Assessing positivity and negativity biases: Emotions and attitudes influence animal cognitive salience in a French National Park

doi:10.21203/rs.3.rs-3997228/v1

Download PDF

Research Article

Assessing positivity and negativity biases: Emotions and attitudes influence animal cognitive salience in a French National Park

https://doi.org/10.21203/rs.3.rs-3997228/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Human representations, attitudes and behaviors are influenced by a range of cognitive biases increasingly understood by neuroscience, psychology and economics. Similarly, how cognitive biases affect people’s relationships with non-human livings is an emerging research topic in conservation biology. Yet, assessing cognitive biases remain a methodological challenge in this field of research that mostly rely on in-vivo methods. This study presents an interview-based approach that links cognitive salience and people’s attitudes and emotions to assess positivity and negativity biases. The approach was tested in the Cevennes National Park in France where we investigated how animal cognitive salience was influenced by people’s attitudes and emotions towards animals. Interviews conducted with 100 respondents combined free-listing tasks of the fauna to assess animal cognitive salience and questions to inform people’s attitudes and emotions towards listed animals. Relationships between cognitive salience and attitudes and emotions were analyzed through multi- and bivariate analyses. Respondents listed a total of 154 animals and mainly associated them with positive attitudes and emotions. A significant relationship was found between animal cognitive salience and people’s attitudes and emotions (ANOVA, p < 0.0001): animals associated with positive emotions were more salient in people’s minds than others, and disliked and negative animals had the second highest cognitive salience score. Aligning with neuroscience advances, results suggested the coexistence of a positivity bias and, to a lesser extent, of a negativity bias. We finally discuss the limitations and transferability potential of this approach for further investigating how cognitive biases may influence biodiversity conservation efforts.

Free-listing

Ethnobiology

Environmental psychology

Human-nature relationships

Mixed-methods

Nature’s contributions to people

After decades of research and practices in conservation biology, it is now undisputed that one of the key challenges to successful conservation stands in the conciliation between ecosystem’s and people’s well-being in so-called ‘social-ecological systems’ (Palomo et al. 2014; Cumming 2016). Under this social-ecological perspective, biodiversity and ecological dynamics are effectively intimately intertwined with people and socio-cultural dynamics. As a result, understanding people’s values, attitudes and behaviors towards non-human livings has become a critical objective in conservation biology (Mascia et al. 2003; Schultz 2011; Christie et al. 2017). This first represents a major epistemological challenge for the discipline due to the barriers to meaningful integration of the social sciences in conservation biology (Bennett et al. 2017). This also is an ontological challenge due to the complexity of the link between people’s values, attitudes and behaviors, which is mediated by a large set of social, cultural and cognitive processes acting at individual and collective levels (Scales 2012; Dietsch et al. 2020; Wallen and Landon 2020).

This paper aims to engage with this second challenge, and in particular with the challenge of better understanding the role of cognitive biases in how people interact with non-human livings. In recent years, several studies have effectively pointed to the influence of cognitive biases on human decision-making and behavior, including their role in the (in)ability of human societies to tackle current sustainability and biodiversity conservation problems (Engler et al. 2018; Cinner 2018; Buijs and Jacobs 2021). For example, research showed that the framing of a message (e.g., positive vs. negative framing) plays a key role in how people receive this message and adjust or not their behavior accordingly, which is called the framing bias (Kusmanoff et al. 2020). The status quo bias is another example that stands for people’s preference for business-as-usual options against options that would require a reform or a change, which could explain why negotiations for a global climate agreement are so tough (Eidelman and Crandall 2012; Engler et al. 2018). Due to the key influence of such cognitive biases on people’s views and behaviors, several authors argued that reaching a better understanding of these has an untapped potential to help towards more effective conservation actions. For example, by being aware of how the framing bias works, conservation practitioners would be able to design more effective messages and better orient people towards pro-environmental behaviors (Byerly et al. 2018; Kusmanoff et al. 2020).

In the face of the large diversity of cognitive biases that are relevant to sustainability and biodiversity conservation issues (Engler et al. 2018), the so-called negativity bias could be seen as a priority. This cognitive bias stands for the fact that humans give greater weight to negative entities such as negative events, objects or traits (Rozin and Royzman 2001). In other words, this bias induces a propensity of humans to attend to, learn from, and use negative information far more than positive information, which is argued to serve critical evolutionarily adaptive functions (Vaish et al. 2008): individuals that better recalled dangerous places and plants were more likely to survive.

While empirical evidences for the negativity bias accumulate, several recent studies have explored this bias in the context of sustainability and conservation sciences. In a meta-analysis, Kansky & Knight (2014) found that the perceived costs of human-wildlife coexistence were more significant than the perceived benefits to explain people’s attitudes towards wildlife. They concluded in the evidence of a negativity bias in people’s attitudes towards wildlife in a context where human-wildlife conflicts were explicit. Similarly, based on a selection of scholars, Blanco et al. (2019) hypothesized that people’s behaviors towards ecosystems might be more influenced by ecosystem disservices (i.e., ecosystem-based detriments) than services (i.e., ecosystem-based benefits). In another perspective, Buijs and Jacobs (2021) advocated for the relevance of positive psychology theory to counteract the negativity bias, in particular in the domain of human-wildlife coexistence. Yet, despite the relevance of the topic, research on cognitive biases in general, and on negativity bias in particular, is only emerging in conservation biology. We argue that this research gap is partly explained by a methodological gap. Indeed, while neurosciences and economics quantify cognitive biases based on experiments (e.g., measuring brain activity or conducting choice experiments under controlled environments), these methods are rarely adapted to conservation social sciences that mostly rely on non-experimental protocols and methods such as structured and semi-structured interviews and observations.

Against this background, this article aims at a twofold objective. First, it aims to devise an interview-based approach to assess the existence and magnitude of a potential negativity bias (or its counterpart, the positivity bias) in the domain of human-wildlife coexistence. The approach relies on a combination of concepts and methods from the social sciences, which we further describe in the background and methods sections. Second, by applying the approach in a case study in South-Eastern France, in the Cevennes National Park (CNP), this paper aims to contribute to filling the knowledge gap on the role of cognitive biases in people’s relationship with the fauna. More specifically, our research question in this case study was: is there any evidence of a negativity bias (or, on the contrary, of a positivity bias) in people’s mind regarding animals of the CNP? Finally, after detailing the results obtained from this case study, we discuss the potentials and limits our approach to inform on people’s cognitive biases in the domain of biodiversity conservation planning and management.

