A direct personal experience of science and nature changes intended behaviours for conservation

ABSTRACT Social and affective empathy may generate future conservation benefits as the consequence of transformed personal attitudes. In this study, we investigated changes in attitudes and intended behaviours about the plight of woodland songbirds before and after participation in science activities and direct interaction with scientists monitoring avian biodiversity at Mount Lofty Ranges (MLR) in South Australia. A total of 55 anonymous adult participants were invited to join a survey before and after participating in two 3-h workshops on avian science plus acoustic data collection. Comparing the survey results before and after the experience, there were significant shifts in self-reported “good” knowledge about woodland songbirds; increased concern about the conservation status of woodland birds; increased concern about cessation of songbird research; and an increased interest to play a role in songbird conservation. Further investigations could aim to elucidate the mechanisms for shifts in attitudes that occur together with experiencing nature. In general, this small-scale study provides evidence that nature-based science activities can represent valuable hands-on experience in science and may contribute to conservation outcomes by fostering environmental awareness and self-reported aims for involvement.


Introduction
Pro-environmental behaviour is expected to enhance conservation outcomes, whereby different forms of engagement are associated with different conservation actions (Alcock et al., 2020;Clayton & Meyers, 2011;Martin et al., 2020).For instance, zoo visits are known to promote affinity to nature (Howell et al., 2019;Powell & Bullock, 2014), whereas watching documentaries rather fosters donation behaviour (Arendt & Matthes, 2016).While the relationship between people and nature is complex (Clayton et al., 2017;Price et al., 2022), there is growing evidence that emotional and prosocial association with nature including volunteering (reviewed in Aknin et al., 2018) promotes the valuation of wildlife.Positive attitudes about wildlife in turn influence proenvironmental behaviours (Gosling & Williams, 2010).Therefore, implementing projects that foster pro-environmental attitudes and behaviour can be a useful component of conservation strategy to redress the current biodiversity and environmental crisis (Reddy et al., 2017;Whitburn et al., 2020).
The promotion of ecological knowledge is a fundamental part of environmental education (Legault & Pelletier, 2000) whereby educational activities play a role in fostering "environmental literacy" (Pitman & Daniels, 2020).Nevertheless, increasing knowledge about conservation and environmental issues does not necessarily lead to behavioural change (Charles et al., 2018;Green et al., 2019;Otto & Pensini, 2017) and may have little effect on pro-environmental behaviour (Otto & Pensini, 2017).There is a growing understanding of the role of affective connectedness, and social interactions, to support pro-environmental behaviour (Barrable & Booth, 2020;Passmore et al., 2020;Žukauskienė et al., 2021).Nature-based science activities can promote awareness of environmental and conservation topics and mitigate the so-called extinction of experience that could arise in highly urban environments (e.g.Dean et al., 2019).Combined with the benefits of mood contagion (Neumann & Strack, 2000;Pérez-Manrique & Gomila, 2022;Volmer, 2012), social contexts with positive peer-group leaders could increase the probability of pro-environmental behaviour (Barsade, 2000;Sy & Choi, 2013;Volmer, 2012).
A lack of environmental awareness around certain topics can be associated with emotional (dis)connectedness and in this way influence individual behaviours (Bamberg & Möser, 2007).The willingness to cooperate in societal issues, for example by adopting pro-environmental behaviour, has been shown to be related to an individual's affective reactions to losses and gains in biodiversity, their competence in both general and topic-specific awareness, and their personal moral norms (Nordlund & Garvill, 2003;White et al., 2020).If people are unaware of current environmental challenges that need to be addressed and/or do not feel emotionally motivated to identify an issue as a challenge, there will be no focused effort to do so.
Given the global, regional and local patterns of rapid biodiversity loss (Cardinale et al., 2012;Kardol et al., 2018;Ross et al., 2021), creative and inspiring ways are needed to redress the vicious cycle of increasing threats to the natural environment and human disengagement (Boehnert, 2012).In fact, there is a strong parallel between precipitous biodiversity decline and human disengagement with environmental issues (Bandura, 2007), which might be context-related (Qin et al., 2022).This phenomenon seems to arise from complex interactions between and within social groups that justify harmful practices and put high costs on the environment (Bandura, 2007).Engaging people to be aware of and think about the problem and identify solutions may be a key to solving complex environmental challenges (Daniels & Good, 2015).In other words, an integration of environmental knowledge and emotional connectedness may forge a path for a sustainable approach to environmental issues, as (human) behaviour is the product of the orchestration of many brain areas, whose functions lead to emotion and cognition (Ardoin et al., 2015;Pessoa, 2008).
