Identifying features within a garden linked to emotional reactions and perceived restoration

ABSTRACT

This study aimed to identify what specific plants and garden features in an intentionally designed wellbeing garden influence emotional reactions and perceived restoration. To test this 142 garden visitors were recruited and asked to complete a survey as they walked around the wellbeing garden. The survey guided participants around the garden instructing them to stop at four specific Zones. They were asked questions concerning how the different Zones of the garden made them feel (sad, happy, excited, calm, and perceived restoration). In addition, the survey included open-ended questions where participants were asked to identify the features such as water elements, plants, views, and structures (sculptures, seating, paving) they felt evoked their emotional response. Using a mixed methods approach with ANOVA and Content Analysis, results found key garden design features linked to emotional reactions and perceived restoration include sounds of flowing water, open seating areas, far-reaching views, and a variety of planting (colour, height, texture, scents, attracting wildlife). This evidence can be used to inform the design of gardens that optimise wellbeing outcomes.

KEYWORDS:

• Garden design
• restorative landscapes
• wellbeing

Introduction

This paper focuses on examining how public gardens could be designed to support wellbeing. This is done by examining emotional experiences in a public horticultural garden designed for wellbeing. Gardens are increasingly becoming acknowledged as places to support overall wellbeing in the general population (Chalmin-Pui et al. 2021, Lampert et al. 2021, Tharrey and Darmon 2022). For example, research to date has shown that spending time in gardens is linked to health and wellbeing outcomes including improvements in physiological and psychological health (Gregis et al. 2021). Designing public gardens for wellbeing may be especially important for urban residents who are exposed to more urban stressors and are less likely to have access to private gardens or green spaces (Triguero-Mas et al. 2021, Han et al. 2022). Gardens or small parks in urban spaces can provide vital resources supporting recovery from stress and mental fatigue among urban residents. Providing appropriately designed public green spaces or publicly accessible wellbeing gardens that support the wellbeing of those who live and work in urban areas could be a beneficial public health strategy (Amano et al. 2018, Reyes-Riveros et al. 2021).

There is a clear evidence base supporting the use of natural landscapes (Callaghan et al. 2021), gardens (Chalmin-Pui et al. 2021) and plants (Hall and Knuth 2019) to support overall wellbeing. Exploring how individuals react emotionally towards different aspects of the garden could provide new insights towards understanding how and why features within a garden support wellbeing, which are the aims of the current study.

Measuring wellbeing outcomes in natural environments

There is substantial evidence that spending time in natural environments is good for human beings because it increases various wellbeing outcomes (Tharrey and Darmon 2022). Wellbeing can be defined and measured in many different ways. Within existing literature on nature and wellbeing, the concept of wellbeing most often refers to aspects associated with the psychological quality of an individual’s life including one’s experience of happiness (MacKerron and Mourato 2013), positive over negative affect (Richardson et al. 2015, Ballew and Omoto 2018), and most often psychological and physiological stress (Berto 2014) and restoration (Mourão et al. 2019). Whilst various measures can and have been used to explore nature and wellbeing, the current study is interested in measuring instant state reactions towards specific garden features. Emotions and perceived restoration are frequently used in nature-based research and are the most sensitive to short-term effects of nature exposure. This includes aspects associated with one’s experience of pleasant emotions (McMahan 2015, Neill et al. 2019, Young et al. 2020), reduced unpleasant emotions (Richardson et al. 2015, Ballew and Omoto 2018) and perceived restoration (Stevenson et al. 2018, Mourão et al. 2019). This study will focus on using the circumplex model of affect and perceived restoration to explore instant state reactions towards different features within a garden.

The circumplex model of affect is a useful indication of state emotional reactions when exposed to garden features. This model highlights how one can experience different emotional reactions from unpleasant to pleasant (e.g. sadness and happiness) and high to low arousal (e.g. excitement and calmness). To measure instant state reactions relevant to wellbeing, this study will focus on measuring emotional reactions and perceived restoration towards different features within a garden. Indeed, many studies have found that contact with the natural environment is linked to perceived restoration (Mourão et al. 2019), can boost pleasant emotions such as happiness (Capaldi et al. 2014, McMahan and Estes 2015) and reduce unpleasant emotions such as sadness (McMahan and Estes 2015), increase feelings of calmness (Kanelli et al. 2023), and in some cases evoke feelings of excitement (Zhang et al. 2023).

Perceived restoration is a common wellbeing measure in nature health research. This stems from Attention Restoration Theory (Kaplan 1995) which states that nature can reduce cognitive fatigue and promote restoration. Many studies have found support for this theory (Ohly et al. 2016), and many measures of perceived restoration have been developed to explore the restorative effects of spending time in natural settings (Han 2018).