There are various definitions and ways of theorizing the concept of cognitive bias according to disciplines. Clarifying which definition to rely on is key due to practical methodological implications on cognitive bias assessment. In neurosciences, cognitive biases are defined as the tendency of the human brain to overreact (or, on the contrary, under-react) to certain stimuli (Norris 2021), requiring a monitoring of brain activity in laboratory to be measured. In psychology, cognitive biases are viewed as the systematic errors of reasoning induced by adopting heuristics, i.e. mental shortcuts or rules of thumb that people use to make decisions and solve problems quickly (Tversky and Kahneman 1974; Das and Teng 1999). In economics, cognitive biases refer to the gaps between observed people’s behaviors and theoretical behaviors as predicted by rational-agent models (Kahneman 2003). Considering this diversity of theoretical backgrounds and underlying hypotheses, we will here refer to cognitive biases as understood by neurosciences, i.e., as an overemphasis of certain stimuli rather than other.

Following this definition, the concept of cognitive bias is proximate to the concept of cognitive salience that is used in several disciplines from the social sciences, including linguistics (e.g., Flecken et al. 2015), neurosciences (e.g., Caduff & Timpf 2008) and anthropology and ethnobiology (e.g., Smith and Borgatti 1997; Blanco and Carrière 2016; Robbins et al. 2017). Based on these different disciplines, cognitive salience can be defined as the prominence or attention-grabbing nature of certain information or stimuli in an individual’s cognitive processing (see Caduff & Timpf 2008). When something is cognitively salient, it stands out and captures a person’s attention more readily than other, less salient information. By influencing which information is more prominent and accessible in a person’s mind, cognitive salience therefore contributes to shaping people’s perception, memory, and decision-making. In sum, differences in cognitive salience between different items such as different plants or animals could indicate potential cognitive biases towards certain items.

There are different methods to assess cognitive salience, but a standard and widely used one is free-listing or free-recall tasks (Albuquerque et al. 2005; Quinlan 2005; Thompson and Juan 2006; Dany et al. 2015). The method consists in asking people to cite as much as items they can relative to a certain topic, for example the animals living in a given region or the medicinal plants they know. Free-listing tasks rely on two assumptions about the cognitive salience of cited items: frequently cited items are more prominent to interviewed people than unfrequently cited items; and items ranked first in the lists are more familiar to people than items ranked last (Quinlan 2005).

On this basis, we posit that cognitive salience and its assessment through free-listing tasks could be used to reveal and quantify cognitive biases regarding animals. In particular, we considered that if animals who are negatively-judged by people have significantly higher cognitive salience scores than positively- or neutrally-judged animals, then negatively-judged animals will capture people’s attention more readily than other animals. This would suggest the existence of a negativity bias towards animals, in the sense of an overemphasis of people towards negative animals. Reversely, higher cognitive salience scores for positively-judged animals would suggest the existence of a positivity bias. Due to the characteristics of free-listing tasks, such an approach could yet only capture cognitive biases at collective level and would be unsuited to the assessment of cognitive biases at individual level.

On the basis of these prerequisites and assumptions, we therefore developed a methodology that allows to assess positivity and negativity biases regarding animals by linking (i) animal cognitive salience (assessed through free-listing tasks) and (ii) people’s attitudes and emotions towards cited animals (assessed through ad-hoc questions).

3.1. Study area

Cevennes National Park (CNP) is the oldest French National Park (created in 1970) located in the southern part of the Massif Central Mountain range. This inhabited park covers an area of approximately 3,000 square kilometers – including a core zone of 938 km² – overlapping four biogeographical areas (the Causses, Cévennes, Aigoual and Mont Lozère). Because of its location in the transition zone between Mediterranean and Atlantic climates and a wide range of altitudes (from 200 to 1,500 meters above sea), CNP hosts a rich and unique biodiversity with a high rate of endemism.

Human activities, in particular pastoralism and crop cultivation in terraces, have shaped CNP landscapes over centuries. As such, CNP has been a UNESCO World Heritage site since 2011, with an explicit focus on preserving ecosystems and human activities that are part of CNP landscapes (PNC 2021). Consequently, even in the core area of CNP, the only one with permanent inhabitants in France, persevering so-called ‘traditional’ human activities such as hunting and certain agricultural practices are part of the park’s management plan (Larrère 2009). Finally, CNP is a key touristic area where people can enjoy different outdoor activities (e.g. hiking, canyoning, biking), learn about non-human entities (e.g. botany, zoology, geology), and get more familiar with the history, livelihoods and culture of CNP residents (e.g. silk, chestnut and onion cultivation, agrotourism, literature and gastronomy).

For this study, we focused on the Val d’Aigoual commune which is located in the southern part of the CNP buffer zone (aire d’adhésion in French; Fig. 1). The valley is characterized by its steep slopes and deep gorges, which have been carved over time by the Hérault and Tarn River and their tributaries. Its hilly landscapes comprise hamlets, pastoral areas, forests, terraced agricultural lands, and inhabited areas. The Aigoual Massif includes the highest peak in the Cevennes range, Mont Aigoual, which culminates at 1,567m. While the region used to be dominated by open grazing areas and is known for its silk industry (Cabanel 2013), it has been subject to a dynamic of forest encroachment since the end of the 19th century, caused by two joint phenomena. First, since 1882, different reforestation plans have been enforced by the National Forest Office, which resulted in the plantation of resinous tree species. Second, rural exodus has led to a dynamic of spontaneous afforestation in formerly cultivated terraces and grazed areas, which are now dominated by chestnut, holm oak and beech trees depending on the local environmental conditions (Cornu 2003). Even though sheep and goat-based pastoral systems persist thanks to local goat cheese markets and the promotion of the region's touristic and heritage value (Napoléone et al. 2015), CNP area harbors today a higher proportion of forests than in the middle of the 20th century.

3.2. Sampling design and respondent selection

Using convenience and quota sampling strategies, we recruited a total of 100 respondents to capture a diversity of profiles in terms of age, gender, place of residence and origin. First, we went to different locations, including villages, hiking trails, roads and rivers, which allowed us to recruit 75 respondents. As interviews progressed, respondents were selected to obtain a balanced sample of age and gender. Yet, this approach was not suited to grasp local residents. As a consequence, we also went to farms and other habitations, which allowed us to recruit 25 additional respondents.

