Children’s perception of biodiversity in their school grounds and its influence on their wellbeing and resilience

ABSTRACT Evidence increasingly demonstrates nature engagement benefits. However, little evidence shows nature engagement provides children with a reflective perception of biodiversity, nor whether perception influences children’s wellbeing and resilience. We explored the impact of weekly nature engagement in schoolgrounds on 509 eight-to-eleven-year-olds’ biodiversity perception, wellbeing, and resilience. At the start and end of the academic year, wellbeing and resilience questionnaires were completed, and biodiversity perception was assessed by children drawing what they thought was in their schoolgrounds. Children initially perceived few organisms within easily visible taxa, and perceived more vertebrates compared to invertebrate species. After, children were more aware of taxa, resulting in a more reflective biodiversity perception. Children with initially low scores in wellbeing and resilience increased in these measures, and a positive association was found between increased invertebrates and vertebrates perception and improved resilience. Nature engagement within curriculum could reduce children’s extinction of experience and improve wellbeing and resilience.


Introduction
Engagement with the natural environment has been identified as having benefits to wellbeing and resilience in young people (Chawla & Gould, 2020;Flouri, Midouhas, & Joshi, 2014;Harvey et al., 2020).Knowing this, nature conservationists have emphasised the wider societal importance of conserving biodiversity in local areas (i.e.supporting positive wellbeing).However, little is known about how children's perception of biodiversity around them can affect their wellbeing and resilience.Here we assess whether children's changes in perception of local biodiversity lead to improvement in their wellbeing and resilience.
The United Kingdom has been identified as one of the most nature depleted countries in the world with 56% of all species listed in the 'State of Nature' report found to have declined over the past four decades (Hayhow et al., 2016;WWF, 2018).This reduction in biodiversity may intensify our already disconnected lifestyle from the natural world through the extinction of experience (Pyle, 1993), with children in the UK identified as having low nature connection (Hughes, Richardson, & Lumber, 2018) and poor ability to identify common species (Balmford, Clegg, Coulson, & Taylor, 2002;White, Eberstein, & Scott, 2018).
Despite most children between the ages of 5-18 years spending a large part of their week in school, and school grounds serving as an untapped resource for biodiversity (Harvey, Gange, & Harvey, 2019), the opportunity for children to explore native species within school lessons has reduced over recent generations (Barker, Slingsby, & Tilling, 2002;Lindemann-Matthies, 2002).Fieldwork in environmental sciences has particularly decreased in recent years (Tilling, 2018) with more textbook-based education leading to children developing a prototypical view of nature (Shepardson, Wee, Priddy, & Harbor, 2007;Tay, 2016).This view of nature does not encompass green areas in urban environments but is purely for plants and animals with no human interaction.Children growing up today, therefore, have an unclear perception of nature surrounding their places of living (Payne, 1998;Rickinson, 2001).
The decline in mental health and psychological wellbeing in young people has also been of great concern in recent years (Bird, 2007;Pitchforth et al., 2018) and is strongly related to other health and developmental issues in young people such as lower educational attainment and obesity (Patel et al., 2007;Sutaria, Devakumar, Yasuda, Das, & Saxena, 2019).It is increasingly recognised that nature is an asset and acts as a buffer to long-term mental and physical health implications from environmental diseases caused by anthropogenic activities (Dearborn & Kark, 2010;Frumkin, 2001;Shanahan, et al., 2015b).It has been suggested that contact with the natural world allows the synergistic health benefits of nature contact and increasing the possibility of social interaction to buffer mental illhealth and support mental restoration, whilst promoting low level activity for good physiological health (Cox, et al., 2017b;Dallimer et al., 2012;Han, 2017).
Nature focussed studies have also suggested a reduction in depression, anxiety, and stress when adults who visit green areas perceive high numbers of plant species and bird song (Cox & Gaston, 2018;Dallimer et al., 2012;Schipperijn et al., 2010).Moreover, green space has also been found to positively influence emotional and behavioural resilience in adolescents (Flouri et al., 2014).In fact, outdoor education has been shown to have a positive impact on the psychological wellbeing of children (Harvey et al., 2020;Pirchio, Passiatore, Panno, Cipparone, & Carrus, 2021), at a time when average life satisfaction in children in England aged 10-15 has been decreasing (Department of Education, 2019).
In this study, conducted over a full academic year, we extended previous work to assess whether children's perceptions of local biodiversity within their school grounds reflects what is present and whether this impacts their wellbeing and resilience.Uniquely, we recorded whether children's perception of biodiversity became a more accurate representation of what is present in their school grounds after taking part in our nature-engagement programme in their schools grounds across twenty-one hourly sessions (hereafter referred to as 'intervention group') by evaluating their drawings and assessing how this related to the biodiversity present in their school grounds through experimenter grounds' surveys.Looking to understand the role of engaging with biodiversity for benefitting wellbeing and resilience, we sought to recruit large groups of 8-to-11-year-olds.This allows us to assess the impact of the nature-engagement programme at ages around when it has been shown that children's life satisfaction is in decline (ages 10-15;DoE, 2019).Importantly, due to the scope of this work we focussed on gaining whole class involvement in the programme of work, and to enhance generalisability we focused on recruiting from a number of state and private schools.
Validated questionnaires were also used to gain an understanding of the children's wellbeing and resilience before and after the programme intervention.Wellbeing and resilience were investigated together as wellbeing captures a psychological state at a given time, whereas resilience considers past and future experiences to cope with a problem (Mguni, Bacon, & Brown, 2012).
It was hypothesised that children would have a poor perception of biodiversity in their school grounds.Specifically, compared to actual biodiversity, it was expected that children would perceive the abundance of mainly 'visible' taxa such as flora fairly consistent with actual abundance, while they would under-represent the abundance of invertebrate species and over-represent the abundance of vertebrate species (such as birds and mammals), before the activities.However, we hypothesised that after the intervention, children who participated in the programme would have a more reflective perception of biodiversity (corresponds better with observed levels) and as a result, would perceive more invertebrate species and fewer vertebrate species.Further, it was hypothesised that children who participated in outdoor education would receive positive benefits to their wellbeing and resilience.To ensure that any changes in perceptions and feelings were as a result of our intervention, we had a group of children who did not take part in the activities (the control group), who took the before and after measures with similar time gaps.Finally, it was hypothesised that children who participated in the activities would show a positive relationship between changes in wellbeing and resilience with increased biodiversity perception.

