Nature Through Young Eyes: Exploring Children’s Understanding of Nature in Urban Landscapes in Beijing, China

Abstract

Rapid urbanization over the last few decades has resulted in children growing up in cities increasingly disconnected from the natural world. A better understanding of children’s conceptions of nature in cities is needed to try to address this unfortunate trend. This study was conducted within four core districts of Beijing, China, and involved interviewing 81 children aged 8 to 12 years to explore their general, conceptual understanding of nature and accessible nature and how it manifests in their daily lives. Some of these children (N = 54) were involved in a follow-up photo Q-sort task aimed at investigating their perceptions of nature in the city, which required them to sort 33 stimuli photographs in piles according to their perceived “naturalness”. The interview data were content analyzed, while the photo-sorting data were subjected to various statistical analyses, including Multidimensional Scaling (MDS). These analyses identified three distinct ways children perceive nature in the city, reflecting ecological, emotional, and visual values for nature in the city. The findings highlight the influence of personal experiences, education, the media, and evolutional predispositions on children’s constructs of nature. The findings can guide environmental design and education actions aimed at integrating nature in urban settings that align better with the children’s conceptions of nature.

1. Introduction

Children growing up in cities today are increasingly losing their sensitivity to and connection with the natural world, which is being driven primarily by rapid urbanization and land use changes, resulting in reduced access to natural environments [1,2]. Changes in lifestyles, and the transformative effects of technology on children’s conceptions of nature, also play an important role. Children’s leisure time has increasingly shifted from direct experiences with nature gained through outdoor play and exploration, an effective way for children to develop meaningful connections with nature, to more sedentary, indoor activities. This shift has been correlated with a decline in children’s physical and psychological well-being, as well as their cognitive and social development—what has been referred to as “Nature-Deficit Disorder” [3]. The potential consequences of this trend highlight the necessity for a better understanding of the relationship between children and nature—how they understand, perceive, and interact with nature, especially within built urban environments where direct access to nature may be limited, in order to reverse this trend. The study reported here is an effort toward achieving this understanding.

Many studies have explored the relationship between children and nature. A key focus has been the impact of nature exposure on children’s health and well-being, aligning with the notion of biophilia, which represents an innate emotional connection between humans and other living organisms [4]. Two major theories have then been proposed in the literature of environmental psychology—Stress Recovery Theory (SRT) and Attention Restoration Theory (ART). SRT proposes that (visual) exposure to nature could quickly elicit positive emotions and block negative ones, thereby reducing blood pressure and stress hormones [5]. ART, pioneered by Kaplan, distinguishes two types of attention—directed and involuntary attention. Directed attention is needed to maintain a high level of performance in people’s day-to-day functioning and decreases when overused, while involuntary attention requires no effort. In this theory, nature is often regarded as rich in (soft) fascination that quickly elicits involuntary attention and simultaneously allows directed attention to be recharged [6]. Subsequent research has employed quantitative indexes of nature exposure derived from satellite, parent-reported, or self-reported data or by evaluating the quality of natural features through on-site measurements, providing empirical support for the positive impacts of passively viewing, immersing oneself in, or actively interacting with environments rich in natural elements on children’s cognitive development and psychological well-being [7,8,9,10,11,12]. However, the increasing disconnection of children from nature in cities has deprived them of this source of soft fascination and the benefits it provides.

Children’s sense of environmental stewardship is another important area of research. Some studies have examined the impact of natural/environmental education on children’s pro-environmental behaviors, with many educational programs incorporating hands-on, nature-based activities [13]. Some research has also explored the influence of connection to nature and the mediating effects of environmental knowledge and attitudes in these processes, with varying effects ranging from medium to weak being identified [2,14,15,16]. These variables have primarily been assessed using standardized scales that include abstract statements, such as “I often feel a sense of oneness with the natural world around me”, in assessing children’s connectedness with nature [2]. However, few studies have investigated how children conceptualize and interpret the notion of nature itself—how they define the concepts of “nature” or “natural”, how they reason about nature, and how they make sense of human–nature interactions.

There is also a lack of research on the impacts of urbanization and human activities on children’s perception and understanding of nature. As Orr [17] (p. 282) put it, children “on average can recognize over 1000 corporate logos but only a handful of plants and animals native to their places”, reflecting a worrisome trend toward what has been termed “biotic impoverishment” [18]. For many children living in cities today, direct exposure to nature is primarily limited to nearby parks, green/blue spaces, and other types of places where “urbanized nature” can be found, including artificial and semi-natural environments that nonetheless may possibly be perceived as natural [19]. However, opportunities for virtual contact with imagined and exotic manifestations of nature have increased dramatically due to technology, leading to a neglect of perceptual, affective, cognitive, and behavioral interactions with actual and local sources of contact with nature, or what has been called the “extinction of experience” [20]. This can impact children’s cognitive and affective responses toward nature, which, in turn, can influence their motivation for engaging in nature-based activities and adopting pro-environmental attitudes and actions [21].

Reintegrating nature into children’s daily lives in cities is essential for reversing this trend, but first, it is necessary to better understand how children conceptualize and perceive nature in urban contexts. The environmental psychologist Wohlwill [22] (p. 7) succinctly defined the notion of “nature” as broadly dealing

“… with the landscape rather than with the built environment. It includes the world of rock and sand, of shoreline, desert, woods, mountains, etc., and the diverse manifestations of plant and animal life that are encountered there. It excludes the man-made world: our cities and towns, our houses and factories, along with the diverse implements devised by mankind, for transport, recreation, commerce, and other human needs.”