Our final sample of 100 informants was well-balanced in terms of gender and status (local residents vs tourists), while variables such as age class, people’s origin (native vs non-native from Cevennes) and professional activity were not homogenously distributed (Table 1). Yet each modality was sufficiently represented to allow statistical analyses.

Table 1

Description of respondents’ socio-demographic characteristics.
Socio-demographic variables	Variable modalities	No. of respondents
GENDER	Woman	47
GENDER	Man	53
AGE CLASS (years)	18–29	24
	30–44	15
	45–59	31
	60–74	22
	75–91	8
STATUS	Local	52
STATUS	Tourist	48
ORIGIN	Native from Cevennes	19
ORIGIN	Not native from Cevennes	81
JOB	Farmers	4
	Craftsmen, traders, and entrepreneurs	7
	Executives and higher intellectual professions	9
	Intermediate professions	15
	Employees	11
	Workers	5
	Retirees	23
	People with no formal job	26

3.3. Interview procedure and ethics

Interviews were structured around three successive steps that aimed at (i) assessing animal cognitive salience, (ii) collecting respondents’ emotions and attitudes towards cited animals, and (iii) gathering explanations from respondents regarding their emotions and attitudes. We relied on free-listings to pursue the first objective; the two other objectives were pursued through semi-structured interviews (Fig. 2).

We initiated the free-listing with the following statement: "I would like you to list all the animals that come to your mind and that you associate with Cevennes landscapes." No time limit was set and the listing was stopped when respondents considered they had no more animals to add. On this basis, the 100 free-lists we obtained were used to calculate an index of cognitive salience for each cited item.

Following the free-listing, a semi-structured interview (see SI. Interview guideline) was meant to gather respondents’ attitudes and emotions towards animals, as well as explanations about these (Fig. 2). For attitudes, we aimed at taking into account the multidimensionality of people judgement about an animal, i.e., the fact that an animal can be seen as positive for some aspects (e.g., wolves are positive for tourism) and negative for other (e.g., wolves are negative for livestock farming). As a consequence, we asked respondents to identify from their list the animals they considered beneficial, detrimental and neutral to five material and non-material domains that are central to CNP landscapes and management policies: (i) crop cultivation, (ii) livestock farming, (iii) ecosystem functioning and preservation, (iv) CNP touristic interest and (v) CNP heritage value. Respondents were asked to explain their responses for each animal and domain. ‘No opinion’ answers were welcome. To collect emotions induced by animals, respondents were asked about how they would feel to cross or observe them. Here again, we welcomed all kinds of emotions as expressed by people.

Finally, we asked respondents for socio-demographic details, including their age, gender, profession, origin and status regarding CNP (i.e., permanent resident, temporary resident, or tourist). Interviews were strictly anonymous as we did not collect people’s names and did not record them. Consequently, no personal data as defined in the European General Data Protection Regulation was collected. After explaining the study’s context and objectives, we nonetheless obtained the Free Prior and Informed Consent of respondents orally. Depending on people’s willingness to discuss and availability, interviews lasted between 20 and 40 minutes. All interviews were conducted by the first author from August to September 2021.

3.4. Free-list analysis and cognitive salience

All statistical analyses were conducted under the R environment (R Core Team 2022). First, we cleaned free-lists by checking for synonyms and spelling mistakes, and translating the cited items from French to English. Instead of using scientific taxonomy, we kept the items as cited by people to remain as close as possible to their representations. For example, items such as ‘stag’ (cerf in French) and ‘hind’ (biche) were kept separate instead of merging them into a unique ‘deer’ item corresponding to Cervus elaphus.

Second, we classified cited items to facilitate result interpretations. A first grouping (animal status) was meant to distinguish wild vs domestic animals. A second grouping (animal group) was meant to distinguish amphibians, arthropods, birds, fishes, mammals, reptiles and fungi.

Third, we processed the 100 free-lists with FLARES (Free List Analysis under R Environment using Shiny, Wencelius et al. 2017). We calculated the most common cognitive salience indexes that take into account both item citation frequency and rank of citation, including Smith index (Smith and Borgatti 1997), Sutrop index (Sutrop 2001) and B’ score (Robbins et al. 2017). Because the three indices were highly correlated, we solely rely in this paper on the Smith cognitive salience index which is given by the following formula:

$$Sa=\frac{{\sum }_{i=1}^{N}\frac{Li-Rai+1}{Li}}{N}$$

S _a = Smith cognitive salience index for animal a

L _i = length of the list of respondent i

R _ai = rank of citation of animal a in the list i

N = total number of lists (or respondents)

Finally, we used the Henley index to calculate the distance between the different items based on the average difference of citation ranks for every pair of items (Henley 1969). Multidimensional scaling was processed using the distance matrix to explore item citation patterns. Because of analytical limitations regarding the management of rare items, this analysis focused on the items mentioned by at least 5% of the respondents.

3.5. Analyzing attitudes, emotions, and cognitive bias

Based on semi-structured interviews, each emotion and attitude expressed by each respondent for each animal was classified as positive (e.g., pleasure, interest, curiosity, amazement) negative (e.g., annoyance, fear, disgust), or neutral. For attitudes, we also had a ‘no opinion’ category. We then summarized the dataset at animal level by summing the number of respondents based on their answers for each domain.

In a preliminary step, we calculated from this dataset an index of positivity for each animal (ratio between positive opinions and the number of respondents asked about the animal) and for each domain (i.e., emotions, crop cultivation, livestock farming, ecosystems, tourism, heritage value). Yet, this led to a dataset with many zero values because for a specific domain (e.g., tourism), many animals were considered neutral (e.g., beetles, spiders, slow worms).

To solve this problem, we decided to switch to a categorical approach and classified each animal as positive, neutral or negative for a given domain based on the most frequent respondents' attitudes and emotions. As a result, each animal was characterized by six categorical variables (each variable corresponding to a domain: emotions, crop cultivation, livestock farming, ecosystems, tourism, heritage), with three possible modalities for each variable: positive, negative, and neutral.

On the basis of this dataset, we computed a Multiple Correspondence Analysis (MCA) and a Hierarchical Clustering on Principal Components (HCPC) with the FactoMineR package (Lê et al. 2008). The MCA was used to synthesize the information gathered on people’s emotions and attitudes towards animals and better understand the most discriminating emotions and attitudes. The HCPC then allowed identifying five clusters of animals based on people’s emotions and attitudes. To do so, we relied on the Manhattan distance associated with Ward’s agglomeration method, we calculated the number of clusters as the one with the higher relative loss of inertia, and we applied 1,000 iterations for the k-means consolidation.

