Ij

SD

a

KDEP

I

ie1rt2feeo2o1ifta

1h

Urban Forestry & Urban Greening 12 (2013) 163– 170

Contents lists available at SciVerse ScienceDirect

Urban Forestry & Urban Greening

jou rn al h omepage: www.elsev ier .de /ufug

nfluence of environmental preference and environment type congruence onudgments of restoration potential

tephanie Wilkie ∗, Andri Stavridouepartment of Psychology, St. Peter’s Campus, University of Sunderland, Sunderland SR6 0DD, UK

r t i c l e i n f o

eywords:irected attentionnvironmental psychologylace identity

a b s t r a c t

This study (N = 120) explored whether congruence between environmental preference and environmenttype influenced judgments of likelihood for directed attention restoration. The study differs from previousresearch by operationally defining preference as a representation of place identity from environmentalpsychology rather than as the attractiveness of the location. Preference was also treated as an influence onjudgments of restoration potential instead of a consequence of directed attention fatigue. Persons expe-riencing images congruent with their environmental preference were expected to rate similar locationshighest in restoration potential compared to incongruent images; however, this was not fully supported.Those with a nature preference perceived congruent environments as the most potentially restorative,but this was higher than, rather than equal to, the restoration expectations of those in the urban congruentcondition. As expected, nature preferences influenced judgments of urban settings; this incongruence was

rated least restorative. Persons with an urban preference perceived both nature and urban environmentsequal in restoration potential; this may explain circumstances when environment choice appears incon-sistent to cognitive restoration goals. Water quality in the settings also influenced restoration judgments;pleasant waterscapes resulted in the best expectations. The findings illustrate the potential contributionof place identity from environmental psychology to the study of attention restoration outcomes andfurther support the growing literature endorsing the positive potential of urban settings.

ntroduction

Directed attention refers to ‘attention that requires effort ands susceptible to fatigue’ (Berto et al., 2010, p. 494). It is consid-red a necessary but exhaustible resource (Cohen and Spacapan,978; Driver, 2001); and depletion of directed attention storesesults in diminished cognitive performance and executive con-rol (Van der Linden et al., 2003; Berto, 2005; Kaplan and Berman,010). However, restoration of this attention capacity and recoveryrom directed attention fatigue (DAF) is well documented in naturalnvironments, particularly in comparison to urban settings, withxposure to nature resulting in improved performance on a varietyf cognitive tasks (e.g., Hartig et al., 1991; Berto, 2005; Berman et al.,008). One theory used to understand natures’ positive impactn directed attention is attention restoration theory (ART, Kaplan,995). According to ART, nature is superior to urban environments

n the recovery from DAF because natural environments possess

our specific characteristics which encourage involuntary atten-ion: fascination, being away, extent or coherence (i.e., richness),nd compatibility with restoration goals (Kaplan, 1995). In loca-

∗ Corresponding author. Tel.: +44 191 515 2601; fax: +44 191 515 2781.E-mail address: stephanie.wilkie@sunderland.ac.uk (S. Wilkie).

618-8667/$ – see front matter © 2013 Elsevier GmbH. All rights reserved.ttp://dx.doi.org/10.1016/j.ufug.2013.01.004

© 2013 Elsevier GmbH. All rights reserved.

tions which possess these qualities, directed attention disengageswhile involuntary attention takes over, thus resulting in restorationof directed attention capacity.

Individuals, it seems, do anticipate nature’s positive influenceon directed attention, rating it higher for perceived likelihood ofrecovery from DAF (Herzog et al., 1997; Laumann et al., 2001;Hartig and Staats, 2006) and recommending it to others in timesof stress (Grahn and Stigsdotter, 2003). Studies have also foundnature was more preferred by those experiencing cognitive fatigue(Hartig and Staats, 2006) and a visit to a preferred place resultedin better restoration outcomes (van den Berg et al., 2003; Korpelaand Ylén, 2009). Proponents of ART argue it is the need fordirected attention restoration and the perceived likelihood forthat restoration to occur in a particular location which createsa general preference for nature over urban environments (e.g.,Laumann et al., 2001; Herzog et al., 2003; van den Berg et al.,2003; Hartig and Staats, 2006). This account implies the perceivedlikelihood for restoration is what drives a universal nature prefer-ence that subsequently results in choosing nature when seekingrestoration from DAF. However, we believe there are two potential

limitations with the implied causal link between perceived like-lihood for restoration and environmental preference proposed byART.

