ARTICLE IN PRESS

Journal of Environmental Psychology 28 (2008) 109120 www.elsevier.com/locate/jep

Engaging with the natural environment: The role of affective connection and identity
Joe Hinds, Paul Sparks
Department of Psychology, University of Sussex, Falmer, East Sussex BN1 7QH, UK Available online 17 November 2007

Abstract Research has shown pro-environmental behaviour to be positively associated with the strength of emotional connection towards the natural environment. The present study (N 199) investigated the predictive utility of an extended model of the theory of planned behaviour (TPB) [Ajzen, I. (1991). The theory of planned behaviour. Organizational Behaviour and Human Decision Processes, 50, 179211] for peoples intentions to engage with the natural environment. It was hypothesised that both affective connection and identication with the natural environment would contribute signicantly to the prediction of peoples intentions. A secondary hypothesis was that participants who had grown up in rural areas would report more positive orientations towards engaging with the natural environment than would urban participants. The research found that affective connection was a signicant independent predictor of intentions to engage with the natural environment. Environmental identity was only a signicant predictor in the absence of affective connection in the regression model. As predicted, rural and urban participants differed signicantly along the measured variables. r 2007 Elsevier Ltd. All rights reserved.
Keywords: Natural environment; Affective connection; Attitudes; Identity; Experience

1. Introduction Modern lifestyles have, for much of the developed world, created psychological and physical divisions between human inhabitants and the natural world. Since the 1850s, the majority of Britains population have made their work and home in towns and cities (Thomas, 1983). It has been argued that the result of this trend has been for many people to no longer experience the natural world directly but rather indirectly or vicariously (Kellert, 2002; Schultz, 2002). Reduced direct contact with the natural world has been labelled the extinction of experience by Pyle (1978), which, he claims, leads to a cycle of apathy and a lack of concern with ecological issues, the natural environment and the wildlife within it. A mitigation of this modern trend, however, could have positive outcomes for the environment: Experiences in the natural environment have been found to have signicant correlations with proenvironmental behaviour, such as recycling, signing petiCorresponding author. Tel.: +44 012 7367 8059.

E-mail address: J.hinds@sussex.ac.uk (J. Hinds). 0272-4944/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.jenvp.2007.11.001

tions in favour of environmental protection and using public transport (Finger, 1994). Nord, Luloff, and Bridger (1998) found strong correlations between frequency of visits to forest areas and self-reported pro-environmental behaviours such as contributing money to environmental organisations and environmentally conscious consumerism (see also Teisl & OBrien, 2003). Similarly, intimate contact with the natural world, especially during childhood, has been suggested to be essential in forming meaningful bonds with, and promoting positive values towards, the natural environment (Chawla, 2002; Horwitz, 1996; Kellert, 2002). For instance, Bunting and Cousins (1985) found that the inclination to positively respond to nature was signicantly stronger in rural children than for their urban counterparts. Their ndings also revealed that childrens self-reported activity preference differed between these groups: Children higher in what they term pastoralism were more likely to undertake activities such as hiking, camping and taking care of animals, whilst children scoring higher on urbanism were signicantly associated with just one activity: Watching television.

ARTICLE IN PRESS
110 J. Hinds, P. Sparks / Journal of Environmental Psychology 28 (2008) 109120

This general pattern of greater experience with the natural environment leading to more pro-environmental attitudes is compatible with the view that when people have direct experiences of an object, their evaluations of that object tend to be more affectively based than they are for those people who have only indirect experience (Millar & Millar, 1996). There is also evidence that direct experience of an attitude object facilitates stronger attitudebehaviour consistency (Fazio & Zanna, 1981). Repeated exposure to an attitude object may also be instrumental in the growth of positive affective connections with that object (cf. Zajonc, 2001, on the mere exposure effect). As well as promoting more positive environmental attitudes and behaviour, experience of the natural environment has been reported to have various health and wellbeing benets (Kaplan & Talbot, 1983). For example, lower rates of sick-calls for prisoners with greater access to natural environment views have been reported (Moore, 1981) as has an increase in recovery rates for hospital patients with windows facing hospital gardens (Ulrich, 1984). Positive social and cognitive outcomes may also accrue as a result of contact with natural places: Improved cognitive functioning has been reported for children who have moved from urban environments to environments considered to be characterised by higher levels of naturalness (Faber Taylor, Kuo, & Sullivan, 2002; Wells, 2000). Moreover, regarding the natural environment as a place of leisure, and using it for restorative or respite experiences, can create a time for self-reexivity (Kaplan & Kaplan, 1989; Kaplan, 1995). This, it has been suggested, may lead to positive change in psychological well-being (Herzog, Black, Fountaine, & Knotts, 1997). 1.1. Affective connection The importance of affect in the context of human relationships with the natural environment has been proposed by several commentators (Kals & Maes, 2002; Kals, Schumaker, & Montada, 1999; Kaplan & Kaplan, 1989). Empirical research has found, for instance, that engendering greater empathy towards nature tends to increase the level of connectedness people feel towards it (Schultz, 2000). It has also been shown that affect can be both an important predictor of environmental attitudes and rated as more important by participants, relative to cognitions, in forming their attitudes to environmental issues such as logging native forests (Pooley & OConner, 2000). Similarly, Kals et al. (1999) have demonstrated that an emotional afnity with nature is able to predict nature protective behaviour, such as public commitments to environmental organisations and the use of public transport. They also showed emotional afnity to be predicted by past and present exposure to the natural environment. These empirical data have helped support Wilsons (1984, 1993) biophilia hypothesis, which has suggested that, as a species, we have an inherent afliation to the natural environment. Wilson posited that the natural world

continues to inuence the human condition through our previous close and enduring evolutionary relationship with it. Essentially, the argument is that our technological development has been so rapid that our evolutionary adaptation to modern environments has yet to develop substantially. Therefore, according to Wilson (1993), there is still a need to be with nature: we have an y innately emotional afliationyto other living organisms (p. 31). There are, however, some advocates of the biophilia hypothesis who have suggested that the genetic bond may well be a weak one, requiring the addition of learning, culture and experience of nature to optimise biophilic tendencies (Kellert, 2002; see also Kahn, 1997). Therefore, it might be expected that affective connection would play an important part in predicting intentions to engage with the natural environment. Moreover, it may also be expected that those people with greater experience of the natural environment may express greater affective connections with it than with those with lesser experience. 1.2. Environmental identity As well as eliciting emotional bonds, experiences of the natural environment may also foster place-identity (Manzo, 2003; Proshansky, 1978). Proshansky (1978) denes placeidentity as the dimensions of self that dene the individuals personal identity in relation to the physical lonneau (2004) has even coined environment (p. 155). Fe the term topological identity (p. 45) to refer to the degree to which one feels an emotional connection with a place and the people associated with it (see also Lalli, 1992). The importance an identication with, or sense of connection to, the environment has only recently been recognised, broadening the mainstream concept of identity formation to include, for example, how people see themselves in relation to the natural world (Clayton & Opotow, 2003). For instance, environmental identity, dened as the meanings that one attributes to the self as they relate to the environment has been found to have both a direct effect on environmental behaviour and an indirect effect through environmental attitudes (Stets & Biga, 2003, p. 406). Furthermore, these researchers have found that the stronger the environmental identity, the more positive the attitudes towards the environment. Similarly, a high correlation has been reported between a measure of environmental identity and self-reported environmental behaviours, such as energy efciency (Clayton, 2003) and recycling (Mannetti, Pierro, & Livi, 2004). Therefore, taken together, affective connection and environmental identity are potentially important explanatory concepts within environmental psychology research. Exposure to the natural environment may facilitate the development of emotional bonds and identication with it, which may in turn lead to positive psychological well-being and to the formation of positive attitudes and behaviours towards the natural environment.

