the predic-tion of these dangers along with worries of getting lost may cause a sense of fear (Andrews & Gatersleben, 2010; Bixler, Carlisle, Hammitt, & Floyd, 1994). When individuals do not feel safe in a landscape, they are likely to experience negative emotions that may spoil their experience in natural envi-ronments. Therefore, a sense of safety (or conversely, perceived danger) remains a consistent predictor of landscape preference (Hagerhall, 2000; Herzog & Kropscott, 2004; Herzog & Kutzli, 2002).

Feelings of danger in nature may be reinforced by the physical layout of a setting (Herzog & Kirk, 2005). Based on Appleton’s (1975) prospect-refuge theory, research findings indicate that an urban environment is perceived as less dangerous when it offers prospect, a quick escape route for a potential victim and few hiding places for a potential offender (Fisher & Nasar, 1992; Nasar & Jones, 1997). Andrews and Gatersleben (2010) applied this theory to natural environments and found a similar relationship between physical lay-out and perceptions of a sense of safety. This description of a less dangerous space is also consistent with the work of Herzog and Kutzli (2002), who reported that visibility and the ability to move through a space are predictors of a sense of danger in a particular natural setting.

In addition, there is reason to believe that human factors, such as gender (Jorgensen, Hitchmough, & Calvert, 2002), age (Jorgensen & Anthopoulou,




Tang et al. 5

2007), and experience in nature (Bixler & Floyd, 1997), are predictors of the sense of safety in natural settings. As a human characteristic, connection to nature describes both psychological identification with and physical enjoy-ment of nature, and this closeness between human and nature should help individuals find comfort in nature and feel that potential danger is not as great when in contact with nature. Therefore, people with deeper connections to nature are expected to retain a sense of safety across different natural landscapes.

Preference Matrix PredictorsA 2 × 2 preference matrix was proposed by S. Kaplan (1987) to introduce a set of predictors of landscape preference. Across numerous studies examin-ing outdoor environment preferences, the authors found two content-inde-pendent predictors. One predictor, understanding, refers to the ability to make sense of environmental information, and the other predictor, exploration, refers to the propensity to acquire new environmental information. In addi-tion to these two themes, the preference matrix has a binary dimension that addresses whether the information in the scene is immediately available or is predicted or promised in the larger world. Together, these two dimensions define four predictors in the preference matrix.

The four predictors in the preference matrix are coherence (immediate understanding), complexity (immediate exploration), legibility (inferred understanding), and mystery (inferred exploration). Coherence refers to an immediate sense of order that contributes to one’s ability to understand a set-ting. Complexity refers to the intricacy and richness of a scene (R. Kaplan & Kaplan, 1989). Both coherence and complexity refer to information that is immediately available in a scene. Legibility refers to the possibility of build-ing a mental map of the larger environment to assist in understanding and finding one’s way. Mystery refers to any aspect of a setting that encourages a person to explore the setting and gain further knowledge of the scene (R. Kaplan & Kaplan, 1989). Both legibility and mystery are associated with information that is inferred from the setting; thus, the experiences and imagi-nation of the perceiver shape how the individual reacts to a scene in terms of legibility or mystery.

Because individuals with greater connections to nature are expected to view natural landscapes as more attractive and fascinating, they are expected to seek fascinating information in natural landscapes in an active manner. As a result, they should be able to perceive the four types of preference-related environmental information, including coherence, complexity, legibility, and mystery, more easily from natural landscape. In particular, connection to





nature may especially have a greater influence on perceptions of the two inferred information variables, legibility and mystery, because the perception of these two environmental qualities depends not only on the ability to gather the information that appears in the visible landscape but also on the ability to imagine and organize environmental information regarding the larger world.

Attentional RestorativenessModern humans experience mental fatigue on a regular basis and often seek natural settings as sites for the restoration of their attentional resources. Empirical studies have suggested that the restorative potential of natural environments, at least partially, leads us to prefer natural environments over human-made environments (Hartig & Staats, 2006; Purcell et al., 2001; van den Berg, Koole, & van der Wulp, 2003).

According to the Attention Restoration Theory (ART; S. Kaplan, 1995), four environmental characteristics promote attention restoration: (a) being away, a sense of being away from typical thoughts and pressures; (b) fascina-tion, a state of effortless attention; (c) extent, a feeling of being engaged in something larger than yourself—in another world; and (d) compatibility, a sense that your surroundings are suited to your activities and purposes. When a person perceives some or all of these qualities in a setting, attention restora-tion is likely to occur.

S. Kaplan (2001) has suggested that being resonant with a setting is a cru-cial component of the restorative process. Because individuals with greater connections to nature are expected to view natural landscapes as more attrac-tive and fascinating, they are also expected to be spontaneously resonant with natural environment. They should seek environmental restorativeness, a type of fascinating information, in natural landscapes in an active manner. In this study, connection to nature is predicted to encourage a perception of restor-ativeness in natural settings.