1 try & U

erttaeatHMtjswtmD(aroserrewdrsd

tAdea22eacaaevttpbsFhUcpeapra1ap2

64 S. Wilkie, A. Stavridou / Urban Fores

The first limitation is the assertion that environmental pref-rence is primarily based on the need for directed attentionestoration. This view is based on ART’s evolutionary positionhat our preference for nature is because of its usefulness inerms of survival (Joye and van den Berg, 2011); and as a result,ttention restoration theorists have treated environmental pref-rence as a consequence of the perceived likelihood exposure to

nature environment would reduce attention fatigue and facili-ate directed attention recovery (e.g., Galindo and Rodriguez, 2000;erzog et al., 2003; Staats et al., 2003; Hartig and Staats, 2006;artens et al., 2011). Yet the findings of several studies contradict

he evolutionary cause and effect relationship between restorationudgments and preference suggested by ART. In some instances,cenes perceived as highly restorative were not necessarily thosehich were highly preferred (Peron et al., 2002); nor was restora-

ion found to predict environmental preference (Han, 2010). Evenore curiously, nature is not always chosen for restoration fromAF, even when perceived as the best option to achieve this goal

Herzog et al., 2002). Together, these findings raised a questionbout whether the type of environment, perceived likelihood forestoration, and environmental preference may interact in waysther than those predicted by ART when individuals make deci-ions on where to fulfill their attention restoration goals. This studyxplores one particular interaction, environment type and envi-onmental preference, as an influence on perceived likelihood forestoration. Specifically, the study tests if the congruence betweennvironment type and an individual’s environmental preferencehich has an effect. In testing this relationship, an importantistinction is made between our approach and that of attentionestoration theorists because environmental preference was con-idered an antecedent influence on judgments of an environment’sirected attention restoration potential.

A second limitation was related to the definition of environmen-al preference implemented in ART research. Even though earlyRT research considered environmental preference as a multi-imensional construct (Kaplan, 1977), preference has been almostxclusively defined as either liking the environment or finding itttractive (Herzog, 1985; Herzog et al., 2003; van den Berg et al.,003; Hartig and Staats, 2006; Han, 2010; White and Gatersleben,011). We found this definition too restrictive. Environmental pref-rence is more than an aesthetic judgment of the environment;nd we believe most preference studies using this definition do notonsider the potential diversity of person–place relationships. Onespect of person–place relationships which is under-researched inttention restoration studies is place identity. In the present study,nvironmental preference is viewed as a characteristic of the indi-idual’s place identity (Lalli, 1992; Drosletis and Vignoles, 2010) orhe ‘dimensions of self that define the individual’s personal iden-ity in relation to the physical environment. . .’ (Proshansky, 1978,. 155). Several studies support our belief that place identity maye relevant to environmental perceptions which could ultimatelyhape location choice when seeking directed attention restoration.or example, Kyle et al. (2004a, 2004b) found individuals withigher place identity associated with the Appalachian Trail in the.S. perceived setting density and greater problems of overuse,rowding, and encroachment. Place attachment, a component oflace identity (Lalli, 1992; Drosletis and Vignoles, 2010), also influ-nced location choice for recreation and stress recovery (Adevind Grahn, 2011). In addition, people also develop bonds withlaces not typically considered restorative because people-placeelationships are inextricably linked to the individual’s identity,s well as a multitude of goals beyond cognitive restoration (Lalli,

992; Drosletis and Vignoles, 2010; Scannell and Gifford, 2010);nd it is these personal ties to particular location types whichrovide a basis for one’s environmental preference (Kyle et al.,004; Drosletis and Vignoles, 2010). Therefore, studies of attention

rban Greening 12 (2013) 163– 170

restoration should explore if environmental preference defined ina manner which accounts for an environment-related identity pro-cesses effects perceived likelihood for restoration judgments andpotentially location choice for these goals.

This research is also conducted as part of the growing challengeto the view of urban environments as non-restorative. Althoughthere is little dispute that exposure to nature can have posi-tive outcomes for cognition, mood, and mental health conditionssuch as depression (e.g., van den Berg et al., 2003; Berto, 2005;Berman et al., 2012); there is also evidence urban green envi-ronments can offer a range of similar benefits (Hernández andHidalgo, 2005; Karmanov and Hamel, 2008; Martens et al., 2011).Further understanding location choices in situations of DAF mayhelp to encourage engagement with places suited to both the indi-vidual’s restoration goals, their place-identity related preferences,and maximize their restoration outcomes and enhance generalwell-being. Exploring the potential influence of environmentalpreference on decision-making in circumstances of DAF may alsoaid in overcoming negative expectations associated with placeswhich are incongruent with the individual’s environmental pref-erence, for example in the case of persons’ with strong natureidentities who find themselves in urban settings which they con-sider to be non-restorative.

Study aim and hypotheses

To address two perceived limitations in the environmental pref-erence research grounded in ART, the current research differedboth in terms of the hypothesized direction of the relationshipbetween environmental preference and perceived likelihood forrestoration; and by the operational definition of environmen-tal preference. Few directed attention restoration studies haveoperationally defined environmental preference as anything otherthan a general aesthetic rating of the environment; and, to ourknowledge, none defined it as a representation of place identity.Nor have prior studies determined how environmental prefer-ence may potentially drive perceived likelihood for restorationjudgments in situations of DAF. The aim is to explore whethercongruence between environmental preference and environmenttype influenced views of restorative potential; and propose it is theinteraction between these two variables which may explain thecounter-intuitive behavior of an individual failing to choose natureeven when aware it is generally the most suited to DAF recoverygoals. The following hypotheses were tested:

• H1: The highest restoration potential ratings would be expectedfor images congruent with the preferred environment comparedto incongruent ones.