ARTICLE IN PRESS
J. Hinds, P. Sparks / Journal of Environmental Psychology 28 (2008) 109120 111

1.3. The theory of planned behaviour The theory of planned behaviour (TPB) proposes that people act or behave in accordance with their intentions towards particular behaviours providing they perceive some degree of control over implementing those behaviours (Ajzen, 1991). Intentions, seen as proximal predictors of behaviour, are in turn dependent upon three separate factors: attitudes towards the behaviour, subjective norms and perceived behavioural control (PBC) (Ajzen, 2001). Thus, intentions to behave in a particular way may be predicted by strength of attitude towards that behaviour, the extent to which the behaviour is perceived to be compatible with perceived social approval and the degree to which the behaviour is perceived to lie within ones personal control. Within the TPB, attitudes towards a given behaviour are said to be determined by salient beliefs about behavioural outcomes (behavioural beliefs) and the evaluations of those behavioural outcomes (outcome evaluations). An estimate of attitude is obtained by multiplying each behavioural belief with the corresponding outcome evaluation and summing the resulting products (Ajzen, 1988). It is also claimed that when behavioural beliefs and outcome evaluations are taken together they may represent an indirect measure of attitude that has been shown to have high correlations with more direct measures of attitudes (Ajzen, 1988, 2002a). The TPB has been applied to a wide range of behavioural domains (Ajzen, 2002b) including those that relate to environmental issues. For example, it has been used to good effect in predicting adoption of green energy and increased bus use (Bamberg, 2003; Bamberg, Ajzen, & Schmidt, 2003), intentions to recycle (Mannetti et al., 2004; Terry, Hogg, & White, 1999), and buying organic foods (Sparks & Shepherd, 1992). Although the TPB has demonstrated its effectiveness as a model, the inclusion of additional predictors should be acceptable providing they explain additional variance over and above the theorys existing variables (Ajzen, 1991). For example, several researchers have shown that the TPB may benet, in terms of its predictive utility, from the inclusion of variables such as identity and affect (see Conner & Armitage, 1998, for a review), moral obligation, selfidentity, and past behaviour (Eagly & Chaiken, 1993; Manstead & Parker, 1995). The role of attitude measurement (within the TPB framework), in particular in relation to the distinction between affect and cognition within attitudes, has provoked much debate (Breckler & Wiggins, 1989a; Tramow & Sheeran, 1998). It has been proposed that current methods of measuring attitudes within the TPB have a tendency to capture cognitive, rational, evaluations but not affective dimensions of attitudes (Conner & Sparks, 1995; French et al., 2005; Manstead & Parker, 1995; van der Pligt, Zeelenberg, van Dijk, de Vries, & Richard, 1998). For example, Ajzen and Timko (1986) found that attitudes

towards generally recommended health practices, measured with a 20-item semantic differential scale, comprised two distinct factors. The rst was deemed to be reective of participants cognitive evaluations and consisted of bipolar adjective-pairings such as harmfulbenecial and wisefoolish. The second was deemed to be reective of more affective judgements with item pairings such as pleasantunpleasant and interestingboring. Interestingly, there was not a signicant correlation between these two factors, each of which was argued to be measuring discrete aspects of the same attitudes. Moreover, the affective dimension of attitudes was found to be a highly signicant predictor of health behaviour independently of the impact of the more cognitive dimensions (Ajzen & Timko, 1986). Consequently, Ajzen (2002a) has proposed that the direct method for assessing attitudes, namely the semantic differential, should elicit both cognitive and affective aspects of attitudes. Although it has been argued that it is very unlikely that there can ever be pure cognitive and affective elements of attitudes (Eagly, Mladinic, & Otto, 1994), there is a plausible assumption that, for some attitudes towards some behaviours or objects, either affect or cognitive elements may predominate (Tramow & Sheeran, 1998). For example, it has been shown that affective responses rather than cognitive responses have the stronger relationship with global attitudes towards blood donation (Breckler & Wiggins, 1989a). There is also an inuential line of argument in the literature that suggests that attitudes are often essentially affectively based (Wilson & Dunn, 1986; Wilson, Dunn, Kraft, & Lisle, 1989). Therefore, the present study also undertook to discover if the prediction of environmental attitudes could be improved using measures of affective connection in addition to behavioural belief and outcome evaluation product terms, the determinants of attitudes within the TPB (Ajzen, 1988). 1.3.1. Affect and the TPB Affective factors (often assessed in the form of anticipated negative affective reactions) have been demonstrated in several studies to add signicantly to the TPB model (Parker, Stradling, & Manstead, 1996; Richard, de Vries, & van der Pligt, 1998; Richard, van der Pligt, & de Vries, 1995). They also tend to be used when researching topics that are heavily affectively laden (Conner & Sparks, 1995, p. 149) such as leisure activities (Ajzen & Driver, 1991) and safe sex practices (Chan & Fishbein, 1993; Richard et al., 1998). Moreover, Richard et al. (1998) demonstrated that when anticipated affective reactions and attitudinal measurements remain distinct factors in questionnaire protocols, the attitude measure may be made redundant in its predictive capacity, whereas the anticipated affective measure may account for a signicant amount of the variance in behaviour. However, Richard et al. (1998) note that greater use should be made of specic affective reactions when assessing outcome evaluations and behavioural