Research QuestionThis study addresses the role of connection to nature in the perceptual evalu-ation of several preference-related informational variables of rural forest landscapes. We suggest that an individual’s connection to nature may help individuals to retain a sense of safety, to take an active role in seeking posi-tive environmental information, and consequently, to perceive more restor-ative and preference-related qualities of natural settings.

The current research focused on four key hypotheses based on these con-siderations (see Figure 1). We tested the hypothesis that when viewing images of rural forest landscapes, a number of informational variables, including




Tang et al. 7

sense of safety, coherence, complexity, legibility, mystery, and attentional restorativeness, will be independent positive predictors of preference (Hypothesis 1 [H1]). Participants who show a stronger connection to nature should also exhibit a greater sense of safety when viewing images of rural forest landscapes even after accounting for landscape type and familiarity (Hypothesis 2 [H2]). After accounting for landscape type and familiarity, people with higher levels of connection to nature should perceive greater amounts of coherence (Hypothesis 3a [H3a]), complexity (Hypothesis 3b [H3b]), legibility (Hypothesis 3c [H3c]), and mystery (Hypothesis 3d [H3d]) when viewing images of rural forest landscapes. After accounting for land-scape type and familiarity, participants who show a stronger connection to nature should also perceive greater potential for attention restoration from rural forest landscapes (Hypothesis 4 [H4]).

Method

ParticipantsThe participants were chosen using a convenience sampling procedure and were invited from a commercial area close to the campus of the University of

Perceptual evaluationsof the rural forest

Sense of safety

Coherence

Legibility

Restorativeness

Complexity

Mystery

Connection to nature

Landscape type

Familiarity

Controls:

H2

H3a

H3b

H3c

H3d

H4

Preference

H1

Figure 1. Hypothesized model of the relationship between connection to nature and perceptual evaluations of the rural forest, using landscape type and familiarity as controls.





Illinois at Urbana–Champaign in early summer. Contact information was given to potential participants so they could make and confirm appointments. Each participant received 10 dollars at the end of the test in return for his or her participation. The total sample consisted of 77 participants, including 40 males (52%) and 37 females (48%). Their ages ranged from 19 to 65 years, with a mean of 28.4 years (SD =11.2 years). The participants were predomi-nately young (60% were between 19 and 25 years of age) and had resided in suburban (63%) or rural (17%) areas as teenagers. No data were missing or excluded from the analysis.

The Natural SettingA rural forest was chosen as the natural setting in this study. A combination of trees massed together in the context of grassland provides a good variety of landscape structure for the study. Three types of landscape structures were chosen to represent different levels of openness of a rural forest: (a) an area inside the forest, (b) an area on the edge of the forest, and (c) an area outside the forest. Two representative black and white images were shown for each landscape setting to give participants an idea of the surrounding conditions. Black and white images were used to avoid the influence of color (e.g., colors of flowers on the grassland) on perceptual evaluations. None of the scenes contained water or people.

To validate the assumption that different landscape structures may con-tain different levels of openness, a test was conducted with 33 respondents consisting of landscape research students and alumni. Respondents were asked to rate the six representative images in response to level of prospect and the level of refuge for a potential offender (Fisher & Nasar, 1992) by three questions: prospect, “The extent your view is unobstructed to allow your field of vision to extend deep into the scene”; the number of hiding places, “The number of potential hiding places and opportunities for con-cealment”; and accessibility, “The ease in which you can move through the scene” (Andrews & Gatersleben, 2010). When an environment offers less prospect, more hiding places for a potential offender, and less accessibility, the place exhibited lower levels of openness. The results showed that the inside-the-forest, less prospect, t(32) = −10.40, p < .001, Cohen’s d = 1.81, more hiding places, t(32) = 6.67, p < .001, d = 1.16, and less accessibility, t(32) = −11.23, p < .001, d = 1.95, and on-the-edge, less prospect, t(32) = −4.59, p < .001, d = .80, and less accessibility, t(32) = −8.90, p < .001, d = 1.55, areas exhibited significantly lower levels of openness than the outside-the-forest area (see Table 1).




Tang et al. 9

ProcedureTo avoid any disturbance to the specific settings under study, the study was conducted in a laboratory. The participants were tested individually. Initially, the participants were given a general overview of the experimental procedure and were briefly shown all six images that were to be presented in the study. We grouped the six images into three pairs according to their landscape struc-tures and then projected the images onto a large screen in random order. We projected each image for 20 s and asked each participant to relax and imagine that she or he was within the environment. Immediately after viewing one pair of two images, the participants rated the landscape in terms of perceived environmental information, including their sense of safety, preference matrix predictors, restorativeness, familiarity, and preferences in a computerized questionnaire. After providing ratings for each of the three pairs of landscape images, the participants reported their level of connection to nature using the NR scale. Finally, the participants provided their age, gender, teenage resi-dential location (“As a teenager, in which kind of setting did you spend most of your time?” urban/suburban or rural), and leisure activity information (“Thinking of the past year, how often, during an average month, would you

Table 1. Mean Ratings (Standard Deviations), Cohen’s d, and Paired-Sample t-Test of Prospect-Refuge Conditions of Three Landscape Types (N = 33).