Participants
Children, ages 8-11 years old, from 10 schools (seven state and three private) across South East England (Surrey, Berkshire, and Middlesex) were recruited for the study.In total, 509 children participated in the program and 243 children were in the control group (intervention group, ̄ age = 8.81, SD = 0.68; control group ̄ age = 8.94, SD = 0.86).Schools in the three counties we were recruiting from were contacted initially with information about the project via heads of school.For those schools that agreed to take part in the research, information letters were sent to parents/ guardians via the school with consent forms.Informed consent was obtained for every child participating in the study, from both children and gatekeepers, as approved by the institutional ethical review process.Datasets for each child were anonymised.

Outdoor education programme
The outdoor education programme, completed by the intervention group, aimed to teach children about the biodiversity in their school grounds through a series of hands-on activities focussing on major taxonomic groups including trees, other plants (i.e.plants that are not trees), lichens and fungi, mammals, birds, insects, other invertebrates (i.e.invertebrates that are not insects), and amphibians.The activities were focussed around discovering and monitoring species (see Table A in Appendices for activities).The activities were developed to be appropriate for all participating schools, regardless of their size and ecological quality and linked to the primary science curriculum.Rather than covering species in blocks, we rotated through species groups to maintain interest and explore various taxa across the seasons.
The activities took place across one academic year (September 2018-June 2019) and comprised seven sessions per term, over three terms, equating to twenty-one sessions.Each session took one hour, and every school followed the same programme of activities each week across the full academic year.The sessions commenced with a short PowerPoint presentation delivered in the classroom to engage the children with the topic of the week and to ensure the delivery of the programme was standardised across activity leaders, who were the researchers, before going outside.