Natural stimuli linked to emotional reactions and perceived restoration

Kaplan (1995) and Ulrich (1999) imply certain properties within the landscape can provide people with experiences that result in beneficial outcomes such as improved restoration and reduced stress. There is evidence to support the idea that certain properties within large natural landscapes are preferred (Stoltz and Grahn 2021). For example, Peschardt and Stigsdotter (2013) found that public urban green spaces were perceived as more restorative when people felt that the spaces provided serene and social opportunities. They also found that stressed individuals preferred green spaces that were more natural. A study exploring the perceived restoration of designed gardens found more natural-looking gardens were perceived to have greater restorative potential than less natural gardens (Twedt et al. 2016). Hoyle et al. (2017) surveyed 1411 members of the public who walked through planting of varying structure, species characters and flower cover. Results showed that gardens perceived to be beneficial for insects were rated more attractive, and natural planting with a flower cover of at least 27% was rated most restorative. These studies provide important information for generic preferences within green spaces and gardens, but there is little understanding of the psychological reactions towards different features within the garden context.

To date, there is no empirical evidence that has specifically explored state emotional reactions towards different features within a garden. The focus of the current study is to examine reported emotional experiences while spending time in different Zones of a designed wellbeing garden and to explore how perceptions of different planting and design features in those garden Zones can help explain different emotional experiences.

This study is not concerned with systematically varying exposure to different features specifically, but rather understanding state emotional reactions and perceived restoration responses towards various features present within a public horticultural garden. However, to understand examples of aspects that could provoke different reactions, this section will outline existing research on psychological reactions towards different natural stimuli including plant form and colour, scents, and sounds.

Choice of planting colour within garden design could be important for influencing emotional reactions and perceived restoration. For example, a cross-sectional survey of over 600 UK residents found that cool colours like blue and white to be linked with lower arousal emotions such as relaxation and warm colours like orange to be linked with higher arousal emotions such as excitement. Another study explored the effects of colour hue and saturation on emotional arousal (Zhang et al. 2023). Results found that participants were more responsive to bright saturated colours than to darker low saturated colours, and after controlling for saturation and brightness, red was the most arousing hue (Wilms and Oberfeld 2018). Contrastingly, another study used EEG to monitor brain waves of students exposed to different plant colour stimuli (Jang et al. 2014). It was found that cognitive restoration increased in response to green plants compared to white, yellow, pink, or red.

Research in aromas suggests that certain fragrances can increase pleasant emotions and perceived restoration, specifically, calmness, and alertness (Reynolds and Card 2016, Franco et al. 2017). This is because the olfactory system is linked to the emotional limbic system and hippocampus (Chen and Haviland-Jones 2000), which can affect mood, memories, and behaviour (Goel and Grasso 2004). For example, one study exposed participants to the aroma of rosemary in different concentrations whilst completing cognitive tasks (Moss and Oliver 2012). They found that cognitive performance significantly improved with higher concentrations of rosemary. Additionally, recent studies have found essential oils, such as lavender and rosemary to be related to decreased anxiety and stress (Donelli et al. 2019, Rahbardar and Hosseinzadeh 2020).

Finally, it is suggested that sounds in nature are also linked with emotional reactions and perceived restoration. Specifically, sounds like birdsong, running water, and gentle winds have been linked with increased sensory stimulation, restoration, and relaxation (Ratcliffe 2021). For example, a study exploring soundscapes and wellbeing found that the majority of participants enjoyed the sound of birdsong and running water as it evoked lower arousal emotions such as calmness (Bates et al. 2020). Importantly, the authors suggested that preference for natural sounds may be linked to individual and cultural experiences highlighting potential cross-cultural differences in garden preferences.

In summary, there is research linking colour (Jang et al. 2014, Hůla and Flegr 2016, Wilms and Oberfeld 2018, Zhang et al. 2023), scents (Moss and Oliver 2012, Donelli et al. 2019, Rahbardar and Hosseinzadeh 2020), and sounds (Alvarsson et al. 2010, Bates et al. 2020) to different psychological reactions. However, the outcome variable and findings across the studies are inconsistent. Additionally, many gardens contain a huge diversity and combination of colours and scents and it is not known how individuals may experience these sensory aspects of the garden. Whilst it is beneficial to systematically explore psychological perceptions towards specific features, it is not replicable of real-life experiences, where individuals are exposed to many features at the same time. As such, very little is known about how these findings relate or can be applied to the actual garden context. This study intends to bring these different perspectives together by examining emotional responses to different garden Zones in a garden specifically designed to evoke different emotional reactions. This approach has higher ecological validity than studying separated features in controlled laboratory settings as previous research has done. Furthermore, to explore the impact of different features within the garden, the current study will explore what features individuals naturally notice. Again, this approach is thought to be more ecologically valid as what individuals notice may be more beneficial to wellbeing than simply exposing them predetermined features within a lab setting (Richardson et al. 2022).

The RHS wellbeing garden

This research is based at the Royal Horticultural Society (RHS) Wisley’s purposely designed and built wellbeing garden to understand how garden design can optimise wellbeing outcomes. The garden is situated outside the RHS Home of Gardening Science, Hilltop Building, Surrey, UK. The wellbeing garden was designed by Matt Keightley in collaboration with scientists across a number of disciplines using different design principles from their expertise to promote wellbeing. The garden is divided into several Zones, with each Zone designed to stimulate different senses and evoke various emotional reactions. This is done with use of planting, colour, scent, and different shapes, textures, and forms of plants (Figure 1).

Figure 1. Overview of RHS wisley wellbeing garden zone including locations 1–4 used in this study. RHS/Oliver Dixon.