In particular, birds appear to be of significant public interest, as human interactions with birds (e.g.bird watching and interactions with birds found in one's garden) have a direct impact on human emotions, and generally, humans are more concerned about their negative impacts on avian populations than say on invertebrate populations (Ainsworth & Burns, 2020;Castillo-Huitrón et al., 2020).A recent study shows that natural soundscapes, i.e. birdsong, foster human well-being in protected areas contributing to mental health (Ferraro et al., 2020).Additionally, music and sounds in general, including language and animal sounds, are well-known factors that evoke emotions (see Koelsch, 2015;Panksepp & Bernatzky, 2002).
In the present study, we assess whether an experience with a conservation issue involving woodland songbirds and several interactions with scientists change the selfreported attitudes and intended conservation behaviours of the participants towards conservation.For this purpose, we invited 55 participants to engage in two 3-h workshops to learn more about avian communication and to collect data on bird behaviour with researchers in the field in the Mount Lofty Ranges (MLR; South Australia) using bird watching and playback experiments.
The MLR in South Australia is a biodiversity hotspot, yet the area has been cleared by ~90% for human activity since European colonisation (Bradshaw, 2012;Daniels & Good, 2015;Ford et al., 2001;Szabo et al., 2011).Even this remaining 10% creates uncertain conditions for population and species persistence because multiple ecological processes tend to negatively impact wildlife in small isolated fragments (Ford, 2011).Consequently, the MLR has a high risk of imminent woodland species extinctions (Szabo et al., 2011).By 2000, the MLR had lost 8 of 120 species of the pre-European fauna with an estimated looming extinction debt of another 40 species within 10 years (Daniels & Good, 2015;Ford & Howe, 1980;Garnett & Crowley, 2000;Garnett et al., 2003Garnett et al., , 2011;;Szabo et al., 2011).From 2000 onwards, population trends showed marked annual declines in more than half of woodland bird species, particularly the insectivorous songbirds (Prowse et al., 2021;Szabo et al., 2011Szabo et al., , 2012)).Predatory birds and brood parasitic cuckoos, in contrast, increased (Ford, 2011;Ford et al., 2001).
Analysis of acoustical networks among Australian woodland birds has shown that different species learn the alarm call of sympatric species and respond to the alarm calls of other species with adaptive behaviour.Even when broadcast from a speaker and in the absence of any actual predator, several Australian songbirds show different behaviours depending on the type of alarm call they hear (Magrath et al., 2020).Songbirds can be described as living in a "vigilant neighbourhood" and can interpret the warning calls of different species to different types of threat.The New Holland Honeyeater (Phylidonyris novaehollandiae), for instance, produces alarm calls that use referential signalling, whereby the alarm call structure is associated with a particular predator type (McLachlan, 2019).Therefore, honeyeaters can be considered sentinels of predation threats to endemic bird communities (Magrath et al., 2009;McLachlan, 2019).Additionally, the New Holland Honeyeater is an important endemic pollinator species in Australia (Paton, 1982) and is very well known to local people.Therefore, because of its key ecological function for pollination and its sentinel role for endemic avian communities, the New Holland Honeyeater is a suitable flagship species for conservation purposes, as for instance to highlight the vulnerability of the ecological system of MLR.
We conducted surveys before and after the experience, during which we tested participants' self-reported knowledge about the conservation status of woodland birds, concern for the state of woodland birds and for songbird research, and intended activities to acquire more information about woodland birds.We expected the experience (and the gained knowledge) to result in significant changes in self-reported attitudes about woodland conservation and to motivate participants to further engage in activities around woodland conservation.

Ethical aspects
The participants were recruited through posters and then attended an introductory lecture where they received the letter of information (in accordance with Flinders University Social and Behavioural Research Ethics Committee SBEC6933) and gave their consent to participate.This research complies with the ethical guidelines of the Flinders University Animal Ethics Committee (E370) and the Flinders University Social and Behavioural Research Ethics Committee (SBEC6933).The science permit was approved by the Department for Environment and Water of South Australia (Z24699-14).

