Land use planners and decision makers aiming to implement socially sustainable management of social-ecological systems (SESs) need to understand how the landscape and its diverse uses are perceived, interpreted, and understood (Buijs et al. 2011, Barreteau et al. 2016). Indeed, these perceptions, interpretations, and understanding shape the attitudes of the public and land users toward the landscape and its uses (Raymond et al. 2014, Plieninger et al. 2015).
To study these perceptions, several concepts have been developed, including ideology, attitude, mental models, and social representations (SRs). During the last 50 years, SRs have increasingly been used by scholars around the world, especially in Europe and South America, and have become a recognized field within social, cultural, and political psychology (Sammut and Howarth 2014), as well as sociology, anthropology, history, and geography (Flament and Rouquette 2003, Valence 2010). An SR is a “sociopsychological construct that performs a symbolic role, representing something - an object - to someone - a person or group. While doing so, the representation actually substitutes the object it represents, and therefore becomes the object itself, for the person or group that refers to it” (Wachelke 2011:730). The social representation theory (SRT) is a socio-psychological theory that focuses on the content and production of both common sense and knowledge, i.e., how people understand the world around them and the meanings they attach to that world (Moscovici and Marková 2000). The SRT aims to explain how different social groups develop different understandings of an issue based on their particular values, ideas, knowledge, metaphors, beliefs, and practices (Moscovici 1976, Sammut et al. 2015a).
The SRT has, for instance, proven useful to explain how social-ecological changes are perceived by nonexperts, and to overcome the opposition between an objective and a subjective perception of changes (Krien and Michel-Guillou 2014). Several studies have demonstrated the relevance of SRT in understanding inhabitant’s or users’ perceptions of change in SESs, such as changes in water quality in lagoons (Audouit et al. 2019), flooding risks in coastal areas (Lemée et al. 2019), ecosystem services in periurban areas (De Vreese et al. 2019), land use in rural landscapes (Anderson et al. 2017), conflicts in national parks (Buijs et al. 2011), rural development in alpine grassland landscapes (Quétier et al. 2010), and adaptive comanagement in recreational mountain landscapes (Lai et al. 2016). These studies show how existing SRs of the landscape and/or the SES dynamics shape beliefs about land use (Anderson et al. 2017).
SRs are organized and socially developed opinions that reflect common knowledge about a given object; they are based on experience and social interactions with others (Moscovici 1976). The assumption that all SRs are developed through communication is central to SRT (Buij et al. 2011). And, in return, SRs shape communication. For instance, SRs of SES shape communication about valuations of SES dynamics, strategic decisions, and actions (Quétier et al. 2010, Buijs et al. 2011, Blicharska and Van Herzele 2015). SR functions range from taking ownership of the environment through symbolic elements specific to the groups to which people belong (Roussiau and Bonardi 2001) to justifying positions and behaviors (Abric 2001a, Michel-Guillou and Moser 2006). When an object is new in the social arena and/or represents a new issue for a social group, the group develops SRs of this object (Moliner and Abric 2015). SRs mediate social groups’ place-dependent relationships to their social and natural environment (Luginbühl et al. 2015). They contribute to building a consensual representation of the environment, and drive changes in worldviews and behaviors (Moscovici and Marková 1998). Consequently, SRs are key to understanding different stakeholders’ viewpoints in order to achieve efficient land use policies (Gilg 2009).
Although SRs of SESs are increasingly studied in sustainability sciences and land use sciences (e.g., Conrad et al. 2011, Anderson et al. 2013, 2017, Moloney et al. 2014), our understanding of SRs dynamics is still scarce, either in time, e.g., in a dynamic SES, or in space, e.g., between places with contrasting land management policies. Yet, such understanding is key for scientists focusing on natural resource management, biodiversity conservation, or sustainability sciences. Indeed, they need to understand not only the social-ecological dynamics and the construction of knowledge, but also the dynamics of SRs to understand the social-political constructions, how they (re)construct social relations, and offer possibilities for different social-ecological trajectory. In order to gain this understanding, comparative approaches between localities or over time are necessary. However, such comparative approaches require being able to elicit SRs and compare them in a standardized and efficient way. In a previous article, we elicited and made a comparative analysis of individual interviews to explore relationships between farmers’ individual mental models of the agricultural landscape and their land management practices in a case study located in Southern France (Vuillot et al. 2016). The present article builds on the previous one and proposes a methodological approach to elicit and compare SRs between places. First, we provide a critical introduction to the SRT, which briefly explores its utilities and weaknesses and then focuses on the structural approach of SRs. Second, we propose an innovative standardized and modular approach to studying SRs from words collected from free listing tasks. This approach consists in (i) identifying the content of SRs through a standardized threshold definition protocol and (ii) analyzing the structure of SRs through a rank-frequency analysis. It allows comparing SRs content, structuration degree, i.e., consensus level, and structure, i.e., identifying core and periphery. Third, we illustrate the value of such a methodological approach to compare farmers’ SRs of the landscape among four contrasted rural SESs in France. We show that it helps understanding the spatial dynamics of SRs and identifying social-ecological factors that influence the structuration degree and the content of SRs. Finally, we discuss our methodological progress and the implications of our results for public policies aimed at managing SESs, and in particular agricultural landscapes, for conservation.
The concepts of ideology, attitude, mental models, and SRs are widely used in social psychology, sociology, human geography, or environmental sciences to study social phenomena, behaviors, and social practices of individuals or groups of actors. However, these four concepts have been developed separately, and only poorly integrated to each other. We believe that the lack of integration between these concepts may have undermined the development of SRs.