�

(a) Inside the forest M (SD)

(b) On the edge M (SD)

(c) Outside the forest M (SD)

Prospect 2.61 (0.95)*** 3.97 (0.83)*** 4.81 (0.90) Cohen’s d 1.81 0.80 —Hiding place 5.36 (1.00)*** 3.52 (0.98) 3.50 (1.13) Cohen’s d 1.16 — —Accessibility 2.69 (0.98)*** 3.59 (0.93)*** 5.05 (0.82) Cohen’s d 1.95 1.55 —

***p < .001 when compared with (c) outside the forest.





say you went to a natural area?” on a scale of 1 to 5, with 1 = never and 5 = occurring 10 or more times).

MeasuresThe NR scale. The NR scale was developed by Nisbet et al. (2009) to depict an individual’s relationship with nature. As the authors stated, this relation-ship should be relatively stable over time and across situations. Although both a three-factor and a one-factor structure were proposed by the authors, the one-factor structure was considered more promising in later studies (Howell, Dopko, Passmore, & Buro, 2011; Nisbet, Zelenski, & Murphy, 2011; Tam, 2013). In the current study, an overall NR score was used. The NR scale measures three aspects of an individual’s connection to nature: the first aspect, NR-Self, reflects an internalized identification with nature (e.g., “My relationship to nature is an important part of who I am”); the second, NR-Perspective, reflects an external, nature-related worldview (e.g., “The state of nonhuman species is an indicator of the future for human”); and the third, NR-Experience, reflects a physical familiarity with the natural world (e.g., “I enjoy being outdoors even in unpleasant weather”). The NR scale consists of 21 items, and some of the measurement items were reverse coded. The participants used a 5-point scale to indicate the extent to which the given statements described their general relationship with the natural environment (1 = strongly disagree, 5 = strongly agree). The mean score on the NR scale was 3.83 (SD = 0.65) and the internal consistency among the items was good (Cronbach’s α = .89).

Familiarity. The participants were asked to indicate their familiarity with each of the three landscape types by answering the question “How familiar are you with places like this?” using a 5-point scale (1 = not at all, 5 = very much).

Sense of safety. The participants were asked to indicate their feelings of safety in each of the three landscape types by rating the question “How safe would you feel here?” on a 5-point scale (1 = not at all, 5 = very much).

Preference matrix predictors. According to the contents of their preference matrix, Herzog and Kropscott (2004) developed a set of questions to define the four predictors of the preference matrix, including coherence, complexity, legibility, and mystery. After comparing the set of questions with the four predictors defined by R. Kaplan and Kaplan (1989), we used the following questions in this study: coherence, “This place seems well-integrated”; com-plexity, “This place has a lot of variety”; legibility, “In this place, I would




Tang et al. 11

easily find my way around” and “In this place, I would not get lost”; and mystery, “In this place, I would be able to explore and learn more” and “In this place, I would discover more if I walked around.” Responses were pro-vided using 5-point scales (1 = strongly disagree, 5 = strongly agree).

The Perceived Restorativeness Scale (PRS-Short). Based on ART, Korpela and Hartig (1996) developed the PRS to measure how much restorative qualities the individual can perceive from the environment. Because PRS is a percep-tual evaluation of the four environmental characteristics, the scale is place-dependent. Participants rate what they perceive depending on the contents of the sites. A short version of the PRS was proposed by Berto (2005). After comparing the short PRS scale with ART, we selected the following ques-tions from the short PRS scale to measure the perceptions of four restorative qualities: being away, “This is a place away from everyday demands where I would be able to relax and think about what interests me”; fascination, “This place is fascinating; it is large enough for me to discover and be curious about things”; extent, “This place seems large; it is a world of its own”; and com-patibility, “In this place, it is easy to find my way, move around, and do what I like.” Responses were provided using 5-point scales (1 = not at all, 5 = very much). The internal consistencies among the items across three types of land-scape structures were all acceptable (Cronbach’s α > .71).

Preference. The participants were asked to indicate their preference for each of the three landscape types by answering the question “How much do you like this setting?” on a 5-point scale (1 = not at all, 5 = very much).

Analytic ApproachThe current study adopted a three-step analytic approach. First, to ensure that all informational variables in this study were valid indicators of positive per-ceptual experiences, a series of bivariate associations between the preference and perceptual variables in three different rural forest landscapes were assessed.