Research questions

1. Are emotional reactions and perceived restoration influenced by different Zones of the wellbeing garden?

2. How are perceptions of garden features associated with emotional reactions and perceived restoration?

Four different Zones were used for this study and selected to incorporate specific features aimed at evoking different senses and emotional responses (Figure 1). Hypotheses are based on the design of the wellbeing garden Zones and previous research on sensory aspects of nature linked to pleasant emotions and perceived restoration. It is hypothesised that Zones 2 and 3 will be linked to pleasant and lower arousal emotions including happiness and calm and may be perceived as more restorative due to features such as cooler colours, scent of lavender and water elements (Donelli et al. 2019, Rahbardar and Hosseinzadeh 2020, Zhang et al. 2023). Zones 1 and 4 will be linked to higher arousal emotions including excitement due to specific features such as brighter colours, rosemary, and curving pathways (Ulrich 2002, Zhang et al. 2023).

Method

Participants

Participants (n = 142) were visitors to RHS Wisley over the age of 18. Participants were asked if they were regular visitors to RHS Wisley, to which the majority reported yes. Participant demographics included mostly White British (n = 122), with only a small number were Black (n = 5), Asian (n = 5) and Mixed ethnic (n = 2). Of these, 93 participants were female and 42 male, 7 did not specify. The mean age was 53.6 years, with the oldest participant being 85 and the youngest 19 (SD 14.7).

This study was advertised to potential participants with a physical sign located at the entrance of the wellbeing garden. Access to the survey was made available via a QR code that was located on this physical sign (Appendix 1). Visitors were able to scan the QR code on their mobile devices to take part. A paper version of the survey was also available to participants in case of technical difficulties or preference. The survey was first released between April and July 2022 to collect responses for Spring/Summer. Over these months, the weather was generally dry, warm, and sunny. The survey was released again between October and January 2022–23 to collect responses for Autumn/Winter. Over these months, the weather was generally sunny, but colder, and breezy. On four days of each data collection period (two weekdays and two weekend days), the researcher actively stood in the wellbeing garden with paper-based surveys to recruit any RHS visitors who approached the wellbeing garden. This method of recruitment had the highest response rate recruiting 120 of the total participants. Data collection was slower during the Winter months and the researcher spent more time actively recruiting during this period.

Design

Four different Zones within the RHS wellbeing garden were used for this study and selected to incorporate specific features aimed at evoking different emotional responses (Figure 1; Tables 1 and 2). Zone 1 incorporated brighter planting schemes such as oranges and reds aimed at promoting higher arousal emotions such as excitement (Wilms and Oberfeld 2018). Zone 2 included a running water feature and Zone 3 included planting with cooler colours such as lilac and white along with calming scents like lavender aimed at promoting lower arousal emotions such as calmness and restoration (Bates et al. 2020, Rahbardar and Hosseinzadeh 2020, Zhang et al. 2023). Finally, Zone 4 included dwarf pines and rosemary. This Zone was designed to be arousing due to the scent of pine and rosemary (Moss and Oliver 2012).

Table 1. Examples of prominent features within selected zones 1–4 during spring/summer.

Zone	Examples of features within the Zone	Images
1	Bright red and orange planting [Kniphofia ‘Alcazar’; Crocosmia ’Hellfire’ & ‘Emberglow’] Large, wooden curved bench seating View of RHS Hilltop Building
2	Running water into a small pond Tall circular planting [Rudbeckia laciniata ‘Herbstsonne’; Pinus mugo ‘Pumilio’; Chamaecyparis obtusa ‘Nana Gracilis’; Iris ‘Silver Edge’] View of RHS Hilltop Building
3	Expansive view of distant trees and hills White scented planting [Lavandula angustifolia ‘Arctic Snow’; Achillea ptarmica ‘The Pearl’ (clonal); Oenothera lindheimeri]
4	Curved path leading up to the garden. Green shrubs edging the path [Pinus mugo subsp. mugo; Pimus mugo ‘Mops’; Salvia rosmarinus ‘Miss Jessopp’s Upright’] View of RHS Hilltop Building

Table 2. Examples of prominent features within selected zones 1–4 during autumn/winter.

Zone	Examples of features within the Zone	Images
1	Red planting [Salvia confertiflora; Rosa ‘Munstead Wood’] Large, wooden curved bench seating View of RHS Hilltop Building
2	Running water into a small pond Tall circular planting [Rudbeckia laciniata ‘Herbstsonne’; Pinus mugo ‘Pumilio’;Chamaecyparis obtusa ‘Nana Gracilis’; Iris ‘Silver Edge’]
3	Expansive view of distant trees and hills White scented planting [Lavandula angustifolia ‘Arctic Snow’; Achillea ptarmica ‘The Pearl’ (clonal); Oenothera lindheimeri]
4	Curved path leading up to the garden. Green shrubs edging the path [Pinus mugo subsp. mugo; Pimus mugo ‘Mops’; Salvia rosmarinus ‘Miss Jessopp’s Upright’] View of RHS Hilltop Building

The survey guided participants around the garden to four different Zones in a randomized order as they completed responses. To control for the order of effects, surveys were completed in random orders with starting points at different Zones. This included position orders 1–4 (N = 49), 4–1 (N = 47), and anti-clockwise 2–3 (N = 46).