However, children’s understanding of nature might differ due to their cognitive development, direct experiences of nature, and the type of environment in which they grew up. Personal construct theory, initially proposed by the psychologist George Kelly, provides a valuable lens in this regard, emphasizing that individuals actively interpret and make sense of the world through a system of personal constructs developed through experiences, typically through templates of bipolar dimensions of connotative meaning, such as, for example, nasty–nice, clean–dirty, natural–unnatural, and other dimensions of meanings [23]. Rather than relying on fixed definitions of nature, children develop their own constructs associated with their understanding of what nature is and is not. From this perspective, two research questions are addressed in this study:

(1) How do children growing up in urban areas conceptualize nature, including conceptual nature (nature in the abstract) and accessible nature (that they encounter in their daily lives that they consider to be natural)?

(2) How do children perceive the nature that is accessible to them in their daily lives?

In exploring these questions, this study sought to address the abovementioned gaps and deepen the understanding of how children, particularly in urban environments, conceptualize and perceive nature by employing both verbal (interview) and perception-based (photo-sorting) methods. In addition to the academic contribution, this study also provides insights that can inform environmental design actions and educational interventions for children to foster more meaningful connections with the natural world.

2. Methods

2.1. Research Site and Sampling

This study focused on children aged 8 to 12 years living in four core administrative districts of Beijing—Haidian, Chaoyang, Dongcheng, and Xicheng. These areas had the highest level of Gross Domestic Product (GDP, Figure 1a) and density of primary school children (Figure 1b) while also having the lowest amounts of public open green spaces per capita (Figure 1c), according to the Beijing Statistical Yearbook 2021. This disparity underscores tensions between urbanization, economic development, population density, and children’s access to nature. Children involved in this study are in an age group where they increasingly become curious about their home range environments and acquire better comprehension and communication skills and, for many children, an increasing appreciation for the natural world [24]. These factors also enabled them to participate more effectively in interviews and photo-sorting tasks. The children were recruited via social media advertisements, posters, and a snowball sampling approach through personal networks. Parental consent was obtained before involving the children in data collection, and the study protocol was approved by the University of Melbourne’s Human Research Ethics Committee (Reference No. 2021-22465-24141-3).

A total of 81 children participated in the first part of the study, which was focused on exploring their understanding of conceptual and accessible nature in their daily lives. The child respondents included a nearly even balance of genders (42 girls, 39 boys) and were evenly distributed by age. For anonymity, each child was assigned a unique identification number prefixed with “C”. In total, 54 children (32 girls and 22 boys) who participated in the first part of the study also participated in the second part. Figure 1d shows the geographic locations of the research site and the child participants.

2.2. Data Collection and Analysis

2.2.1. Part 1—Understanding of Nature

This part of the research investigated children’s understanding of nature by allowing them to freely express themselves about what nature is to them without the researchers imposing any predefined definitions of the notion of nature. Data collection involved two tasks incorporated in one-on-one semi-structured interviews, where children were firstly asked to close their eyes and visualize what nature is or what elements of the natural world they have in their minds—conceptual nature. The interview questions included, for example, “Can you describe what you think of when you hear the word nature?” and “What kinds of places or things are part of nature?”. Probing techniques for interviews proposed by Zeisel were also used, which have been widely used in environmental and behavioral research [25]. These probes were aimed at encouraging the children to fully express themselves and elaborate on their responses (depth probes) without the researchers biasing their answers. For example, when a child mentioned “forest”, the researcher would follow up with questions, such as “What can you find in the forest?” or “Can you think of anything else?”, to encourage the child to provide more detailed and in-depth responses.

The second task required them to describe what, among the many things they encounter in their daily lives, they consider to be nature—accessible nature. These questions included, for example, “Do you think there is anything in your daily life that you consider to be part of nature?”, “Are there any things or places around your home or school that you would call nature?”, and “When was the last time you felt like you were in nature and what was there?”. A similar probing technique was also adopted as used in the first task.

The interviews took approximately 15 to 20 min per participant. Qualitative data obtained through interviews were input into the NVivo 20 program for conducting content analysis using inductive and deductive coding processes to identify similarities and differences between these two aspects.

2.2.2. Part 2—Perceptions of Nature in the City

A photo-based Q methodology was used to explore the second research question concerning children’s perception of nature in their daily life. The Q methodology has primarily been employed in small-sample studies to explore people’s attitudes, perceptions, and values [26,27,28]. It was traditionally used by psychologists, performed using a set of verbal statements (the Q set), in which participants were asked to sort them along a continuum of the construct of interest being tested, with their responses analyzed to identify patterns. Pitt and Zube [29] were the first to adapt this method for use in assessing people’s perceptions of landscapes using photographs as stimuli, with the reliability and validity of the results being well-supported by subsequent studies [30,31,32,33,34]. This photo-sorting also has the advantage of eliciting personal constructs and related connotative meanings associated with people’s perceptions of complex landscapes, which aligns well with the purpose of this study [35].