Finally, to analyse the relationships between (i) animal cognitive salience and (ii) people’s emotions and attitudes towards animals, we processed a one-way ANOVA between Smith cognitive salience indexes and the clusters identified by the HCPC. Before processing the ANOVA, we operated a logarithmic transformation of the Smith cognitive salience index to meet the normality assumption. The normal distribution of the new log(S) variable was confirmed with a Shapiro test (Fig. S1, p = 0.43), and the homogeneity of variance between groups was checked with Levene tests (p = 0.27). Lastly, we computed Tukey post-hoc analyses that allowed to test the significance of the difference in log(S) between all pairs of animal clusters. The ANOVA and Tukey analyses were computed with the rstatix R package (Kassambara 2023).

4.1. Cognitive salience and proximity of CNP animals

Respondents cited a total of 154 animals, of which 53 were mentioned by at least 5% of the respondents (Fig. 3). On average, one respondent cited 9.2 ± 5.3 (SD) animals, with a maximum of 26 animals and a minimum of 3 animals. Data saturation, i.e., the minimum number of respondents required to get the full list of 154 animals, was obtained after 36 respondents (Fig. S2). The most cognitively salient animals were ‘wild boar’ (S = 0.55), ‘bird’ (S = 0.34), ‘fox’ (S = 0.28), ‘roe deer’ (S = 0.22) and ‘goat’ (S = 0.18). Yet, we note that cited animal names referred to different taxonomic categories: names either referred to a taxonomic class (e.g. ‘bird’, ‘insect’), family (e.g. ‘eagle’, ‘ant’), genus (e.g. ‘buzzard’, ‘earthworm’), species (e.g. ‘Bonelli’s eagle’, ‘boxwood borer’), or to an infraspecific category (e.g. ‘stag’ and ‘hind’ distinguishing the male and female deer).

The most represented groups of animals were birds (N = 56 items), mammals (N = 48), and arthropods (N = 34), while reptiles (N = 7), amphibians (N = 5) and fishes (N = 3) were the least represented. Smith’s cognitive salience index significantly differed between these groups (ANOVA, p = 0.015). Tukey post-hoc analyses revealed that the difference was significant between mammals (S = 0.056 ± 0.094 SD) and birds (S = 0.023 ± 0.052 SD) (Fig. S3). In addition, item-by-item proximity analyses focused on the 53 most-cited animals revealed that people tended to cite mammals together and arthropods together (Fig. S4). This was not the case for birds and reptiles that were overlapping with other animal groups.

Finally, respondents listed 138 wild animals and 16 domestic animals. We found no significant effect of animal status on Smith’s cognitive salience index (ANOVA, p = 0.54, Fig. S3).

4.2. Respondents’ emotions and attitudes toward CNP animals

We collected data on people’s attitudes and emotions for 153 animals (out of the 154 cited by respondents). Overall, respondents mainly expressed neutral and positive emotions towards cited animals (Fig. 4). An equal number of 114 animals was associated with positive and neutral emotions by at least one respondent, whereas 31 animals were associated with negative emotions (Fig. 4a). For example, wild boars were associated with positive, neutral, and negative emotions by 22, 25 and 32 respondents, respectively. Foxes were associated with positive, neutral, and negative by 26, 26 and 4 respondents, respectively. On the contrary, animals such as raptors, fishes, and rabbits were never associated with negative emotions; and animals such as rodents, scorpions, and earthworms were never associated with positive emotions.

Animal classification according to the most frequent emotion (Fig. 4b) revealed 94 animals that dominantly generated positive emotions, 52 that dominantly generated neutral emotions and seven that dominantly generated negative emotions. The latter included maggot, mole cricket, mosquito, rodent, scorpion, tick, and wild boar.

For all domains, respondents mainly had no opinion on the role of cited animals, except for ecosystem functioning that was dominated by positive attitudes (N = 122 animals, Fig. 4a). In addition, neutral attitudes were well represented, at a similar level as positive attitudes except for livestock farming and ecosystem functioning. Negative attitudes were largely minoritarian, and no respondent considered any animal negative for CNP heritage value (Fig. 4a). Finally, animal classification based on the most frequent attitudes revealed a similar pattern, with a dominance of positive animals for all domains and a limited number of negatively-judged animals (Fig. 4b).

Crop cultivation was the domain for which the largest share of negative animals was identified (N = 18 negative animals based on dominant respondents’ attitude, or 11.8% of all cited animals, Fig. 4b). Among the most cognitively salient animals, wild boar, roe deer, stag, badger, and mosquito were dominantly reported as negative for crop cultivation. On the contrary, goat, fish, sheep, insect and eagle were the animals with the highest cognitive salience seen as positive for crop cultivation. For livestock farming, a total of 13 animals were considered negative by at least one respondent (Fig. 4a), and eventually, seven of them were dominantly considered as negative (Fig. 4b). The latter included fox, mouflon, vulture, wolf, fly, Asian hornet and pig. On the contrary, among the most salient animals, goat, raptor, dog, rabbit, and sheep were dominantly reported as positive for livestock farming. Interestingly, several animals were considered negative for ecosystem functioning: 14 negative animals were cited by at least one respondent, while negative attitudes dominated for three of these (including cattle, elephant, and gorilla that all come from other regions and/or continents). Finally, no animal was dominantly considered as negative to CNP touristic attractiveness and heritage value (Fig. 4b). Yet, some respondents considered dog (N = 1 respondent), wild boar (N = 3) and wolf (N = 1) negative for tourism.

Considering these different results, wild boar (sus crofa, or sanglier in French) appeared to be a particularly salient and problematic species in CNP, dominantly associated with negative emotions and a negative influence on agriculture. No other cognitively salient animal received so negative viewpoints. For many respondents, the situation with wild boar was effectively alarming: "The wild boar destroys terraces and enters into home gardens” (respondent 80) and is "very harmful for agriculture because it uproots everything and eats earthworms" (respondent 12). Consequently, most respondents owning a home garden reported they already witnessed wild boars turning over the soil, and farmers generally fence their fields to protect crops specifically from this animal. Some respondents yet tempered this alarming diagnosis, highlighting that even if wild boars were "developing at great speed" (respondent 15) because of the absence of " predators apart from man" (ibid.), they are not necessarily problematic as long as local hunting allows to regulate populations properly. Similarly, some respondents pointed out that “it [wild boar] does a bit of damage, but it’s nice to see a wild boar” (respondent 3). Finally, wild boar was mentioned positively in rare cases, particularly for the services it provides as a game reserve – it was seen as “useful as a source of food for people far away” (respondent 2) – or as an emblematic animal of CNP: “Coming back from an evening out, there is much chance of running into a wild boar. The boar is part of us” (respondent 43).

