Cultural probes for environmental education: Designing learning materials to engage children and teenagers with local biodiversity

questions (Fig 7). The collected data suggest that the CP technique promoted productive ques- tioning concerning the surrounding nature-rich environment. Overall, differences between age groups are significant (χ2(1) = 5.839; p = .016), and the number of ’strategic’ questions is much smaller in the older group. Nonetheless, both groups generated a higher number of ’productive’ questions; 84% (78 out of 93) for the younger par- ticipants, 96% (68 out of 71) for the older participants. Additionally, we crossed the questions posed with their corresponding activities to determine whether some activities prompted more ’strategic’ questions than others. For the younger participants (Fig 8), the distribution of ’productive’ and ’strategic’ questions by activity is not homogenous (χ2(5) = 12.666; p = .027). Activities 3 ("The smells of the forest") and 6 ("Who lives here?") raised a more significant number of ’strategic’ questions when compared to other activities. For the older group (Fig 9), ‘strategic’ questions represent only three out of 71, corresponding to just two activities. Thus, the distribution of ‘productive’ and ‘strategic’ questions was not significantly different between activities (χ2(5) = 7.541; p = .183). When distributing questions by question tokens used in their production (Fig 10), we observed that participants used all of the tokens provided. However, their selection is not homogenous; it is evident that the tokens ‘‘Explain why. . .?” (27%; n = 44) and ‘‘Why is. . .

important?” (21%; n = 34) were preferred. In the first token, the focus is on possible causes– e.g., ‘‘Explain why invasive plants exist?”–and the purpose of something–e.g., ‘‘Explain why nature needs moss, fern and trees?”. In the second token, the focus is on the relevance of the purpose–e.g., ‘‘Why is preserving nature important?”. Tokens related to comparison, e.g., "How are. . . and. . . similar" and "How do. . . and. . . dif- fer", as well as those associated with the utility of something, e.g., "How could. . . be used to. . .?" were used less often. We further observed that only 7% of the questions (n = 11) were constructed without the use of tokens–e.g., "Do plants choose their natural environment?". Thus, content-free question tokens appear to benefit participants’ questioning. Additionally, the distribution of ’productive’ and ’strategic’ questions is significantly different between Fig 7. Distribution of valid participant questions (n = 164) according to the approach adopted in questioning by age group. In total, younger participants generated 93 questions, and older participants generated 71 questions. https://doi.org/10.1371/journal.pone.0262853.g007 PLOS ONE Cultural probes for environmental education PLOS ONE | https://doi.org/10.1371/journal.pone.0262853 February 10, 2022 17 / 30