Study design
This study was carried out during August 2015 at Flinders University and Cleland Wildlife Park, South Australia.A group of 55 volunteers participated in two activities: (1) a three-hour workshop on avian communication that was hosted at Flinders University and 2 weeks later (2) a three-hour guided-excursion to Cleland Wildlife Park to observe and record bird behaviour in the wild.Using a survey, we scored the beliefs, attitudes, and intended behaviours of each participant about woodland bird conservation before and after six hours of experience with the researchers.

Participants
The participants were recruited via on-campus posters and online advertisement from among the student body at Flinders University as well as the broader community, as many cafeteria staff, administration staff and international visitors regularly stroll across the campus.The call for participation was open for 2 weeks.The rule of acceptance was first come, first served.The number of participants was limited to 55 as this was the maximum room capacity for the workshop.After reaching the capacity number, the call was closed and further requests for participation could not be accommodated.There was no type of compensation (monetary or services) linked with participation, which was fully voluntary and anonymous, however volunteers did receive a free bus ride and guided experience at the wildlife park.The age of the participants ranged from 18 to 74 and spanned 10 nationalities (Table S1).Per gender, the number of participants was 38 females, 16 males, 1 non-binary.Most (93%) had either part-time or full-time employment, and some (7%) were full-time students.

Experience of science: activities
Activity 1: Participants were invited to attend an introductory meeting and complete a survey.Immediately thereafter, they were retained for all activities and joined the threehour workshop (Activity 2).
Activity 2: Three-hour workshop on avian communication.Participants were invited to attend a three-hour workshop listening to birds calls at Flinders University, where they had the possibility to be introduced to acoustic software used in avian research (e.g.RavenLite, Cornell Lab of Ornithology Bioacoustics Research Program, Ithaca, New York).Spectrograms of New Holland Honeyeater alarm calls were prepared in advance by the researchers, and the call structure of the alarm calls in response to different predator types (e.g.fox and bird of prey) was explained to the participants.Alarm calls with a broader frequency "chirp" are used for terrestrial predators, and the narrower frequency "whistles" are used for aerial predators (Jurisevic & Sanderson, 1994;Magrath et al., 2020).The conservation significance of alarm calls as well as the behavioural response of the songbirds at Cleland Wildlife Park to predator threat was put into ecological context through lectures and impromptu questions/answers about the state of the Mount Lofty Ranges, the high proportion of declining woodland species (Szabo et al., 2011) and the increase in native avian predatory birds in human-altered environments (Chace & Walsh, 2006).
Activity 3: Three-hour field trip to observe the response to alarm calls.Two weeks after the initial meeting to listen to and visually inspect alarm calls in the lab, the researchers organised a bus and travelled with 27 and 28 participants, respectively, on two different days to Cleland Wildlife Park.The participants added their names to the date of their choice on a list.We spent a total of 3 hours in the park; ~1.5 h was spent walking around the park conducting playback trials to test neighbourhood response patterns to the terrestrial alarm call of New Holland Honeyeater, ~1 hr was devoted to presentations by each group summarising their observations; ~30 min was used for the final survey.
To create a manageable group size for experimentally observing birds' responses to the broadcast of pre-recorded alarm calls of the New Holland Honeyeater, we divided the ~28 participants into five groups of five to six participants.Each group was led by an experienced ornithologist (Class A or Class R bander, PhD student or Professor in Ornithology) who had been trained in all research methods and who was listed on the animal welfare ethics permit to conduct teaching experiments using alarm calls.The research scientist placed a speaker in a known avian territory and broadcasted alarm calls.The playback trial consisted of 60 New Holland Honeyeater alarm calls broadcast as 10 alarm calls (one call every 0.2 s) every 10 s for 1 min.This is within the range of the normal alarm calling of the species (Jurisevic & Sanderson, 1994).During the minute of alarm calls and the following minute of silence, the lead scientist recorded the species' identities and the participants aided in noting down the behaviour of all birds; behaviours included "approach the speaker within 10 m", "vocalisations" and/or "crosses over the speaker".One participant was designated as the data recorder, one had a stopwatch, and four watched in all directions to make sure we observed all activity.At the end of the 2 min, the participants and the lead research scientist recapped the main observations as a group and made sure they agreed on what had been seen.Then, the group moved on to the next territory.Each group conducted five experimental trials.Participants gained first-hand experience observing animal response.Each group of five participants summarised their observed response patterns in a presentation to the whole participant group in the third hour.In general, between one and five different species responded to the playback, approached the speaker, and called to the playback of the alarm call of a New Holland Honeyeater.The participants could observe the neighbourhood effect of vocalisation among woodland birds.
Activities 1 to 3 were led and managed by a team of students and researchers led by SK.