In the second step, we used generalized linear mixed-effects models (GLMMs) to examine the relationship between an individual’s connection to nature and perceptual evaluations of the rural forest, including his or her sense of safety, coherence, complexity, mystery, legibility, and attentional restorativeness, while using landscape type and familiarity with the natural settings as controls. This method was chosen because of the ability to address repeated measures. We began by modeling the intercept random effect on full fixed-effect models. Furthermore, we introduced a slope random effect to





allow the effect of landscape type to vary between participants. We used the Akaike information criterion (AIC) to compare the models to find the best random- and fixed-effects structures. All statistical analyses were computed in the R statistical computing environment (R Development Core Team, 2012) using the LME4 and NLME libraries for mixed-effect models.

In addition to the general relationship between an individual’s connection to nature and perceptual evaluations of the rural forest, we examined the per-ceptual response patterns in each landscape structure. We conducted a series of bivariate associations between NR and perceptual evaluations in each landscape structure.

Results

BackgroundsWe first investigated the relationship between the participants’ backgrounds and their personal connections to nature. Personal connection to nature was found to be correlated significantly with age (r = .27, p = .017). Those who reported a higher connection to nature also reported that they had made more visits to natural areas during the past year (r = .47, p < .001). There were no significant differences for gender and teenage residential location.

We then investigated the relationship between the participants’ back-grounds and their landscape perceptions. Age only correlated significantly with perceptions of sense of safety of rural forest landscapes (r = .30, p = .007). There were no significant associations between the frequency of natu-ral visits and any of the landscape perceptions of rural forests. There were no significant differences for gender and teenage residential location either.

Informational Variables and PreferenceWe assessed the associations between the preference and informational vari-ables in the three rural forest landscapes. As shown in Table 2, the preference for each of three landscape types was found to be correlated significantly with most of the environmental information perceived, except for legibility in the edge landscape. In the current study, a deeper preference for rural forest landscapes was found to be related to a higher perceptual evaluation of envi-ronmental information, consistent with H1.

Connection to Nature and Sense of SafetyAs shown in Table 3, when viewing images of rural forest settings, a connec-tion to nature significantly predicts a sense of safety after controlling for




Tang et al. 13

landscape type and individual familiarity with the natural setting (β = .406, p < .001). The result is consistent with H2. In relation to familiarity, connec-tion to nature has a significant additional effect on one’s sense of safety in rural forest settings. Furthermore, an individual’s connection to nature is cor-related significantly with his or her sense of safety in each landscape struc-ture (see Table 2).

Table 2. Mean Ratings (Standard Deviations) and Pearson’s Correlations for Sense of Safety, Coherence, Complexity, Legibility, Mystery, and PRS in Three Landscape Types (N = 77).

Safety Coherence Complexity Legibility Mystery PRS

(a) M (SD) M (SD) M (SD) M (SD) M (SD) M (SD)

Forest 33.40 (1.18) 3.73 (1.11) 3.94 (1.02) 2.31 (0.92) 4.44 (0.72) 4.03 (0.74) Edge 3.51 (1.17) 3.17 (1.11) 2.77 (1.09) 3.49 (1.07) 3.76 (1.03) 3.33 (1.01) Outside 3.87 (1.07) 3.78 (0.96) 3.05 (1.01) 3.65 (1.10) 3.96 (0.91) 3.92 (0.84)(b) Preference Forest 0.55*** 0.46*** 0.31** 0.27* 0.65*** 0.72*** Edge 0.53*** 0.35** 0.37*** 0.19 0.33** 0.54*** Outside 0.57*** 0.40*** 0.60*** 0.23* 0.58*** 0.75***(c) Connection to nature Forest 0.38*** 0.14 0.14 0.32** 0.47*** 0.37*** Edge 0.23* −0.01 0.11 0.43*** 0.27* 0.18 Outside 0.47*** 0.18 0.06 0.36*** 0.18 0.33**(d) Sense of safety Forest — 0.31** 0.09 0.54*** 0.52*** 0.56*** Edge — 0.37*** 0.30** 0.40*** 0.32** 0.52*** Outside — 0.41*** 0.24* 0.37*** 0.20 0.38***

Note. PRS = perceived restorativeness.*p < .05. **p < .01. ***p < .001.

Table 3. Parameter Estimates of the Mixed-Model Regression on Sense of Safety.

Fixed effect E SE t value Pr(>|t|)

Intercept 0.446 0.423 1.052 .294LT edge/forest −0.034 0.164 −0.206 .837LT outside/forest 0.253 0.167 1.518 .130Familiarity 0.393 0.726 5.411 <.001***Connection to nature 0.406 0.112 3.634 <.001***

Random effect Variance SD

Intercept <0.001 <0.001

Note. LT = landscape type.***p < .001.





Table 4. Parameter Estimates of the Mixed-Model Regression on Coherence.