Biodiversity perception
Children's perception of the biodiversity in the school grounds was recorded at the beginning and end of the study.We discussed what nature meant with the children first, suggesting it was plants and animals, but not giving examples.The children were given five minutes to complete a labelled pencil sketch of the living things that they thought were in their school grounds.This time was chosen, in discussion with teachers, since in pilot studies it was found that this age group became distracted if a longer period was given.At the start of the activity, it was stressed that there were no right or wrong answers to the task, and that we wanted the children to draw what they thought was there.This same task was repeated approximately nine months later, at the end of the academic year, once the outdoor education programme activities had been completed by the intervention group, to assess if the children's perception had changed.Figure 1 displays an example of two children's drawings (one from the intervention group and one from the control group) of their perception of biodiversity in their school grounds, before and after the programme.
During the process of classifying the organisms in the children's drawings, grass was excluded, as well as in the biodiversity survey of school grounds since it might be affected by mowing and fertilisers, which undermines the representation of the true biodiversity.The number of items were counted within each category to account for species diversity.Drawing methodologies are quick, simple, and enjoyable widely used methods to evaluate children's perception of the environment (e.g.Drissner, Haase, Wittig, & Hille, 2014;Moula, Walshe, & Lee, 2021;Prokop & Tunnicliffe, 2008).

Biodiversity surveys of school grounds
Surveys of the school grounds were carried out using direct and indirect monitoring methods to record different taxa by the research team (as described in Table B of Appendices).These were completed within one month of the children's biodiversity perception measures (i.e.their before and after drawings), to estimate the biodiversity present for each school, allowing comparison between the research team's survey findings and the children's biodiversity perception drawings.To ensure the biodiversity surveys of the schools' grounds were comparable to the children's drawings, species and abundance within each category were recorded with the use of biodiversity monitoring techniques used by the children throughout the academic year during the programme.This ensured the children were exposed to all taxa identified within the surveys.The biological categories selected for monitoring were trees, other plants, fungi and lichens, mammals, birds, insects, other invertebrates, and amphibians, allowing for study across the academic year.
The KS-27 questionnaire was completed to gather an understanding of the children's perception of their physical and psychological wellbeing, autonomy, relationship with parents and peers, their social support and school environment, and finally overall life satisfaction at a given time.The KS-27 is a series of 27 questions on a five-point Likert scale from strongly disagree (1) to strongly agree (5).
The CYRM-12 explores the resources (individual, relational, communal, and cultural) available to aid in the children's resilience by overcoming adversity and sustaining wellbeing over time.It is a series of 12 questions on a three-point scale: (1) no; (2) sometimes; (3) yes.

Data analysis
All questionnaire measures were scored using the existing protocols as detailed in section 2.3 and 2.5 for both control and intervention groups.Before and after scores for biodiversity perception, wellbeing and resilience were compared in the intervention and control groups.
There was some attrition due to children not being available during our visit and respondents were able to skip questions, meaning some children did not fully complete the measures.For the wellbeing questionnaire (27 items), only children who responded to at least 80% of the items, where we replaced their missing values with the rounded mean of the other items ( ̄ completed items = 26.0;SD = 1.4), were included in the analyses (Vélez et al., 2016;Ronen et al., 2014).For the remaining measures, resilience questionnaires (12 items) and the biodiversity perception drawings from both time points, only children with 100% completion were included in analyses.
All statistical analysis was carried out using R 3.5.3(R Core Team, 2019) and graphs were made on GraphPad Prism 5 (GraphPad Software, 2010).The biodiversity perception data were analysed with Wilcoxon signed-rank tests with Bonferroni correction (to control for multiple comparisons) to look at the number of drawn organisms within each investigated category before and after the activities.The difference in the children's perception of biodiversity compared to surveyed biodiversity before and after the activities was examined (perception likeness).Mann-Whitney-Wilcoxon tests were used to explore the difference between before and after results of the control and intervention groups.Spearman rank correlations were also used to explore whether there was any association between the children's wellbeing or resilience and the taxa groups (i.e.trees, other plants, lichens and fungi, mammals, birds, insects, other invertebrates, and amphibians) they perceived within their school grounds.
The effects of school, age cohort and collection point (before and after) on children's wellbeing and resilience questionnaires were examined with generalised linear mixed models to account for influential random effects of individuals with Poisson error structure using the R package lme4 (Bates, Maechle, & Bolker, 2014).Interaction and statistical significance were gathered by removing terms sequentially and testing the significance between the models until a minimum model was achieved.Tukey post-hoc tests were used to test the significance of each pairwise comparison.Additionally, Wilcoxon signed-rank tests were performed on the data of children who had low wellbeing and resilience baseline scores.Finally, Spearman rank correlation coefficient was used to investigate whether there was an association between the categories of organisms investigated and their scores from each questionnaire before and after the activities.