According to Rosanvallon and Viveret (1977, as cited in Jodelet 1991), it is often not the world that changes but rather its representation that constitutes a potential obstacle to the implementation of a new public policy. The concept of ideology (beliefs, values, norms) would then be insufficient to account for the relationship that is established between practice and SRs. Ideology is more and more often replaced by the concept of political culture or social values, i.e., “standards that people use to evaluate goals, entities, behaviors, and state of affairs as desirable, non-desirable, acceptable etc.” (Tsirogianni et al. 2014:1809). It refers to a mixture of theoretical, practical, event-driven, and mythical elements organized around a dominant representation of social change. Aiming to describe the production of meaning and the relationships of meanings existing between different dimensions, i.e., economic, political, family, religious, etc., researchers define these elements as representations and seek to link them to effective practices.
The concept of attitude, i.e. the positive or negative orientation of the individual regarding the object, is used by action-oriented social sciences because it allows for the evaluation of an object in its affective, cognitive, and conative dimensions (Bergamaschi 2011). An attitude is the psychological evaluation of an object according to a degree of favor or disadvantage that integrates social values, i.e., which describe our ideas about a good life. It shapes behaviors and possible changes in it. However, many authors consider that the origin of attitudes is external to the individual (Deconchy 2003). Thus, attitudes and SRs are closely correlated variables, so it is within SRs that the individual’s evaluations of social phenomena—his or her attitudes—are shaped. The joint study of SRs and attitudes is likely to anchor the latter in the web of social relations that characterizes a given context (Bergamaschi 2011).
Exploring human cognition dimensions that could explain preferences, actions, and behaviors is also a growing field in natural resource and SES management (Lynam and Brown 2011). The concept of mental model, as internal representation of external reality, is extensively used with a high diversity of meanings and methodologies as well as an important concept across a variety of academic disciplines (Jones et al. 2011). In this paper we seek to add to the synthesis of Jones et al. (2011), what we started to explore a few years ago (see Mathevet et al. 2011, Lynam et al. 2012, Vuillot et al. 2016): considering the mental model concept as an SR because the latter is developed by an historical social process, communication, and dissemination of knowledge systems (Moscovici 1988, Wagner and Hayes 2005). Socio-cultural knowledge, shared by a community, forms the common basis of all SRs of all groups in that community. Each group can develop specific knowledge based on the ideology, the social practices, and discourses of its group, which are contextualized and specified in mental models. These mental models are the mental constructs that control discourse, interaction, and other social practices. Reciprocally, it is through mental models that discourses can influence and reproduce SRs (Van Dijk 2006). When using mental models, different dimensions or sets of statements are stressed according the different disciplines (Barsalou 2009, Jones et al. 2011), and as we suggested earlier (Lynam et al. 2012) cognitive dimensions of social-ecological systems are key and little-discussed components of these complex systems, and the structuralist theory of SRs (Abric 2001b, see above) is beneficial in making sense of the multidimensional subject area associated with the concept of mental model. Therefore gaining a better understanding of how SRs represent complex and dynamic SES and how they change according place and time could allow researchers and practitioners to develop more efficient management and use of natural resources and also more effective valuations of our social-ecological interdependencies (Mathevet et al. 2016).
Overall, the work of social psychology showed that (i) a cognitive perspective, i.e., based on individual mental models, that is too “intra-individual” does not allow the articulation to social and historical contexts; (ii) the concepts of ideology or attitude under which some may classify SRs fails to provide the framework for a detailed reflection on social cognition (Jodelet 1991). It is this deficiency that Moscovici (1976) wanted to overcome by articulating these concepts and promoting the concept of SR as we will see in more details thereafter.
SRs are about “processes of collective meaning-making resulting in common cognitions which produce social bonds uniting societies, organisations and groups” Höijer (2011:3). As a social psychology theory, SRT links individual and society, managers and users, society and environment, and helps to understand what is perceived as common sense. Developed on the basis of Durkheim’s seminal work at the end of the 19th century, i.e., Durkheim calls “collective” rather than “social” representations “the way in which that special being, which is society, thinks about its own experience” (Durkheim 1968:621), the theoretical framework of SRs was mainly nurtured by sociology, social psychology, and anthropology. The very concept of SR has been the object of several definitions since the 1960s (Moscovici 1972). SRs are the “outcomes of processes of communication that represent reality for a given people, and once in existence they constitute social reality sui generis” (Sammut 2015:104). We will keep in mind that an SR is “... an organized set of opinions, attitudes, beliefs and information referring to an object or a situation. It is determined both by the subject himself (his history, his experience), by the social and ideological system into which he is inserted, and by the nature of the links that the subject maintains with the social system” (Abric 1989:188).
An SR is then the result of an interaction between the individual, the social, and ecological world that surrounds him and the position he occupies in this world (Abric 2011). According to Flament and Rouquette (2003), social thinking can be described by four more or less integrated dimensions with a high degree of intra- and interindividual variability. These dimensions are the following: ideologies, social representations, attitudes, and opinions, i.e., view or judgement formed about something, not necessarily based on any fact or knowledge. But because the limits between what we think, what we know, and what we believe is often blurred, an SR is an organized and structured set of cognitive elements (opinions, information, beliefs) relating to a social object (Moliner et al. 2002). The conditions for the existence of an SR of a given object are not always met. The object must be important to individuals, it can be a potential threat or a particular interest, and it must be abstract or generic (Moliner 1993). Furthermore, the object must be present in the relations and communications between members of the social group, in a way that it carries a stake and participates in the social dynamic (Flamant and Rouquette 2003).
For Jodelet (2003) an SR entails three dimensions: information, field, and attitude. Information consists of an individual’s knowledge about the object, this knowledge influences both individuals’ image of a particular situation and individuals’ behavior in reaction to this situation. Field corresponds to the way individuals organize their knowledge about the object. Attitude determines both the selection of information and its hierarchy in the field, shaping action and behavior (Sammut et al. 2015b).
Two mechanisms contribute to SR development, i.e., allow the social to transform an object and then the representations to transform the social. The “objectivizing process” selects the information that will be retained after individuals’ beliefs and values filtration. The selected elements form what is called a figurative core. At the end of the transformation, the core acquires the status of evidence and becomes reality, a “common thought.” The “anchoring process” makes it possible to integrate the object that is represented into a pre-existing system of thought. The social anchoring then confers a specific meaning, on what is familiar, on the elements (Herzlich 1972, Jodelet 1991, Wagner et al. 1999).