4.3. Relationships between animals’ cognitive salience and people’s emotions and attitudes

MCA and HCPC led to grouping animals into five clusters (Fig. 5). Along its first dimension, the MCA tended to distinguish (i) animals associated with neutral attitudes concerning their contribution to cropping and livestock activities and CNP touristic attractiveness and heritage from (ii) animals seen as positive to crop cultivation and associated with neutral emotions. The second dimension was mainly structured by people’s negative emotions and negative attitudes related to the agricultural domain (Fig. 5). This distinction was well-evidenced by the HCPC that distinguished five clusters of animals (detailed cluster description available in Table S1):

Cluster 1 – exogenous animals (N = 3 animals): this cluster gathered the three animals that were perceived as negative to ecosystems, which were the three exogeneous/exotic animals cited by respondents;
Cluster 2 – neutral animals (N = 14): regrouped the animals that were considered as neutral to ecosystem functioning and maintenance;
Cluster 3 – positive animals to CNP landscapes (N = 86): this cluster regrouped the largest number of animals, and was characterized by respondents’ positive attitudes towards them, including for cropping and livestock activities, and CNP tourism and heritage value. This cluster was neither characterized by positive or negative emotions;
Cluster 4 – negative and disliked animals (N = 18): gathered animals that generated negative emotions to respondents, as well as animals considered as negative to agriculture. This cluster tended to over-represent arthropods;
Cluster 5 – neutral and liked animals (N = 32): this last cluster gathered animals that were considered as neutral to CNP landscapes by respondents, but that still generated positive emotions to respondents.

In sum, the HCPC revealed that animals that received positive attitudes (Cluster 3) differed from those that people would appreciate to cross and/or observe (Cluster 5). On the contrary, animals considered as negative to agriculture were also animals that people disliked (Cluster 4).

Finally, the ANOVA revealed a significant relationship between people’s attitudes and emotions, and animals’ cognitive salience (p < 0.0001, Fig. 6). In particular, animals in Cluster 5 (neutral and liked animals) were significantly more salient (S = 0.052 ± 0.057 SD) than animal in clusters 2 (neutral animals, S = 0.010 ± 0.015 SD) and 3 (positive animals to CNP landscapes, S = 0.025 ± 0.047 SD). Tukey post-hoc analyses revealed no other significant difference between groups (Fig. 6). Yet, we note that animals in Cluster 4 (negative and disliked animals) had the highest salience index on average (S = 0.058 ± 0.134 SD).

The approach and results presented in this study resonate with former research in the domain of ethnobiology while adding a connection with the body of research on cognitive biases that is emerging in conservation biology. As we will further discuss below, our results emphasize the complexity of navigating cognitive biases in non-experimental research designs and stress the need for further research on this critical topic for conservation social sciences.

5.1. Animal cognitive salience and people’s attitudes

By combining methods from ethnobiology and environmental psychology, this work revealed that permanent residents and temporary visitors in Cevennes were prone to cite a consistent number of animals and express a range of attitudes and emotions towards these.

Free-listings, in particular, offered somehow contrasting results according to the level considered. At coarse level, respondents cited more animals belonging to bird, mammal and arthropod groups than to amphibian, fish and reptile groups. This result echoes with the findings of similar studies (Czech and Krausman 2001). It is also consistent with the ecological apparency hypothesis – that assumes that people are more familiar with apparent species (de Lucena et al. 2012; Blanco and Carrière 2016). However, at item level, cited reptiles, amphibians and fishes did not have a higher cognitive salience index than cited birds, mammals and arthropods (Fig. S3). The different taxonomic levels at which people named animals could explain this gap between the two analytical levels. For example, the ‘fish’ and ‘insect’ items were cited by 18 and 16 respondents, respectively, while they designated entire animal groups. This situation leads to a low number of items in these groups (explaining coarse-level results) but a relatively high salience index for the few cited items (explaining item-level results). To our knowledge, such a problem affects all studies based on free-listing tasks and is not properly handled by existing salience indexes. It can be addressed in two ways: (i) processing a two-level analysis as we did here; or (ii) homogenizing cited items based on scientific taxonomy using, for example, the generic level that is the most commonly known by non-specialists (Atran 2003). While this second approach performs a more standardized coding, it does not accurately depict respondents’ answers and representations. Furthermore, homogenization at generic level would not allow distinguishing respondents who cited a genus (e.g., ‘eagle’) from respondents who cited a species (e.g., ‘golden eagle’) or several species within a genus (e.g., ‘golden eagle’ and ‘Bonnalli’s eagle’).

Our results on people’s emotions and attitudes also evidenced an analytical level effect. At coarse level, we effectively found that people were more likely to associate species with positive and neutral emotions and attitudes than with negative ones, resulting in a higher number of positively-perceived animals across all the domains investigated (Fig. 4). This first result suggests the dominance of positive views of people regarding non-humans. However, and as we will discuss in the next subsection, this result is to be nuanced by the results obtained when focusing on individual animals (Fig. 6).

5.2. The complexity of navigating cognitive biases

As mentioned, a coarse analytical level focus revealed that positive emotions and attitudes largely dominate negative ones in people’s representations of CNP animals. However, item-level results showed that (i) animals seen as neutral to CNP landscapes but generating positive emotions (animals in Cluster 5) were significantly more salient than animals that generated no particular emotions to respondents (Clusters 2 and 3); and (ii) animals seen as negative and disliked (Cluster 4) were more salient than other animals even though the difference was not statistically significant.

Therefore, at fine analytical level, a positivity bias seemed to influence animal cognitive salience but this bias was rather linked to positive emotions than positive attitudes. This emphasizes the importance of emotions compared to more analytical thinking in how people relate with non-humans and their conservation (Schaffner et al. 2015). This also supports recent calls in sustainability science to pay more attention to people’s affects, emotions and sensory experience to better understand their behavior (Heinrichs 2019; Ives et al. 2020; Woiwode et al. 2021).

Results also suggested that a negativity bias could also be at play, supported by the higher cultural salience of negative and disliked animals. While seemingly contradictory, both results are in fact consistent with neuroscience and experiment-based findings. In a study where participants were asked to place a (un)pleasant stimulus on a pleasure-displeasure gradient, Norris et al. (2011) found evidence of the coexistence of positivity and negativity biases. Similarly, both very positive and very negative messages about wildlife conservation were found to be effective on people reactions measured through their heart rate (Jain et al. 2019). Thus, it appears that our results are aligned with existing literature.