Measures

The circumplex model of affect was used to examine emotional reactions (Russell 1980). To reduce the length of this questionnaire, four specific emotions were selected by choosing one emotion from each quadrant of the circumplex model (Russell 1980). This included happy and sad from the valence dimension and excited and calm from the activation dimension. Participants were asked ‘How do you feel whilst taking in the surroundings in front of you? Please answer below how true each emotion is of you right now’. They were able to rate the extent they experienced each emotion (Sad, Happy, Calm, and Excited) on a 5-point Likert scale ranging from −2, ‘Very untrue of me’ to 2, ‘Very true of me’ (Appendix 2).

Self-rating restoration scale (SRRS; Han 2003) was used to examine perceived restoration of different Zones within the garden. Specific questions related to the cognition and behaviour dimensions were used for this survey. Participants were asked ‘How do you feel about the views in front of you? Please answer below how true each statement is of you right now. Participants were able to rate the extent they experienced each facet of the scale with a 5-point Likert scale ranging from −2, ‘Very untrue of me’ to 2, ‘Very true of me’ (Appendix 3).

Open-ended questions were included to answer the research question ‘What specific plants and garden features elicit emotional reactions and perceived restoration?’ Participants were asked, ‘Please identify any specific feature/s in your immediate surroundings you think is/are linked to your feelings of this space’. Participants were able to freely respond with written text answers which were analysed using Content Analysis.

Results

Are emotional reactions and perceived restoration influenced by different zones of the wellbeing garden?

Data was exported from Qualtrics and screened in Jamovi (version 2.3.13). A total of 142 survey responses were used for data analysis (79 for Spring/Summer; 63 for Autumn/Winter). To answer the first question, ‘How do different planting schemes and garden features influence emotional reactions and perceived restoration?’ a repeated measures ANOVA was carried out in Jamovi to explore differences in emotional and restorative responses in each of the four Zones across the seasons. Data were also tested for assumptions of repeated measures ANOVA including normality and sphericity. In cases where Mauchly’s test indicated that the assumption of sphericity had been violated, Greenhouse-Geisser was used to correct the F-value.

Spring/summer results

There was a statistically significant effect of Zone on the emotion sad (F(3, 234) = 4.93, p < .005, η2p = .06). Post-hoc tests showed that the emotion sad was rated as somewhat untrue of me more in Zone 1 than in zone 2, which was mostly rated as very untrue of me (p = .04). Sad was also rated as somewhat untrue of me more in Zone 4 than in Zone 2 (p = .002) and Zone 3 (p = .007) which mostly rated as very untrue of me.

There was a statistically significant effect of zone on the emotion happy (F(3, 234) = 9.32, p < .001, η2p = .11). Post-hoc tests showed that the emotion happy was rated as somewhat true of me more in Zone 1 than in Zone 4 which was mostly rated as somewhat untrue of me (p < .001). Happy was also rated as somewhat true of me more in Zone 3 than Zone 4 (p < .001) and Zone 2 which was mostly rated as neutral (p = .03).

There was a statistically significant effect of Zone on the emotion excited (F(2.76, 215) = 2.85, p = .04, η2p = .04). Post-hoc tests showed that the emotion excited was rated as neutral more in Zone 1 than in Zone 2 (p = .03) and Zone 4 (p = .01) which were mostly rated as somewhat untrue of me.

There was a statistically significant effect of Zone on the emotion calm (F(2.66, 207) = 26.3, p < .001, η2p = .25). Post-hoc tests showed that calm was rated as somewhat true of me more in Zone 2 than in Zone 1 (p < .001) and Zone 4 (p < .001) which were mostly rated as neutral. Calm scores were also rated as somewhat true of me more in Zone 3 than in Zone 1 (p < .001) and Zone 4 (p < .001). Calm ratings were also significantly different between Zone 1 and Zone 4 (p = .03).

There was a statistically significant effect of Zone on perceived restoration (F(2.56, 199) = 22.1, p < .001, η2p = .22). Post-hoc tests showed that restoration was rated significantly higher in Zone 2 than in Zone 1 (p = .02), Zone 3 (p = .008) and Zone 4 (p < .001). Restoration was also rated higher in Zone 3 than in Zone 1 (p < .001) and Zone 4 (p < .001). Restoration was also rated significantly higher in Zone 1 than in Zone 4 (p = .002) (See Table 3).

Table 3. Means and standard deviations for each emotion relevant to garden zones for spring/summer.

Emotion	Zone 1	Zone 2	Zone 3	Zone 4
Sad	M = −1.27	M = −1.53	M = −1.49	M = −1.09**
	SD = 1.08	SD = 0.85	SD = 0.95	SD = 1.02
Happy	M = 0.89***	M = 0.70	M = 0.98**	M = −1.09
	SD = 0.89	SD = 0.97	SD = 1.06	SD = 0.98
Excited	M = 0.06**	M = −0.22	M = −0.05	M = −0.26
	SD = 0.95	SD = 0.90	SD = 0.95	SD = 1.01
Calm	M = 0.65*	M = 1.27***	M = 1.34***	M = 0.27
	SD = 1.12	SD = 0.94	SD = 0.89	SD = 1.11
Perceived	M = 0.35**	M = 0.68**	M = 1.07***	M = −0.06
restoration	SD = 0.94	SD = 1.04	SD = 1.12	SD = 0.98

Sample size N = 79. Mean scores are from −2 (Very untrue of me) to 2 (very true of me).