• H2: A nature preference/urban environment incongruence wouldresult in the lowest levels of perceived restoration potential.

Method

Sample characteristics

A predominantly female (80%) sample of students (N = 120) froma northeastern English university received class credit for partici-pation. The mean age of the sample was 22.91 years (SD = 7.09). Thestudy followed British Psychological Society (BPS) ethics guidelines(BPS, 2010) and was approved by the university ethics committee.Participants indicated if they considered themselves ‘a city per-

son or a country person’ as a method of capturing environmentalpreference as a representation of their place identity. A similarurban/country person distinction has been used in place iden-tity research (Knez, 2005). ‘Country persons’ were categorized as

try & U

hu

E

mdet2pOiuwaso

HripHw(eoCt1ng

D

sT(taippS

stuaaeDfs

cEi2(2Rs

S. Wilkie, A. Stavridou / Urban Fores

aving a nature preference (n = 48) and city persons as having anrban preference (n = 65).

xperimental stimuli and materials

As a control, water type (pleasant, unpleasant, no water) wasanipulated in both environments resulting in six stimuli con-

itions (Fig. 1). Previously presence of water influenced bothnvironmental preference ratings (i.e., attractiveness) and atten-ion restoration outcomes (Laumann et al., 2001; White et al.,010), while perceived water quality significantly influenced theerceived likelihood for restoration in coastal locations (Hipp andgunseitan, 2011). The pleasant water condition included water

magery such as coastlines or well-maintained canals, whilst thenpleasant water condition included imagery of water featureshich were poorly maintained, with murky water, or established

lgae blooms. To increase ecological validity, images were cho-en to be relatively familiar with some including people, modesf transportation, or physical structures.

The 12-item version of the Perceived Restoration Scale (SRPRS,artig et al., 1997) measured the extent to which exposure to envi-

onments similar to the stimuli were considered likely to resultn directed attention restoration. Although other measures of theerceived likelihood for restoration exist (Laumann et al., 2001;an, 2003; Hartig et al., 2003; Pals et al., 2009), the PRS has beenidely used and has established reliability and validity information

e.g., Hartig et al., 1997; Han, 2003). The full scale contained16 Lik-rt items (1 = not at all to 7 = completely) representing four aspectsf restorative places (Being Away, Fascination, Coherence/extent,ompatibility). Evidence indicated these items represent two fac-ors, coherence and general restorativeness (Hartig et al., 1996,997). Only the 12 SRPRS items representing general restorative-ess were used; and Cronbach’s alpha indicated these items hadood internal consistency in the present study (.81).

esign and procedure

The study implemented a quasi-experimental design con-isting of a congruence independent variable with four levels.hese levels were created based on environmental preferencenature/urban) and environment type (nature/urban). Two condi-ions were congruent: nature preference/nature imagery (n = 23)nd urban preference/urban imagery (n = 30); and two werencongruent: nature preference/urban imagery (n = 25) and urbanreference/nature imagery (n = 35). The dependent variable oferceived restoration potential was calculated as the mean of theRPRS items.

After a study briefing and provision of consent to participate,tudents were randomly allocated to one of the six imagery condi-ions (see Fig. 1). The 20-minute session took place during a normalniversity day in the second academic term. Others have used dailycademic demands as an indicator of cognitive fatigue (e.g., Hartignd Staats, 2006; Karmanov and Hamel, 2008); therefore, it wasxpected the participants would be experiencing a similar level ofAF. They received course credit for participation but could choose

rom a large number of projects; therefore, participation was con-idered voluntary.

Participants viewed a slide show presented using E-prime. Itonsisted of 10 images specific to their experimental condition.ach image was presented for 15 s on a continuous loop last-ng 7 min consistent with previous studies, (van den Berg et al.,003; Berto, 2005). Other studies have used slide shows effectively

Ulrich, 1979; Peron et al., 2002; Herzog et al., 2003; Staats et al.,003; Berto, 2005; Hartig and Staats, 2006; Berman et al., 2008;yan et al., 2010), with little difference between real-world expo-ure, video or slide-shows of images (Hartig et al., 1997; Kjellgren

rban Greening 12 (2013) 163– 170 165

and Buhrkall, 2010). This method was chosen over actual environ-mental exposure to reduce participant burden during the wintermonths when data collection took place; and, because engagementwith the environment would result in ratings of actual restora-tion instead of anticipated restoration likelihood. After exposure tothe experimental stimuli, participants completed the demographicquestions and were instructed to consider environments like thosethey had just viewed in the slide show when completing the SRPRS.Finally, all participants viewed a humorous video of a baby laugh-ing to counteract any negative effects from viewing non-preferredenvironments or unpleasant water images.