The photo Q set. To develop the set of photographs used as stimuli for the Q methodology, the researchers initially took numerous photographs across diverse sites in Beijing, including residential areas, city and national-level parks, and a variety of other types of environments children might encounter in the course of their daily lives. All photographs were taken with the same Canon EOS 200D II camera (Canon Inc., Tokyo, Japan) under similar weather and lighting conditions (clear sky between 10 a.m. and 11 a.m. or 3 p.m. and 4 p.m.) in May 2023. The photo selection criteria included context relevance and diversity of depicted content elements to capture a range of environmental settings the children might be likely to encounter. The researchers also consulted various environmental psychology and landscape design experts in selecting the final set of 33 photographs that were used as stimuli (labeled S) in the study. All of the photographs were printed in color and mounted on 13 cm × 9 cm cards—large enough for children to observe details yet easy enough to handle and be manageable to administer in the field.

Procedure. The 54 child participants were asked to independently sort the 33 photographs into nine piles arranged from natural to unnatural. The numbers of photographs in each pile were prescribed to reflect a normal distribution [29]. A diamond-shaped normal distribution template was adopted, where the children were asked to place the most natural environments at the top and the least natural ones on the bottom [36]. The template, which was printed on durable fabric, defines a specific number of photographs (1, 2, 4, 6, 7, 6, 4, 2, 1) to the nine piles, which are ranked in values from −4 to +4. Before each interview, the photographs were randomly distributed in front of the child respondent, and they were instructed to initially categorize the 33 photographs into three groups: most natural, most unnatural, and neutral. Next, they had to place the photographs in the different piles, working from the top and bottom ends to the center. They were then asked to verbally explain the environments depicted in the seven photographs they had placed in the top and bottom places that corresponded to what they thought were the most and least natural environments depicted in the photographs.

Analysis. Multidimensional Scaling (MDS) was used to analyze the naturalness rating data from photo Q-sorting method. MDS is an exploratory analytical method well-suited for analyzing sorting task data, as it can reveal latent perceptual structures without imposing predefined categories. It also provides a visual representation through plotting spatial configurations of the stimuli elements that helps with visualizing cognitive maps of children’s environmental concepts in a way that reflects both proximity and relational meaning compared to clustering methods, which often create discrete, non-overlapping categories [37,38]. Therefore, MDS was considered well-suited for exploring the underlying dimensions of meaning, or personal constructs, in which the children discriminated between various environmental stimuli settings as depicted in the photographs and associated attributes [30,39]. The data matrix created from the photo-sorting data was subjected to Classical MDS (CMDS) using the alternating least squares scaling algorithm (ALSCAL) in SPSS 22 software [40]. Because the data in the matrix are ordinal, the Euclidean distance was used. Several MDS models, from 2 to 4 dimensions, were initially generated. An examination of the fitness values of each model suggested that the three-dimensional solution was the best based on its low stress value of 0.051, high RSQ of 0.989, and better interpretability [41].

The data matrix was also subjected to a hierarchical cluster analysis using the between-group method and the stimuli photographs as variables. Examination of the cluster analysis dendrogram allowed the photographs of environmental settings to be categorized into groups of similarly perceived elements. Six meaningful clusters were identified based on interpretability and labeled as “closed vegetative landscapes”, “open vegetative landscapes”, “nature in orderly frames”, “mixed built landscapes”, “urban leftover spaces”, and “human-dominated landscapes”. Combined with content analysis of the after-sorting interviews, it further facilitated the interpretation of the underlying perpetual dimensions apparent in the MDS stimulus configuration plot [30,42].

3. Results

3.1. Understanding of Nature—Conceptual and Accessible

This section reports the results of the first part of the study, which examined how the children conceptualize and understand nature in general (conceptual) and nature in the city (accessible). Their answers were divided into living and non-living manifestations of nature and scenarios.

3.1.1. Living Organisms

Almost all of the children mentioned living organisms when describing both conceptual nature (N = 78) and accessible nature (N = 79). Their descriptions of living organisms were categorized into three groups, flora, fauna, and fungi, broadly corresponding to the roles of producers, consumers, and decomposers within ecosystems (

Table 1. The proportions of child participants who mentioned each category of nature.

Categories	Sub-Categories		Conceptual Nature	Accessible Nature
Living organisms	Total
	Flora	General flora
		Specific flora
	Fauna	General fauna
		Specific fauna
	Fungi
Non-living entities	Total
	Natural elements
	Natural phenomena
Scenarios	Total
	Human-involved
	Nonhuman

Note: = 0%, = (0%, 12.5%], = (12.5%, 25%], = (25%, 37.5%], = (37.5%, 50%]; = (50%, 67.5%], = (67.5%, 75%], = (75%, 87.5%], = (87.5%, 100%].

).

Flora. The children most frequently mentioned plants when defining nature, including both conceptual (N = 74) and accessible nature (N = 75). Interestingly, “tree” was the most frequently mentioned feature, both for conceptual (N = 54) and accessible nature (N = 64). They typically used terms for plants/trees more generally in their descriptions, with few of them being able to name particular plant types or species. This, to some extent, reflected a form of plant blindness [43], especially when compared to their understanding of animals, as discussed below. When describing conceptual nature, only four children mentioned specific plants, such as apple and pine trees. Interestingly, one child referenced the Corpse Flower (Rafflesia arnoldii), an exotic species with unique biological, ecological, and sensory properties. In comparison, 21 children named specific plant types in describing accessible nature, including willows, poplars, pine trees, gingko, and other trees commonly seen in Beijing. They also identified environmental settings where trees grew, such as at their schools, residential areas, or parks, with which they were familiar. Some children described experiences of climbing trees, collecting leaves, and other activities, or simply seeing them, bringing them into contact with nature. As C67 (a 10-year-old girl) said: “I especially like the ginkgo woods in the park. Every autumn, the ginkgo trees turn yellow and fall. The earth seems to be covered with a blanket, a thick yellow blanket. I always want to pick up ginkgo leaves in the park.”