Survey questions
Each participant completed two surveys: one before and one after the 6 h of activity (split across two meetings).The first survey was done within the first half hour following arrival as part of activity 1 (before the laboratory experience of listening to and visualising alarm calls), and the second survey was done during the last 30 min of activity 3 two weeks later (after the field trip to Cleland Wildlife Park).During the first survey, the participants were given a survey ID number; they were told to remember the ID (to store the number on their phone, for example) for 2 weeks so the survey questionnaires could be analysed anonymously but each pair of responses could also be analysed per person.
We conducted a baseline survey with 16 questions (Supplementary material, Table S2).In addition to age, gender, and nationality, five questions were analysed for the purposes of this study (12) how did you gain your current knowledge of woodland songbird conservation status?Please select all that apply (Television, Radio, News, Publications, Books, Friends, Conservation groups, Volunteer groups, School Teachers, University Lecturers, Zoo visit, Cleland visit, Other), (13) How would you describe your level of knowledge about songbirds?(poor, average, good, very good), ( 14) How concerned are you about the conservation status of woodland songbirds?(not at all, a little concerned, fairly concerned, very concerned), (15) If research or conservation efforts towards woodland songbirds stopped tomorrow how concerned would you be? (not at all, a little concerned, fairly concerned, very concerned), and ( 16) Other Comments?Please state.
After the experience in the field, we conducted a treatment survey (Supplementary material, Table S3).We analysed the following questions: (6) How would you describe your level of woodland songbird knowledge (poor, average, good, very good), (7) How concerned are you about the conservation status of woodland songbirds (not at all, a little, fairly concerned, very concerned), (8) If research or conservation efforts towards the woodland songbirds stopped tomorrow how concerned would you be (not at all, a little concerned, fairly concerned, very concerned), (9) How interested would you be in learning more about woodland songbirds, and (10) If you are interested in learning more about woodland songbirds, how would you like to receive information?Please select all that apply (Join Birds SA, Join Friends Group, Join Volunteer Group, Brochure, Country show or stall, Email newsletter, Field trip, Museum visit, Research paper, Research/ update, presentation, Website, Zoo visit, Higher education, Other -please state).

Statistical analysis
Data were analysed with R version 4.0.2(R Core Team, 2020) and SPSS version 20.1.To test whether the science experience affected the response scores (response variables: (1) knowledge about songbirds, (2) awareness about songbird conservation status, (3) concern about research), we conducted Cumulative Link Mixed Models (CLMMs) with an adaptive Gauss-Hermite quadrature approximation (number of quadrature points = 3) using the R function clmm2 (package ordinal, Christensen, 2019).Response scores were ordered for each response variable (e.g.poor < average < good < very good, not at all < a little concerned < fairly concerned < very concerned).We included surveys before and after as a fixed effect.To control for their potential effects, we included gender and age group as further fixed effects.The ID of each participant was included as a random intercept effect.As an overall test of the effects of the survey before and after we conducted a full-null model comparison (likelihood ratio test) to avoid multiple testing.The null model lacked the fixed effect of the survey but was otherwise identical to the full model including the control fixed effects (i.e.gender and age group).To obtain more accurate p-values for the effects of individual predictors, we applied likelihood ratio tests (comparison of the full model with reduced models; Barr et al., 2013).To test for a measure of association between our "experience" and the self-ranked response scores per question, we used Cramer's V.

