Nature benefit hypothesis: Direct experiences of nature predict self‐reported pro‐biodiversity behaviors

Human activities are damaging the world's ecosystems, posing a serious threat to life on Earth, including humanity. To address this situation, widespread and significant changes in human behavior are necessary. Direct experiences of nature can encourage individuals to adopt positive actions towards biodiversity (hereafter pro‐biodiversity behavior), but this relationship has not been well studied. Using a large sample of Japanese adults, we demonstrate that both recent and childhood frequencies of nature experiences are associated with an increased likelihood of exhibiting pro‐biodiversity behaviors. This association was found to be consistent across various forms of behaviors, including purchasing ecofriendly products, reducing pesticide use in domestic gardens, and donating to conservation organizations. However, our research also reveals a declining trend of childhood experiences of nature in Japan, resulting in an “extinction of experience.” Our results suggest that enhancing people's personal experiences with nature could help promote desired behavioral change to halt biodiversity loss.


INTRODUCTION
The continued loss of biodiversity due to human activities is a pressing concern in the 21st century, despite increased conservation efforts (Johnson et al., 2017;Tilman et al., 2017). To mitigate and reverse this trend, significant and widespread changes in individual behavior are necessary Nielsen et al., 2021;Perino et al., 2022). Everyday activity of individuals, regardless of its extent, contributes to the impact on biodiversity either directly or indirectly Maynard et al., 2020;Nielsen et al., 2021;Selinske et al., 2020). Thus, even a slight increase in the proportion of individuals who adopt This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2023 The Authors. Conservation Letters published by Wiley Periodicals LLC.
actions that benefit biodiversity (hereafter pro-biodiversity behavior) within a population can result in substantial collective impact on the success of biodiversity conservation (Selinske et al., 2022). The significance of human behavior in protecting biodiversity is now well-recognized, and researchers and policymakers are seeking ways to promote pro-biodiversity behaviors. Personal experiences with nature are considered as a key driver of pro-biodiversity behavior Soga & Gaston, 2022). This idea is based on evidence that experiences with nature, especially during childhood, can shape and reinforce positive attitudes and beliefs towards nature (Cheng & Monroe, 2012;Soga & Gaston, 2016;Wells & Lekies, 2006). It is believed that enhancing personal experiences with nature can lead to support for pro-biodiversity policies and actions, which can have a positive impact on conservation efforts Prévot et al., 2018;Richardson et al., 2020), the so-called nature benefit hypothesis (Soga & Gaston, 2022). However, the validity of this hypothesis is yet to be fully established. While research has linked direct experiences with nature to "pro-environmental" behaviors such as waste reduction, water conservation, and energy efficiency (Alcock et al., 2020;Broom et al., 2017;Collado et al., 2015;Lin & Lee, 2020;Martin et al., 2020;Richardson et al., 2020;Rosa et al., 2018), the relationship between nature experiences and pro-biodiversity behavior remains uncertain Prévot et al., 2018).
Using survey data from a large sample of the adult population of Japan, we aimed to understand better the relationships between direct experiences with nature and self-reported pro-biodiversity behavior. To address this goal, we quantified the link between the frequency of nature experiences (recent and childhood) and the likelihood of adopting various pro-biodiversity behaviors (see Table 1) (Study 1). Given the evidence showing that nature experiences are associated with increased pro-biodiversity behaviors, we then examined how the frequency of nature experiences that people had during their childhoods changed over time in Japan to infer the possible consequences for biodiversity conservation (Study 2). We show evidence for a gradual increase in the number of people who had almost no personal experience with nature during their childhood, that is there has been an "extinction of experience" (Pyle, 1993;Soga & Gaston, 2016). With these results in mind, we discuss the importance of enhancing people's direct experiences with nature in the midst of an unprecedented biodiversity crisis.

Survey design
In February 2022 we surveyed 20,000 Japanese adults via an online questionnaire administered by Cross Marketing Ltd. The respondents were representative of Japan's population in terms of age, gender, and home location (prefecture). The survey covered three topics: (1) self-reported pro-biodiversity behaviors, (2) frequency of nature experiences, and (3) sociodemographic and personal information (see Table 2). Full participant information can be found in Table S1. The study was approved by the ethics committee of the Graduate School of Agricultural and Life Sciences at the University of Tokyo (H21-022).