Fauna. Animals were another frequently mentioned category in describing conceptual (N = 68) and accessible nature (N = 67). Children often used more general terms to describe animals in their notions of conceptual nature, while significantly more children could identify different types of animals when describing daily accessible nature (N = 60). Birds, mammals, and insects were the most frequently mentioned animals, although there were subtle differences in mentions of animals between conceptual and accessible nature.

Birds are visually appealing to children and contribute to sensory experiences of nature (e.g., the sounds of birdsongs), creating a multisensory connection to nature. The birds mentioned most in describing accessible nature were sparrows, magpies, woodpeckers, swallows, ducks, and pigeons, which were accessible to them in their everyday environments. This is in contrast to the more abstract depiction of conceptual nature, in which penguins, hummingbirds, and other exotic animals were mentioned, in addition to the more commonly seen birds.

Insects were mentioned in the descriptions of conceptual (N = 20) and accessible nature (N = 27). Half of the children identified specific types of insects, and some reported experiences of catching butterflies, dragonflies, and other insects and expressed a fondness for certain insects. As C74 (an 8-year-old boy), for example, said: “I like insects because they’ve been on the earth for over a billion years, and there are so many of them. I often observe insects… I especially like praying mantises because they can capture insects much larger than themselves, even birds.” In contrast, some children lacked knowledge about insects or feared them, sometimes feeling disgust toward insects, a biophobic response that may be rooted in an evolutionarily innate predisposition to avoid harmful creatures or negative experiences with them (e.g., painful bites or stings) [4,44]. Their first-hand observations and sensory experiences likely contribute to the perception of insects as integral parts of nature.

Mammals stood out as the category of animals where children most diverged when describing conceptual versus accessible nature. While slightly more children mentioned mammals when describing accessible nature, the diversity of mammals referenced in conceptual nature was three times greater. When describing accessible nature, they frequently mentioned animals common in the city—cats and dogs (primarily kept as pets), as well as other urban wildlife, including hedgehogs, weasels, squirrels, and rabbits that can be seen in parks and open spaces in residential neighborhoods. In contrast, larger mammals, such as elephants, rhinos, dolphins, and whales, were frequently mentioned in their descriptions of conceptual nature, which contained iconic types of “wild animals” that they are likely exposed to through the internet, cultural narratives, educational materials, and various forms of media that portray a version of nature that is more diverse, larger, and wilder than what is accessible to the children in Beijing. The children can experience some of the animals they mentioned in zoos and museums, as well as by visiting national parks and wild areas.

Some of the children described reptiles. Only one child mentioned turtles in accessible nature, while several referenced crocodiles and snakes, and one child even mentioned dinosaurs when describing conceptual nature, highlighting the potential influence of media and education on their perceptions of what nature means to them.

Fungi. Fungi are often relatively small, primarily found underground or low to the ground, and visible at certain times of the year; hence, they are not easily observed or as captivating as plants and animals. Only one child mentioned fungi when describing conceptual nature.

3.1.2. Non-Living Entities

The children also mentioned non-living entities when describing conceptual (N = 67) and accessible nature (N = 45), categorized into natural elements and phenomena. Natural elements are relatively stable and long-lasting, such as the sun and mountains. Natural phenomena, such as weather, involve dynamic processes or transient events. More of the children mentioned natural elements than natural phenomena.

Natural elements. Non-organic natural elements were mentioned by 66 and 44 children in describing conceptual and accessible nature, respectively. These included land-based geomorphological features (e.g., mountains and deserts) and water-based elements (e.g., rivers and seas). These are visually distinctive features, and while they were frequently mentioned, they would not be easily accessible to the children in their everyday experiences in the areas in Beijing that were the focus of this study. For example, as C71 (a 10-year-old girl) mentioned, “I absolutely love the ocean. It gives me a very unique and dreamy feeling, though I haven’t had the opportunity to visit the seaside.” Their identification of these landscape features was likely due to the influence of education and media on their conceptions of nature. Nonetheless, many children also mentioned direct experiences with some of these natural elements through visiting a forest, hiking in the mountains, or playing by a river, given that Beijing is surrounded to the west by mountain ranges featuring lakes and rivers. There are some large parks in Beijing, such as the Olympic Forest Park and the Summer Palace, in addition to many small and medium-sized community parks. They can also explore other cities, including those in coastal areas, when on vacation, which may shape their conceptions of nature.

The children also mentioned atmospheric/celestial elements, with the sky being the most frequently mentioned in their descriptions of both conceptual and accessible nature, followed by the sun, clouds, and the moon. This might be explained by the visual dominance of these elements and the sensory experience they can afford, e.g., the vast expanse of the sky, the vibrant hues and transformations of clouds, the brilliance and warmth of the sun, and the luminosity and phases of the moon. This was illustrated by C61’s (an 8-year-old boy) vivid description of his experiences of observing clouds: “Sometimes during free play in PE class, I daydream and watch the clouds. Initially, it looked like a dinosaur, but slowly it turned into a cat.”