ARTICLE IN PRESS
112 J. Hinds, P. Sparks / Journal of Environmental Psychology 28 (2008) 109120

beliefs in order to fully appreciate the independence of affective beliefs from more cognitive beliefs. Although this research area tends to be dominated by the assessment of negative affective reactions, some applications with positive affective reactions have been reported (Richard, van der Pligt, & de Vries, 1996). Therefore, because of the relative scarcity of an assessment of affect (positive affect in particular) in the literature, it is important to try and clarify the importance of affective connection, both within a natural-environment research focus and within the methodological framework of the TPB. Drawing on previous research (e.g., Kals et al., 1999; Mayer & Frantz, 2004) we dene affective connection with the natural environment as the subjective experience of an emotional attachment with the natural environment. 1.3.2. Identity and the TPB Self-identity often seems to represent an important addition to the TPB (Sparks & Shepherd, 1992; Terry et al., 1999; see also Charng, Piliavin, & Callero, 1988). For instance, Sparks and Shepherd (1992) found that identifying strongly as a green consumer contributed signicantly and independently to intentions to consume organic vegetables. It has been noted, however, that the effect of identity as a predictor may vary according to the target behaviour in question (Conner & Armitage, 1998). Moreover, there has been some criticism that measures of identity merely act as proxies for past behaviour (Charng et al., 1988; Sparks & Guthrie, 1998; Stets & Biga, 2003) or intentions (Fishbein, 1997). In order to further inform the natural-environment psychological literature regarding the importance of environmental identity (cf. Clayton & Opotow, 2003) a different kind of measure of identication with the natural environment was included in the present study. The aim of this measure was both to avoid items that could be construed as proxy indicators of intentions and behaviours and to represent aspects of identity that are more personal than many of the more role-based forms of identity that are often addressed in this kind of research (cf. Mannetti et al., 2004). 1.4. The present study From the research literature it appears that experience of the natural environment may elicit positive environmental attitudes and behaviours, as well as facilitating positive psychological well-being. Therefore, identifying the antecedents of intentions to engage with the natural environment may be seen as a useful contribution to the literature. It would also appear that environmental identity and affective connection combined with key TPB variables may be important predictors of intentions to engage with the natural environment. On the basis of the above considerations, the present research focussed on three key hypotheses. First, we expected that a sense of affective connection with the

natural environment would be found to be a signicant independent predictor of participants intentions to engage with the natural environment (because such affective experience is not well represented within standard TPB variables). Second, and for similar reasons, we expected that environmental identity would also be found to be a signicant independent predictor of participants intentions. Finally, we expected that participants from rural backgrounds, because of their potential greater exposure to the natural environment, would be distinguished from urban participants by having signicantly more positive ratings for behavioural intentions, attitudes, subjective norm, PBC, identication, and affective connection. 2. Method 2.1. Participants Participants (N 199; female 166, male 33) were undergraduate social science students at the University of Sussex, UK. Participants represented a convenience sample, in that that they agreed to participate in the study in return for course credits as part of the requirements of their degree programme. Their mean age was 21.7 years (range 1853). 2.2. Materials All participants received an identical four page questionnaire concerning Attitudes and the Natural Environment. Following the instructions, a working denition of Engaging with the natural environment was presented as being in and actively participating in areas and settings produced by nature, such as woodland, hills, lakes, valleys, coastal areas, mountains, rivers and forests. The questionnaire assessed the central concepts of the TPB as well as a number of additional measures, some of which are not reported here. All responses were recorded on seven-point Likert-type scales (response scale end points are indicated in parentheses). Reverse coding of variables was carried out where appropriate. 2.2.1. Childhood location Following questions relating to age and gender, participants were asked to indicate the type of location in which they grew up: In which area did you spend most of your childhood? with three possible responses, Urban (n 71), Suburban (n 90), and Rural (n 36). 2.2.2. Behavioural beliefs and outcome evaluations Eight behavioural beliefs about engaging with the natural environment were constructed from an initial belief elicitation pilot study, conducted according to Ajzens (2002a) guidelines, with a sample of undergraduate students (N 30). The most frequently reported responses were included in the main questionnaire. These were: My engaging with the natural environment wouldy allow

ARTICLE IN PRESS
J. Hinds, P. Sparks / Journal of Environmental Psychology 28 (2008) 109120 113

me to experience beautiful scenery, make me feel happy, be too isolating, help me escape the stresses of life, give me a sense of connection with nature, be an uncomfortable experience, be inconvenient for me, and help promote environmental awareness (extremely unlikely to extremely likely). These were followed by the eight corresponding outcome evaluation questions, which simply asked participants to evaluate each of the outcomes mentioned in the behavioural belief items, Please evaluate each of the followingy (extremely bad to extremely good). 2.2.3. Attitudes Attitudes were assessed using a ve-item semantic differential measure. Responses were recorded to the statement, For me, engaging with the natural environment would bey (extremely bad to extremely good; extremely harmful to extremely benecial; extremely foolish to extremely wise; extremely unpleasant to extremely pleasant and extremely unenjoyable to extremely enjoyable), anchored at the end points only. The mean of these items (a 0.88) was used to form a composite measure of attitudes. 2.2.4. Subjective norm Subjective norm was measured with two items: Most people who are important to me probably think that I should engage with the natural environment (strongly disagree to strongly agree) and If I were to engage with the natural environment, most people who are important to me would probablyy (disapprove strongly to approve strongly). The mean of these items (r 0.49) was used to form a composite measure of subjective norm. 2.2.5. Behavioural intentions Behavioural intentions were measured with the three items I shall try to engage with the natural environment within the next two weeks (denitely shall not try and denitely shall try), I shall make an effort to engage with the natural environment in the next two weeks (denitely false and denitely true), and I intend to engage with the natural environment in the next two weeks (strongly disagree to strongly agree). The mean of these items (a 0.93) was used to form a composite measure of intentions. 2.2.6. Perceived behavioural control PBC was measured with the items How much control do you have over whether or not you engage with the natural environment (no control and complete control) and It is mostly up to me whether or not I engage with the natural environment (strongly disagree to strongly agree). The mean of these items (r 0.63) was used to form a composite measure of PBC. 2.2.7. Affective connection Affective connection was measured with the items (adapted from Thompson & Barton, 1994), Sometimes when I am unhappy I nd comfort in nature, It makes me sad to see natural environments destroyed, Being out

in nature is a great stress reducer for me and I need time in nature to be happy (strongly disagree to strongly agree). The mean of these four items (a 0.77) was then used to form a measure of affective connection.1 2.2.8. Environmental identity Identication with the natural environment was measured with the three items I see myself as someone who empathises with the natural environment, For me, engaging with the natural environment gives me a greater sense of who I am, and I identify with the natural environment (strongly disagree to strongly agree). The mean of these items (a 0.75) was used to form a measure of environmental identity. 3. Results Means, standard deviations and inter-correlations for the TPB constructs, behavioural intentions, attitude, subjective norm and PBC, and the added constructs of environmental identity and affective connection (see Table 1), indicate that although some correlations are deemed to be high (Cohen, 1988) there was only one extreme case, namely that between affective connection and identication (r 0.80). However, an examination of the collinearity statistics revealed that each predictor variable fell within the acceptable boundaries of tolerance (40.3) and the VIF coefcient (o10), thus ruling out any substantive multi-collinearity (cf. Field, 2000). Additionally, although the sample was predominantly female, a series of t-tests found no differences between males and females on the extended TPB variables. 3.1. Predicting behavioural intentions A hierarchical regression was carried out of intentions to engage with the natural environment on attitudes (step 1), subjective norms (step 2), PBC (step 3), environmental identity (step 4) and affective connection (step 5). The model was a signicant predictor of intentions to engage in the natural environment, R 0.69, po0.001. Results show that there were signicant percentage change in the variance explained by the inclusion of attitudes, F change 98.91, po0.001; subjective norm, F change 15.65, po0.001; environmental identity, F change 20.00, po 0.001 and affective connection, F change 9.80, po0.002. There was no signicant incremental contribution to the model from PBC (Table 2). Final beta values show signicant independent predictive effects for attitude (b 0.23, p 0.002), subjective norm (b 0.22, p 0.001) and affective connection (b 0.28, p 0.002) but not for PBC (b 0.08, p 0.12) nor
1 It must be noted that the original intention was for these items to measure aspects of ecocentric attitudes (Thompson & Barton, 1994). However, given their clear affective content, we consider them good indicators of affective connection to the natural environment.