Perception of biodiversity
Initially, the intervention children had a poor awareness of biodiversity present within their school grounds, drawing on average 3.07 (SD = 1.47) organisms (Figure 2).Additionally, the children perceived more trees than any other species, and more vertebrates (mammals, birds, and amphibians) than invertebrates (insects and other invertebrates) (Figure 2).After the intervention, children who participated in the activities drew significantly more organisms for each explored taxa, except for trees and other plants (displayed in Table 1) and the number of invertebrates (insects and other invertebrates) perceived increased by 297%, while vertebrates increased by 184% (mammals, birds and amphibians).
As displayed in Figure 2, although the control group had a significantly greater overall baseline (W = 21,534, p < 0.001) due to drawing more easily visible taxa (e.g.trees and other plants), the intervention group drew significantly more organisms than the control group at the end of the study (W = 11,882, p = 0.01).In the baseline data, the control group perceived more organisms within the individual taxa groups of trees, other plants, birds, and amphibians than the intervention group.Yet, after the intervention this remained only the case for plants (W = 19,740, p < 0.001).Importantly, following the programme, the intervention group drew significantly more insects, other invertebrates, lichen and fungi, and mammals than the control group, whilst there was no difference between the more obvious visible taxa of trees, birds and amphibians.Additionally, there was a significant increase in the total number of species that the intervention group drew (V = 1750, p < 0.001) within all taxa categories, except trees and other plants, which showed no change.No change was seen in the control group overall (V = 1479.5,p > 0.05), nor within individual taxa groups (see Table 1).
Children's perceptions of biodiversity are compared to the research team's biodiversity survey findings (detailed in section 2.4) to identify perception likeness.Table 2 showed that the intervention group's perceptions significantly improved following the intervention (see Figure 3(a)).There were no such improvements for the control group (Table 2).
A positive association between wellbeing and perception of trees was found in the intervention group children who had heightened perception of tress in their schools grounds after the programme (rho = 0.22, p < 0.05).Children who perceived fewer insects (rho = -0.20,p < 0.05) and whose perception likeness score was low for insect distribution (rho = 0.015, p < 0.05) after completing the set of activities had more positive wellbeing.Comparably, although no relationship was found for those with initially low wellbeing scores, children who were identified to have low resilience initially had a positive association between their improvement in resilience and increased perception of other invertebrates (rho = 0.60, p < 0.05) and mammals (rho = 0.51, p < 0.05) after the intervention.