Self-reported pro-biodiversity behaviors
The survey asked participants to report the frequency of their participation in 18 pro-biodiversity behaviors over the past year, as listed in Table 1. The selection of behaviors was based on previous research (Alcock et al., 2020;Cooper et al., 2015;Prévot et al., 2018;Selinske et al., 2020) and input from two experts in Japanese conservation issues. The purpose of the study was to investigate the relationship between nature experiences and "pro-biodiversity" behaviors, so behaviors aimed at addressing environmental issues other than biodiversity conservation (e.g., energy saving or recycling) were excluded. Nine questions focused on behaviors in specific situations (e.g., food purchasing, garden management). Participants who reported not having experienced those situations over the past year were excluded from the analysis for each of the nine probiodiversity behaviors. Participants' self-reported levels of pro-biodiversity behaviors were recorded on a continuous scale, but were later dichotomized as "adopted at least once" or "not adopted at all".

Nature experiences
Following previous studies (Soga et al., 2018(Soga et al., , 2020, the frequency of recent and childhood nature experiences was measured by asking participants to report their frequency of visiting natural environments (e.g., greenspaces, woodlands, grasslands) during childhood and the past year, using a six-point Likert scale (Table 2).

Sociodemographic and personal characteristics
We collected sociodemographic and personal circumstance data that could potentially influence the adoption of pro-biodiversity behaviors (see Table 2).

Statistical analyses
To examine the relationship between nature experiences and pro-biodiversity behaviors, we used multiple logistic regression, calculating the odds ratio (OR) of adopting each of the 18 pro-biodiversity behaviors based on the TA B L E 1 List of 18 pro-biodiversity behaviors considered in this study. They are classified into six categories: consumption, social, donation, lifestyle, advocacy, and stewardship

Category ID Behavior
Consumption 1* When buying products, choose environmentally friendly products (e.g., papers certified with FSC label) 2* When buying vegetables or fruits, choose organic ones 3* When buying fish, avoid choosing fish species with relatively high extinction risk (e.g., tuna, eel) 4* When purchasing ornamental plants, avoid choosing wild plant species with relatively high extinction risk (e.g., cactus, orchid) 5* When purchasing furniture or clothes, avoid choosing products from rare wild animals and plants (e.g., wooden furniture, fur coats, leather products) Discuss the actions and practices that benefit biodiversity conservation with other people (e.g., family members, friends) 7 Encourage other people (e.g., family members, friends) to adopt biodiversity-friendly behaviors 8 Tell other people (e.g., family members, friends) how wonderful nature is (e.g., share beautiful nature images on social media) 9 Take other people (e.g., family members, friends) to natural environments 10 Share (e.g., on social media) official announcements and messages from conservation organizations with other people (e.g., families or friends)

11
Volunteer for activities that preserve the natural environment 12* When managing a domestic garden, avoid using chemical herbicides or insecticides 13* When visiting wild areas (e.g., protected areas), remove soil on footwear to prevent the introduction and spread of alien plants 14* When visiting wild areas (e.g., protected areas), avoid picking wild flowers and damaging vegetation frequency of nature experiences, which were grouped into six categories (never; yearly; quarterly; monthly; weekly; daily). The OR was adjusted for the five sociodemographic and personal variables (Table 2) to control for confounding factors. To account for multicollinearity, separate models were built for recent and childhood nature experiences, instead of using both variables in a single model. These analyses were conducted using the epiDisplay package in R version 3.4.1.
To validate the results, generalized additive models (GAMs) controlled for covariates (five sociodemographic and personal variables) were used to explore the rela-tionship between nature experiences and pro-biodiversity behaviors. The frequency of nature experiences was treated as a continuous variable, and the five sociodemographic and personal variables were used as fixed factors. The analysis and plotting were done using the gamm4 and ggplot2 packages in R.
After preliminary analysis, no significant difference was found in the relationship between nature experiences and pro-biodiversity behaviors among groups based on the five sociodemographic and personal variables (e.g., age, gender). We therefore analyzed the combined data of all respondents. TA B L E 2 Two measures of experience with nature and five sociodemographic and personal variables collected in the questionnaire

Category and variable
Answer scale

Study 2: Temporal changes in childhood nature experiences
In October 2020 we surveyed 6300 Japanese adults aged 20s to 70s, equally distributing male and female participants across six age categories (20s, 30s, 40s, 50s, 60s, 70s) through an online questionnaire administered by Cross Marketing Ltd. The survey asked respondents to rate the frequency of their participation in eight common nature-based activities during childhood (visiting urban greenspace; visiting natural environments in the countryside; visiting remote natural environments; exploring natural environments; touching wild flowers and trees in natural environments; observing wild birds in natural environments; catching/observing insects in natural environments; catching/observing aquatic organisms in natural environments). Responses were recorded on a sixpoint Likert scale, ranging from 1 (never) to 6 (daily). The survey was approved by the ethics committee of the Graduate School of Agricultural and Life Sciences at the University of Tokyo (H20-008).