Natural phenomena. A few of the children also mentioned natural phenomena in describing conceptual (N = 14) and accessible nature (N = 6). The majority mentioned weather events, including rain, rainbows, thunder, lightning, and snow, and a few identified invisible or indirectly observable phenomena, such as the wind, which engages multiple senses, providing them with sensory-rich experiences. As described by C39 (a 10-year-old boy), he could “see the leaves swaying and feel the wind on the face”, and C06 (a 9-year-old girl) could “hear the sound of rain, feel the touch of raindrops and smell the scent of rain”. Only two children mentioned the seasons when describing conceptual nature. C39 provided a remarkably vivid description of the changing seasons: “In spring, there are swallows and newly grown grass, creating a lively scene with clear skies, a gentle breeze, and a sense of vitality. In summer, trees are lush with leaves, accompanied by the sound of cicadas. Occasionally, it might rain, and the weather tends to be hot and humid. Autumn brings crisp air, yellowing leaves, and a chilly breeze on the streets. Winter arrives with snowfall and opportunities for building snowmen and engaging in snowball fights.” These cyclical natural events reflect their understanding of temporal and spatial changes associated with nature, although the number of children who mentioned these phenomena was limited.

3.1.3. Scenarios—Involving Humans and Nonhumans

As previously mentioned, the children typically used isolated words when describing nature, such as a tree, grass, rabbit, mountain, etc., rather than more lengthy narratives. This could be due to their inability to express these concepts verbally. However, it may also suggest that their concepts of nature are disjointed and fragmented rather than a systematic, coherent collection. Some children, however, talked about the interconnectedness of natural attributes by describing nature scenarios primarily composed of plants, animals, and their habitats, with some also including details about the sky, sunlight, weather, and other non-biological features. For example, C54 (a 9-year-old boy) described conceptual nature: “The first image that comes to me is a vast grassy field with trees, wildflowers, and insects like bees. I also see some areas of the grass being slightly darker due to the sunlight being blocked by the trees, and there might be birds on the trees. Beyond the grassy field, there might be some hills.” C61 (8-year-old boy) described accessible nature and said that “In the middle of the park, there are numerous trees, abundant grass, and a serene lake where a few ducks gracefully drift. The lake is adorned with lotus leaves and blossoms, and beneath its surface dwell fish and some aquatic plants. The water shimmers with ripple.”

However, when describing scenarios of conceptual nature, many children incorporated dramatic elements beyond the scope of their everyday experiences, such as C36 (an 11-year-old boy) saying that “leopards and rhinos run across the savanna”, while another interesting example comes from C72 (a 9-year-old girl): “The real nature, where the grass remains untouched and untrimmed, is quite wild. If there are wild animals in that area, you might observe a lion hunting antelope, giraffes grazing on leaves high up in tall trees. Snakes could slither nearby, observing birds. Elephants and hippos may be spotted drinking water, while crocodiles might rest at the riverbed.”

Beyond everyday and dramatic nature, the question arises as to what “real nature” is, as highlighted by C72, or, conversely, what is “fake nature”. This was revealed in how the children perceive the presence of humans or human traces. The number of children mentioning human-involved versus nonhuman scenarios was therefore calculated, revealing that a significantly higher proportion of children depicted nonhuman scenarios than human-involved scenarios with regard to conceptual nature. In contrast, the opposite was true for accessible nature. Wild, poetic, sublime scenarios undisturbed by humans were conceptually conceived of as “real nature”. Ordinary, accessible natural scenarios with traces of humans were more likely to be perceived as “fake nature”.

3.2. Perception of Nature in the City

This section reports the results of the second part of the study, which explored how the children perceive the types of nature they encounter in their everyday city lives. As mentioned, a photo Q-sort was used to test how they perceive the degree of naturalness displayed in 33 photographs of urban environmental settings.

3.2.1. Perceived Naturalness

Of all 33 stimuli photographs displaying various environmental settings, S10 (mountains and streams), S1 (reeds by the water), and S7 (dense forest) received the highest mean aggregate naturalness ratings, compared to S3 (an amusement park), S16 (fitness equipment), and S18 (a residential parking area), which received the lowest ratings.

The children primarily considered environments rich in natural elements, especially those dominated by trees, more natural, while perceiving environments dominated by artificial structures and signs of human intervention as unnatural. For example, C18 (a nine-year-old boy) said: “A place that has water, trees, rocks, and all these natural elements is the most natural”. There were, however, subtle differences among children’s perceptions, with some children believing that to be natural, environments should be suitable habitats for animals, emphasizing the importance of biodiversity. Some children considered natural environments to be beautiful, bright, lush, and full of life, “but when it withers, it feels less natural”, as mentioned by C09 (a nine-year-old girl), who focused on the aesthetic qualities of nature. Some children viewed planted vegetation as unnatural if there were signs of human intervention. These varied perspectives reveal how children interpret and value different aspects of natural environments and settings that contain natural elements, as displayed in the photographs of urban scenes. The following section reports the quantitative data collected from the photo Q-sort interviews to delve into the children’s perceptions of nature by combining the statistical and content analyses from the photo-sorting task.

3.2.2. Neighborhood Clusters

Six distinct neighborhood groupings of the environmental settings displayed in the stimuli photographs were identified using hierarchical cluster analysis (Figure 2). The environments (photographs) contained within each of the clusters are perceived to be somewhat similar to each other and distinct from those in other clusters, possibly due to the (dis)similarity of the biophysical characteristics apparent in the environments depicted and/or the meanings the children attributed to these settings. Below is a description of these six neighborhood clusters.