* = Significant differences with higher scores for emotional reaction or perceived restoration *p < .05. **p < .01. ***p < .001.

Autumn/winter results

There was a statistically significant effect of Zone on the emotion sad (F(2.66,162) = 3.26, p = .03, η2p = .05). Post-hoc tests showed that sad was rated as somewhat untrue of me more in Zone 4 than in Zone 2 (p = .005) and Zone 3 (p = .02) which were rated as very untrue of me.

There was a statistically significant effect of Zone on the emotion happy (F(3,183) = 4.47, p < .005, η2p = .68). Post-hoc tests showed that the emotion happy was mostly rated as somewhat true of me more in Zone 3 than Zone 1 (p = .02) and Zone 4 (p < .001) which were mostly rated as neutral. Happy was also rated higher in Zone 2 than in Zone 4 (p = .03).

There was no significant difference between the Zones for the emotion excited (F(2.57,183) = 0.61, p = .6).

There was a statistically significant effect of Zone on the emotion calm (F(2.30, 140) = 12.0, p < .001, η2p = .16). Post-hoc tests showed that calm was rated as somewhat true of me more in Zone 2 than in Zone 1 (p = .003) and Zone 4 (p < .001) which were mostly rated as neutral. Calm was also rated as somewhat true of me more in Zone 3 than Zone 1 (p < .001) and Zone 4 (p < .001).

There was a statistically significant effect of Zone on perceived restoration (F(2.68, 163) = 14.0, p < .001, η2p= .19). Post-hoc tests showed that restoration was rated significantly higher in Zone 2 than in Zone 1 (p < .001) and Zone 4 (p < .001). Restoration was also rated higher in Zone 3 than in Zone 1 (p < .001) and Zone 4 (p < .001) (See Table 4).

Table 4. Means and standard deviations for each emotion relevant to garden zones for autumn/winter.

Emotion	Zone 1	Zone 2	Zone 3	Zone 4
Sad	M = −1.13	M = −1.39	M = −1.40	M = −1.08**
	SD = 0.98	SD = 0.82	SD = 0.82	SD = 0.92
Happy	M = 0.57	M = 0.71*	M = 0.85***	M = 0.40
	SD = 0.90	SD = 0.95	SD = 0.88	SD = 0.85
Excited	M = −0.29	M = −0.24	M = −0.39	M = −0.38
	SD = 0.88	SD = 0.94	SD = 0.95	SD = 0.95
Calm	M = 0.60	M = 1.19***	M = 1.27***	M = 0.64
	SD = 0.98	SD = 0.88	SD = 0.70	SD = 0.85
Perceived restoration	M = 0.31	M = 0.8***	M = 0.98***	M = 0.17
	SD = 0.98	SD = 0.83	SD = 0.72	SD = 0.96

Sample size N = 63. Mean scores are from −2 (Very untrue of me) to 2 (very true of me).

* = Significant differences with higher scores for emotional reaction or perceived restoration *p < .05. **p < .01. ***p < .001.

How are perceptions of garden features associated with emotional reactions and perceived restoration?

To answer the second research question, ‘How are perceptions of garden features associated with emotional reactions and perceived restoration?’, Content Analysis of the written responses was carried out to identify the garden features associated with the reported emotions and perceived restoration within each Zone. Content Analysis is a systematic way of identifying and coding characteristics within written text (Krippendorff 2018). This was deemed an appropriate way of interpreting the written responses to identify the specific features mentioned by participants as being linked to their emotional reactions. To do this, specific features mentioned by participants were coded, and the frequency of a specific feature being mentioned was recorded and calculated into an overall percentage for each Zone. Specific features that were only noted by one or two participants have not been reported here but can be seen in Appendix 5.

When asked what features were linked to their feelings of Zone 1, participants mostly commented on the colourful red planting. In Spring/Summer, 68% (n = 54) of participants identified a specific red plant called Kniphofia ‘Alcazar’ (Red Hot Poker) (Appendix 6) as the feature that stood out to them. Similarly, in Autumn/Winter, 24% (n = 15) of participants identified the bold pops of red planting but did not note specific plants. Additionally, in Autumn/Winter 16% (n = 10) of participants suggested that Zone 1 lacked planting. The lack of colour in this Zone compared to Spring and Summer may explain why no significant association was found with the emotion excited in the Autumn and Winter months. Also, 23% (n = 32) of participants commented on the curved bench seating across the seasons (Appendix 6) where participants suggested it was an interesting place to sit because of the curved shape and large social space.

When asked what features were linked to their feelings of Zone 2, across the seasons 68% (n = 97) of participants specifically identified the sound of running water as calming (Appendix 7). Additionally, across the seasons 18% (n = 25) of participants generally commented on the tall planting structures surrounding the water (Appendix 7). Participants commented on the various textures and shades of greenery in the planting and suggested that it provided a sense of privacy and seclusion which added to the sense of calmness.