ARTICLE IN PRESS
114 J. Hinds, P. Sparks / Journal of Environmental Psychology 28 (2008) 109120 Table 1 Descriptive statistics and inter-correlations for the extended TPB variables (N 199) 1 1. 2. 3. 4. 5. 6. Behavioural intentions Attitude Subjective norm Perceived behavioural control Environmental identity Affective connection
po0.05. po0.01. po0.001.

2 0.58

3 0.45 0.43

4 0.22 0.22 0.12

5 0.56 0.61 0.33 0.16

6 0.61 0.62 0.37 0.17 0.80

M 4.86 5.74 5.18 5.40 5.04 5.51

SD 1.48 0.84 1.03 1.11 1.10 1.00

Table 2 Hierarchical regressions of intentions to engage with the natural environment (N 199) Step 1. 2. 3. 4. 5.
po0.01. po0.001.

Predictor Attitudes Subjective norm Perceived behavioural control Environmental identity Affective connection

R 0.58 0.62 0.63 0.67 0.69

R2 0.33 0.38 0.39 0.45 0.48

Increment to R2 0.33 0.05 0.01 0.06 0.03

F change 98.91 15.65 2.84 20.00 9.80

Final b 0.23 0.20 0.08 0.12 0.28

environmental identity (b 0.12, p 0.20). Moreover, these results remained largely unaffected after controlling for childhood location. Although able to account for an additional 1.5% of the variance in the model (Fchange 4.03, po0.019), the effect of childhood location on intentions as a independent signicant predictor was marginal (b 0.13, p 0.09).2 Therefore, it appears that the more positive participants attitudes, the greater their intentions to engage with the natural environment. Moreover, affective connection and subjective norm had a positive relationship with behavioural intentions towards engaging with the natural environment: those with a strong affective connection with the natural environment reported being more likely to engage with the natural environment, as did participants who perceived social approval about engaging with it. Thus, the rst hypothesis was supported in that affective connection was found to be a signicant, independent predictor of intentions. However, the second hypothesis regarding the predictive utility of environmental identity was only supported when affective connection was not included in the regression analysis. 3.2. Predicting attitudes Means, standard deviations and inter-correlations for attitudes, behavioural belief and outcome evaluation
Dummy variables were computed using Rural as the baseline measure with this beta weight representing the effect of dummy variable one: the difference between Rural and Urban on intentions.
2

product terms,3 and affective connection, indicated that there were no overly high inter-correlations between predictor variables (see Table 3). As an additional analysis, a hierarchical regression was also carried out of attitudes to engage with the natural environment on all of the beliefs (step 1) and affective connection (step 2). Beliefs and affective connection accounted for 54% of the variance in attitudes. Results show that affective connection (b 0.30, po0.001), the allow me to experience beautiful scenery belief (b 0.15, p 0.027), the be inconvenient for me belief (b 0.13, p 0.023) and the help promote environmental awareness belief (b 0.14, p 0.011) were each signicant independent predictors of attitudes (Table 4). 3.3. The inuence of childhood location A 3 (childhood location: urban, suburban, rural) 6 (extended TPB variables: behavioural intentions, PBC, subjective norm, attitudes, affective connection, and environmental identity) MANOVA carried out to determine if there were differences on the extended TPB variables according to childhood location revealed a signicant overall effect of childhood location, l largest 0.11, F(6, 191) 3.43, p 0.004. Univariate one-way ANOVAs are reported in Table 5 for childhood location for each of the variables. As the prediction for this comparison is directional, the given p values have been corrected by a factor of 2 to
3 For ease of exposition, behavioural belief and outcome evaluation product terms will be referred to as beliefs for the remainder of the paper.

ARTICLE IN PRESS
J. Hinds, P. Sparks / Journal of Environmental Psychology 28 (2008) 109120 Table 3 Inter-correlations and descriptive statistics for attitudes, beliefs and affective connection (N 199) 1 1. Attitudes 2. Beautiful scenery 3. Make me feel happy 4. Be too isolating 5. Help me escape the stresses of life 6. Sense of connection with nature 7. Be an uncomfortable experience 8. Be inconvenient 9. Promote environmental awareness 10. Affective Connection
po0.05. po0.01. po0.001.

115

2 0.46

3 0.57 0.51

4 0.06 0.09 0.14

5 0.50 0.46 0.48 0.08

6 0.51 0.45 0.50 0.01 0.49

7 0.24 0.26 0.31 0.31 0.12 0.15

8 0.30 0.11 0.30 0.24 0.16 0.09 0.30

9 0.35 0.25 0.19 0.15 0.28 0.37 0.08 0.01

10 0.62 0.44 0.58 0.08 0.50 0.57 0.08 0.21 0.31

M 5.74 5.42 5.79 0.85 4.26 3.08 3.98 1.71 1.82 5.51

SD .84 3.12 2.73 3.69 3.57 3.23 3.45 3.34 3.35 1.00

Table 4 Multiple regression of attitudes towards engaging with the natural environment on beliefs (step 1) and affective connection (step 2) (N 199) Step 1. Predictors Beautiful scenery Make me feel happy Be too isolating Help me escape the stresses of life Sense of connection with nature Uncomfortable experience Be inconvenient Promote environmental awareness Affective connection
po0.05. po0.001.

R 0.70

R2 0.49

Increment to R2 0.49

F change 22.52

Final b 0.15 0.13 0.04 0.11 0.07 0.08 0.13 0.14 0.30

2.