Results

The analysis sample was reduced to 113 due to single itemmissing values and identification of one case as an outlier on thedependent variable. Data was analyzed using SPSS version 19.0.Prior to the main analysis, several checks for possible confoundswere conducted. Restoration potential ratings did not differ by gen-der, t (112) = −0.05, p = 0.96; or by age, F (1, 109) = 2.12, p = 0.13.They also did not did not differ solely by environmental prefer-ence (t = 0.36, df = 111, p = 0.72); and indicates that both participantswith a nature preference and those with an urban preference wereequally able to perceive an environment’s for restoration potential.

Effect of environment and water type

The results of a 2 × 3 (Environment × Water Type) between-subjects ANOVA on restoration potential indicated a significantmain effect of both environment (F (1, 109) = 17.23, p < 0.001,�2

p = 0.14) and water type (F (1, 109) = 12.04, p < 0.001, �2p = 0.18).

Nature environments were rated higher in potential for restora-tion (M = 5.78, SE = 0.13) than urban ones (M = 4.81, SE = 0.13). Posthoc analyses of the main effect of water type using bonferroniadjustment indicated the pleasant water condition was perceivedas significantly more restorative (M = 5.76, SE = 0.16) than both theno water (M = 5.18, SE = 0.16, p = 0.03) and unpleasant water con-ditions (M = 4.64, SE = 0.16, p < 0.01); while presence of unpleasantwater was rated lower than no water (p = 0.05). Effect sizes for themain effects were small to moderate (Ferguson, 2009). The inter-action between environment and water type was not significant, F(1, 109) = 1.86, p = 0.16.

Effect of environmental preference/environment type congruence

The primary analysis of interest used a quasi-independentvariable created based on the congruence between environmen-tal preference and environmental type. A between-subjects 4× 3 (environmental preference/environment type congruence ×water type) ANOVA was conducted and post hoc comparisonsused bonferonni adjustment. The interaction between the pri-mary independent variable and water was non-significant (F (6,101) = 0.72, p = 0.64). Fig. 2 illustrated the effect of environmentalpreference/environment type congruence on perceived restora-tion potential (F (3, 101) = 6.32, p = 0.001, �2

p = 0.16). A naturepreference/nature imagery congruence resulted in significantlyhigher perceived restoration potential ratings than an urban prefer-ence/urban imagery congruence (p = 0.02). The difference betweenthe congruent nature preference/nature imagery and incongruenturban preference/nature imagery conditions was non-significant(p = 0.72). Essentially, exposure to nature images, irrespective oflocation, resulted in the highest expectations of restoration. Per-

sons with an urban preference also viewed the congruent urbanand incongruent nature images as equal in restorative potential(p = 0.63). There was, however, a difference in restoration poten-tial ratings between incongruent preference/imagery conditions

166 S. Wilkie, A. Stavridou / Urban Forestry & Urban Greening 12 (2013) 163– 170

F ure wU

(pu

D

rirt

ig. 1. Examples of environmental stimuli. (A) Nature with pleasant water. (B) Natrban with unpleasant water. (F) Urban with no water.

p = 0.01); those in the nature preference/urban imagery conditionrovided the lowest perceived restoration ratings compared to therban preference/nature imagery condition as predicted.

iscussion

The study aim was to explore if congruence between envi-

onmental preferences, defined here as a representation of placedentity, and environment type influenced judgments of the envi-onment’s restorative potential. Specifically, it was hypothesizedhe congruence between environmental preference and the type

ith unpleasant water. (C) Nature with no water. (D) Urban with pleasant water. (E)

of environment influenced perceptions of the likelihood for recov-ery from directed attention fatigue (DAF). Congruent environmentswere expected to be perceived as most restorative, irrespective ofwhether they were nature or urban ones; and the incongruencebetween a nature preference/exposure to urban locations to resultin the lowest ratings.

The expected effect of environmental preference/environment

type congruence was only partially supported since congru-ence did not always result in higher perceived restorationpotential. Those in the congruent nature preference/nature envi-ronment imagery condition anticipated the greatest likelihood for

S. Wilkie, A. Stavridou / Urban Forestry & Urban Greening 12 (2013) 163– 170 167

F ment

a

rthgiuwppwners

iaotNtign

M

acpvswg

ig. 2. Mean perceived restoration potential by environmental preference/environt all’ to ‘completely’.

estoration from similar locations; but this level was higher thanhe corresponding congruent urban condition contrary to the firstypothesis. Perceived likelihood for restoration ratings in the con-ruent nature condition also did not differ significantly from thencongruent nature condition, which was a mismatch between anrban preference and nature environment, thus indicating natureas considered the more restorative location generally. However,eople exposed to images congruent to their urban preferenceerceived similar locations as equal in restoration potential to thoseith the same urban preference exposed to incongruent images ofature settings. As the second hypothesis predicted, the incongru-nce between a preference for nature and exposure to urban imagesesulted in the lowest perceived restoration potential ratings forimilar urban locations.