SRs occupy a place both upstream and downstream of action (Abric 2011), they achieve four functions in social dynamics and practices: (i) a knowledge function, to understand and explain reality by taking into account its values and framings ; (ii) an identity function, to characterize and preserve the identity of the group, by situating the group’s membership and its members in the social field; (iii) a guidance function, to guide behavior, practices, and action; (iv) a justification function, to justify positions and behavior a posteriori.
Different methods of data collection do exist for the elicitation and study of SRs (Table 1). Discourse-based methods, e.g., semidirective interviews, questionnaires, scenarios, images, and drawings, are usually used, but methods specifically developed to study SRs have become widespread in the last two decades, such as free listing methods (Abric 2003, Flamant and Rouquette 2003, Moliner and Abric 2015).
Overall, the concept of SR is increasingly used in many disciplines and is becoming for some authors both a “buzzword” or a boundary-concept that allows different disciplines to work on a common topic. Despite its constant attraction for social scientists, the concept is controversial, like any developing theory. Criticisms often put forward the lack of an operational definition of SR, as well as the lack of an adapted methodology to study it (Farr 1991). However, over the last two decades, several studies have clarified the definition of SR (Abric 2011), and developed specific methods to study it (Guimelli and Rouquette 1992, Doise et al. 1993, Moliner and Abric 2015). Readers are invited to consult Voelklein and Howarth (2005) and Sammut et al. (2015b) to explore in detail how researchers have engaged a critical dialogue to improve the theoretical framework. For the sake of brevity, only two important points from these debates are presented below.
First, most criticisms are often based on misunderstandings about concepts because they have different meanings in different disciplines or different languages and cultures. Several authors have suggested that there is a misunderstanding about the concept of SR because “social” can be understood as a form of consensus shared by members of a group about a particular object (Voelklein and Howarth 2005). However, seminal works have shown that representations are “social” because they are created and validated collectively, through communication and social interactions. Their content and specific form are influenced by historical and economic contexts as well as by social practices and cultural context (Moscovici and Marková 2000).
Second, critics tend to reduce the SRT to the influence of society on the individual, i.e. the effects of culture on cognition, therefore overlooking the fact that the SRT emphasizes the symbiotic relationship between culture and cognition. Many criticisms focus on the difficulty in understanding and integrating the complex, dynamic, and dialectical relationship between individual agency and social structure that constitutes the core of the SRT. According to Moscovici (1976), ideology, science, and worldview cannot capture the psychological organization of knowledge produced by society. Therefore, in an attempt to integrate sociological and psychological aspects, he developed the concept of SR, which is distinct from notions such as opinion, attitude, and stereotypes (Jodelet 2003, Voelklein and Howath 2005). Other criticisms focus on the complex and dynamic relationship between social structure and individual agency that he has put forward in theory. It is this dialectical concept of social life and social cognition that contrasts sharply with the Cartesian dualism that structures social psychology (Moscovici and Marková 2000, Farr 2003).
According to Voelklein and Howarth (2005), these criticisms make the SRT difficult to integrate into American and British social psychology, in which the relationship between the psychological and the social is analyzed through a separation of individual perception and cognition on the one hand, and culture and social context on the other. The unusual position of SRs between individuals and societies in which they live has led to critiques of social determinism and cognitive reductionism (Howarth 2006). This is why, for the majority of authors using the SRT, the latter constitutes the opposite reflexive pole to the dominant social psychology, which is said to be individualistic, behaviorist, and experimental (Potter and Wetherell 1998, as cited in Voelklein and Howarth 2005).
Among the various approaches available to study SRs, the structural approach represents one of the main developments of the theory (Wachelke 2011, Lo Monaco et al. 2017). In contrast to more anthropological socio-psychology approaches (e.g., Jodelet 1989), the structural approach has led to the development of semiquantitative methodologies. Although most studies and books have been published in French, this approach to SRs is increasingly available to an international English-reading audience (Wachelke 2011). In this approach, SRs lie in the symbolic and semantic universe associated with a subject that is shared by people within a social group. Studying SRs requires defining what are the symbols and meanings associated with a subject consensual enough to be part of the SR.
One of the most frequently used methods to elicit the content of SRs relies on free association tasks (Dany et al. 2015) as they allow for accessing the latent semantic universe associated with the studied subject (Abric 2003, 2011, Dany et al. 2015). Respondents are asked to give the first words that come spontaneously to their minds when hearing an inductor item, in the order that these words come to their mind until it becomes necessary to search for words. Some studies set a priori a total or a maximum number of items to be uttered (e.g., Dany et al. 2015) and may ask the respondent to rank the items after the spoken phase. Uttered items can be words or short groups of words but must not be sentences.
To identify the content of SRs, we must be able to distinguish uttered words that belong to the shared semantic universe among people in the studied social group, hereinafter called “consensual words,” from words that reflect individual or smaller subgroup viewpoints. Measuring the proportion of consensual words among all the words uttered allows (i) assessing the degree of structuration (Jeziorski 2014) of SRs and (ii) evaluating whether the SR is emerging, stable, or challenged by new socio-environmental conditions (Moliner et al. 2002). To our knowledge, there is no agreement in the literature on how to define the appropriate frequency threshold that allows for differentiating consensual words from other words (Alessio et al. 2011). This was identified as a major methodological challenge by Lo Monaco et al. (2017) and remains an obstacle to comparative analyses of studies based on free-listing tasks and rank-frequency analysis.