0.73

0.54

0.04

17.44

give one-tailed values (cf. Field, 2000). Levenes test for homogeneity is assumed unless otherwise stated. There were signicant effects for behavioural intentions, F(2,195) 5.51, p 0.005, for PBC, F(2,195) 5.32, p 0.0064 and, marginally, for subjective norm, F(2,195) 2.46, p 0.088, environmental identity, F(2,195) 2.77, p 0.065 and affective connection, F(2,195) 2.41, p 0.093. There was no main effect for attitudes, F(2,195) 2.02, p 0.136. However, the contrast between urban and rural participants revealed that for all variables rural participants gave signicantly higher ratings than did urban participants (see Table 5). 4. Discussion The present ndings concur with natural-environment focused research regarding the importance of the inclusion of measures of affect (Kals & Maes, 2002; Kals et al., 1999; Pooley & OConner, 2000). Although Kals et al. (1999) suggest that past and present experience of the natural environment predicts positive emotional afnity with it, in
4

Equal variances not assumed.

the present study, this observation is extended by indicating that the more one has an affective connection with the natural environment, the greater ones intentions to engage with it. Therefore, the present ndings support Kals and Maes (2002) and Kals and colleagues (1999) work regarding the importance of affective connection in natural-environment issues. The present ndings are also congruent with the research literature that addresses the importance of the development of meaningful bonds with the natural environment during childhood (Bunting & Cousins, 1985; Chawla, 2002; Horwitz, 1996; Kellert, 2002). The hypothesis that participants from a rural childhood would differ signicantly from urban participants in terms of the research variables was clearly supported. Participants from rural childhoods reported more positive affective connections, stronger identication, stronger behavioural intentions, more positive attitudes, more acceptable subjective norm, and greater PBC about engaging with the natural environment than did participants with urban childhoods. Thus, the ndings also support both the Kals et al. (1999) argument that past experience of the natural environment has an important role to play in the formation of positive affective relationships

ARTICLE IN PRESS
116 J. Hinds, P. Sparks / Journal of Environmental Psychology 28 (2008) 109120 Table 5 Means and standard deviations for the extended TPB variables by childhood location showing contrasts signicant levels with Rural as the referent group Rural (N 36) Behavioural intentions M 5.36 (SD) (1.35) Attitudes M 5.99 (SD) (0.78) Subjective norm M 5.51 (SD) (0.95) Urbana (N 71) 4.48 (1.39) 5.67 (0.88) 5.11 (0.99) Suburbanb (N 90)

4.98 (1.49) 5.68 (0.80) 5.08 (1.07) 5.47 (1.11) 5.02 (1.03) 5.45 (1.03)

Perceived behavioural control M 5.82 5.12 (SD) (0.65) (1.20) Environmental identity M 5.37 (SD) (1.09) Affective connection M 5.83 (SD) (0.90)
a b

4.85 (1.15) 5.41 (0.99)

Signicance levels refer to Contrast 1 (Rural vs. Urban) 1 tailed. Signicance levels refer to Contrast 2 (Rural vs. Suburban) 2 tailed. po0.06. po0.05. po0.01. po0.001.

with it and more theoretical work on the role of direct and repeated exposure to an attitude object and affect (Millar & Millar, 1996; Zajonc, 2001). Moreover, the results are informative regarding the utility of the TPB as a theoretical framework within a natural environment research focus. The present study revealed that our extended TPB model could account for almost half of the variance in intentions to engage with the natural environment, which is relatively high compared to many previous TPB ndings (see Armitage & Conner, 2001). The study also nds support for a signicant and independent role of affective connection in predicting peoples intentions to engage with the natural environment. The results show that affective connection explains an additional 8% of the variance in intentions to engage with the natural environment, when not only the standard TPB predictors of intentions but also when a measure of environmental identity are included in the regression model. In fact this nding along with the predictive effect of affective connection on attitudes indicates both a direct and an indirect effect of this variable on peoples intentions. There is, therefore, an argument from these ndings for the inclusion of an assessment of additional affective factors within an extended TPB model, thus supporting previous suggestions in this regard (Chan & Fishbein, 1993; Parker et al., 1996; Richard et al., 1998,

1995). The signicant contribution of affective connection to the attitude object, such as the natural environment in the present study, suggests that it should perhaps be considered as an additional predictor for the TPB model, following Ajzens (1991) comments regarding the openness of the model to additional predictor variables. However, the breadth of behaviours for which such an additional variable may be a useful predictor clearly merits empirical exploration, as has been advised by Ajzen and Fishbein (2005). Although the inclusion of affective connection had a signicant independent predictive contribution, the inclusion of environmental identity into an extended TPB model as a predictor of intentions to engage with the natural environment failed to explain any unique variance once affective connection was included. This may be explained by the particularly high correlation between affective connection and environmental identity in the model. As affective connection was a signicant predictor it would seem that elements of environmental identity, as measured here, could have been encapsulated by the affective measure. Certainly, the suggestion that people often identify with what they care about (Frankfurt, 1988) would be entirely consistent with this interpretation. Interestingly, previous researchers who have utilised measures of identity and found them to be signicant predictors in various models have done so without the inclusion of explicit measures of affect (Charng et al., 1988; Mannetti et al., 2004; Sparks & Guthrie, 1998; Sparks & Shepherd, 1992; Stets & Biga, 2003; Terry et al., 1999). Moreover, the framing of the environmental identity items used for the present study may require critical attention. It has been remarked, for instance, that identication with the natural environment lies along a social continuum with some environmental identities evidencing far fewer social associations than others (Clayton & Opotow, 2003). People may form environmental identities that have a strong social component, such as belonging to and actively participating with environmental groups and community environmental programmes. Other forms of environmental identity may reect minimal social inuence or engagement, such as experiencing individual emotional connections to place (Clayton & Opotow, 2003). The measure of identication incorporated in the present study reects a more personal form of identication rather than the kinds of social, or role-based identity which have frequently been used in previous TPB work (however, see Mannetti et al., 2004). It might be suggested therefore that the inclusion of other forms of identity, such as social or role identity, might explain additional variance in the model over and above that explained by the TPB and by affective connection. It is also possible that the strength of an environmental identity may well be different for participants with differing degrees of previous exposure lonneau, 2004; Manzo, to the natural environment (Fe 2003). This was borne out in the current ndings, with

ARTICLE IN PRESS
J. Hinds, P. Sparks / Journal of Environmental Psychology 28 (2008) 109120 117

participants from rural backgrounds having stronger identication with the natural environment than did participants from urban backgrounds. There are some caveats to be considered with regarding the ndings generally. Our sample was predominantly female. Bunting and Cousins (1985) reported that females were more likely to respond favourably to the natural environment than were males, a nding that has received some degree of support elsewhere (Stern, Dietz, & Kalof, 1993; Zelezny, Chua, & Aldrich, 2000). However, as noted by Stern et al. (1993), there is a degree of inconsistency in the literature in this regard. Some studies have found little in the way of gender differences in regard to concern about pollution (Lyons & Breakwell, 1994), and undertaking responsible environmental behaviours (Hines, Hungerford, & Tomera, 1986/1987). However, despite no differences being found between males and females on the extended TPB variables, the present sample was predominately female and there remains the possibility that this asymmetry may have inuenced the pattern of results. The present study also used a rather rudimentary measure of environmental experience, viz. childhood location. A more detailed assessment of environmental experience is warranted in order to make more assured statements about the relationship between experience of the natural environment and positive perceptions about it (see Nord et al., 1998). The degree to which the present ndings might be generalised to a wider population should also be considered. The study sample consisted of undergraduate social science students, not a representative sample of the general population. Therefore, care should be taken not to extrapolate these results beyond our sample until further research can determine the extent to which the present ndings have broader applicability. There may also be some concern regarding the intercorrelations between predictor variables described in the present study. As pointed out by Cohen (1988), correlations between variables exceeding 0.50 in social psychological research tend to indicate high levels of intercorrelation. However, there remains within the TPB framework a large body of research that suggests that correlations of this magnitude are relatively common-place (Bamberg, 2003; Bamberg et al., 2003; Norman, Clarke, & Walker, 2005; Sparks & Guthrie, 1998). In particular, the relationship between attitudes and behavioural intentions is such within the TPB that meta-analyses reveal consistently high inter-correlations (e.g., r 40.50) (Sheeran & Taylor, 1999). Coupled with the multi-collinearity analysis which indicated no overly problematic overlap between variables, the high correlations reported here raise little cause for concern. The current ndings support research that has shown the importance of both beliefs and affect in predicting environmental attitudes (Pooley & OConner, 2000) through the demonstrated contribution that affective connection and beliefs can have in predicting attitudes