Although not a specific study aim, the presence of water didmpact judgments of restoration potential. Locations with pleas-nt waterscapes were considered potentially more restorative thannes containing negative imagery or those without water. In addi-ion, there was a significant main effect of environment type.ature settings were judged as being potentially more restorative

han urban settings; however, it is important to note the urban sett-ngs which include images of both general urban vistas and urbanreen spaces were also rated as potentially restorative rather thanon-restorative; but just not to the same extent as nature locations.

ethodological considerations

Several study limitations related to the sample, measurement,nd general design which should be addressed. One methodologi-al issue was that the sample was comprised of students and wasrimarily female in composition. The use of a quasi-independent

ariable for environmental preference also resulted in unequalample sizes for this variable, with lower representation of thoseho preferred nature in the study. Together, these issues limit the

eneralizability of the findings.

type congruence. Restoration Potential was rated on a seven-point scale from ‘not

Two measurement issues also merit discussion. Our methodof capturing environmental preference, considered to be repre-sentative of the individual’s place identity, differed from moststudies focused on preference in the attention restoration theory(ART) literature. Some might criticize our operational definition ofenvironmental preference because it was a forced choice optionrather than rating the degree of preference using a Likert scale;but our dichotomous method of categorization has previouslydifferentiated people amongst several place identity processes,particularly when associated with preference-congruent locations(Knez, 2005). Yet, we acknowledge there are still potential limi-tations with a forced-choice item as a measure of environmentalpreference more generally. Subsequent studies should confirm thechoice between being a ‘country person’ or a ‘city person’ does mea-sure environmental preference and that it represents place-relatedidentity by establishing convergent validity with other identitymeasures. This may be accomplished by determining if city andcountry persons differ on existing measures of place identity (e.g.,Drosletis and Vignoles, 2010) and in the level of emotional connec-tion to with nature (e.g., Perrin and Benassi, 2009). Studies couldalso confirm these categorizations represent environmental pref-erence by asking the respondent where they prefer to spend theirtime.

Another measurement issue relates to use of perceived likeli-hood for restoration using a Likert scale of general restorativeness(SRPRS, Hartig et al., 1996, 1997). This scale, although widely used,has been criticized for its inability to consistently represent thefour aspects of restoration predicted by attention restoration theory(Laumann et al., 2001). Although a valid concern in studies test-ing the theoretical predictions of ART, we believe the measure wassuited to the aim of the current study which was to test how pref-erence and environment interacted to influence perceptions of an

environment’s restoration potential. However, the perceived like-lihood for restoration did rely on self-report ordinal data which isnot without limitations. Since this study focused on subjective per-ceptions of an environment’s suitability for restoration, we believe

1 try & U

oaadaa

mitro(2epianlAvMhgte

oudstotis2a

Ece

eDreiTaetefmspmma

bpT

68 S. Wilkie, A. Stavridou / Urban Fores

rdinal data was the most suitable to the research aim. We dolso recognize that anticipation of restoration is not the same aschieving recovery from DAF. Future studies should determine ififferences of perceived likelihood for restoration translate intoctual differences in restoration outcomes (e.g., improved directedttention capacity) in preferred environments.

The use of images rather than experience in the environ-ent may be considered a weakness based on the argument that

mmersion within the environment, referred to extent in atten-ion restoration theory (Kaplan, 1995), is important to successfulestoration. We counter this concern in two ways. First, a numberf studies indicated imagery can have similar restorative benefitsBerman et al., 2008; Berto et al., 2010; Kjellgren and Buhrkall,010) and simple visualization of an environment also had thisffect (Ryan et al., 2010). We also felt that imagery was more appro-riate to understanding judgments of restoration potential as an

nfluence on making the decision about whether to engage withn environment or not. If the respondents had been immersed inature or urban contexts, estimation of restoration potential and

ocation choice become irrelevant since participants are in situ.nother critique relates to specific image content, which was notaried as in other studies (e.g., Herzog, 1985; Berto et al., 2010).anipulation of qualities such as coherence or fascination would

ave introduced a further layer of complexity outside the scope ofeneral preference; and images were selected which were likelyo be encountered on a typical visit to the location to maximizecological validity.

Other studies also controlled for time of the day and day of weekf participation (Hartig and Staats, 2006; Martens et al., 2011); butnfortunately, this information was not recorded. However, ran-om assignment to experimental conditions at varying times/dayshould have precluded this from being a confounding factor. Finally,his was an independent subjects design without baseline measuresf DAF fatigue because such fatigue was presumed based on the par-icipant situation. Participants were students who were mid-waynto a second term going about their normal academic routines sohould have been experiencing cognitive fatigue (Hartig and Staats,006) but this should be explicitly measured in future studies using

repeated measures design.

nvironmental preference and environment type congruence as aompliment to attention restoration theory’s account ofnvironmental preference

This study originated from two perceived limitations of ART’sxplanation of our environmental preference for nature in times ofAF. The first limitation related to the dominant view of attention

estoration theorists that environmental preference is a functional,volutionary response to nature; and preference results from thendividual’s need for restoration of directed attention resources.he second limitation was that environmental preference haslmost exclusively been defined as an aesthetic judgment of annvironment. We acknowledge the considerable body of evidenceo support both assertions; however, our concern was the pref-rence/restoration literature rarely, if ever, considered individualactors beyond the experience of directed attention fatigue which

ay also contribute to judgments of these environments (and sub-equent location choice). To counter these limitations, this studyrovided evidence environmental preference does influence judg-ents of restoration potential and that environmental preferenceay represent something more than ‘liking’ a location or finding it

ttractive.