After selecting consensual words that represent the content of the SR, the second step consists in exploring its structure. The structure of SRs can be broken down into two components: a “central core” and a “peripheral system” (Abric 1994). The central core includes a limited number of words that correspond to the most stable elements of an SR, giving it its long-term continuity (Elands and Wiersum 2001) and its stable and most widely shared meaning (Dany et al. 2015). In contrast, the peripheral system includes words that are more sensitive to contextual variations; it allows an SR to be dynamic and to adapt through space and time (Moliner and Abric 2015). Focusing on the structural approach of SRs we present below a methodology to allow a robust analysis and comparative approach of SRs according time and SESs.
To distinguish consensual words from other words, researchers most of the time use arbitrary relative thresholds, e.g., words uttered by 10% or more of the interviewed people (Alessio et al. 2011, Wachelke and Wolter 2011). However, this approach does not allow for robust comparisons of the content and structuration degree of SRs between studies.
In order to determine the frequency threshold above which words are considered consensual in a quantitative and absolute way, and to make studies rigorously comparable, we propose to use an adapted version of the binomial test developed by Salès-Wuillemin et al. (2011). The binomial test determines for each word whether it is more frequently cited than it would have been if respondents had randomly picked up words from within the intra-site pool of words associated to the inductor word (with a type I error α = 5%). We can simulate the random distribution of uttered words by adapting Salès-Wuillemin et al. (2011)’s model to interview designs where there is no limitation to the number of words a respondent can utter (details on the model are provided in Appendix 1). We then test each word to determine whether its citation frequency is greater than the (1-α) quantile of the model (consensual word) or lower (nonconsensual word). In contrast to previous studies, we strongly advise checking for false positive detection rates, i.e., the risk of identifying a word as consensual when it is not, using Holm-Bonferroni p-value adjustment technique for multiple test procedure (Holm 1979).
This method is based on a very conservative definition of consensus and, consequently, when sample sizes are low and/or when SRs are not well structured (for example when they are emerging), the number of consensual words can be very low, even nil. Considering this, we propose a modular approach that distinguishes two options that remain comparable thanks to a few common indicators (Fig. 1).
The first option is well adapted to cases where the sample studied is large and/or the data exploration shows that the social representation might be highly structured. Researchers can then focus only on consensual words to study the structure of the representation (e.g., Salès-Wuillemin et al. 2011). In that case, we suggest displaying and discussing other words (except hapax, i.e., words uttered only once) in an appendix to allow for qualitative comparisons with other studies with lower sample sizes.
The second option might be necessary in a case where the sample studied is small and/or the data exploration shows that the social representation might not be very structured, e.g., an emerging social representation. Then, the SR contains very few consensual words and researchers need to study the less frequently uttered words that may reveal trends in the development of the representation. In this case, following several authors (e.g., Vergès 1992, Morlot and Salès-Wuillemin 2009, Alessio et al. 2011, Jeziorski and Ludwig-Legardez 2013), we propose not only to consider consensual words defined through the binomial test but also to include all items uttered by at least 10% of respondents within a social group and on a site in order to study the structure of the representation. However, among these items, consensual words defined through the binomial test have to be clearly identified in order to allow for comparison between studies.
In both cases, two indicators can be used to measure the structuration degree of the social representation (Salès-Wuillemin et al. 2011): the proportion of consensual words and the proportion of hapax, i.e., words uttered by one single respondent (Lebart and Salem 1988). The definition of a hapax is straightforward and the proportion of hapax can therefore be unambiguously compared between studies. This proportion is however seldom mentioned (but see Salès-Wuillemin et al. 2011). We strongly recommend calculating and displaying both indicators. These indicators can be statistically compared between studies using Chi square tests or Fisher’s exact tests for count data.
After consensus and variability levels have been assessed, the structure and content of social representations are most commonly studied though prototypical analysis followed by a categorical analysis of the words (Vergès 1992, Vergès et al. 2002). Prototypical analysis aims to distinguish within consensual words those that belong to the central core from those belonging to the periphery of the social representation according to their evocation frequency (the number of times a word is cited in a given study site) and utterance ranking. Higher frequency indicates a highly shared word, whereas first rankings indicate a word with a high cognitive availability (Dany et al. 2015). Crossing these two indicators allows for identifying salient words more frequently cited than the median evocation frequency and cited by respondents with a smaller rank, on average. These words are likely to be part of the central core of the social representation. In contrast, other combinations of levels of frequency and rank define words belonging to the periphery. Within these, frequent and later ranked words are considered unstable and prone to change part of the representation, whereas less frequent and earlier ranked items are likely to be redundant or to specify the central core. Finally, less frequent and later ranked items are not very important in the current representation but likely to become more so in case of adaptation to a change (Moliner et al. 2002, Michel-Guillou and Moser 2006).
Categorical analysis consists in grouping together semantically close words in broader themes. This step is defined by Vergès as a “merger between the researcher’s own categorization system and what seems to emerge from the data” (Vergès 1992:204 [author’s translation]). This step involves the researcher’s subjectivity in defining the themes and can be quite difficult because some words can intersect several themes (Vergès et al. 2002, Conrad et al. 2011). This stage must be seen as an interpretative step that is essentially part of the analysis and allows an easier comparison between sites. The categorical evaluation grid has to be the same across study sites to allow for comparing representations across sites. It should be established for all items by the same researcher and crosschecked by at least one other person for consistency. To make meta-analyses feasible, a table showing the categories and the words they regroup must be displayed in all articles using categorical analysis (see Appendix 1, Table A1.1).
As highlighted in the European Landscape Convention, our society increasingly considers landscapes as SESs. Agricultural landscapes are the visible outcomes of long-term interactions between human activities, mainly agriculture, and the environmental action (Gibon 2005). Today, the search for sustainable agriculture and wise use of natural resources raises questions on the best ways to conciliate food production and biodiversity conservation in farmed landscapes (Tscharntke et al. 2005, Mathevet et al. 2014, Pe'er et al. 2014, Plieninger et al. 2015, Sirami et al. 2019). The European Landscape Convention (European Council 2000) advocates public participation in defining public policies for ecologically sound agricultural landscapes. In this context, research on farmers’ social representations of the agricultural landscape promises to improve our way of thinking about and designing conservation policies in farmed landscape.