When asked what features were linked to their feelings of Zone 3, during Spring/Summer 53% (n = 42) of participants commented on the Lavandula angustifolia ‘Arctic Snow’ (lavender) (Appendix 8). This was because of the scent, and the fact that it attracted wildlife. In Autumn/Winter, 38% (n = 24) of participants commented on the tall white flowers Oenothera lindheimeri (Gaura) because of its delicate and gentle movements (Appendix 8). Importantly, the lavender was more prominent in Spring/Summer and had died back by the Autumn when the Gaura became more prominent. This may explain why participants noted different plants across the seasons. Additionally, across the seasons 42% (n = 60) of participants commented on the distant views (Appendix 8). Participants suggested that the views were open and expansive, giving them a sense of freedom.

When asked what features were linked to their feelings of Zone 4, across the seasons 24% (n = 34) of participants suggested the curving path looked interesting to walk up and explore (Appendix 9). Additionally, across the seasons 42% (n = 60) of participants specifically commented on the Pinus mugo (dwarf pines) (Appendix 9). However, there was mixed opinion among these participants. Whilst 16% (n = 23) commented positively on the texture or green colour of the pines, 26% (n = 37) suggested that there was a lack of variety and colour due to the planting being predominantly pines (See Table 5).

Table 5. Examples of garden features related to emotional and restorative responses.

Emotional response	Identified features in the RHS Wisley garden	Examples for garden design
Excited	Bright red and orange planting [Kniphofia ‘Alcazar’; Crocosmia ’Hellfire’ & ‘Emberglow’]	Bright planting with bold contrasting colours (e.g. Red, orange). Ensure to consider seasonal changes and include plants with such colour in the Winter seasons.
	Large, wooden curved bench seating	Large areas suitable for social activities. Such as open seating or larger pattio spaces.
Calm	Sounds of running water	Create calming spaces with use of sound such as running water features, wind chimes, tall planting that rustles in the breeze or by encouraging birds into the garden.
	Tall planting that gives a sense of seclusion.	Include tall planting that encloses spaces to give a sense of privacy.
Happy	Seeing insects including bees and butterflies.	Include a variety of pollinator-attracting plants. Position these in spaces where garden users can easily see and watch the wildlife.
	Variety of planting colours and textures.	It’s important to consider variety and texture in the planting palettes to prevent disinterest or boredom.
High perceived restoration	Open and expansive views.	Orientate the garden to encapsulate nearby views. Where possible position seating to overlook open views.

Discussion

This study aimed to explore emotional and restorative responses to different Zones in the wellbeing garden at RHS Wisley. Support was found for the first hypothesis that Zones 2 and 3 would be linked to the emotion calm and higher perceived restoration due to features such as cooler colours (Zhang et al. 2023), scent of lavender (Donelli et al. 2019) and water elements (Rahbardar and Hosseinzadeh 2020). Support was also found for the hypothesis that Zone 1 would be linked to excitement, due to bright colours (Zhang et al. 2023). No support was found for the hypothesis that Zone 4 would be linked to excitement. This study is the first to explore emotional reactions and perceived restoration as individuals experience different Zones of an intentionally designed wellbeing garden. The key findings highlight how different features within a garden influence emotional reactions and perceived restoration. Namely, bright colours and social spaces evoke excitement, whilst sounds of running water, delicate planting palettes, and open views support a feeling of calm and restoration. A variety of planting that attracts wildlife can evoke happiness. These findings help build a wider picture of the lived experiences of garden users and how their wellbeing may be influenced by specific garden features. These findings are intended to support design guidelines of gardens and public green spaces concerning human wellbeing. As this paper is concerned with applying the findings to garden design recommendations, this section will be focused on discussing specific design features and their link to different emotional reactions and perceived restoration.

Planting

This study identified planting as a key feature linked to emotional reactions and perceived restoration. This included different coloured plants such as bold and light colours, scented planting, variety, and biodiversity of planting.

Participants in this study identified features such as red planting suggesting that it was linked to excitement, which is a higher arousal emotion. The planting stood out to participants because of the bright orange-red colours and participants described this as vibrant, bold, and attention-grabbing. Attention restoration theory suggests that effortless attention should be captured for the environment to stimulate fascination and have restorative benefits (Kaplan and Kaplan 1989). It could be that the red planting grabbed attention due to the colour contrast between red and green (Palmer and Schloss 2010). This supports the findings of previous studies suggesting that people are responsive to bright colours such as red that can evoke higher arousal emotions such as excitement (Wilms and Oberfeld 2018).

Secondly, participants noted the scent of lavender, suggesting it was linked to happy and calm, which are pleasant and lower arousal emotions (Appendix 8). Whilst the aromas of nature are largely neglected in nature-health theories, both garden design literature and current research suggest that fragrant plants are an important design aspect. Garden design recommendations suggest using multi-sensory planting including scented plants to help evoke the senses and pleasant emotions (Bengtsson and Grahn 2014, Lau et al. 2014). Research in aromas suggests that fragrant plants can increase calmness, alertness, or mood (Franco et al. 2017). Furthermore, essential oils, such as lavender, have been shown to decrease anxiety and stress and increase feelings of contentment (Moss et al. 2003, Donelli et al. 2019). The white and lilac colours of the lavender could also be linked to the calming emotions experienced, supporting results found in the recent study by Zhang et al. (2023).