Study 1: The relationship between nature experiences and pro-biodiversity behavior
There was a positive association between nature experiences, both in childhood and recent, and the likelihood of adopting 18 pro-biodiversity behaviors (Figures 1 and 2; see also Table S2 for the results of the logistic regressions). Those who had recent nature experiences at least once a year were more likely to adopt all of the 18 pro-biodiversity behaviors compared to those who never had such experiences ( Figure 2). Childhood experiences of nature, on the other hand, were not significantly linked to adopting four of the 18 behaviors when experienced yearly (behavior #3, 4, 12, and 13 in Figure 1). For the majority of the behaviors, the ORs showed a peak at a weekly frequency for both childhood and recent nature experiences (Figures 1 and 2).
The results of GAMs supported the findings from the multiple logistic regressions and indicated a steady F I G U R E 1 The odds ratios and 95% confidence intervals of the likelihood of adopting 18 pro-biodiversity behaviors as a function of frequency of experience with nature during childhood (1: never [reference category]; 2: yearly; 3: quarterly; 4: monthly; 5: weekly; 6: daily). Note: Controlled for age, gender, income, level of education, and the degree of urbanization of the home environment. The numbers shown in each panel indicate the pro-biodiversity behavior ID (Table 1).

F I G U R E 2
The odds ratios and 95% confidence intervals of the likelihood of adopting 18 pro-biodiversity behaviors as a function of frequency of experience with nature during the previous 12 months (1: never [reference category]; 2: yearly; 3: quarterly; 4: monthly; 5: weekly; 6: daily). Note: Controlled for age, gender, income, level of education, and the degree of urbanization of the home environment. The numbers shown in each panel indicate the pro-biodiversity behavior ID (Table 1).
increase in the likelihood of adopting pro-biodiversity behaviors up to about a weekly frequency, with no further marked increase (Figures S1 and S2). The relationship between nature experiences and pro-biodiversity behaviors showed a similar pattern for both childhood and recent experiences and across the 18 behaviors ( Figures S1 and S2).

Study 2: Temporal changes in childhood nature experiences
For most of the eight nature-based activities, there were age-related differences in childhood experience frequency (Figure 3). Younger age categories had fewer participants F I G U R E 3 Proportion of participants who gave each of the six answers to the questions asked to assess their levels of experiences with nature during childhood (A: never; B: yearly; C: quarterly; D: monthly; E: weekly; F: daily). The plots show each of eight common nature-based activities (1: visiting urban greenspace; 2: visiting natural environments in the countryside [e.g., woods]; 3: visiting remote natural environments [e.g., national parks]; 4: exploring natural environments; 5: touching wild flowers and trees in natural environments; 6: observing wild birds in natural environments; 7: catching/observing insects in natural environments; 8: catching/observing aquatic organisms in natural environments). who frequently (i.e., more than once a month) experienced nature-based activities compared to older categories (Figure 3e and f). Younger age categories also had more participants who had almost no childhood experiences of nature (i.e., nonnature users) (Figure 3a and b). Despite the diversity of nature-based activities, the overall age-related changes in experience frequency were similar across the eight activities (Figure 3).