Discussion
In line with other research (Harvey et al., 2020;Tillmann, Tobin, Avison, & Gilliland, 2018), this study has shown that engaging children with nature is associated with benefits to children.The current research demonstrates that gains in awareness, through a hands-on outdoor education programme, of the biodiversity in the school grounds is associated with improvements in children's wellbeing and resilience for those who have initially low scores and therefore, most in need of support.This work uniquely demonstrates that the benefits of such an intervention go beyond this to improving pupil awareness of biodiversity, and that such knowledge can be accrued by spending a short period of time outside in school grounds, over the period of an academic year.
Our results show that children generally have a poor perception of the presence and abundance of local flora and fauna, with most children mainly drawing more visible tree and charismatic vertebrate species.However, once children have participated in outdoor learning, they have a greater awareness of the various taxa and have a more reflective perception (a greater correspondence with observed levels) of biodiversity found in the local natural environment; while the control group did not.
Within the intervention group drawings, the perception of birds did not change but their awareness of nocturnal mammals rose, in part, no doubt due to our use of camera traps placed in the grounds; the control group's perception remained largely unchanged.The change in the intervention group's perception demonstrates for the first time that children show little awareness of what they cannot see and that it is important to include methods to facilitate observation of 'invisible' local biodiversity within programmes.Moreover, initially, the intervention children's perceptions of the biodiversity in their school grounds poorly represented what was actually present, despite children spending long periods of each day within the school grounds.This strengthens the case that children have a poor awareness and understanding of local flora and fauna (Shwartz, Turbé, Simon, & Julliard, 2014;White, Eberstein, & Scott, 2018).Yet, after the intervention, children's perceptions were more representative of local biodiversity composition, with a greater variety of species, including the less 'visible' insects, other invertebrates, and amphibians.Therefore, utilising school grounds throughout their curriculum is a valuable resource to tackle children's extinction of experience.
Importantly, our work highlights that nature engagement intervention programmes have an impact on children who initially have low wellbeing and resilience.Previous studies have identified a similar trend that various contact with nature can positively influence young people's selfconfidence, self-esteem, stress, and attention restoration (Flouri et al., 2014;Roberts, Hinds, & Camic, 2019); thus, they are likely to positively impact wellbeing and resilience.Many of the previous studies have involved assessing the impact of children engaging in activities centred around adventure games, focussing on the presence of neighbourhood nature, as opposed to engaging with the biodiversity in their environment.Additionally, many were not supported by control group data and did not look at relationships with changes in wellbeing and resilience together (Berger, 2008;Doucette, 2004;McArdle, Harrison, & Harrison, 2013).Our work builds the evidence base of the benefits of nature through directly exploring how increasing children's awareness (perception) of the biodiversity within their environment, with hands-on engagement with biodiversity, is related to positive changes in wellbeing and resilience.
Notably, those who initially have low wellbeing and resilience scores also showed significant improvements in their perception of biodiversity.This was achieved over a short time frame: only twenty-one hourly sessions across the academic year.Further, for those with initially low levels of resilience, there was a positive association between their increased perception of other invertebrates and mammals and improvements in resilience scores.These findings are some of the first to identify that certain groups of taxa may be beneficial in supporting the wellbeing and resilience of children and that a varied biodiversity is necessary to ensure individuals can have both supported.
Although children had a more reflective perception of biodiversity after the intervention, the charismatic larger taxa of mammals and birds were still over-represented compared to local abundance; and insects and other invertebrates remained under-represented.This finding supports previous investigations that identified children perceive larger animals, such as vertebrates, as more prevalent compared to invertebrates (Snaddon, Turner, & Foster, 2008).Yet, within the current research, the distribution of perceived vertebrates changes to encompass more nocturnal vertebrates, such as badgers, and demonstrates that providing environmental education opportunities can reduce the lack of awareness children have for the local environment and help reduce the extinction of experience (Pyle, 1993), which may have wider implications for the environment and conservation.Although a greater awareness of the natural environment may not encourage proenvironmental behaviours, poor awareness and perception of biodiversity can make conservation efforts challenging (Balmford et al., 2002).An understanding and awareness of local flora and fauna are crucial to foster the next generation of environmentalists (Richardson et al., 2019).
The finding that children focus on organisms that they can see within nature may also be a reflection of what they see in the media and in books from an early age (Genovart et al., 2013;Titley et al., 2017).Children's books show that 60% of animal books have mammals and 18% have birds within the title (More, 1979).Clayton and Myers (2011) have argued that perceptions are formed throughout a lifetime and are strongly reliant on memory as well as the present and future experiences people have.Therefore, these past childhood experiences can enforce the distorted perception children have of the local environment (Clayton & Myers, 2011).It is therefore vital, if children are to become engaged with their local wildlife and in particular the more invisible invertebrates and nocturnal species, that they experience it within their daily lives by actively incorporating outdoor learning experiences into the curriculum (Waite, 2010).It is in this way that they can gain the wellbeing and resilience as well as the knowledge and engagement of local biodiversity that will endure (Richardson et al., 2019).
As the children's perception of insects and other invertebrates increased after the activities, we identified that children with initially low resiliency benefited from learning about these species which may influence their future wellbeing (Mguni et al., 2012).The results in the present study suggests that interacting with nature and perceiving higher species richness of both vertebrates and invertebrates can have psychological benefits.Importantly, enhancing biodiversity in school grounds and providing opportunities for contact with it can support children with poor resilience; building resilience may benefit wellbeing in the longer term due to previous research identifying resilience to positively impact subjective wellbeing (Liu, Wang, & Lü, 2013).
Previous studies have identified that high species richness of plants can influence the wellbeing of green space visitors (Fuller, Irvine, Devine-Wright, Warren, & Gaston, 2007) and a level of 20-30% vegetation coverage in urban areas can reduce stress and anxiety prevalence (Cox & Gaston, 2017).Extending the aforementioned earlier work, we identified that children who perceived higher levels of trees than present tended to have higher wellbeing and resilience scores.This supports previous findings that the presence of trees and green space can support mental health and wellbeing of adults (Guan, Wei, He, Ren, & An, 2017) and behavioural resilience in adolescents (Flouri et al., 2014).
This paper contributes to the gap in knowledge as to whether a reflective perception of biodiversity influences children's wellbeing and resilience and which taxa influence them.The findings support the concept that merely offering a green space (such as a playing-field for exercise) may not support children's wellbeing and health adequately.In consideration of the key strengths and potential limitations of this work, it must first be noted that we have used novel approaches to assessing biodiversity and children's perceptions of this.We assessed children's recognition of biodiversity within their school environment through drawings, which was an engaging task for the children and allowed us to understand what they perceive in an open way.One consideration is that as children were given a limited time and space to draw their perception of local biodiversity, some children may have drawn what came first to their minds and may not have completed the task fully.However, this time limit was suggested by teachers and was supported by pilot work, carried out by us but not reported here.Moreover, we saw that there were differences in the intervention and control groups thus, indicating that children did attend to the task at hand.To confirm and extend our findings, follow-up interviews or the use of a quantitative scale may be appropriate in future research.Examining the children's drawings more fully to explore the anatomical detail of their diagrams and detail of species names would also benefit future research, exploring further children's knowledge and awareness of local biodiversity composition.An additional consideration to be explored in future work would be the role of nature connectedness as a potential mediator of the relationship we have shown here between biodiversity perception and wellbeing and resilience.Research has supported that there is a relationship between connection to nature and feelings of happiness (see the meta-analysis by Capaldi, Dopko, & Zelenski, 2014).