How would you describe your level of knowledge about songbirds?
Overall, survey before and after, gender and/or age group clearly influenced self-ranked responses across categories of knowledge (full-null model comparison: χ 2 = 44.870,df = 8, p < 0.001).The difference in response across categories (poor, average, good, very good) of self-ranked "knowledge about songbirds" was significantly different before and after the experience (comparison of full model with reduced model not comprising survey: χ 2 = 35.141,df = 1, p < 0.001, Table S4; Figure 1).There was a higher percentage of responses to "good" and "average" knowledge after the experience and a higher percentage of responses to "poor" before the experience.The mean ± se self-reported rank in "knowledge" was lower before (1.6 ± 0.07) than after (2.2 ± 0.07) the experience.Self-ranked responses differed between age groups (comparison of the full model with the reduced model not comprising the age group: χ 2 = 12.189, df = 5, p = 0.032), whereby participants aged 35-44 showed lower knowledge of songbirds (Table S4).Gender had no obvious impact on self-ranked responses across categories of knowledge (comparison of the full model with a reduced model not comprising gender: χ 2 = 0.790, df = 2, p = 0.674).There was a strong positive association between "experience with the scientists" and "knowledge" (Cramer's V = 0.497, p < 0.001).

How concerned are you about the conservation status of woodland songbirds?
Overall, survey before and after, gender and/or age group clearly influenced self-ranked responses across levels of concern (full-null model comparison: χ 2 = 39.419,df = 1, p < 0.001).The difference in response across levels of concern (none, low, moderate, high) of self-ranked "concern about the conservation status of woodland songbirds" was significantly different before and after the experience with the scientists (comparison of the full model with reduced model not comprising survey: χ 2 = 39.049,df = 1, p < 0.001, Table S4; Figure 2).There was a higher percentage of responses to "high concern" after the experience with the scientists (50.9%) versus before the experience (10.9%).The mean ± se self-reported rank in "concern about the conservation status of woodland songbirds" was lower before (2.6 ± 0.09) than after (3.5 ± 0.08) the experience.Gender (comparison of the full model with reduced model not comprising gender: χ 2 = 1.904, df = 2, p = 0.386) and age group (comparison of the full model with reduced model not comprising age group: χ 2 = 6.879, df = 5, p = 0.230) had no obvious impact on self-ranked responses across levels of concern.There was a strong positive association between "experience with the scientists" and "concern about conservation status" (Cramer's V = 0.552, p < 0.001).

If research or conservation efforts towards woodland songbirds stopped tomorrow, how concerned would you be?
Overall, survey before and after, gender and/or age group clearly influenced self-ranked responses across levels of concern (full-null model comparison: χ 2 = 33.557,df = 1, p < 0.001).The difference in response across levels of concern (none, low, moderate, high) of self-ranked "concern about stopping research into woodland songbirds" was significantly different before and after the experience with the scientists (comparison of full model with reduced model not comprising survey: χ 2 = 33.557,df = 1, p < 0.001, Table S4; Figure 3).There was a higher percentage of responses to "high concern" after the experience (81.7%) versus before (34.5%).The mean ± se self-reported rank in "concern about stopping research into woodland songbirds" was lower before (3.2 ± 0.09) than after (3.8 ± 0.05) the experience.Gender (comparison of the full model with reduced model not comprising gender: χ 2 = 3.250, df = 2, p = 0.197) and age group (comparison of the full model with reduced model not comprising age group: χ 2 = 6.692, df = 5, p = 0.245) had no obvious impact on self-ranked responses across levels of concern.There was a strong association between "experience with the scientists" and "concern about stopping research" (Cramer's V = 0.494, p < 0.001).

Activities to acquire information about woodland bird conservation
Participants increased their selected number of intended activities to acquire information about woodland songbird conservation after the science experience (Figure 4).For this question, we compared the number of activities respondents ticked in response to the baseline Q1 "How did you gain your current knowledge of woodland songbird conservation status?" against the number of activities the participants ticked in response to the Self-reported concern about the conservation status of woodland birds Figure 2. The percentage response of 55 participants to the question "How concerned are you about the conservation status of woodland songbirds?"before (black bars) and after (grey bars) joining the project on woodland songbird behaviour (see methods).Participants significantly increased their selfreported rank of "attitude towards conservation status" after the experience (see results).

Level of concern
treatment Q4 "If you are interested in learning more about woodland songbirds, how would you like to receive information?".The mean ± se number of activities during the baseline survey was 2.3 ± 0.3, and during the treatment survey as 3.7 ± 0.3 (t-test on logtransformed data: t = 3.447, p = 0.001).Inspection of Figure 4 shows that the distribution of responses changed: during the baseline survey, the response distribution was skewed left (Shapiro-Wilk p < 0.001) with most respondents answering "zero activities"; whereas during the treatment survey, the response distribution was normal (Shapiro-Wilk p = 0.064).