Intercept 2.731 0.499 5.476 <.001***LT edge/forest −0.653 0.153 −4.258 <.001***LT outside/forest −0.095 0.157 −0.604 .546Familiarity 0.269 0.076 3.528 <.001***Connection to nature 0.010 0.131 0.073 .942


Intercept 0.190 0.436


This result suggests that despite the influence of landscape structure, a stronger connection to nature promotes the perception that the dangers in a particular setting are not as great—and therefore, there is less to fear—rela-tive to individuals with a weaker connection to nature. This interpretation holds not only in highly open natural environments but also in less open conditions. We further investigated the relationship between sense of safety and other environmental information in the three rural forest landscapes. Sense of safety was found to be significantly correlated with most of the environmental information perceived (see Table 2). In particular, sense of safety is correlated significantly with the perception of attentional restor-ativeness in each landscape structure. In conclusion, a stronger connection to nature appears to give people a sense of safety and enhance their perceptual enjoyment of natural settings.

Connection to Nature and Preference Matrix PredictorsCoherence and complexity. Table 4 shows that connection to nature did not significantly predict perceptions of coherence when viewing images of rural forest landscapes; thus, the results do not support H3a. Similarly, a connec-tion to nature did not significantly predict perceptions of complexity from the rural forest (see Table 5), also rejecting H3b. Further investigation of the associations between connection to nature and coherence also showed no sig-nificant correlation in any of the three landscape types (see Table 2). The correlations between connection to nature and complexity were similarly insignificant (see Table 2).

A sense of coherence or complexity is immediately available in a setting, and the perceptions of these two qualities are based on automatic, instanta-neous judgments regarding the content and arrangement of the elements




Tang et al. 15

within a site. These results show that connection to nature had no influence on our participants’ perceptions of the coherence or complexity of the forest landscape.

Legibility and mystery. Does one’s connection to nature predict scores for the two inferred characteristics of the preference matrix? As shown in Tables 6 and 7, individuals who reported a stronger connection to nature also per-ceived greater levels of legibility (β = .452, p < .001) and mystery (β = .353, p < .001) from the rural forest landscape when we controlled for landscape type and individual familiarity with the natural setting. These findings are consistent with H3c and H3d.

Table 5. Parameter Estimates of the Mixed-Model Regression on Complexity.


Intercept 3.264 0.500 6.529 <.001***LT edge/forest −1.174 0.154 −7.602 <.001***LT outside/forest −0.892 0.158 −5.652 <.001***Familiarity 0.016 0.077 0.203 .840Connection to nature 0.161 0.132 1.219 .224




Table 6. Parameter Estimates of the Mixed-Model Regression on Legibility.


Intercept −0.288 0.463 −0.622 .535LT edge/forest 1.090 0.135 8.057 <.001***LT outside/forest 1.205 0.138 8.712 <.001***Familiarity 0.243 0.069 3.544 <.001***Connection to nature 0.452 0.122 3.703 <.001***








We further investigated the perceptual response patterns in each landscape structure. Table 2 shows that the individual connection to nature was signifi-cantly correlated with the perception of legibility for each landscape type. However, the individual connection to nature was only significantly corre-lated with the perception of mystery in the inside-the-forest and on-the-edge settings, the two types of landscapes with lower levels of openness. This result may suggest that connection to nature influences people to perceive more mysterious qualities in less open conditions. Further research may be needed to clarify this relationship.

Connection to Nature and Perceived RestorativenessOur final hypothesis predicted that individuals with a stronger connection to nature would report that forest settings offered greater attentional restorative-ness. We found that a connection to nature significantly predicted perceptions of restorativeness from a rural forest setting after controlling for landscape type and familiarity with the natural setting (β = .303, p = .002, see Table 8). Participants who showed a stronger connection to nature also imagined that being in a rural forest would produce greater attention restoration after con-trolling for familiarity and landscape type, confirming H4.

To further examine the extent to which a personal connection to nature was correlated with the perceived restorativeness, we conducted a series of Pearson’s correlations. Connection to nature was found to be significantly correlated with perceived restorativeness in the inside-the-forest and outside-the-forest settings (see Table 2). These results may suggest that a deeper

Table 7. Parameter Estimates of the Mixed-Model Regression on Mystery.


Intercept 2.446 0.323 7.570 <.001***LT edge/forest −0.738 0.133 −5.525 <.001***LT outside/forest −0.571 0.129 −4.415 <.001***Familiarity 0.178 0.053 3.367 .001***Connection to nature 0.353 0.086 4.111 <.001***


Intercept 0.211 0.459 LT edge/forest 1.102 1.050 LT outside/forest 0.980 0.990





Tang et al. 17

personal connection to nature may generally promote stronger perceptions of restorativeness from rural forest settings, but this relationship is influenced by other environmental factors, such as landscape types or visual qualities of the settings. Further research might be needed to clarify this relationship.