towards engaging with the natural environment. According to Ajzen (2002a, 2002b), the summed products of behavioural beliefs and corresponding outcome evaluation should be signicant determinants of attitudes towards behaviours. Although this was indeed the case in the reported research, the addition of affective connection explained a further 10% of the variance in attitudes. These ndings also add weight to the literature on the importance of the study of affect in environmental psychology where attitudes towards environmental issues, it has been argued, are more affectively driven compared to other attitudes (Pooley & OConner, 2000; Schultz, 2000). Moreover, at a more theoretical level, the decit in the predictive utility of beliefs, highlighted here by contribution of affective connection, may be an artefact of the framing of the initial belief elicitation study (French et al., 2005). Participants are asked, based upon the TPB guidelines, to list the advantages and disadvantages of engaging with the natural environment (Ajzen, 2002a). As has been contended by Wilson et al. (1989), asking people to consider their reasons, rather than focussing on feelings towards the target object, may tap more cognitive aspects of attitudes. Eliciting reasons for, or, as in the present study, advantages and disadvantages of engaging with the natural environment, may be especially suboptimal ways of assessing the antecedents of peoples attitudes in such cases (Tramow & Sheeran, 1998; Wilson & Dunn, 1986; Wilson et al., 1989; Zanna & Rempel, 1988). Similarly, Manstead and Parker (1995) suggest that belief-based attitudes towards a behaviour may represent more deliberative reactions compared with more automatic reactions associated with direct measures of attitudes (cf. Jessop, Churchill, & Sparks, unpublished manuscript; Manstead & Parker, 1995). Automatic or gut reactions associated with direct measures of attitudes may therefore tap aspects of attitudes that are more affective in nature and which represent important fundamental dimensions of attitudes towards some attitudinal objects. Research in the domain of the natural environment, for example, has indicated a positive relationship between implicit or subconscious connections with nature with explicit environmental attitudes (Schultz, Shriver, Tabanico, & Khazian, 2004). On a related theme, it is interesting that the role of unconscious processes in judgement and decision making has attracted much recent attention within the discipline (Gigerenzer, 2007; Wilson, 2002). It is noteworthy that the affective connection items used here which comprised items such as, being out in nature is a great stress reducer seem to tap aspects of affective wellbeing. There remains, therefore, the possibility that the quality of affective connection that motivates people towards engaging with the natural environment might be benecial for affect-based psychological well-being (Herzog et al., 1997; Kaplan & Talbot, 1983; Ulrich, 1983) as well as having the potential for engendering positive environmental attitudes and behaviour.

ARTICLE IN PRESS
118 J. Hinds, P. Sparks / Journal of Environmental Psychology 28 (2008) 109120 Armitage, C. J., & Conner, M. (2001). Efcacy of the theory of planned behaviour: A meta-analytic review. British Journal of Social Psychology, 40, 471499. Bamberg, S. (2003). How does environmental concern inuence specic environmentally related behaviours? A new answer to an old question. Journal of Environmental Psychology, 23, 2132. Bamberg, S., Ajzen, I., & Schmidt, P. (2003). Choice of travel mode in the theory of planned behavior: The roles of past behavior, habit, and reasoned action. Basic and Applied Social Psychology, 25, 175187. Breckler, S. J., & Wiggins, E. C. (1989a). Affect versus evaluation in the structure of attitudes. Journal of Experimental Social Psychology, 25, 253271. Breckler, S. J., & Wiggins, E. C. (1989b). On dening attitude and attitude theory: Once more with feeling. In A. R. Pratkanis, S. J. Breckler, & A. G. Greenwald (Eds.), Attitude structure and function (pp. 407427). New Jersey: Lawrence Erlbaum. Bunting, T. E., & Cousins, L. R. (1985). Environmental dispositions among school-age children: A preliminary investigation. Environment and Behavior, 17, 725768. Chan, D. K.-S., & Fishbein, M. (1993). Determinants of college womens intentions to tell their partners to use condoms. Journal of Applied Social Psychology, 23, 14551470. Charng, H.-W., Piliavin, J. A., & Callero, P. L. (1988). Role identity and reasoned action in the prediction of repeated behaviour. Social Psychology Quarterly, 51, 303317. Chawla, L. (2002). Spots of time: Manifold ways of being in nature in childhood. In P. H. Kahn Jr., & S. R. Kellert (Eds.), Children and nature: Psychological, sociocultural and evolutionary investigations (pp. 199225). Cambridge, MA: The MIT Press. Clayton, S. (2003). Environmental identity: A conceptual and operational denition. In S. Clayton, & S. Opotow (Eds.), Identity and the natural environment: The psychological signicance of nature (pp. 4565). Cambridge, MA: MIT Press. Clayton, S., & Opotow, S. (2003). Introduction: Identity and the natural environment. In S. Clayton, & S. Opotow (Eds.), Identity and the natural environment: The psychological signicance of nature (pp. 124). Cambridge, MA: MIT Press. Cohen, J. (1988). Statistical power analysis for the behavioural sciences (2nd ed). New Jersey: Lawrence Erlbaum Associates. Conner, M. J., & Armitage, C. (1998). Extending the theory of planned behaviour: A review and avenues for further research. Journal of Applied Social Psychology, 28, 14291464. Conner, M. J., & Sparks, P. (1995). The theory of planned behaviour and health behaviours. In M. Conner, & P. Norman (Eds.), Predicting health behaviour: Research and practice with social cognition models (pp. 121162). Buckingham: Open University Press. Eagly, A. H., & Chaiken, S. (1993). The psychology of attitudes. Fort Worth, TX: Harcourt College Publishers. Eagly, A. H., Mladinic, A., & Otto, S. (1994). Cognitive and affective bases of attitudes towards social groups and social policies. Journal of Experimental Social Psychology, 30, 113137. Faber Taylor, A. F., Kuo, F. E., & Sullivan, W. C. (2002). Views of nature and self-discipline: Evidence from inner city children. Journal of Environmental Psychology, 22, 4963. Fazio, R. H., & Zanna, M. P. (1981). Direct experience and attitude behaviour consistency. Advances in Experimental Social Psychology, 14, 161202. lonneau, M.-L. (2004). Love and loathing of the city: Urbanophilia and Fe urbanophobia, topological identity and perceived incivilities. Journal of Environmental Psychology, 24, 4352. Field, A. (2000). Discovering statistics using SPSS for windows: Advanced techniques for beginners. London: Sage Publications. Finger, M. (1994). From knowledge to action? Exploring the relationship between environmental experiences, learning and behavior. Journal of Social Issues, 50, 141160. Fishbein, M. (1997). Predicting, understanding and changing socially relevant behaviours: Lessons learned. In C. McCarty, & S. A. Haslam