Discussion
A six-hour experience of participation in science involving personal contact with the researchers and personal experience about woodland birds resulted in significant shifts in the self-reported attitudes and intended behaviours of 55 participants.We suggest that experience with a scientist can promote win-win scenarios for all the participants (scientists and non-scientists; short-term commitment of the participants) and for society (potential long-term engagement) (Joyce et al., 2018).We discuss limitations of this study below and summarise the main findings here.Before the experience, no one (0%) self-reported to have a "good" knowledge of woodland songbirds compared with  22% after; before the experience, 11% were very concerned with the conservation status of woodland birds compared with 51% after; before the experience, 35% were very concerned about cessation of songbird research compared with 82% after.The shifts in participant response were 2-to 5-fold stronger after two experiences across a two-week period of information and outdoor activity with scientists leading the experience.Another marked shift was the change in intended behaviours to engage in acquiring information about woodland songbirds: participants self-reported that they would engage in ~1.5 more activities after the experience as compared to before the experience.Interestingly, the response distribution in intended activities changed from being leftskewed (many participants stated zero current activities, average ~2 activities) to a normal distribution with ~4 planned activities after participating in the overall experience (Figure 4).Nature-based science activities foster opportunities for people to interact with nature and could potentially help transform attitudes about human -nature interactions relevant to redress the biodiversity crisis (i.e.extinction of experience, Schuttler et al., 2018;Soga & Gaston, 2016).