DiscussionThis study examined the extent to which a personal connection to nature predicts a sense of safety, coherence, complexity, legibility, mystery, and per-ceived restorativeness when viewing images of various rural forest settings. The results indicate that a stronger connection to nature is correlated with most of these outcomes. Compared with those who have a weaker connection to nature, individuals with a stronger connection to nature tend to feel safer in forest areas, prefer forest scenes to a greater extent as reflected by their per-ceptions of greater legibility and mystery, and imagine that being in the set-ting would result in greater attention restoration. A personal connection to nature is likely to enhance a person’s perceptual experiences of natural land-scapes. We consider these findings and explore the contributions, implica-tions, and limitations of this work below.

ContributionsThe findings of this study offer three contributions to our understanding of the consequences of having a strong personal connection to nature.

Table 8. Parameter Estimates of the Mixed-Model Regression on Attentional Restorativeness.


Intercept 2.254 0.353 6.381 <.001***LT edge/forest −0.755 0.133 −5.661 <.001***LT outside/forest −0.194 0.113 −1.719 .088Familiarity 0.171 0.056 3.047 .003**Connection to nature 0.303 0.094 3.234 .002**


Intercept 0.302 0.550 LT edge/forest 1.094 1.046 LT outside/forest 0.668 0.817

Note. LT = landscape type.**p < .01. ***p < .001.





The first contribution is to demonstrate that a stronger personal connection to nature predicts a greater sense of safety in rural forest environments. This result remains valid even after accounting for individuals’ familiarity with the natural setting. Although a personal connection to nature is suggested by some research to be developed through direct experience and knowledge of natural environments, the differences in the sense of safety ratings may be explained by the possibility that people with a deeper personal connection to nature are psychologically closer to natural environments. Our results show that despite the influence of familiarity with the encountered landscape, a greater personal connection to nature still contributes to a greater sense of safety across different natural landscapes. The psychological closeness to nature appears to enhance confidence and mitigate fears related to natural settings.

Many natural environments do not provide good visibility or easy escape routes, such as rural woodlands or mountain trails, and as prospect-refuge theory suggests, such a setting could arouse considerable fear. This fear could trigger negative emotions toward nature, interfering with the positive experi-ence of being in a natural environment. The capacity to feel a sense of safety in nature is likely to enhance a person’s experience of nature. A personal con-nection to nature appears to be helpful in overcoming the evolutionary incli-nation to feel afraid in less open settings and in retaining a sense of safety when in natural settings.

The second contribution of this work is that it enriches our understanding of the individual characteristics that influence a person’s reaction to natural settings. The findings demonstrate that a strong personal connection to nature predicts two components of the preference matrix proposed by R. Kaplan and Kaplan (1989): legibility and mystery. These two qualities are categorized as inferred information that is not immediately available at the scene. Perceptions of these two qualities arise not only from the spatial characteristics of a set-ting but also from an individual’s mental map of that type of setting. Individuals who have stronger connections to nature are likely to possess a better understanding of natural environments and are, therefore, likely to organize the physical surrounding with more confidence. This may result in more legibility. Moreover, individuals with higher levels of connection to nature view natural landscapes as more attractive and fascinating and, there-fore, have a stronger propensity to find the mysterious qualities in the natural settings. As the authors have explained (R. Kaplan & Kaplan, 1989, p. 58), “With more Legibility, confidence is enhanced that the setting will continue to be understandable. More Mystery entices one to further exploration.” Thus, psychological connectedness to natural settings appears to play an important role in shaping a person’s reactions to these settings.




Tang et al. 19

In addition, the results show that connection to nature had no influence on our participants’ perceptions of the coherence or complexity of the forest landscape. A sense of coherence or complexity is information immediately available in a setting, and the perceptions of these two qualities are based on automatic, instantaneous judgments regarding the content and arrangement of the elements within a site. The results may suggest that even less con-nected people possess innate ability to make sense of the spatial characteris-tics of forest environments and to perceive the two immediate preference-related qualities from a setting. Further research may be needed to see whether results are malleable in different types of natural environments.

The third contribution of this work is to demonstrate the association between individuals’ connection to nature and their assessment of how restor-ative a natural setting would be. Individuals with a strong connection to nature are significantly more likely to assess a forest setting as having stron-ger restorative potential than their counterparts who have weaker connec-tions. This result remains valid even after we account for individuals’ familiarity with the natural settings used in this study. People with higher levels of connection to nature are more likely to perceive sufficient restor-ativeness while in contact with forest settings and to experience the health-promoting qualities of natural environments. Further research may be needed to see whether results are malleable in different types of natural environments.

ImplicationsHow could individuals’ connection to nature be increased? We examined sev-eral characteristics of the study participants (e.g., age and the number of times per month spent visiting nature) to determine whether they are related to the individuals’ connection to nature. The results are consistent with previ-ous work suggesting that frequent contact with natural settings is a strong predictor of a person’s connection to nature (Mayer & Frantz, 2004; Nisbet et al., 2009). Further research is needed to discover strategies for increasing the frequency and quality of natural contact, especially for urban-dwelling indi-viduals. Such strategies could include improving the accessibility and geo-graphic distribution of green areas near residential areas, encouraging the study of environmental knowledge, adding outdoor activities to environmen-tal education programs, and examining the content of natural activities to ensure that the settings do not evoke fear in some individuals. Further studies of various aspects of natural contact are likely to assist us in understanding how to better promote connection to nature.