Two aspects that were specific to our case study might yet explain why we did not obtain a statistical confirmation for the negativity bias. First, the number of negatively-judged animals was too small compared to positively-judged animals. Second, wild boar was close to an outlier as it was the animal with the highest cognitive salience, but also the most controversial. Thus, this single animal had a leverage effect on the mean cultural salience of Cluster 4. As sample imbalance and leverage effect reduce the power of parametric statistical tests (such as ANOVA), detecting any potential negativity bias was challenging with the methods we used. This stresses the need for more research on cognitive bias applied to human/non-human relationships, including methodological refinements. In particular, further studies with more respondents, a focus on specific species, or a cross-site approach, could be insightful to draw more generalizing conclusions.

Wild boar also allows us to point to another area of complexity around cognitive biases linked to the socio-political, historical and cultural context. In other contexts, some research showed that people’s conflicts with wildlife were linked to their attitudes towards the protected area, its staff member and policies (McElwee 2010). This was not necessarily the case for wild boar that is in a more complex position in the CNP, and in France more broadly. On the one hand, protected areas are blamed for offering a refuge to wild boar, indirectly favoring crop damage and road collisions (Mathevet and Bondon 2022). Yet, unlike other controversial species such as wolf, wild boar is of little interest to naturalists: it is a widespread species that sometimes crossbreeds with domestic pigs and that threatens certain species, especially birds such as capercaillie, greylag geese and purple heron (Mounet 2009; Mathevet and Bondon 2022). Furthermore, wild boar expansion in Cevennes is the symbol of forest encroachment on former agricultural lands, of the decline in the number of hunters and, more broadly, of rural exodus and the decline of historical rural activities (Blanco et al. 2020). As a consequence, the crystallization of people’s attention on a species such as wild boar reflects of a complex entanglement between cognitive processes and people’s positioning on what this species symbolizes. Thus, combining cognitive and socio-political analyses could be insightful of these complex and multidimensional human/non-human relationships (see also, Blanco et al. 2022).

5.3. Limits and perspectives

Our method relies on widely used interviewing techniques, demonstrating their value and transferability to diverse contexts. However, some aspects of the method should be carefully considered before its application elsewhere.

First, while free-listing tasks are useful to a broad spectrum of research in social sciences, they have well-known inherent limitations that should be acknowledged (see Brewer 2002; Dany et al. 2015; Quinlan 2017). We note in our case that the high number of respondents to helps reduce these limitations, but that additional techniques exist to attenuate them further (e.g., Brewer 2002). We invite researchers and practitioners to consider integrating these techniques that could contribute to improving the method and its outcomes.

Second, we note that emotions and attitudes are part of a two-way relationship between people and another entity that is not necessarily accessible in absolute terms. This relationship is diverse according to individual education and experience, highly dynamic and changing according to age and emotional state, among other (Ochsner and Gross 2005). The quantitative orientation of our method does not allow entering such complexity as (i) it would have required in-depth discussions with respondents, implying working with fewer informants and therefore reducing statistical robustness, and (ii) it would have led to an extensive qualitative material not suited to a quantitative analysis.

Another methodological challenge is that emotions are not universal across human societies and cultural groups and are felt and expressed differently in different cultures (Wierzbicka 1992). In this sense, emotions could be considered a cultural product and specific approaches have been proposed to take into account this inter-cultural variation (e.g. Oettingen and Seligman 1990). In particular, verbal language is not necessarily the best entry point to emotions for certain categories of persons or cultures (Crapanzano 1994). Art (e.g., drawing, painting, photography, film, theatre) could be an alternative that could help reveal things that are hardly said (Pagezy et al. 2010; Fache et al. 2022).

Finally, while we focused in this study on positivity and negativity biases by considering all the 100 respondents together, further research aiming at unpacking respondents could be insightful. In particular, knowing that cognitive biases are mediated on people’s gender, age and cultural background (Norris et al. 2011; Clark 2016), exploring potential differences between respondents would be of interest. This enterprise would requires a large sample of respondents, in particular to gather a sufficient amount of data on people’s attitudes and emotions towards animals (Norris et al. 2011). In sum, we argue that the approach presented in this study holds promises to further investigate cognitive biases applied to biodiversity conservation sciences through a quantitative lens. Such research efforts might help towards more effective conservation practices and strategies given the key role of cognitive biases on people’s representations, attitudes and behaviors.

Biodiversity conservation, as other sustainability challenges, is increasingly considered a social issue more than a biophysical one (Mascia et al. 2003). Consequently, a vast range of research in social sciences and the humanities would be relevant to be specifically applied to sustainability and biodiversity conservation challenges. In this perspective, this study aimed at applying research on cognitive biases to the domain of human/non-human relationships. In the Cevennes National Park in France, results suggested a simultaneous influence of negativity and positivity bias in how people relate with the fauna. In addition, emotions were found to be a key driver of cognitive bias that seemed to overcome attitudes. Ultimately, this study adds arguments on the need to reinforce research on the complexity of human cognition and behavior, in particular people’s affects and emotions towards non-human entities, which could help foster more effective biodiversity conservation communication and strategy.

Author Contribution

All authors contributed to the study conception and design. Material preparation were performed by T.D. under the supervision of S.C., C.M. and J.B. Data collection and analysis were performed by T.D. and J.B. The first draft of the manuscript was written by T.D. and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Acknowledgements:

This study was supported by the French National Research Institute for Sustainable Development (Institut de Recherche pour le Développement).