Cluster 1: Closed vegetative landscapes. This cluster received the highest overall rating concerning perceived naturalness, with an aggregate mean value of 1.98. It primarily consists of photographs displaying environments dominated by trees and characterized by closed vegetative landscapes that create a lush and immersive natural setting with minimal to no visible traces of human intervention. Many of these photographs also display other natural elements, including water bodies, rocks, and birds, conveying a sense of thriving and diverse ecosystems. For example, C35 (an 11-year-old girl) said in describing S1 (depicting waterside trees and reeds with birds) that “There are animals, and animals usually live in natural environments. The surrounding shrubs and other vegetation also appear very abundant and natural, along with a waterbody”. Analysis of the open-ended responses suggested that these environmental settings were most often described as being “green”, “wild”, “ecological”, and “like primeval forests”, noting the absence of “artificial traces”.

Cluster 2: Open vegetative landscapes. Two photos depicting scenes of open vegetative landscapes, including a lawn and a field with flowers, are included in this group. The aggregate mean value of 1.88 means they were perceived as highly natural. These environments were often described as spacious, aesthetically pleasing with open views, and full of life. C35, for instance, remarked, “I feel the lawn is spacious, natural, and very beautiful. I like this place very much at first sight”. The children frequently mentioned trees, grass, flowers, and animals, including birds, bees, and butterflies, that they felt contributed to a lively and vibrant atmosphere. While noting the absence of artificial structures, some features, such as neatly arranged flowers, which were artificially planted, were suggested to display “such orderliness [they] would not typically occur in a wild natural setting”, as C26 (an 11-year-old boy) said.

Cluster 3: Nature in “orderly frames”. This group is represented by photographs displaying natural environments and artificially framed elements (e.g., by roads and hedges) defining orderly spaces. The children generally emphasized the natural elements present in these settings, such as trees, water, and stones, but the presence of roads, bridges, and buildings that frame these elements were also apparent in the photographs. These artificial structures do not dominate the natural scenery but suggest thoughtful integration within the environment and what has been referred to as “cues to care”, the presence of human intervention that shows the environment is being maintained and cared for, as described in Nassauer’s seminal article Messy ecosystems orderly frames [45]. As C26 described, “It’s like placing some human-made things within nature”. The children often used terms including “harmonious” and “congruous” to describe these scenes. For example, C53 (an 11-year-old boy) described the pavilion depicted in S23 as follows: “It feels like it belongs there; the scene is very harmonious, so it doesn’t feel out of place”. Similarly, C15 (a 12-year-old girl) described the trees, grass, sunshine, and a kitten in S14 as “a very natural and harmonious scene”.

Cluster 4: Mixed built landscapes. Similar to Cluster 3, this set of photographs depicts scenes where natural and artificial elements are equally represented. However, the key difference between these two clusters lies in the prominence of each element depicted. In Cluster 3, natural elements are the focus, while artificial traces subtly blend into orderly frameworks. In contrast, this cluster has more artificial structures that dominate the scenes, often positioned prominently at the center of the scene, which likely affects their perceptions of their degree of naturalness. As C13 (a 10-year-old boy) commented, “The flower bed artificially confines the flowers, and I feel it would be better if it were more like this photo [P21]”, suggesting a preference for more organic, less controlled arrangements of features where natural elements dominate.

Cluster 5: Urban leftover spaces. This cluster is represented by only one photograph, S19, which depicts an urban leftover space in a residential area that had once been transformed into a play area. Most children noted that artificial structures, including buildings, occupied a large area in the photograph, while the presence of trees was sparse. Although some children acknowledged that wooden play structures were made from natural materials, it did little to influence their perceptions of the naturalness of the scene. Beyond these descriptions of physical features, some of the children also expressed their attitudes toward the depicted environment, describing it as old, desolate, less vibrant, and too cluttered.

Cluster 6: Human-dominated landscapes. This cluster received the lowest aggregate mean rating for perceived naturalness (M = −1.73). It primarily contains photographs displaying environments dominated by human-made elements, including residential areas, playgrounds, sports fields, and other environmental settings the children may encounter in their daily lives in the city. They frequently used the terms “artificial”, “built for human use”, and “unsuitable for animals” to describe the buildings, parking lots, and fitness facilities made of plastic and metal visible in these photographs, while emphasizing the scarcity of greenery, particularly trees. For example, C74 (an 8-year-old boy) said in describing S32 (paved play square): “This place is for playing, and there are many people here… There are only a few trees. Places with many people are not suitable for animals”. Interestingly, despite the limited presence of natural elements, some children remarked that these environments felt familiar to them as they are like those they encounter in their daily lives, evoking a sense of familiarity and comfort similar to feelings associated with home environments. As C11 (a 9-year-old girl) noted when describing S13, S15, and S18, “Places that feel like home are more natural”. This suggests that for some children, the familiarity of their everyday surroundings can influence their perception of naturalness, highlighting the role of personal experiences and emotional connections to known spaces in shaping how children interpret and value the environment.

3.2.3. Underlying Perceptual Dimensions

The results of the MDS analysis revealed three underlying bipolar dimensions of meaning, as revealed in the stimulus configuration plots (see Figure 3 and Figure 4), which visually map the perceptual relationships and distances between the various stimuli elements (photographs of environments). These dimensions represent the key discriminating factors that influence how children perceive the naturalness of the different environments tested in this study and categorize them accordingly in clusters, as previously described.