LimitationsThe study is limited in several ways. First, it did not take place in actual forest settings to avoid uncontrollable disturbance, but people might react differ-ently when walking in an actual forest. In addition, black and white photos were used in this study to avoid the influence of color, but viewing black and white photos might create a very different experience than real life forest viewing on the other side. The results of landscape perceptions in this study might be limited by the images used. Further research that includes color images or visits to actual forest environments might be needed to determine whether results are the same.

With regard to the experimental procedures, it is not ideal to assess a con-nection to nature after participants rated the photos as ratings of connection to nature might be influenced by viewing images of rural forest landscapes. Indeed, the mean of NR in this study was found to be slightly higher when compared with other research (e.g., Howell et al., 2011; Nisbet et al., 2009; Tam, 2013). Future research should consider assessing a connection to nature before viewing images of natural landscapes.

Finally, for the measurement of the study, the statement of coherence, “this place seems well-integrated” was not understandable enough to participants. This might have caused the insignificant correlations between connection to nature and perception of coherence. Future research should avoid this statement.

ConclusionIn this research, we found support for the notion that an individual’s connec-tion to nature is related to his or her sense of safety and perceptual evaluation of various rural forest scenes. We also found that a stronger connection to nature was a significant predictor of an individual’s assessment of the restor-ative properties of a forest setting: Those with stronger connections to nature tended to assess the natural setting as being more restorative. The results sug-gested that an individual’s perceptual evaluations of natural landscapes could be improved by increasing an individual’s connection to nature. Our modern lifestyles have created psychological and physical divisions between human habitats and the natural world (Kellert, 2002; Louv, 2005). More efforts are needed to establish a close relationship between humans and nature. The con-sequences of such efforts will rewardingly enhance our perceptual experi-ences while in contact with nature.

Declaration of Conflicting InterestsThe author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.




Tang et al. 21

FundingThe author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study received financial support of the research project “Analyzing Psychophysiological Benefit of Natural Landscape by fMRI” from the National Science Council of Taiwan [NSC 100-2410-H-002-170-MY2].

ReferencesAndrews, M., & Gatersleben, B. (2010). Variations in perceptions of danger, fear

and preference in a simulated natural environment. Journal of Environmental Psychology, 30, 473-481.

Appleton, J. (1975). The experience of landscape. New York, NY: John Wiley.Berman, M. G., Jonides, J., & Kaplan, S. (2008). The cognitive benefits of interacting

with nature. Psychological Science, 19, 1207-1212.Berto, R. (2005). Exposure to restorative environments helps restore attentional

capacity. Journal of Environmental Psychology, 25, 249-259.Bixler, R. D., Carlisle, C. L., Hammitt, W. E., & Floyd, M. F. (1994). Observed fears

and discomforts among urban students on school field trips to wildland areas. Journal of Environmental Education, 26, 24-33.

Bixler, R. D., & Floyd, M. F. (1997). Nature is scary, disgusting, and uncomfortable. Environment and Behavior, 29, 443-467.

Chawla, L. (1999). Life paths into effective environmental action. Journal of Environmental Education, 31, 15-26.

Davis, J. L., Green, J. D., & Reed, A. (2009). Interdependence with the environ-ment: Commitment, interconnectedness, and environmental behavior. Journal of Environmental Psychology, 29, 173-180.

De Vries, S., Verheij, R. A., Groenewegen, P. P., & Spreeuwenberg, P. (2003). Natural environments-healthy environments? An exploratory analysis of the relationship between greenspace and health. Environment and Planning A, 35, 1717-1731.

Dutcher, D. D., Finley, J. C., Luloff, A. E., & Johnson, J. B. (2007). Connectivity with nature as a measure of environmental values. Environment and Behavior, 39, 474-493.

Fisher, B. S., & Nasar, J. L. (1992). Fear of crime in relation to three exterior site features. Environment and Behavior, 24, 35-65.

Hagerhall, C. M. (2000). Clustering predictors of landscape preference in the tradi-tional Swedish cultural landscape: Prospect-refuge, mystery, age and manage-ment. Journal of Environmental Psychology, 20, 83-90.

Hartig, T., Kaiser, F. G., & Bowler, P. A. (2001). Psychological restoration in nature as a positive motivation for ecological behavior. Environment and Behavior, 33, 590-607.

Hartig, T., Kaiser, F. G., & Strumse, E. (2007). Psychological restoration in nature as a source of motivation for ecological behaviour. Environmental Conservation, 34, 291-299.





Hartig, T., & Staats, H. (2006). The need for psychological restoration as a determi-nant of environmental preferences. Journal of Environmental Psychology, 26, 215-226.