The findings indicated that, in some cases, the congruence

etween environmental preference and type influenced judgmentserceived likelihood for restoration from directed attention fatigue.his congruence was particularly important to nature lovers who

rban Greening 12 (2013) 163– 170

anticipated the best restoration in nature and the least in urbansettings, while those with an urban preference were able to seeequivalent restoration potential in either type of environment.Additionally, nature was considered as the most likely locationto result in cognitive restoration irrespective of preference. Over-all, the results support earlier studies which found nature wasperceived as more restorative than urban locations (Herzog et al.,1997; Laumann et al., 2001; Hartig and Staats, 2006); and withART’s assertion that nature is superior in aid of the restoration goals(Kaplan, 1995; Berman et al., 2008). However, ART would also havepredicted that everyone exposed to urban imagery would havelower expectation for restoration, which was not the case. Thosewith an urban place-related identity (i.e., city persons) viewedboth urban and nature settings as equivalent in their ability toaid recovery from DAF; and, on average, urban locations werejudged as being ‘rather’ likely to result in restoration instead ofnon-restorative which would be expected by ART. This suggeststhat the urban environment is perceived as possessing similarbenefits to nature when urban environments are generally pre-ferred; and supports Karmanov and Hamel’s (2008) findings thatwell-designed urban environments can rival nature for restorationoutcomes.

The results also suggested environmental preference can bedefined in terms of place identity; and preference defined in thisway does have an effect on expectations for attention restora-tion. In doing so, the study outcomes supported ART’s premiseof compatibility, or congruence between location and goal, asimportant to restoration (Kaplan, 1995; Scopelliti and Giluliani,2004). This study indicated identity-compatible locations couldalso have an influence; and in some cases, locations incompati-ble with this preference may preclude restoration because peopleare less likely to see the environment’s restoration potential. Thisis likely a particular component of place identity known as place-congruent continuity (Knez, 2005; Drosletis and Vignoles, 2010),which refers to the locations suitability to the individual’s beliefs.Identity-related compatibility could explain the curious choice ofan environment inconsistent with restoration goals (Herzog et al.,2003). It may have been less-than-suitable location choices forrestoration goals were actually based on reasoned decisions by per-sons with a strong urban-related place identity who anticipated thesame benefit in their preferred urban setting. However, the prefer-ence/environment congruence explanation we proposed here forthis restoration goal/location choice inconsistency is only specula-tive and should be tested in future studies. One way to do confirmthis explanation would be to assess the relative importance ofdirected attention recovery goals compared to personal environ-mental preference when selecting locations for the recovery fromDAF.

Applications for urban design

The information presented may be useful to those responsiblefor urban green space management or health initiatives in urbanlocations, particularly when presented with the challenge of a lackof uptake with local facilities. Korpela and Ylén (2009) establishedfavorite place prescriptions are beneficial to mood and cognitiverestoration and the findings here illustrate a potential means forengaging those with a strong urban-related place identity withurban green spaces. Nisbit and Zelenski (2011) noted people under-estimate nature’s benefit; it also appears persons with a naturepreference may underestimate the benefit of urban environments.Therefore, programs could also be targeted to those with a strong

nature-related identity but residing in urban settings to remindthem of the equivalent restorative benefits to be found in their‘non-preferred’ urban setting; and highlight to people with a naturepreference that they should consider their own expectation of a lack

try & U

os

gcisdof

C

bm2Susatotbrpomiifime

R

A

B

B

B

B

BC

D

D

F

G

G

G

H

H

S. Wilkie, A. Stavridou / Urban Fores

f restoration may be precluding positive outcomes in urban greenpaces.

Another important finding relevant to the development of urbanreen spaces was the clear influence of varying water types on per-eptions of restoration likelihood. White et al. (2010) speculatedt is not simply the presence of water that has this influence. Thistudy confirmed views of water sources which are neglected willetract from user perceptions. This could result in a lack of usagef the space more generally; therefore, it is important that sucheatures are continually maintained.

onclusion

Despite its limitations, this study contributed to the growingody of literature endorsing the potential benefits of urban environ-ents for directed attention restoration (Hernández and Hidalgo,