Dimension 1: organic to inorganic (natural to artificial). In this dimension, Clusters 1 and 2 are positioned on the far right of the plot, indicating that children perceive these environments as the most natural. These clusters are characterized by landscapes dominated by plants with almost no visible signs of artificial elements and were described as “green” and “ecological”, along with references to animals, such as birds and insects. In contrast, Clusters 5 and 6, which represent environments dominated by artificial elements, are located on the far left of the plot, suggesting that Dimension 1 represents a spectrum from organic/natural to inorganic/artificial based on the interpretation of the environments included in these four clusters, highlighting this dimension as mainly related to objective evidence and signs of naturalness. Additionally, the positioning of Clusters 3 and 4 further reinforces this interpretation. Cluster 3 integrates orderly frames within predominantly natural settings and is located closer to the right than Cluster 4, which features artificial structures as the main focus of the environment.

Dimension 2: congruous to incongruous (fitting to non-fitting). The interpretation of Dimension 2, like that of Dimension 1, is revealed by examining the relative positions of the clusters and, in this case, the vertical distribution of photographs and the associated clusters. Cluster 2 is at the top of the plot, for which the children frequently used terms including “spacious”, “comfortable”, and “beautiful” to describe the depicted environments. Similarly, S14 (community abandoned land) and S28 (traditional architecture at the water’s edge) in Cluster 3 are also positioned near the top, with the children describing them as being “harmonious”. In contrast, S9 (sparse forest) was described as a place that felt “dark” and “depressing”, with the stones resembling tombstones and being “incongruous”. Similar descriptions were given of S7 (a dense forest-like setting), which was perceived as being the least natural along this dimension. Interestingly, S15 (community paved square) in Cluster 6 is also positioned near the top, and, in fact, most photographs in this cluster are positioned relatively high or at least around the middle along this dimension compared to the first dimension, where the depicted environments were perceived as the least natural. These photographs predominantly depict residential areas, community parks, and similar environments familiar to children in their daily lives. A common feature among the photographs positioned higher in these clusters is that children described these environments as being “comfortable”, “fitting”, and “harmonious”, suggesting a subjective evaluative dimension associated with emotional responses to the depicted environments.

Dimension 3: colorful to drab (vibrant to dull). By comparing the plot shown in Figure 3 with the one shown in Figure 4, Cluster 2 remains in a relatively high position on the plot; however, S21 (a field of flowers) is positioned higher than S2 (a flowered lawn). In Dimension 3, the photographs in Cluster 4 have generally shifted upwards, with S20 (an artificial flower bed) positioned at the very top. In Cluster 6, S3 (an amusement park) has moved to the highest position. What these photos have in common is that they depict the most colorful environments, suggesting a strong association with vibrancy and visual appeal. In contrast, Cluster 5 shows the most significant change and is rated the least natural along this dimension. Similarly, S14 (community abandoned land) and S18 (a dense residential parking area) were also perceived as being less natural, with a common trait of being perceived as drab and dull in color. This shift indicates that environments with more vivid colors and visually engaging elements tend to be perceived as more natural along this dimension.

4. Discussion

The findings of this study indicate that plants, particularly trees, are the dominant elements of children’s constructs of nature. This is evident from the interviews on what conceptual and accessible nature means to them and the criteria they used to rate the stimuli photographs regarding the degree of perceived naturalness, especially regarding the first dimension that differentiates organic from inorganic. Plants, in general, were found to be highly instrumental in defining and characterizing natural settings and features, especially in urban open space areas, including parks and reserves. Trees are among the most conspicuous natural elements, given their large scale, towering presence, distinct shapes, and seasonal changes, and they also provide habitats and sustenance for other biological organisms (e.g., birds and small mammals), potentially appearing as an independent system to children. Children’s experiences with plants (e.g., observing plant landscapes, climbing trees, and collecting leaves) provide them with the city’s most tangible signs of nature, thereby shaping their constructs of plants/trees as manifestations of nature [46].

The prominence of plants/trees in the children’s conceptualization of nature could also be traced back to human evolution, as early humans relied on trees and other plants as a primary source of food, shelter, and other resources necessary for their survival. For example, trees provide elevated locations to obtain views (prospect) over the landscape and find refuge from predators [47]. The frequently mentioned grass and flowers and the children’s fondness for them might be explained by their resemblance to savanna-type landscapes where early humans are believed to have lived during a crucial period of evolution, and flowers would have also signaled prospective food sources available to our ancient ancestors [48,49]. Non-living entities, especially water bodies, were also frequently mentioned in the interviews and as key criteria that define what are perceived as being natural settings, as gleaned from the sorting tasks, with many children showing fondness for waterscapes, which aligned with existing research [50]. To some extent, this reflects the enduring influence of biophilic learning rules in modern societies, in which human–nature connections are correlated with multiple meanings and expressions of satisfaction—aesthetically, intellectually, cognitively, and even spiritually [4].

The children’s conceptualization and perception of nature could also be influenced by media and educational materials. One example is the notion of “plant blindness”, as suggested by the findings of this study, where few children could name specific plant species, especially when compared to their ability to identify animals, which is consistent with other studies [51,52,53]. Despite the relatively weak visual cues for identifying plants, especially when they are not in bloom or bearing fruit, a possible explanation lies in the influence of media and educational materials that portray plants as more of a collective presence that contributes to the overall ambiance of settings rather than independent plant types and species [54,55], compared to animals, which have more distinctive features and resemblance to human traits (e.g., the ability to move and make sounds) and therefore often play key roles in children’s stories [56]. The differences in children’s descriptions of animals between conceptual and accessible personal constructs of nature may also reflect the impact of media and education, as well as their daily experiences in the city, on their personal constructs. More children mentioned animals commonly encountered in their everyday lives in describing accessible nature, particularly cats, dogs, birds, and insects, compared to conceptual nature, where children referred to a broader range of animals, often iconic, large, distinctive, or endangered types of wildlife, such as elephants, rhinos, and cheetahs.