Albuquerque U, Andrade L, Caballero J (2005) Structure and floristics of homegardens in Northeastern Brazil. J Arid Environ 62:491–506. https://doi.org/10.1016/j.jaridenv.2005.01.003
Atran S (2003) Théorie cognitive de la culture: Une alternative évolutionniste à la sociobiologie et à la sélection collective. lhomme 107–144. https://doi.org/10.4000/lhomme.220
Bennett NJ, Roth R, Klain SC et al (2017) Mainstreaming the social sciences in conservation. Conserv Biol 31:56–66. https://doi.org/10.1111/cobi.12788
Blanco J, Bellón B, Barthelemy L et al (2022) A novel ecosystem (dis)service cascade model to navigate sustainability problems and its application in a changing agricultural landscape in Brazil. Sustain Sci 17:105–119. https://doi.org/10.1007/s11625-021-01049-z
Blanco J, Carrière SM (2016) Sharing local ecological knowledge as a human adaptation strategy to arid environments: Evidence from an ethnobotany survey in Morocco. J Arid Environ 127:30–43. https://doi.org/10.1016/j.jaridenv.2015.10.021
Blanco J, Dendoncker N, Barnaud C, Sirami C (2019) Ecosystem disservices matter: Towards their systematic integration within ecosystem service research and policy. Ecosyst Serv 36:100913. https://doi.org/10.1016/j.ecoser.2019.100913
Blanco J, Moreau C, Guerbois C et al (2020) La biodiversité, une ressource, mais aussi un fardeau ? Intérêt et limites des notions de services et disservices écosystémiques pour repenser les interactions nature-sociétés dans les territoires ruraux. vertigo. https://doi.org/10.4000/vertigo.29079
Brewer DD (2002) Supplementary Interviewing Techniques to Maximize Output in Free Listing Tasks. Field Methods 14:108–118. https://doi.org/10.1177/1525822X02014001007
Buijs A, Jacobs M (2021) Avoiding negativity bias: Towards a positive psychology of human–wildlife relationships. Ambio 50:281–288. https://doi.org/10.1007/s13280-020-01394-w
Byerly H, Balmford A, Ferraro PJ et al (2018) Nudging pro-environmental behavior: evidence and opportunities. Front Ecol Environ 16:159–168. https://doi.org/10.1002/fee.1777
Cabanel P (2013) De la soie au charbon: une province industrieuse. Presses Universitaires de France, Paris cedex 14, pp 65–83
Caduff D, Timpf S (2008) On the assessment of landmark salience for human navigation. Cogn Process 9:249–267. https://doi.org/10.1007/s10339-007-0199-2
Christie P, Bennett NJ, Gray NJ et al (2017) Why people matter in ocean governance: Incorporating human dimensions into large-scale marine protected areas. Mar Policy 84:273–284. https://doi.org/10.1016/j.marpol.2017.08.002
Cinner J (2018) How behavioral science can help conservation. Science 362:889–890. https://doi.org/10.1126/science.aau6028
Clark A (2016) Surfing uncertainty: prediction, action, and the embodied mind. Oxford University Press, Oxford; New York
Cornu P (2003) Déprise agraire et reboisement: Le cas des Cévennes (1860–1970). Histoire Sociétés Rurales 20:173. https://doi.org/10.3917/hsr.020.0173
Crapanzano V (1994) Réflexions sur une anthropologie des émotions. terrain 109–117. https://doi.org/10.4000/terrain.3089
Cumming GS (2016) The relevance and resilience of protected areas in the Anthropocene. Anthropocene 13:46–56
Czech B, Krausman PR (2001) Differences in important species types as ranked by the public. The endangered species act: history, conservation biology, and public policy. JHU, pp 63–66
Dany L, Urdapilleta I, Lo Monaco G (2015) Free associations and social representations: some reflections on rank-frequency and importance-frequency methods. Qual Quant 49:489–507. https://doi.org/10.1007/s11135-014-0005-z
Das TK, Teng B-S (1999) Cognitive Biases and Strategic Decision Processes: An Integrative Perspective. J Manage Stud 36:757–778. https://doi.org/10.1111/1467-6486.00157
de Lucena RFP, de Medeiros PM, Araújo EDL et al (2012) The ecological apparency hypothesis and the importance of useful plants in rural communities from northeastern Brazil: an assessment based on use value. J Environ Manage 96:106–115. https://doi.org/10.1016/j.jenvman.2011.09.001
Dietsch AM, Wallen KE, Clayton S et al (2020) Introduction: New directions in conservation psychology at a critical time. Conserv Biol 34:1335–1338. https://doi.org/10.1111/cobi.13629
Eidelman S, Crandall CS (2012) Bias in Favor of the Status Quo. Social Personality Psych 6:270–281. https://doi.org/10.1111/j.1751-9004.2012.00427.x
Engler J-O, Abson DJ, von Wehrden H (2018) Navigating cognition biases in the search of sustainability. https://doi.org/10.1007/s13280-018-1100-5. Ambio
Fache E, Sabinot C, Pauwels S et al (2022) Encouraging Drawing in Research with Children on Marine Environments: Methodological and Epistemological Considerations. Hum Ecol 50:739–760. https://doi.org/10.1007/s10745-022-00332-6
Flecken M, Carroll M, Weimar K, Von Stutterheim C (2015) Driving Along the Road or Heading for the Village? Conceptual Differences Underlying Motion Event Encoding in French, German, and French–German L2 Users. Mod Lang J 99:100–122. https://doi.org/10.1111/j.1540-4781.2015.12181.x
Heinrichs H (2019) Strengthening Sensory Sustainability Science—Theoretical and Methodological Considerations. Sustainability 11:769. https://doi.org/10.3390/su11030769
Henley NM (1969) A psychological study of the semantics of animal terms. J Verbal Learn Verbal Behav 8:176–184. https://doi.org/10.1016/S0022-5371(69)80058-7
Ives CD, Freeth R, Fischer J (2020) Inside-out sustainability: The neglect of inner worlds. Ambio 49:208–217. https://doi.org/10.1007/s13280-019-01187-w
Jain E, Jacobson SK, Raiturkar P et al (2019) Using Audience Physiology to Assess Engaging Conservation Messages and Animal Taxa. Soc Nat Resour 32:1092–1098. https://doi.org/10.1080/08941920.2018.1556760
Kahneman D (2003) Maps of Bounded Rationality: Psychology for Behavioral Economics. Am Econ Rev 93:1449–1475
Kansky R, Knight AT (2014) Key factors driving attitudes towards large mammals in conflict with humans. Biol Conserv 179:93–105. https://doi.org/10.1016/j.biocon.2014.09.008
Kassambara A (2023) rstatix: Pipe-Friendly Framework for Basic Statistical Tests
Kusmanoff AM, Fidler F, Gordon A et al (2020) Five lessons to guide more effective biodiversity conservation message framing. Conserv Biol 34:1131–1141. https://doi.org/10.1111/cobi.13482