According to Moscovici (1976), social groups build SR of an object when three processes occur: first, when information on the object is dispersed through various channels and there is no orthodoxy on the object’s meanings; second, when the social group needs to focus on part of the semantic universe associated with the object in order to master the object in a way that maintains the identity and cohesion of the social group; third, when there is external pressure on the social group to develop a discourse on the object, leading the group to infer the lesser known aspects of the objects. Therefore, considering (i) the polysemy of the term “landscape,” (ii) the agricultural landscape being increasingly an issue in the public sphere, challenging farmers’ identity and role, and (iii) the evolution over time of its acceptance in public policies (Swaffield 1991, Buijs et al. 2006), we assume that the agricultural landscape is an object of social representations among farmers. Moreover, we consider that regional contexts, especially differences in the local public policies, are likely to influence farmers’ representation of the agricultural landscape.
We sought to determine to what extent the representation of the agricultural landscape was derivative from what is commonly understood about landscape and whether a distinctive representation of agricultural landscape had emerged amongst farmers. It must be noted that, in French, the word paysage means both landscape and scenery, i.e., an aesthetic dimension is embedded in the term whilst not defining it entirely. It is thus important to see what dimension will be emphasized in farmers’ SRs of the agricultural landscape across different contexts.
The differences between microregions in the local importance of issues associated with the agricultural landscape and the local variety of stakeholders involved in such issues may especially have an effect on the content of the representation and the level of consensus. We consider that local landscape management policies are likely to bring farmers closer to the social subject agricultural landscape. Previous research has shown that the distance to the object of representation, i.e., the level of practices, knowledge, and engagement of the individual toward the object of representation (Dany et al. 2015), influences the consensus level, structure, and content of the representation (Galand and Salès-Wuillemin 2009, Morlot and Salès-Wuillemin 2009, Salès-Wuillemin et al. 2011).
Based on these studies, we can hypothesize (H1) that the level of local public policies dealing with agricultural landscape related issues has an impact on the proportion of consensual words as well as on the content and structure of the social representation. In this context, we made a first prediction: (P1) we will find a greater proportion of consensual words and smaller proportion of hapax in study sites where landscape issues are advertised and discussed through local management agencies and local public policies. Considering the inter-regional heterogeneity of agricultural landscapes and biodiversity issues, the intra-regional heterogeneity of farmers (Celio et al. 2014), and the social and contextual nature of SRs (Halfacree 1993), we hypothesize (H2) that the local environment is likely to influence SRs of the landscape. Depending on the area, issues related to the agricultural landscapes can certainly be more or less important in the local social arena and thus trigger differently the different processes through which social representations of an object appear. Consequently, the SR of the agricultural landscape is more or less likely to show a high structuration degree depending on the local context. Thus, we made a second prediction: (P2) some words belonging to the core of the SR will be cited in common by farmers regardless of the site they farm on while the peripheries of the SRs will differ among sites. We expected a decrease in saliency of functional words related to specific practices or precise descriptions when territorial coordination decreases.
Our study took place in four French SESs studied in the FarmLand BiodiVERsA project (https://www.farmland-biodiversity.org/index.php?sujet=1&lang=en): Gascony Valleys and Hills, Armorique, Plaine et Val de Sèvre, and the Camargue (Fig. 2 and Appendix 2). The project aimed at assessing the relationships among landscape heterogeneity, biodiversity, and ecosystem services to provide orientations for more efficient agricultural policies on the landscape scale (Sirami et al. 2019).
In the past 50 years, the four SESs have known agricultural intensification and specialization that has led to a decrease in the number of cultivated crops and of holdings, and to an increase in sizes of fields and farms and consequent decrease in hedges, though these processes varied qualitatively and quantitatively between SESs. However, the degree to which issues related to the local landscape, e.g., restoring the hedgerows network for conserving farmland biodiversity, are the subject of local public policies or locally advocated by various stakeholders varied among SESs (Fig. 2 and Appendix 2). These differences induce differences among SESs in the local dispersal of information or pressure to infer on the social group of farmers. Table 2 provides the main features of each SES and a more detailed description of each context is provided in Appendix 2.
Our sampling design aimed to encompass the local diversity of landscape conditions and farm types in each SES rather than selecting a statistically representative sample of farmers. Within each studied SES, sampling areas were selected along a gradient of landscape heterogeneity (Calatayud et al. 2012, Pasher et al. 2013). We contacted farmers to obtain a sample of 30 farmers per SES located along a large gradient of landscape heterogeneity, i.e., considering the land cover and the biodiversity conservation perimeters of each site, and covering the local range of farming systems. Sample characteristics for each SES are provided in Appendix 2, Table A2.1. The size of the sample (119 farmers distributed across four SESs) was constrained by our interview design and resulted from a trade-off between total time allowed to the study and number of selected SESs on which to conduct our comparative analysis. As highlighted in the introduction, this specific work took place in a Biodiversa project where we elicited and analyzed individual mental models (IMM) of farmers and their relationships to land management practices (Vuillot 2015, Vuillot et al. 2016). The free listing was done during the first 15 min of an in-depth 2-hour, face-to-face interview aiming to elicit the IMM.
All interviews were conducted on the farms or in the homes of the farmers. In contrast to many studies on landscape perceptions and preferences, we chose not to conduct the interviews outside in order to access the farmers’ abstract cognitive representation of the landscape, i.e., what they spontaneously associated with the concepts, rather than their perception of a given scenery. It was explained to respondents that the aim of the study was to learn about their own vision of the local landscape as farmers working there. No photographs were shown; the European Landscape Convention definition of the landscape was given to each respondent and the SES boundaries were indicated beforehand on a map showing only waterways and main cities.