Attitude research has been frequently admonished for insufcient attention to affective inuences on behaviour (Breckler & Wiggins, 1989b; Giner-Sorolla, 1999). While applications of the TPB have certainly not been immune to such criticisms (Manstead & Parker, 1995; van der Pligt et al., 1998; Zanna & Rempel, 1988) there have been few applications in this regard that have offered both theoretical and empirical contributions (but see Jessop et al., unpublished manuscript; Richard et al., 1996). The study thus provides both a useful contribution to discussions addressing affective connection as an important predictive factor in the context of environmental issues and as an additional variable in an extended TPB. We do not doubt that both theoretical and measurement renements would be benecial, and that the inclusion of objective measures of actual behaviour in future research would help strengthen the assessment of the role of a sense of affective connection. Nevertheless, we would suggest that the present research offers a useful indication of the importance of affective factors in natural-environmental research and in attitudebehaviour models generally. Moreover, it offers an insight into the need to examine the nature of inter-relationships between measures of affect, on the one hand, and identity on the other and of the applied importance of seeking to understand environmentrelated behaviour in terms of the inter-relationships between peoples affective connections, their sense of their own identity and their direct experience of the natural environment. Acknowledgements The authors would like to thank the three anonymous reviewers of this article for their valuable constructive comments and insights on an earlier draft of this manuscript. References
Ajzen, I. (1988). Attitudes, personality, and behaviour. Chicago, IL: Dorsey Press. Ajzen, I. (1991). The theory of planned behaviour. Organizational Behaviour and Human Decision Processes, 50, 179211. Ajzen, I. (2001). Nature and operation of attitudes. Annual Review of Psychology, 52, 2758. Ajzen, I., 2002a. Constructing a TPB questionnaire: Conceptual and methodological considerations. Retrieved January 14, 2003, from: /http://www.unix.oit.umass.edu/aizen/tpb.htmlS. Ajzen, I., 2002b. The theory of planned behavior: A bibliography. Retrieved January 14, 2003, from: /http://www.people.umass.edu/aizen/ tpbrefs.htmlS. Ajzen, I., & Driver, B. L. (1991). Prediction of leisure participation from behavioural, normative, and control beliefs: An application of the theory of planned behaviour. Leisure Sciences, 13, 185204. Ajzen, I., & Fishbein, M. (2005). The inuence of attitudes on behavior. In D. Albarrac n, B. T. Johnson, & M. P. Zanna (Eds.), The handbook of attitudes (pp. 173221). Mahwah, NJ: Erlbaum. Ajzen, I., & Timko, C. (1986). Correspondence between health attitudes and behaviour. Basic and Applied Social Psychology, 7, 259276.