However, how nature is presented in children’s books and media is also likely to influence how they perceive nature in their everyday environment. Children tend to spend more time indoors engaging with electronic devices, from which they more frequently encounter exotic species of plants and animals virtually at the expense of awareness of everyday biodiversity in their surroundings; thus, they are likely to be desensitized to everyday nature, which they may find to be ordinary and boring in comparison [18,57]. This phenomenon is reflected in the children’s discrimination of real versus fake nature, as discussed in the results section, which highlights children’s tendency to romanticize nature, and it can lead them to perceive the relationship between humans and nature in a somewhat distorted manner. This may, in turn, further exacerbate their disconnection with nature, impacting their motivation to engage in efforts at conserving nature, and may even develop in them a conservation bias—a selective focus on the conservation of certain charismatic, “cute”, or endangered species at the expense of species they may encounter in their everyday city lives [58].

In addition, it is interesting to note that despite the first dimension of organic to inorganic, an objective evaluative dimension consistent with the interview results and the definition of nature provided by Wohlwill, two subjective dimensions were also identified in children’s perceptions of accessible nature in urban settings tested in this study. These dimensions focused more on non-living and artificial elements and were associated with particular affective responses (congruous to incongruous, feelings of comfort, harmony, and familiarity) and aesthetic qualities (colorful to drab, visually vibrant features) of the environments, indicating the impact of their daily experiences and surroundings on their conceptions of nature. These findings may differ from those of adults or those with specialized knowledge, such as ecologists, who might place more emphasis on biodiversity and ecosystem functions and services. These insights highlight the influence of the design of urban environments and human activities on children’s perceptions of nature, suggesting that both environmental design and environmental education should seek to provide children with more meaningful and direct experiences of nature that they can encounter in their everyday surroundings.

An important step in doing this is to encourage children to spend more time in outdoor natural environments over engaging in indoor activities and the use of digital and electronic devices. It is, therefore, crucial to provide them with accessible natural environments, particularly blue (water) and green (vegetated) spaces in proximity to residential neighborhoods and schools where they live and study. These environments should incorporate a diverse array of native plant species and create suitable habitats that can accommodate common types of urban wildlife, such as birds, insects, and small mammals. Because the children tended to associate vibrant, colorful, and harmonious environments with nature, these environments should also be visually appealing and engaging, and drawing on biophilic design principles (e.g., shapes, colors, and sounds of nature) helps to create immersive environments that can facilitate regular, unstructured outdoor activities and exposure to elements of the natural world. In addition, environmental educational interventions should incorporate elements of local nature and encourage study of nature in the curriculum that involves hands-on interaction with nature, such as bird observation and insect surveys, that emphasize greater understanding of local biodiversity.

There are several limitations that should be acknowledged. Firstly, the sample of children used in the study was not large or varied enough to fully represent the population of a megacity like Beijing, as most participants shared similar socioeconomic backgrounds, which could limit the generalizability of the findings to other populations and geographic contexts. Secondly, while studies on environmental perception often employ a somewhat larger number of photographs as stimuli in using the photo Q-sort method, this study used only 33 photos considering the cognitive and attentional capacities of children compared to adults, which could restrict the scope and variety of environments represented. However, using children as respondents and highly urban environmental settings as stimuli for this method is innovative and helps to fill a gap in the current literature addressing this topic. Future research could, however, address these limitations by including a larger and more diverse sample of child participants and environmental contexts. Comparative studies between urban and rural children across different age groups or cultural contexts may also provide deeper insights into how children understand and perceive nature.

5. Conclusions

The study reported here first explored children’s understanding of conceptual and accessible nature and identified similarities and differences between the two. The results revealed that plants, particularly trees, and vegetated landscapes were perceived by the children to be the most natural elements, followed by animals. They also frequently mentioned non-living natural elements and phenomena, which was consistent across both conceptual and accessible modes of conceptualizing nature. The most notable difference between the two is that when describing nature conceptually, the children tended to romanticize it, considering exotic and dramatic elements as “real nature” while seemingly being desensitized to ordinary, accessible, everyday nature.

Further exploration of children’s perceptions of accessible nature in urban settings shows that they distinguish between natural and non-natural environments primarily through an objective evaluative dimension, contrasting nature with artificial/inorganic elements. Interestingly, the other two dimensions revealed through the MSD analysis—congruous to incongruous and colorful to drab—highlight the influence of the design of urban environments and associated human activities on children’s understanding of nature in cities. This study also highlighted the impacts of personal experiences, education, media, and possibly evolutionarily ingrained predispositions on their personal constructs of nature. This understanding can provide useful insights for urban and landscape design and environmental educational curricula and activities aimed at engaging children in more meaningful contact with nature that will allow them to benefit from cultural ecosystems services, which can both promote their physical and psychological well-being and health and influence them in adopting pro-environmental attitudes, which, in turn, can help to ensure for them a more sustainable future.