Farmers were asked to complete two free association tasks in a row, referring to the term “landscape” (paysage) and then “agricultural landscape” (paysage agricole) as inductor items. The inductor item “landscape” was always given before “agricultural landscape.” This design aimed at disentangling what was specific to the representation of the agricultural landscape, assuming that the representation of the “agricultural landscape” was likely to be partly anchored in the representation of the “landscape” (Doise 1985, 1992, Jeziorski and Ludwig-Legardez 2013). We chose a multiple response free association technique: respondents were asked to utter spontaneously the first words that came to mind when hearing the inductor items with no restriction on the number of words uttered (Flament and Rouquette 2003). The same person (CV) conducted all the interviews. In compliance with French law, we did not collect any personal data.
We analyzed words uttered in each site using the second option of the methodological approach previously described. We first performed a prototypical analysis followed by a categorical analysis. The categorization process was performed by CV and crosschecked by RM on a random sample of words. The details of the categorization grid is provided in Appendix 1. All analyses were performed with R (R Core Team 2014) and the R script we wrote can be found in Appendix 1.
We present below our comparative analysis of farmers’ SRs of the agricultural landscape in our four contrasted SESs. Results regarding our first prediction on the consensus level and variability of the representation are based on the prototypical analyses. Results regarding our second prediction on the content of social representations are based on our categorical analysis; details about the prototypical results can be found in Appendix 3. The results of the free association task with the inductor item “landscape” are displayed in Table 3 and results with the inductor item “agricultural landscape” are displayed in Table 4. Comparing these two inductor items allows for evaluating to what degree the SR of the “agricultural landscape” is embedded in the SR of the “landscape.” In the resulting tables, consensual words are identified with the symbol † (Table 3 and Table 4).
Our results showed a weak intra-SES consensus for all SESs and both inductor items. The proportion of consensual words among all uttered words ranged from 2.0% to 4.2% (Table 5). Consensus levels did not differ significantly between SESs (Fisher’s exact test, p > 0.1). The proportion of hapax, indicating intra-group variability was high in all SESs and for both free-listing tasks. It ranged from 65% to 79% (Table 5). We found no significant differences between SESs regarding intra-SES variability. To sum up, consensus and intra-group variability did not significantly differ between SESs; this does not allow us to validate our first prediction.
Qualitatively, it can be observed, though, that in Camargue and Armorique, where landscape issues are more publicized and discussed in social arenas, the consensus is higher on words associated with the inductor “landscape” than with “agricultural landscape” whereas in Gascony Valleys and Hills and Plaine et Val de Sèvre, the consensus is higher on “agricultural landscape” (Table 5). Camargue is the SES with the lowest number of hapax to describe “landscape” whereas Gascony Valleys is the SES with the lowest number of hapax associated to the “agricultural landscape” (Table 5).
Through the categorical analysis of the social representation of the “landscape” (Table 3) we found that the unique category in common at the central core of the representation in the four different SESs was related to the topographical features of the landscape. Then, seminatural elements (hedges, meadows, or woods) can be found in the central core in all SESs but the Camargue, where natural elements are more salient (wild nature, wetlands). It can be further noted that within the central core, topography is perceived as the most salient element in Gascony Valleys and Hills and Plaine et Val de Sèvre, i.e., the SESs where landscape issues are less publicized and discussed in social arenas (see Fig. 2). By contrast, in Armorique, issues related to agriculture and to seminatural elements (hedges and woods) are the most salient and in Camargue issues related to wild nature and biodiversity are the most salient.
The categorical analysis of the social representation of the “agricultural landscape” (Table 4) also showed very few elements in common in the central core of the representation in the four SESs. The unique category we found in common is made of words related to the agricultural vocation of the landscape. Then, the category referring to the patterns created by agriculture on the landscape, e.g., plots size or plot plans, is present in the central core in all sites except Gascony Valleys and Hills.
Beside these inter-SESs consensual elements, many differences can be observed at the central core and on the peripheries of the social representation among the four SESs. We observe that, if part of the content (type of categories) of the representation is shared among SESs, the structure (the place of these categories within the representation) varies a lot. It is beyond the scope of this illustrative case study to detail here all the fine differences among SESs, but we invite interested readers to refer to Tables 3 and 4 as well as Appendix 3 for an exhaustive view of results. The results highlighted above seem to confirm our second prediction, that some elements would be found in common in the core of the SRs regardless of the SES while the peripheries would differ among SESs, although we found fewer similarities between the four central cores than expected.
The categorical analysis showed also that in Camargue, the category “wild nature and biodiversity” is a core element in the representation of the “landscape” (Table 3) but is not part of the representation of the “agricultural landscape” at all (Table 4). In Plaine et Val de Sèvre, the category is absent from both representations. In Armorique and Gascony Valleys and Hills, “wild nature and biodiversity items” lay at the periphery of both “landscape” and “agricultural landscape” representations. These results show that the representation of the “landscape” was most inclusive of nondomestic nature in Camargue, where many conservation tools are applied on the territory, but the SR of the “agricultural landscape” excluded it. Surprisingly, nondomestic nature is absent from representation in Plaine et Val de Sèvre despite a high level of territorial animation on conservation issues through Natura 2000 and Agro-Environmental Scheme (AES) contracts.
Here, we revisit our two sets of hypothesis/prediction in light of our results and highlight possible consequences for agricultural landscape management and biodiversity conservation policies. We also discuss the benefits of our methodological approach.
Our first hypothesis (H1) was that the level of local public policies dealing with agricultural landscape had an impact on the content and structure of the social representation. Our first prediction (P1) was that we should find a greater proportion of consensual words and smaller proportion of hapax in study sites where landscape issues are advertised and discussed through local management agencies and local public policies.
In all study sites, farmers’ SRs of the landscape and of the agricultural landscape seemed poorly structured because consensus level was low and hapax proportion was high. We found no significant differences in consensus between sites. This would suggest that the “landscape” and “agricultural landscape” are not exactly social objects yet for the social group of “farmers”; the representations are still emerging (Jeziorski 2014). Indeed, according to Moscovici (1976), an SR cannot appear without sufficient “focalization” and “pressure to infer” processes. This means that members of the social group have to be sufficiently exposed to a social object and sufficiently pressured by the social context for them to position themselves on the object as a social group and elaborate a social representation of the object. Both processes might not have been strong enough in our study sites, even in Camargue, despite policy makers’ growing interest in the landscape.