Larrère R (2009) Histoire(s) et mémoires des parcs nationaux. In: Histoire des parcs nationaux. Editions Quæ, p 21
Lê S, Josse J, Husson F (2008) FactoMineR : An R Package for Multivariate Analysis. J Stat Softw. https://doi.org/10.18637/jss.v025.i01. 25:
Mascia MB, Brosius JP, Dobson Ta et al (2003) Conservation and the social sciences. Conserv Biol 17:649–650. https://doi.org/10.1046/j.1523-1739.2003.01738.x
Mathevet R, Bondon R (2022) Sangliers, géographies d’un animal politique, Éditions Actes Sud
McElwee PD (2010) Resource Use Among Rural Agricultural Households Near Protected Areas in Vietnam: The Social Costs of Conservation and Implications for Enforcement. Environ Manage 45:113–131. https://doi.org/10.1007/s00267-009-9394-5
Mounet C (2009) Quel «vivre ensemble» possible avec le loup et le sanglier ? In: L’animal sauvage entre nuisance et patrimoine: France, XVIe-XXIe siècle. ENS Éditions, pp 71–82
Napoléone M, Chazoule C, Fleury P (2015) La ressource fourragère « qui convient »: Conceptions et points de vue d’éleveurs des Cévennes et du Vercors. tc 110–129. https://doi.org/10.4000/tc.7427
Norris CJ (2021) The negativity bias, revisited: Evidence from neuroscience measures and an individual differences approach. Soc Neurosci 16:68–82. https://doi.org/10.1080/17470919.2019.1696225
Norris CJ, Larsen JT, Crawford LE, Cacioppo JT (2011) Better (or worse) for some than others: Individual differences in the positivity offset and negativity bias. J Res Pers 45:100–111. https://doi.org/10.1016/j.jrp.2010.12.001
Ochsner K, Gross J (2005) The cognitive control of emotion. Trends Cogn Sci 9:242–249. https://doi.org/10.1016/j.tics.2005.03.010
Oettingen G, Seligman MEP (1990) Pessimism and behavioural signs of depression in East versus West Berlin. Eur J Soc Psychol 20:207–220. https://doi.org/10.1002/ejsp.2420200303
Pagezy H, Carrière S, Sabinot C (2010) Nature du monde, dessins d’enfants. CTHS, Paris
Palomo I, Montes C, Martín-López B et al (2014) Incorporating the Social–Ecological Approach in Protected Areas in the Anthropocene. Bioscience 64:181–191. https://doi.org/10.1093/biosci/bit033
PNC (2021) Le Parc National des Cévennes. Guide du naturaliste Causses Cévennes: à la découverte des milieux naturels du Parc national des Cévennes, 3e éd. revue et corrigée. Glénat, Grenoble, pp 8–9
Quinlan M (2005) Considerations for Collecting Freelists in the Field: Examples from Ethobotany. Field Methods 17:219–234. https://doi.org/10.1177/1525822X05277460
Quinlan MB (2017) The Freelisting Method. In: Liamputtong P (ed) Handbook of Research Methods in Health Social Sciences. Springer Singapore, Singapore, pp 1–16
R Core Team (2022) R: A Language and Environment for Statistical Computing
Robbins MC, Nolan JM, Chen D (2017) An Improved Measure of Cognitive Salience in Free Listing Tasks: A Marshallese Example. Field Methods 29:395–403. https://doi.org/10.1177/1525822X17726726
Rozin P, Royzman EB (2001) Negativity bias, negativity dominance, and contagion. Personality Social Psychol Rev 5:296–320. https://doi.org/10.1207/S15327957PSPR0504_2
Scales IR (2012) Lost in translation: conflicting views of deforestation, land use and identity in western Madagascar. Geographical J 178:67–79. https://doi.org/10.1111/j.1475-4959.2011.00432.x
Schaffner D, Demarmels S, Juettner U (2015) Promoting biodiversity: do consumers prefer feelings, facts, advice or appeals? J Consumer Mark 32:266–277. https://doi.org/10.1108/JCM-11-2014-1220
Schultz PW (2011) Conservation Means Behavior: Conservation Means Behavior. Conserv Biol 25:1080–1083. https://doi.org/10.1111/j.1523-1739.2011.01766.x
Smith JJ, Borgatti SP (1997) Salience counts-and so does accuracy: Correcting and updating a measure for free-list-item salience. J Linguistic Anthropol 7:208–209. https://doi.org/10.1525/jlin.1997.7.2.208
Sutrop U (2001) List Task and a Cognitive Salience Index. Field Methods 13:263–276. https://doi.org/10.1177/1525822X0101300303
Thompson EC, Juan Z (2006) Comparative Cultural Salience: Measures Using Free-List Data. Field Methods 18:398–412. https://doi.org/10.1177/1525822X06293128
Tversky A, Kahneman D (1974) Judgment under Uncertainty: Heuristics and Biases. Science 185:1124–1131. https://doi.org/10.1126/science.185.4157.1124
Vaish A, Grossmann T, Woodward A (2008) Not all emotions are created equal: The negativity bias in social-emotional development. Psychol Bull 134:383–403. https://doi.org/10.1037/0033-2909.134.3.383
Wallen K, Landon A (2020) Systematic map of conservation social psychology research. 34:1339–1352. https://doi.org/10.31235/osf.io/bjtfn
Wencelius J, Garine E, Raimond C (2017) FLARES
Wierzbicka A (1992) Introduction. Semantics, culture, and cognition: Universal human concepts in culture-specific configurations. Oxford University Press on Demand, pp 3–31
Woiwode C, Schäpke N, Bina O et al (2021) Inner transformation to sustainability as a deep leverage point: fostering new avenues for change through dialogue and reflection. Sustain Sci 16:841–858. https://doi.org/10.1007/s11625-020-00882-y

No competing interests reported.

SICogBiasBiodivCons.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Assessing positivity and negativity biases: Emotions and attitudes influence animal cognitive salience in a French National Park

Status:

Version 1

Abstract

Figures

1. Introduction

2. Background

3. Methods

3.1. Study area

3.2. Sampling design and respondent selection

3.3. Interview procedure and ethics

3.4. Free-list analysis and cognitive salience

3.5. Analyzing attitudes, emotions, and cognitive bias

4. Results

4.1. Cognitive salience and proximity of CNP animals

4.2. Respondents’ emotions and attitudes toward CNP animals

4.3. Relationships between animals’ cognitive salience and people’s emotions and attitudes

5. Discussion

5.1. Animal cognitive salience and people’s attitudes

5.2. The complexity of navigating cognitive biases

5.3. Limits and perspectives

6. Conclusion

Declarations

Author Contribution

Acknowledgements:

References

Additional Declarations

Supplementary Files

Status:

Version 1