Additionally, participants frequently mentioned that they enjoyed watching the bees that were attracted to the lavender and Gaura which evoked pleasant emotions. Studies have found planting that attracts insects such as butterflies and bees are rated as more attractive and restorative (Hoyle et al. 2017). Indeed, garden design literature suggests the use of pollinator-attracting plants so that people can enjoy watching wildlife as they spend time in the garden (Stigsdotter and Grahn 2003, Erickson 2012, Balode 2013). Research findings would also support watching wildlife as a restorative activity. Specifically, research has found that watching wildlife can be a mindful activity that initiates a sense of awe and wonder, similar to the fascination concept mentioned in Attention restoration theory (Bell et al. 2018, Macaulay et al. 2022). It is also important to consider use of plants that attract wildlife throughout the seasons so that people can also enjoy such benefits during the Winter months.

Importantly, areas with low variety of planting were also rated low for perceived restoration. Both Attention restoration and Stress reduction theories suggest that fascination and complexity are important for restorative landscapes. Garden design literature suggests that multi-sensory planting should be used in gardens which contains different colours, scents, and textures to help evoke different senses and engage garden users (Shackell and Walter 2012, Peschardt 2014, Douglas et al. 2017, Harries et al. 2023). It makes sense that a diversity of plants would create more interest especially as research in landscape preference suggests that the psychological benefits associated with nature such as pleasant emotional reactions and perceived restoration increase with high vegetation and more biodiversity (Fuller et al. 2007, Chiang et al. 2017). It is also thought that variety of vegetation may aid stress reduction (Lindemann-Matthies and Matthies 2018). This includes natural environments with medium levels of variety and vegetation (Lindemann-Matthies et al. 2010) and a variety of planting shape, colour, and textures (Wohlwill 1983).

Sounds

When participants were exposed to the sounds of running water, they were most likely to report feeling calm, which is a pleasant lower arousal emotional reaction. Across the seasons, people consistently cited flowing water as the key feature linked to this experience because the water sound gave them a calming feeling (Appendix 7). This finding is consistent with both theory and current literature on natural soundscapes. Stress reduction theory suggests that water in a garden is a key component of stress reduction (Ulrich 2002). Furthermore, systematic reviews of garden design recommend the use of water features to create a serene space (Shukor et al. 2012, Harries et al. 2023). This finding also supports literature suggesting that the sound of water can enhance pleasure, relaxation, pleasant emotions, improve mood, and reduce self-reported stress (Largo-Wight et al. 2016, Ebbensgaard 2017, Ratcliffe 2021).

Spaces

Participants in this study specifically identified the large curved wooden benches as key features that stood out to them, suggesting that the seating area looked inviting (Appendix 6). Literature in garden design suggests seating is an important element (Naderi and Shin 2008, Erickson 2012, Douglas et al. 2017). Additionally, Kaplan and Kaplan (1989) suggest that the restorative environment should be compatible, providing opportunities to meet individual needs. Seating that is suitable for different users can provide the opportunity to sit and relax in the garden as well as provide chances for social interaction when desired (Harries et al. 2023).

Secondly, participants enjoyed the far-reaching views and suggested it evoked pleasant emotions including happiness and calm which supports previous findings. Specifically, one study found that environments with medium to high levels of open landscape and low levels of dense trees or planting were perceived as more restorative (Appleton 1984; Gatersleben and Andrews 2013). The finding may also be applied to the extent and fascination components of Attention restoration theory where the far-reaching views provide sufficient scope to sustain soft fascination (Kaplan and Kaplan 1989). It could also be that seeing vast views evokes the experience of awe, which has been found to support numerous benefits including positive mood and emotions (Piff and Keltner 2015, Scott 2016).

Additionally, some participants cited the curving paths as a key feature because the path made them want to travel up and explore. Both Attention restoration and Stress reduction theories include components related to this finding (Kaplan and Kaplan 1989, Ulrich 2002). Specifically, Attention restoration theory suggests that the environment should have extent with sufficient scope to sustain exploration. Stress reduction theory also includes deflected vistas such as curving paths to engage interest. Garden design literature advises that paths are one of the most important features and should be accessible with a suitable surface to accommodate different users (Ulrich 2002, Hussein 2010, Harries et al. 2023). It is recommended that paths should also link to all areas of the garden to provide a sense of direction and facilitate exploration (Bengtsson and Grahn 2014, Lau et al. 2014, Harries et al. 2023).