Herzog, T. R., & Kirk, K. M. (2005). Pathway curvature and border visibility as pre-dictors of preference and danger in forest settings. Environment and Behavior, 37, 620-639.

Herzog, T. R., & Kropscott, L. S. (2004). Legibility, mystery, and visual access as predictors of preference and perceived danger in forest settings without path-ways. Environment and Behavior, 36, 659-677.

Herzog, T. R., & Kutzli, G. E. (2002). Preference and perceived danger in field/forest settings. Environment and Behavior, 34, 819-835.

Howell, A. J., Dopko, R. L., Passmore, H.-A., & Buro, K. (2011). Nature connected-ness: Associations with well-being and mindfulness. Personality and Individual Differences, 51, 166-171.

Jorgensen, A., & Anthopoulou, A. (2007). Enjoyment and fear in urban woodlands: Does age make a difference? Urban Forestry & Urban Greening, 6, 267-278.

Jorgensen, A., Hitchmough, J., & Calvert, T. (2002). Woodland spaces and edges: Their impact on perception of safety and preference. Landscape and Urban Planning, 60, 135-150.

Kals, E., Schumacher, D., & Montada, L. (1999). Emotional affinity toward nature as a motivational basis to protect nature. Environment and Behavior, 31, 178-202.

Kaplan, R., & Kaplan, S. (1989). The experience of nature: A psychological perspec-tive. New York, NY: Cambridge University Press.

Kaplan, S. (1987). Aesthetics, affect, and cognition: Environmental preference from an evolutionary perspective. Environment and Behavior, 19, 3-32.

Kaplan, S. (1995). The restorative benefits of nature: Toward an integrative frame-work. Journal of Environmental Psychology, 15, 169-182.

Kaplan, S. (2001). Meditation, restoration, and the management of mental fatigue. Environment and Behavior, 33, 480-506.

Kellert, S. R. (2002). Experiencing nature: Affective, cognitive, and evaluative devel-opment in children. In P. H. Kahn Jr. & S. R. Kellert (Eds.), Children and nature: Psychological, sociocultural and evolutionary investigations (pp. 117-151). Cambridge, MA: The MIT Press.

Korpela, K., & Hartig, T. (1996). Restorative qualities of favorite places. Journal of Environmental Psychology, 16, 221-233.

Louv, R. (2005). Last child in the woods. Chapel Hill, NC: Algonquin Books.Mayer, F. S., & Frantz, C. M. (2004). The connectedness to nature scale: A mea-

sure of individuals’ feeling in community with nature. Journal of Environmental Psychology, 24, 503-515.

Nasar, J. L., & Jones, K. M. (1997). Landscapes of fear and stress. Environment and Behavior, 29, 291-323.

Nisbet, E. K., & Zelenski, J. M. (2011). Underestimating nearby nature: Affective forecasting errors obscure the happy path to sustainability. Psychological Science, 22, 1101-1106.




Tang et al. 23

Nisbet, E. K., Zelenski, J. M., & Murphy, S. A. (2009). The nature relatedness scale: Linking individuals’ connection with nature to environmental concern and behavior. Environment and Behavior, 41, 715-740.

Nisbet, E. K., Zelenski, J. M., & Murphy, S. A. (2011). Happiness is in our nature: Exploring nature relatedness as a contributor to subjective well-being. Journal of Happiness Studies, 12, 303-322.

Purcell, T., Peron, E., & Berto, R. (2001). Why do preferences differ between scene types? Environment and Behavior, 33, 93-106.

R Development Core Team. (2012). The R project for statistical computing. Available from http://www.r-project.org/

Tam, K.-P. (2013). Concepts and measures related to connection to nature: Similarities and differences. Journal of Environmental Psychology, 34, 64-78.

van den Berg, A. E., Koole, S. L., & van der Wulp, N. Y. (2003). Environmental preference and restoration: (How) are they related? Journal of Environmental Psychology, 23, 135-146.

van den Berg, A. E., Maas, J., Verheij, R. A., & Groenewegen, P. P. (2010). Green space as a buffer between stressful life events and health. Social Science & Medicine, 70, 1203-1210.

Author BiographiesI-Chun Tang is pursuing a doctoral degree in the Department of Horticulture and Landscape Architecture at National Taiwan University. Her research interests include human–nature relationship and landscape perception and evaluation.

William C. Sullivan is a professor of landscape architecture at the University of Illinois in Urbana–Champaign. His research examines the relationships between land-scapes and human health.

Chun-Yen Chang is a professor in the Department of Horticulture and Landscape architecture at National Taiwan University. His research interests include landscape ecology, health benefits of landscape, and therapeutic landscape.


Environment and Behavior Perceptual Evaluation of © 2014 SAGE Publications Natural...

Documents

Transcript of Environment and Behavior Perceptual Evaluation of © 2014 SAGE Publications Natural...