Conclusion
A poor perception and awareness of the diversity in nature is found in children, largely with an inflated perception of vertebrates compared to invertebrates.This study contributes towards the understanding of the influence outdoor education, centred around nature engagement, can have on children.It highlights the importance of fieldwork during their education to enhance their awareness of local flora and fauna, and how this fieldwork and perception of the environment may support children with low wellbeing and resilience.We have shown that engaging with nature over a short time can positively influence those with low wellbeing and resilience and improve awareness of local biodiversity.Combined with previous findings, we expect that enhancing opportunities to learn about and engage with nature throughout the school curriculum can lead to children having better wellbeing and resilience, particularly for those who are most in need of mental health and wellbeing support.

Figure 1 .
Figure1.Examples of biodiversity perception drawings for two children (both 9 years old), one from the intervention group and one from the control group, before and after the programme of activities.

Figure 2 .
Figure 2. Mean number of drawn taxa before/after the activities by intervention and control groups.Error bars represent Standard Error of the Mean (SEM).

Figure 3 .
Figure 3.The percentage of children's perception of biodiversity compared to actual biodiversity present in school grounds in the (a) intervention group and (b) control group children's drawings before (B) and after (A).Error bars represent SEM.The ± represent the larger proportion of children over/ under perceiving the biodiversity present in school grounds.

Figure 4 .
Figure 4. Mean percentage change of wellbeing and resilience scores in all intervention group children and in intervention group children that had initially low baseline scores.Error bars represent SEM.

Table 1 .
Assessing significant change in the number of drawn organisms within each category for the intervention and control group before and after the intervention.

Table 2 .
Assessing significant change in biodiversity perception likeness to actual biodiversity in school grounds for the intervention and control group before and after the intervention.

Table 3 .
Mean scores of the wellbeing and resilience measures for intervention and control group children.