We found that the SR of “landscape” is more structured than the representation of the “agricultural landscape” in Camargue and Armorique, whereas in Gascony and Plaine et Val de Sèvre, consensus is higher on “agricultural landscape.” This could indicate stronger processes of “dispersal of information,” “focalization,” and “pressure to infer” on the social subject of the “landscape” in Armorique and Camargue. This could be due to, respectively, high visibility of the decrease in the hedgerow density for its impact on landscape aesthetic and tourism in Armorique (Bazin and Schmutz 1994, Burel and Baudry 1995, Perichon 2004, 2005, Thenail and Baudry 2004) and an important communication effort on the remarkable features of the Camargue landscape by the Regional Nature Park (Mathevet 2004). In PVDS and Gascony, the agricultural vocation of those territories might be more anchored so that there are fewer issues about the “landscape,” its aesthetic, “focalization” being mainly on agricultural issues. In addition, we found few collectively shared items across regions even through the categorical analysis suggested very local and social representations of the “landscape” and the “agricultural landscape.”
Regarding the implication for public policies, weak consensus and great variability of representations is likely to be an obstacle to any landscape management public policy. The local land management agency still has to build a shared representation of the landscape with farmers before designing or implementing any policy. Moreover, the very few similarities between regions favor a territorialized approach (context-specific approach) for landscape management policies. Consequently, the first step would be to locally raise the landscape as a concern for farmers, for example, through collective building of a shared representation to design innovative policies (Berthet et al. 2016, De Vreesse et al. 2019) aimed at improving farmers’ stewardship of the landscape (Scherr and McNeely 2007, de Sainte Marie 2014).
Our second hypothesis (H2) was that the local environment is likely to influence SRs of the landscape. Our second prediction (P2) was that some words belonging to the core of the SR will be cited in common by farmers regardless of the site they farm on while the peripheries of the SRs will differ among sites.
Farmers SRs of the “landscape” and the “agricultural landscape” were mainly descriptive and aesthetic at the core, and descriptive, aesthetic, and functional (Buijs et al. 2011) on the periphery. This result seems to be in line with similar studies conducted in different cultural contexts such as Malta (Conrad et al. 2011). Indeed, several studies have shown that the landscape experience is mostly a process of social and cultural mediation (Berque 1995). In addition to its physical dimensions, landscape is a sensitive and emotional experience linked both to its physical shape and to cultural and social norms. That means that the landscape is not only facts or emotions: it is both ontological and ideological (Matless 2003).
We found that SRs of the “landscape” may be more linked to the sensitive interpretation, revealing the main symbolic and cultural norms of the region, whereas SRs of the “agricultural landscape” may be structured more by the practical dimensions of farming, as Baumgart-Getz et al. (2012) showed in their literature review. Information, financial and networking variables better explain why farmers adopt best management practice in the USA. It thus appears that the restriction of the “landscape” through the addition of the adjective “agricultural” leads to changes in several dimensions of the SR: the aesthetic dimension (“scenery” dimension) becomes less important and the functional dimensions are more salient, the emphasis on seminatural component decreases and emphasis on cropped or grazed areas increases. This result is consistent with Rateau (1998) and Gruev-Vintila and Rouquette (2007), which suggested that the descriptive and functional dimensions are dominant in the SRs of an object when individuals practice with this object.
Consequently, we recommend that landscape management policies focus on the “agricultural landscape” as being a coproduction of farming activities and natural features (Gibon 2005), with which farmers can more easily identify, and feel in charge. However, our results suggest that SRs of the “agricultural landscape” show a weak consensus due to poor focalization on this subject in each region. This suggests that research and conservation processes, such as applied sciences, collaborative sciences, or action-researches that have occurred in these regions did not seem to have triggered a higher structuration degree, i.e., consensus level, of the SR. The “agricultural landscape,” and a fortiori the “landscape,” is still the subject of a negotiation process and needs to be collectively discussed before any new conservation or adaptive management policy is elaborated (Mathevet et al. 2014).
Our results suggest that SRs of the “landscape” tend to be overlooked in the existing governance processes of these territories, i.e., weak consensus and great variability of SRs is likely to be an obstacle to any landscape management public policy. The local land management agencies of several SESs still have to shape a shared representation of the “landscape” with farmers in order to implement an efficient policy. This seems problematic because these context-specific SRs could motivate stakeholders to participate in the planning process (De Vresse et al. 2019). We think that we need to change our own ways of thinking about agricultural policy design by developing more bottom-up processes that would frame and integrate farmers’ SRs (Mathevet et al. 2014, Vuillot et al. 2016). Understanding shared and unshared elements is a means to increase mutual comprehension, to create an arena for deliberation on SES’s management and planning, and to frame solutions that deal with divergent visions (Plieninger et al. 2015, Raymond et al. 2014, Mathevet et al. 2018a) and that build landscape stewardship (Bieling and Plieninger 2017). In addition to the SRT, concepts like place attachment (Brown et al. 2019) and ecological solidarity (Mathevet et al. 2016) could help understanding SRs and implementing a sound social-ecological stewardship (Mathevet et al. 2018b, Chapin et al. 2009).