005; Sanesi et al., 2006; Karmanov and Hamel, 2008; Grahn andtigsdotter, 2010; Stigsdotter and Grahn, 2011). Persons with anrban preference certainly perceived this benefit in the currenttudy; and the findings supported earlier work which found urbannd natural environments were equivalent for affective restora-ion (Karmanov and Hamel, 2008). By addressing two limitationsf the attention restoration account of environmental preference,he study also provided an argument for a complimentary approachetween the predominant evolutionary view of restorative envi-onments (Joye and van den Berg, 2011) and one based in theerson–place literature of environmental psychology. Explorationf the influences of place identity and concepts such place attach-ent and place dependence in conjunction with factors established

n ART should provide further insight on location choice when seek-ng restoration from a number of life stressors. Finally, we hope thendings were informative for those involved with the design andanagement of urban green spaces and used to facilitate increased

ngagement with health initiatives in these locations.

eferences

devi, A., Grahn, P., 2011. Attachment to certain natural environments: a basis forchoice of recreational settings, activities, and restoration from stress. Environ-ment and Natural Resources Research 1, 36–52.

erman, M., Jonides, J., Kaplan, S., 2008. The cognitive benefits for interacting withnature. Psychological Science 19, 1207–1212.

erman, M., Kross, E., Krpan, K., Askren, M., Burson, A., Deldin, P., Kaplan, S.,Sherdell, L., Gotlib, I., Jonides, J., 2012. Interacting with nature improves cog-nition and affect for individuals with depression. Journal of Affective Disorders140, 300–305.

erto, R., 2005. Exposure to restorative environments helps restore attentionalcapacity. Journal of Environmental Psychology 25, 249–259.

erto, R., Baroni, M., Zainaghi, A., Bettella, S., 2010. An exploratory study of theeffect of high and low fascination environments on attentional fatigue. Journalof Environmental Psychology 30, 494–500.

ritish Psychological Society, 2010. Code of human research ethics. Leicester, Author.ohen, S., Spacapan, S., 1978. The aftereffects of stress: an attentional interpretation.

Environmental Psychology and Nonverbal Behavior 3, 43–57.river, J., 2001. A selective review of selective attention research from the past

century. British Journal of Psychology 92, 53–78.rosletis, O., Vignoles, V., 2010. Towards an integrative model of place identification:

dimensionality and predictors of intrapersonal-levels place preferences. Journalof Environmental Psychology 30, 23–34.

erguson, C., 2009. An effect size primer: a guide for clinicians and researchers.Professional Psychology: Research and Practice 40, 532–538.

alindo, M., Rodriguez, C., 2000. Environmental aesthetics and psychological well-being: relationships between preference judgments for urban landscapes andother relevant affective responses. Psychology in Spain 2000 (4), 13–27.

rahn, P., Stigsdotter, U., 2003. Landscape planning and stress. Urban Forestry andUrban Greening 2, 1–18.

rahn, P., Stigsdotter, U., 2010. The relation between perceived sensory dimensionsof urban green space and stress restoration. Landscape and Urban Planning 94,264–275.

an, K., 2003. A reliable and valid self-rating measure of the restorative quality ofnatural environments. Landscape and Urban Planning 64, 209–232.

an, K., 2010. An exploration of relationships among responses to natural scenes:scenic beauty, preference, and restoration. Environment and Behavior 42,243–270.

rban Greening 12 (2013) 163– 170 169

Hartig, T., Evans, G., Jamner, L., Davis, D., Garling, T., 2003. Tracking restorationin natural and urban field settings. Journal of Environmental Psychology 23,109–123.

Hartig, T., Korpela, K., Evans, G., Gärling, T., 1996. Validation of a measure of perceivedenvironmental restorativeness. Göteburg Psychological Reports 26, 1–64.

Hartig, T., Mang, M., Evans, G., 1991. Restorative effects of natural environmentexperiences. Environment and Behavior 23, 3–26.

Hartig, T., Korpela, K., Evans, G., Gärling, T., 1997. A measure of restorative qualityin environments. Scandanavian Housing and Planning Research 14, 175–194.

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

Hernández, B., Hidalgo, C., 2005. Effect of urban vegetation on psychological restora-tiveness. Psychological Reports 96, 1025–1028.

Herzog, T., 1985. A cognitive analysis of preference for waterscapes. Journal of Envi-ronmental Psychology 5, 225–241.

Herzog, T., Black, A., Fountaine, K., Knotts, D., 1997. Reflection and attentional recov-ery as distinctive benefits of restorative environments. Journal of EnvironmentalPsychology 17, 165–176.

Herzog, T., Chen, H., Primeau, J., 2002. Perception of the restorative potential ofnatural and other settings. Journal of Environmental Psychology 22, 295–306.

Herzog, T., Maguire, C., Nebel, M., 2003. Assessing the restorative components ofenvironments. Journal of Environmental Psychology 23, 159–170.

Hipp, J., Ogunseitan, O., 2011. Effect of environmental conditions on perceived psy-chological restorativeness of coastal parks. Journal of Environmental Psychology31, 421–429.

Joye, Y., van den Berg, A., 2011. Is love for green in our genes? A critical analysis ofevolutionary assumptions in restorative environments research. Urban Forestryand Urban Greening 10, 261–268.

Kaplan, R., 1977. Patterns of environmental preference. Environment and Behavior9, 195–216.