Limitations and future research

Although this study found statistical significance with medium to large effect sizes, the sample size was still relatively small with a limited demographic representation. Most visitors reported regularly visiting RHS Wisley more the once a month. The demographic of this study consisted mostly of white female participants with an average age of 53.6 years. Only 30% of the participants were male, and whilst this study was not concerned with addressing the perceived differences between sexes, it is important to acknowledge the limited demographic representation of this sample. It is not known how different sexes, ages, or sociocultural backgrounds may experience the garden and the features within, which future studies should address. Indeed, culture can have influences on environmental preferences, and there are very few comparative studies on cultural differences in garden preferences (Buijs et al. 2009). It is also important to consider how personality may impact one's experience or emotional reactions towards specific features within natural environments (Ambrey and Cartlidge 2017). Additionally, participants were asked if they were regular visitors of RHS Wisley or had professional horticultural experience. Whilst most were regular visitors, nobody reported professional horticultural experience. Participants in this study likely had a high affinity to gardens in general. Both horticultural professionals or individuals not regularly exposed to gardens may experience different emotional reactions and preferences towards the garden features. Future studies could examine a different demographic to compare the findings to the current study and add more knowledge of individual and cultural differences towards garden preferences.

Additionally, there may be methodological issues with the experiment being conducted outside. As with any nature-based research, the seasons, time of day and weather may influence experience (Duan and Li 2022). Participants in this study were asked if the weather impacted their experience, to which most reported yes, but this was a positive influence as the data was collected on mostly dry, sunny days. No participant completed the survey in the rain. Data collection was more challenging during the Autumn and Winter months, particularly as the weather got colder and visitors were reluctant to spend too long standing outside. Future research could specifically explore the differences in emotional reactions towards garden exposure in the Summer and Winter months. Additionally, this research was carried out in a botanical garden in the countryside. To expand this research and make it more specific to the design of green spaces in cities, future research could explore the possibility of an experimental garden in a public green space within a more urban setting.

Finally, this study used a small sample of specific emotions based on the circumplex model of emotions. This was to encapsulate each quadrant of the model whilst keeping the survey short. This limited selection of emotional responses may not have been a true reflection of what participants were feeling. Future research could test the results of the current findings by exploring a more diverse range of emotional reactions toward garden features.

Conclusions: informing garden design for wellbeing

The findings from this study provide a more detailed understanding of the psychological reactions linking specific characteristics within gardens to wellbeing. It has also provided detailed examples of features within a garden that may enhance pleasant emotional reactions and perceived restoration. The emotional reactions examined in this study are thought to contribute towards wellbeing aspects underlying the relationships between gardens and wellbeing. It is intended that this evidence will help support current design guidelines for wellbeing and therapeutic gardens. It is important to acknowledge that designing for human wellbeing is important but it is of course equally important to maintain consideration for nature when deciding what plants to include in the garden. Indeed, many of the aspects within gardens found to support human wellbeing, such as variety of planting and encouraging pollinators and wildlife into the garden, are also beneficial to local biodiversity (Waylen 2006).

Current findings suggest that key features in gardens linked to pleasant emotional reactions and perceived restoration include planting, sounds, and spaces. Specifically, planting is an important feature for evoking pleasant emotions and perceived restoration. Whilst plenty of greenery can be nice to see, it is important to include a variety of multisensory planting. This includes selecting planting pallets with different textures, colours, and scents (Chiang et al. 2017, Donelli et al. 2019, Harries et al. 2023). It could be that specific areas of the garden are dedicated to evoking specific senses and emotions. For example, if one intends to create a purely calming and relaxing garden, then one could avoid bright colours which may evoke excitement and instead focus on cooler planting pallets. Importantly, plants that attract wildlife are also essential for evoking pleasant emotional reactions as well as supporting the ecological integrity of the garden (Hoyle et al. 2017). However, it is important to consider this throughout the seasons to ensure that there is plenty of variety to arouse such emotions even in the Winter months. Additionally, incorporating the sound of running water for example, with use of streams or fountains, is an important feature to create serene and calming spaces (Largo-Wight et al. 2016, Harries et al. 2023). Another important feature identified was seating, which invites people to sit, rest, and enjoy the garden (Kok et al. 2013, Harries et al. 2023). Seating could be positioned next to planting designed to evoke the senses, near water features so one can enjoy the relaxing sounds or where possible to overlook views. Finally, the use of pathways that bend and curve around the space may help facilitate interest and encourage garden visitors to walk around and explore (Ulrich 2002, Harries et al. 2023).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/23748834.2023.2300235

Additional information

Funding

This work was supported and funded by Royal Horticultural Society [TV8399] and University of Surrey [KF5051B].

Notes on contributors

Bethany Harries

Bethany Harries is a postgraduate researcher in Environmental and Health Psychology. Her current research focuses on the pathways between nature, health, and wellbeing.

Lauriane Suyin Chalmin-Pui

Lauriane Suyin Chalmin-Pui is a postdoctoral research fellow at the Royal Horticultural Society and at the University of Sheffield.

Birgitta Gatersleben

Birgitta Gatersleben is a professor of Environmental Psychology at the University of Surrey. Her research focuses on sustainability, nature engagement and wellbeing.

Alistair Griffiths

Alistair Griffiths is the Director of Science and Collections at the Royal Horticultural Society. His research focuses on sustainability and human health through gardening.

Eleanor Ratcliffe

Eleanor Ratcliffe is a Lecturer in Environmental Psychology at University of Surrey. Her research focuses on restorative environments, including diverse sensory experiences, and place attachment.