Level of concern
There are several limitations to this study that we fully acknowledge.We cannot state if the change in self-reported attitudes or intended activities was specifically associated with the experience we organised or with other events across the same time period.Future research should include a control group.The room capacity for the three-hour workshop prior to the field trip constrained the number of participants and contributed to the small sample size.Also, the rule of acceptance "first come, first served" did not allow us to address either the cultural background of the participants or potential gender bias among participants, which could have influenced the decision to participate (Paleco et al., 2021).Since social, educational and health conditions can affect participation in physical and group activities (Hernán et al., 2004), we acknowledge that our data set may be skewed because participants disproportionately possessed traits affecting participation that could be linked with the measured outcomes.However, the age range in this study (18 to 74) likely reflects at least some breadth of participant traits and life experiences.Notably, half the participants were from Aus/NZ, and therefore the results could reflect a response especially for this demographic.Future research could explore the role of participant characteristics for response strength in more detail.Finally, we acknowledge that the pre-survey Q1 "How did you gain your current knowledge of woodland songbird conservation status?" addresses where participants collected information prior to the activity, and the post-survey Q4 "If you are interested in learning more about woodland songbirds, how would you like to receive information?" potentially measures communication preferences and that the two variables are therefore not directly comparable.However, with caution, one could make the case that the behaviour change we have addressed involves a potential shift in sources of information acquisition that could be associated with different behaviours required to engage with different information sources.
The possibility of sharing knowledge and discussing a common topic of interest with other volunteers or scientists is often considered a building block that can lead to further active engagement for environmental and/or conservation issues (Jordan et al., 2011;Oberhauser & Prysby, 2008;Toomey & Domroese, 2013).Crall et al. (2013) evaluated participants involved in monitoring invasive species and showed differences in their current and planned behaviour following the project's training.In our study, participants reported that their way of thinking about nature had changed as a synergistic consequence, perhaps, of close contact with scientists and participation with science in nature (Crall et al., 2013).Toomey and Domroese (2013) suggested that participation itself leads to the perception of having done something positive for the environment ("greener sense of self"), thus leading to more positive attitudes towards conservation behaviours in general, and thereby reinforcing the motivation that originally drove the participants to join the project (Maund et al., 2020).Nature experience has been shown to positively affect mental health and may increase feelings of optimism (Dallimer et al., 2012).In our study, the self-stated willingness to further engage in pro-environment activities was expressed just after the nature experience and therefore could be considered an ephemeral signal.Ideally, we would have conducted a third survey several weeks or months later to test for the persistence of a long-term commitment to the topic, as suggested by other studies (e.g.Ardoin et al., 2015).Since the study design does not allow us to fully disentangle the single effects achieved by the informative session (Activity 2), the outdoor activity per se (Activity 3) or the potential role of the people during the outdoor activity (data recorder, bird observer) (Activity 3), we only report on the shift in attitudes and intended behaviours overall.
Creating opportunity for emotional connectedness with conservation issues could result in more conservation behaviour.Research into learning by experience (i.e.effects of handson experience on students' interests and attitudes) has a long history in the field of science education (Bergin, 1999).There is a general consensus that hands-on experiences foster students' learning effectiveness (Satterthwait, 2010).However, the performance of various hands-on activities can influence students' interests differently and is context dependent (Holstermann et al., 2010;Sjøberg & Schreiner, 2005).Furthermore, there is empirical evidence that collaboration between education and research enhances motivation for out-of -school learning (Miczajka et al., 2015;Scheuch et al., 2018) and increases the acquisition and retention of non-traditional information compared to classroom-based curriculum learning (Hirschenhauser et al., 2019).Another example of altered understanding about environmental issues is provided by the "Austrian roadkill project"; participants that monitored dead animals along roads had measurably increased understanding of wildlife conservation issues and were more willing to share their awareness with others (Heigl & Zaller, 2014).The volunteers in the study acted as knowledge multipliers and contributed to raising public awareness about wildlife conservation, and of science, in general.Engaging the public in the scientific process may contribute to an increased appreciation of the wider contexts in which science operates, including as part of culture and policy (Bonney et al., 2009;Mitchell et al., 2017), and as a potential promoter of environmental empathy through engagement (Schuttler et al., 2018).However, there are additional costs when involving volunteers in research (e.g. from training to data quality) that must be considered during project planning if the aim is also to generate publishable data with data collected by volunteers (Frigerio et al., 2018;Thornhill et al., 2016;Vohland et al., 2021).
Conservation outcomes are urgently needed to safeguard rapidly declining biodiversity.In this particular study, the condition of the remaining woodlands within the southern Mount Lofty Ranges is highly management-dependent to avoid further deterioration and loss of species (Paton et al., 2010).While governmental and nongovernmental agencies are actively involved in managing the threats affecting the woodlands, spatial coverage is often inadequate and management actions need to be continuously re-applied (Paton et al., 2010).There is also an urgent need to ensure that management actions benefit woodland birds (Belder et al., 2018).This study shows that there could be an appetite for engaged citizen scientists to contribute to the ongoing management of the Mount Lofty woodlands (see also Fritz et al., 2019), perhaps through nature experience, and perhaps also through encouragement of citizen involvement in local policy decisions (Bonney et al., 2009(Bonney et al., , 2014)).

Conclusion
The geographical scale of biodiversity management required to inform conservation policy will rely on networks of people working together.Our study suggests a measurable impact of nature-based science activities on the participants.Participants increased their self-reported knowledge about the conservation status of woodland birds and reported a change in attitude about the value of woodland conservation programmes.Further research should address the types of knowledge (factual and processrelated) that might be related to environmentally protective outcomes.In combination with their self-reported motivation to become more involved in woodland conservation activities, it is possible that participants may have experienced personal transformation in line with pro-conservation behaviour as the result of participating in activities with a professional in the natural sciences.It is essential to monitor such changes over the long-term, in order to investigate to what extent such shifts persist and result in tangible actions.

Figure 1 .
Figure1.The percentage response of 55 participants to the question "How would you describe your level of knowledge about songbirds?"before (black bars) and after (grey bars) the experience with scientists about woodland songbirds (see methods).Participants significantly increased their selfreported rank of "knowledge of songbirds" after the experience (see results).

Figure 3 .
Figure 3.The percentage response of 55 participants to the question "If research or conservation efforts towards woodland songbirds stopped tomorrow, how concerned would you be?" before (black bars) and after (grey bars) joining the project on woodland songbird behaviour (see methods).Participants significantly increased their self-reported rank of "attitude towards conservation research" after the experience (see results).

Figure 4 .
Figure 4.The number of different activities participants are currently doing or intend to engage in to acquire information about woodland songbirds (a) before and (b) after joining the project on bird behaviour.The mean number of self-reported intended activities increased after the data collection experience, and the distribution pattern in response changed from skewed (before) to normal (after) (see results).