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

Kaplan, S., Berman, M., 2010. Directed attention as a common resource for executivefunctioning and self-regulation. Perspectives on Psychological Science 5, 43–57.

Karmanov, D., Hamel, R., 2008. Assessing the restorative potential of contemporaryurban environment(s): beyond the nature versus urban dichotomy. Landscapeand Urban Planning 86, 115–125.

Kjellgren, A., Buhrkall, H., 2010. A comparison of the restorative effect of a naturalenvironment with that of a simulated natural environment. Journal of Environ-mental Psychology 30, 464–472.

Knez, I., 2005. Attachment and identity as related to place and its perceived climate.Journal of Environmental Psychology 25, 207–218.

Korpela, K., Ylén, M., 2009. Effectiveness of favorite-place prescriptions: a fieldexperiment. American Journal of Preventive Medicine 36, 435–438.

Kyle, G., Graefe, A., Manning, R., Bacon, J., 2004a. Effect of activity involvement andplace attachment on recreationists’ perceptions of setting density. Journal ofLeisure Research 36, 209–231.

Kyle, G., Graefe, A., Manning, R., Bacon, J., 2004b. Effects of place attachment on users’perceptions of social and environmental conditions in a natural setting. Journalof Environmental Psychology 24, 213–225.

Kyle, G., Mowen, A., Tarrant, M., 2004. Linking place preferences with place mean-ing: an examination of the relationship between place motivation and placeattachment. Journal of Environmental Psychology 24, 439–454.

Lalli, M., 1992. Urban-related identity: theory, measurement, and empirical findings.Journal of Environmental Psychology 12, 285–303.

Laumann, K., Garling, T., Stormark, K.M., 2001. Rating scale measures of restor-ative components of environments. Journal of Environmental Psychology 21,31–44.

Martens, D., Gutscher, H., Bauer, N., 2011. Walking in ‘wild’ and ‘tended’ urbanforests: the impact on psychological well-being. Journal of Environmental Psy-chology 31, 36–44.

Nisbit, E., Zelenski, J., 2011. Underestimating nearby nature: affective forecast-ing errors obscure the happy path to sustainability. Psychological Science 22,1101–1106.

Pals, R., Steg, L., Siero, F., van der Zee, K., 2009. Development of the PRCQ: a measureof the perceived restorativeness of zoo attractions. Journal of EnvironmentalPsychology, 441–449.

Peron, E., Berto, R., Purcell, T., 2002. Restorativeness, preference, and theperceived naturalness of places. Medio Ambiente y Comporamiento Humano 3,19–34.

Perrin, J., Benassi, V., 2009. The connectedness to nature scale: a measure of emo-tional connection to nature? Journal of Environmental Psychology 29, 434–440.

Proshansky, H., 1978. The city and self-identity. Environment and Behavior 10,147–169.

Ryan, R., Weinstein, N., Bernstein, J., Brown, K., Mistretta, L., Gagné, M., 2010. Vitaliz-ing effects of being outdoors and in nature. Journal of Environmental Psychology30, 159–168.

Sanesi, G., Lafortezza, R., Bonnes, M., Carrus, C., 2006. Comparison of two differ-ent approaches for assessing the psychological and social dimensions of greenspaces. Urban Forestry and Urban Greening 5, 121–129.

Scannell, L., Gifford, R., 2010. Defining place attachment: a tripartite organizingframework. Journal of Environmental Psychology 30, 1–10.

Scopelliti, M., Giluliani, M., 2004. Choosing restorative environments across the life-span: a matter of place experience. Journal of Environmental Psychology 24,423–437.

1 try & U

S

S

U

v

89–98.

70 S. Wilkie, A. Stavridou / Urban Fores

taats, H., Kievet, A., Hartig, T., 2003. Where to recover from attentional fatigue: anexpectancy-value analysis of environmental preference. Journal of Environmen-tal Psychology 23, 147–157.

tigsdotter, U., Grahn, P., 2011. Stressed individuals’ preference for activities andenvironmental characteristics in green spaces. Urban Forestry and Urban Green-ing 10, 295–304.

lrich, R., 1979. Visual landscapes and psychological well-being. Landscape Research4, 17–23.

an den Berg, A.E., Koole, S., Van der Wulp, N., 2003. Environmental preference andrestoration: (How) are they related? Journal of Environmental Psychology 23,135–146.

rban Greening 12 (2013) 163– 170

Van der Linden, D., Frese, M., Meijman, T., 2003. Mental fatigue and the control ofcognitive processes: effects on perseveration and planning. Acta Psychologia113, 45–65.

White, E., Gatersleben, B., 2011. Greenery on residential buildings: does it affectpreferences and perceptions of beauty? Journal of Environmental Psychology,

White, M., Smith, A., Humphryes, K., Pahl, S., Snelling, D., Depledge, M., 2010. Bluespace: the importance of water for preference, affect and restorativeness rat-ings of natural and built scenes. Journal of Environmental Psychology 30 (4),482–493.