Does increasing biodiversity in an urban woodland setting promote positive emotional responses in humans? A stress recovery experiment using 360-degree videos of an urban woodland

producing two semi-continuous variables ranging from 5 to 25. The internal reliability, mea- sured as Cronbach alpha, was good for both scales (Cronbach α > 0.8). The I-PANAS-SF was chosen as it has a short completion time, is validated for non-native English speakers and has good reliability for representing genuine emotions [54]. Affective states, which include emo- tions, stress responses and mood, are important indicators of mental health [55]. Short-term affective responses have been linked to longer-term mental health indicators, such as life satis- faction and depression [56, 57], thus making the I-PANAS-SF a useful proxy for determining potential health outcomes related to nature-based interactions. Perceptions of biodiversity were measured at the end of stage 3 (i.e. after watching the video). The participants were asked to rate the location shown in the video in terms of “value for plants and wildlife” [58, 59]. The rating ranged from “very bad” (1) to “very good” (5). Those who rated the environment as “good” or “very good” were also asked to share what they noticed in the video that corresponded to this response (open question). This provided a proxy of their perceptions of biodiversity without mentioning the word directly, reducing the risk of biased responses. Participants were asked to provide their demographic characteristics (Gender, Age, Ethnic- ity), to verify the balance of the randomisation (i.e. all condition groups should have similar demographics). To check for the novelty effect participants were also asked if they had ever

watched a 360-degree video before. Along with their demographics, participants were asked to rate their Nature Connectedness via the Inclusion of Nature in Self scale (1–7 scale) [60] and to recall the amount of time they spent outdoors during their childhood, ranging from “none” (1) to “a lot” (4). Nature Con- nectedness, one’s extent of affective affiliation with nature [61], has been shown to act as a moderator of the well-being benefits derived from being in nature [62]. Statistical analysis All analyses were carried out on SPSS (version 26) for Windows. Metadata (completion time, IP location) were used to check the quality of the data. Responses meeting one or more of the following criteria were excluded: 1) Incomplete response; 2) Overtime completion (over 30 minutes); 3) Invalidating comments (e.g. the participant admitted clicking through the proce- dure); 4) Multiple submissions (same IP and demographics). Only the first valid observation was kept; 5) All I-PANAS-SF ratings, at all stages, were the same number (e.g. all 1). One-way analysis of variance (ANOVA) was used to compare the condition groups in terms of the demographics, at baseline. Changes in Negative Affect (NA) and Positive Affect (PA) over the three stages of the experiment and between the condition groups were examined via repeated measure analysis of variance (RM-ANOVA). Since the procedure was identical for all participants except for the video, the RM-ANOVA model included an interaction term between the stage of the procedure and the between-subjects variable. Where the RM-ANOVA

assumption of sphericity was violated, probability (P) values were calculated with the Huynh- Feldt correction [63]. Where post-hoc comparisons were performed, the Bonferroni correc- tion for multiple comparisons was applied. Video condition was selected as the main independent variable (between groups). Sepa- rately, the perceived biodiversity of the site, the participants’ Nature connectedness and time spent outside during childhood were also used as independent variables. To identify the effect of individual elements of nature on the affective states, comments from participants who had evaluated the environment as “good” or “very good” were analysed via Content Analysis [64], generating a list of codes (S1 Appendix). All codes were used as dummy variables (between groups) in a series of exploratory RM-ANOVA analyses. PLOS ONE Does increasing biodiversity in an urban woodland setting, promote positive emotional responses in humans? PLOS ONE | https://doi.org/10.1371/journal.pone.0297179 February 7, 2024 7 / 20