Key findings
The findings of this study highlight the importance of direct experiences with nature as a predictor of probiodiversity behavior. Indeed, the association between pro-biodiversity behaviors and nature experiences was similar to, and sometimes stronger than, associations with other sociodemographic and personal variables (see Table S2). Given the well-established link between proenvironmental attitudes and behaviors and these other factors (Gifford & Nilsson, 2014), we interpret the magnitude of the relationship with nature experiences to be meaningful in terms of the potential implications for behavioral change, although more research on causality is needed. Our results align with recent studies showing a positive association between nature experiences and proenvironmental behaviors (e.g., Alcock et al., 2020;Lin & Lee, 2020;Martin et al., 2020;Richardson et al., 2020;Rosa et al., 2018). This is noteworthy given the differences between pro-environmental and pro-biodiversity behaviors in terms of their formation processes and environmental impacts . The consistency of the association of nature experiences with these two types of behaviors suggests that, if there is causality, enhancing personal direct experiences with nature might contribute simultaneously to the alleviation of various environmental issues we face today (e.g., climate change, resource depletion, and loss of biodiversity). This could be vitally important at a time of unprecedented environmental destruction when humans need to consider how best to solve multiple major environmental issues in a comprehensive manner.
The 18 pro-biodiversity behaviors we studied are likely to be interconnected. For instance, behaviors in the "social" domain (Table 1) are expected to have positive associations with other behaviors, as they may drive their adoption. Additionally, some behaviors, particularly those in the "social" and "advocacy" domains, have the potential to encourage others to engage in pro-biodiversity behaviors. Our results, considering these spillover effects (Maki et al., 2019), indicate that nature experiences could have far-reaching positive impacts within a society.
Our finding of nonlinear relationships between nature experiences and pro-biodiversity behaviors has potentially important implications for the design of behavior change interventions and reveals new hypotheses that warrant further attention. Our results, shown through GAMs (Figures 1 and 2), suggest that there may be a threshold at which positive outcomes of nature experiences emerge. They also indicate that there may be an optimal level of nature experience that can enhance pro-biodiversity behaviors. Interestingly, we found a downward trend in odds ratios of adopting pro-biodiversity behaviors with daily nature experiences compared to weekly experiences. One possible explanation is that individuals who experience nature daily may be professionals working in natural settings (e.g., farmers). Indeed, work-related nature experiences have been found to have less impact on pro-environmental behaviors than recreational activities (Collado et al., 2015). Additionally, these individuals may be more likely to perceive negative impacts of nature on health and the economy (i.e., ecosystem disservices), which may reduce their willingness to protect it (Soga et al., in press). Further research is needed to understand the applicability of our results in different cultural, social, and environmental settings.
There is growing concern among the academic community that many people (especially children) are losing their connection to nature, known as the extinction of experience, which can have adverse impacts on the conservation of biodiversity (Ives et al., 2018;Miller, 2005;Soga & Gaston, 2016, 2023. Our findings partially support this apprehension, as they demonstrate a rising trend of individuals who have not experienced nature during their childhood in Japan ( Figure 3) and a lower likelihood of participation in pro-biodiversity behaviors among those with limited childhood nature experiences (Figures 1  and 2). The decline in nature experiences observed in our study is likely a result of both the reduction in opportunities and motivations for interaction with nature. However, due to limited data, we were unable to delve deeper into this aspect. Further research involving experimental approaches and long-term monitoring is necessary to thoroughly comprehend the extent and underlying causes of the extinction of experience and its impacts on biodiversity.

Limitations
While this study makes some important contributions, there are also some limitations. First, the cross-sectional design of our study precludes the determination of causality. This may be particularly relevant for "stewardship" and "social" behaviors (Table 1) because these behaviors can potentially enhance an individual's opportunities, motivations, and capabilities for interaction with nature (Soga & Gaston, 2022). It might also be possible that there exists a third factor that could drive both nature experiences and pro-biodiversity behaviors. Second, our study relied on self-report surveys, which may lead to reporting bias, such as under-or overreporting of the actual frequency of participating in pro-biodiversity behaviors or nature experiences. Furthermore, our measure of nature experiences during childhood was obtained through retrospective questions, which may be prone to recall bias, particularly among elderly participants who have a longer interval between childhood and the present. Third, we did not examine the impact of the quality of nature experiences on pro-biodiversity behaviors. Obviously, the quality, as well as the quantity, of an individual's experiences varies substantially among people, and this can affect their attitudes and behavior towards biodiversity. Lastly, we did not consider all of the possible factors that can affect pro-biodiversity behaviors. Although we included key sociodemographic and personal variables as covariates in the analysis, there are potentially other important ones, such as knowledge, values, religious affiliation, and social norms (Gifford & Nilsson, 2014). While we analyzed our data at the level of the individual person, it is likely that an individual's decision to participate in pro-biodiversity behaviors is influenced by the social context in which they are situated.

CONCLUSION
In summary, our study has demonstrated a consistent positive association between direct experiences of nature and self-reported pro-biodiversity behaviors in Japan. While it is too early to make specific policy recommendations based on our findings, they suggest that enhancing people's experiences with nature might be one approach to encouraging desired behavioral change to halt biodiversity loss. If this is true, then future conservation policy likely benefits from expanding efforts to reconnect people to nature. Current policies often focus on the creation and preservation of urban green spaces as a means to mitigate the loss of local biodiversity. Given the wider impact of lifestyle behaviors on the protection and restoration of biodiversity, our results suggest that these places may also contribute to biodiversity conservation by fostering a connection between urban dwellers and the natural world and facilitating more positive actions towards biodiversity.

A C K N O W L E D G M E N T S
We thank M. Akasaka and K. Tsuchiya for their helpful comments.

C O N F L I C T O F I N T E R E S T S TAT E M E N T
The authors declare that they have no competing interests.