ARTICLE IN PRESS
J. Hinds, P. Sparks / Journal of Environmental Psychology 28 (2008) 109120 (Eds.), The message of social psychology (pp. 7791). Oxford, UK: Blackwell. Frankfurt, H. G. (1988). The importance of what we care about: Philosophical essays. Cambridge, UK: Cambridge University Press. French, D. P., Sutton, S., Hennings, S. J., Mitchell, J., Wareham, N. J., Grifn, S., et al. (2005). The importance of affective beliefs and attitudes in the theory of planned behaviour: Predicting intention to increase physical activity. Journal of Applied Social Psychology, 35, 18241848. Gigerenzer, G. (2007). Gut feelings. London: Allen Lane. Giner-Sorolla, R. (1999). Affect in attitude: Immediate and deliberative perspectives. In S. Chaiken, & Y. Trope (Eds.), Dual-process theories in social psychology. New York: Guilford. Herzog, T. R., Black, A. M., Fountaine, K. A., & Knotts, D. J. (1997). Reection and attentional recovery as distinctive benets of restorative environments. Journal of Environmental Psychology, 17, 165170. Hines, J. M., Hungerford, H. R., & Tomera, A. N. (1986/1987). Analysis and synthesis of research on responsible environmental behaviour: A meta analysis. Journal of Environmental Education, 18, 18. Horwitz, W. A. (1996). Development origins of environmental ethics: The life experience of activists. Ethics and Behavior, 6, 2954. Jessop, D. C., Churchill, S., & Sparks, P. (unpublished manuscript). High fat food consumption: The role of implicitly measured attitudes. Kahn, P. H. (1997). Developmental psychology and the biophilia hypothesis: Childrens afliation with nature. Developmental Review, 17, 161. Kals, E., & Maes, J. (2002). Sustainable development and emotions. In P. Schmuck, & W. P. Schultz (Eds.), Psychology of sustainable development (pp. 97122). Norwell, MA: Kluwer Academic Publishers. Kals, E., Schumaker, D., & Montada, L. (1999). Emotional afnity toward nature as a motivational basis to protect nature. Environment and Behavior, 31, 178202. Kaplan, R., & Kaplan, S. (1989). The experience of nature: A psychological perspective. Cambridge: Cambridge University Press. Kaplan, S. (1995). The restorative benets of nature: Toward an integrative framework. Journal of Environmental Psychology, 15, 169182. Kaplan, S., & Talbot, J. F. (1983). Pyschological benets of a wilderness experience. In I. Altman, & J. F. Wohlwill (Eds.), Human behavior and the environment: Advances in theory and research: Behavior and the natural environment (Vol. 6, pp. 163203). New York: Plenum Press. Kellert, S. R. (2002). Experiencing nature: Affective, cognitive, and evaluative development in children. In P. H. Kahn Jr., & S. R. Kellert (Eds.), Children and nature: Psychological, sociocultural and evolutionary investigations (pp. 117151). Cambridge, MA: The MIT Press. Lalli, M. (1992). Urban-related identity: Theory, measurement and empirical ndings. Journal of Environmental Psychology, 12, 285303. Lyons, E., & Breakwell, G. M. (1994). Factors predicting environmental concern and indifference in 13- to 16-year-olds. Environment and Behavior, 26, 223238. Mannetti, L., Pierro, A., & Livi, S. (2004). Recycling: Planned and selfexpressive behaviour. Journal of Environmental Psychology, 24, 227236. Manstead, A. S. R., & Parker, D. (1995). Evaluating and extending the theory of planned behavior. In W. Stroebe, & M. Hewstone (Eds.), European review of social psychology (Vol. 6, pp. 6996). Chichester, UK: Wiley. Manzo, L. C. (2003). Beyond house and haven: Toward a revisioning of emotional relationships with places. Journal of Environmental Psychology, 23, 4761. Mayer, F. S., & Frantz, C. M. (2004). The connectedness to nature scale: A measure of individuals feeling in community with nature. Journal of Environmental Psychology, 24, 503515. Millar, M. G., & Millar, K. U. (1996). The effects of direct and indirect experience on affective and cognitive responses and the attitude behavior relation. Journal of Experimental Social Psychology, 32, 561579. Moore, E. O. (1981). A prison environments effect on health care service demands. Journal of Environmental Systems, 11, 1734. 119 Nord, M., Luloff, A. E., & Bridger, J. C. (1998). The association of forest recreation with environmentalism. Environment and Behavior, 30, 235246. Norman, P., Clark, T., & Walker, G. (2005). The theory of planned behaviour, descriptive norms, and the moderating role of group identication. Journal of Applied Social Psychology, 35, 10081029. Parker, D., Stradling, S. G., & Manstead, A. S. R. (1996). Modifying beliefs and attitudes to exceeding the speed limit: An intervention study based on the theory of planned behaviour. Journal of Applied Social Psychology, 26, 119. Pooley, J. A., & OConner, M. (2000). Environmental education and attitudes. Environment and Behavior, 32, 711723. Proshansky, H. M. (1978). The city and self-identity. Environment and Behavior, 10, 147169. Pyle, R. M. (1978). The extinction of experience: A loss of neighborhood species endangers our experience of nature. Horticulture, 56, 6467. Richard, R., de Vries, N., & van der Pligt, J. (1998). Anticipated regret and precautionary sexual behavior. Journal of Applied Social Psychology, 28, 14111428. Richard, R., van der Pligt, J., & de Vries, N. (1995). Anticipated affective reactions and prevention of AIDS. British Journal of Social Psychology, 34, 921. Richard, R., van der Pligt, J., & de Vries, N. (1996). Anticipated affect and behavioural choice. Basic and Applied Social Psychology, 18, 111129. Schultz, P. W. (2000). Empathizing with nature: The effects of perspective taking on concern for environmental issues. Journal of Social Issues, 56, 391406. Schultz, P. W. (2002). Inclusion with nature: The psychology of humannature relations. In P. Schmuck, & W. P. Schultz (Eds.), Psychology of sustainable development. Norwell, MA: Kluwer Academic. Schultz, P. W., Shriver, C., Tabanico, J. J., & Khazian, A. M. (2004). Implicit connections with nature. Journal of Environmental Psychology, 24, 3142. Sheeran, P., & Taylor, S. (1999). Predicting intentions to use condoms: A meta-analysis and comparison of the theories of reasoned action and planned behaviour. Journal of Applied Social Psychology, 29, 16241675. Sparks, P., & Guthrie, C. A. (1998). Self-identity and the theory of planned behaviour: A useful addition or an unhelpful a artice? Journal of Applied Social Psychology, 28, 13931410. Sparks, P., & Shepherd, R. (1992). Self-identity and the theory of planned behaviour: Assessing the role of identication with green consumerism. Social Psychology Quarterly, 55, 388399. Stern, P. C., Dietz, T., & Kalof, L. (1993). Value orientations, gender, and environmental concern. Environment and Behavior, 25, 322348. Stets, J. E., & Biga, C. F. (2003). Bringing identity theory into environmental sociology. Sociological Theory, 21, 398423. Teisl, M. F., & OBrien, K. (2003). Who cares and who acts? Outdoor recreationists exhibit different levels of environmental concern and behavior. Environment and Behavior, 35, 506522. Terry, D. J., Hogg, M. A., & White, K. M. (1999). The theory of planned behaviour: Self-identity, social identity and group norms. British Journal of Social Psychology, 38, 225244. Thomas, K. (1983). Man and the natural environment: Changing attitudes in England 15001800. London: Penguin Books. Thompson, S. C., & Barton, M. A. (1994). Ecocentric and anthropocentric attitudes toward the environment. Journal of Environmental Psychology, 14, 149157. Tramow, D., & Sheeran, P. (1998). Some tests of the distinction between cognitive and affective beliefs. Journal of Experimental Social Psychology, 34, 378397. Ulrich, R. S. (1983). Aesthetic and affective response to natural environment. In I. Altman, & J. Wohlwill (Eds.), Human behavior and environment: Behavior and the natural environment (Vol. 6, pp. 85125). New York: Plenum. Ulrich, R. S. (1984). View through a window may inuence recovery from surgery. Science, 224, 420421.

ARTICLE IN PRESS
120 J. Hinds, P. Sparks / Journal of Environmental Psychology 28 (2008) 109120 Wilson, T. D., Dunn, D. S., Kraft, D., & Lisle, D. J. (1989). Introspection, attitude change, and attitudebehavior consistency: The disruptive effects of explaining why we feel the way we do. Advances in Experimental Social Psychology, 22, 287342. Zajonc, R. B. (2001). Feeling and thinking: Closing the debate over the independence of affect. In J. P. Forgas (Ed.), Feeling and thinking: The role of affect in social cognition (pp. 3158). Cambridge, UK: Cambridge University Press. Zanna, M. P., & Rempel, J. K. (1988). Attitudes: A new look at an old concept. In D. Bar-Tel, & A. W. Kruglanski (Eds.), The social psychology of knowledge (pp. 315334). Cambridge, UK: Cambridge University Press. Zelezny, L. C., Chua, P.-P., & Aldrich, C. (2000). Elaborating on gender issues in environmentalism. Journal of Social Issues, 56, 443457. van der Pligt, J., Zeelenberg, M., van Dijk, W. W., de Vries, N. K., & Richard, R. (1998). Affect, attitudes and decisions: Lets be more specic. In W. Strobe, & M. Hewstone (Eds.), European review of social psychology (Vol. 8, pp. 3366). Chichester, UK: Wiley. Wells, N. M. (2000). At home with nature: Effects of greenness on childrens cognitive functioning. Environment and Behavior, 32, 775795. Wilson, E. O. (1984). Biophilia. Cambridge: Harvard University Press. Wilson, E. O. (1993). Biophilia and the conservation ethic. In S. R. Kellert, & E. O. Wilson (Eds.), The biophilia hypothesis (pp. 3141). Washington, DC: Island Press. Wilson, T. D. (2002). Stranger to ourselves: Discovering the adaptive unconscious. Harvard: HUP. Wilson, T. D., & Dunn, D. S. (1986). Effects of introspection on attitudebehavior consistency: Analyzing reasons versus focusing on feelings. Journal of Experimental Social Psychology, 22, 249263.