Our previous work on individual mental models suggested that farmers’ ways of thinking and ways of farming were linked (Vuillot et al. 2016) but also that farming practices under strong policy and technical or economic constraints, influence farmers’ representations (see also Baumgart-Getz et al. 2012). Thus, building shared representations by increasing farmers’ awareness of social-ecological interdependencies may not be sufficient to change practices or to promote any new landscape management policy because technological and economic path-dependency processes continue to shape the acceptance on the grounds of any biodiversity friendly landscape recommendations (Pe'er et al. 2014, Sirami et al. 2019). Although SRs do not suffice to explain practices, they provide a better understanding of how contexts, institutional, and organizational dimensions guide, if not determine, practices. It was not the purpose of this study to report on the interview findings outside of free-listing activities, but it is obvious that measuring SRs should be articulated with broader observations that more fully explain how SRs are socially constructed, i.e., in our case study, SRs are elaborated and are fundamentally part of social interactions that were mainly characterized in each SES by the intensity of local territorial animation on landscape and biodiversity conservation (see Fig. 2 and Appendix 3). An interdisciplinary method coupling SRs’ elicitation, interviews, mental models, and farming systems approaches provides an opportunity to enhance our understanding of the relationships between farmers’ worldviews and their practices, in order to design better biodiversity conservation policies in farmed landscapes (Vuillot 2015).
Finally, SRs are embedded in dialogue and debate processes, media, and scientific discourses (Moscovici and Marková 2000, Höjier 2011). Thus SRs are also shaped by power relations (Vorlklein and Howarth 2005). Various representations of the same social objects, here the landscape, are operated to achieve different ends (Howarth 2006). Different social groups or stakeholders/individuals have different levels of access to the construction of the social reality of any SES processes within the public arena, and thus to the legitimization of knowledge systems; they, therefore, experience different levels of social marginalization or inclusion processes (Voelklein and Howarth 2005). SRs are dynamic systems of knowledge, they are always open to negotiation (Howarth et al. 2014), and powerful groups can shape social reality by actively participating in the construction of SRs, i.e., political participation being defined as “the power to construct and convey particular representations over others” by Howarth et al. (2014:21). As these authors stated, hegemonic representations could be normative and shape ways of thinking and engaging with others (Moloney and Walker 2002). In our case studies, we observed that no SR of any SES is dominated by a hegemonic vision of nature conservation nor intensive farming worldviews. Different knowledge systems compete for the social construction of meaning, i.e., nature conservation, intensive farming, organic farming, traditional, hedonistic, economic worldviews (Mathevet 2004, Vuillot et al. 2016). Thus, how knowledge is legitimized and reified is an important research area. The SRT, despite some theoretical criticism, may provide ways to critically analyze the SES management and landscape planning. By allowing rigorous data collection and comparative analysis process, the method we have provided and this type of analysis could facilitate the study of the social-ecological consequences of SRs and the relationship between SRs, power relations, and social conflicts.
In this paper, we highlighted the value of studying SRs and proposed a methodological approach to elicit SRs and compare them in a standardized and efficient way. We applied this approach to the study of SRs of the “landscape” and the “agricultural landscape” among distinct social-ecological contexts. Our case study demonstrated how this standardized approach allowed us to detect subtle differences both among social-ecological contexts and objects. Moreover, we showed that this approach can be used to compare SRs across studies with various sample sizes and various consensus levels. This type of comparison, among social groups or localities, is crucial to understand processes that influence the content and structuration level of SRs. Our methodological approach represents a trade-off between reliability and feasibility, especially for nonspecialists, because it is easy to implement and allows comparisons between space, time, social groups, and subjects of representation, provided a unified methodological approach is adopted. We therefore hope that this method will convince more people to use the SRT in SES studies. This method should also facilitate meta-analysis across various studies on SRs of SESs, which will contribute to a better understanding of the dynamics of SRs in space and time, and their contribution to adaptive management of SES (Lai et al. 2016, Bouamrane et al. 2017).
This research was funded by the ERA-Net BiodivERsA and the French National Research Agency (ANR-11-EBID-0004), German Ministry of Research and Education, German Research Foundation, and Spanish Ministry of Economy and Competitiveness, part of the 2011 BiodivERsA call for research proposals. We acknowledge the farmers who contributed to this study and were willing to share their own vision of the landscape, and for their time and knowledge. We are also grateful for the insightful comments of two anonymous reviewers and Ecology and Society editors.
Data Availability Statement
The data/code that support the findings of this study are available on request from the corresponding author, RM.
Abric, J.-C. 1989. L’étude expérimentale des représentations sociales. Pages 187-203 in D. Jodelet, editor. Les représentations sociales. Presses Universitaires de France, Paris, France.
Abric, J.-C. 1994. L’organisation interne des représentations sociales: système central et système périphérique. Pages 73-84 in C. Guimelli, editor. Structures et transformation des représentations sociales. Delachaux et Niestlé, Neuchâtel, Switzerland.
Abric, J.-C. 2001a. Pratiques sociales, représentations sociales. Pages 217-238 in J. C. Abric, editor. Pratiques sociales et représentations. Presses Universitaires de France, Paris, France.
Abric, J.-C. 2001b. A structural approach to social representations. Pages 42-47 in K. Deaux and G. Philogène, editors. Representations of the social: bridging theoretical traditions. Blackwell, Oxford, UK.
Abric, J.-C. 2003. L’analyse structurale des représentations sociales. Pages 375-392 in S. Moscovici and F. Buschini, editors. Les méthodes des sciences humaines. Presses Universitaires de France, Paris, France.
Abric, J.-C. 2011. Pratiques sociales et représentations. Presses Universitaires de France, Paris, France.
Agreste. 2000. Recensement Agricole – 2000. Actualité et statistique agricole. Agreste, Paris, France. [online] URL: http://188.8.131.52/enquetes/structure-des-exploitations-964/recensement-agricole-2000/
Alessio, R. L. D. S., T. Apostolidis, M. de Fátima de S Santos, and L. Dany. 2011. Représentations sociales et embryon humain: une étude comparative Brésil/France. Cahiers Internationaux de Psychologie Sociale 92:371-395. https://doi.org/10.3917/cips.092.0371
Anderson, N. M., R. M. Ford, and K. J. H. Williams. 2017. Contested beliefs about land-use are associated with divergent representations of a rural landscape as place. Landscape and Urban Planning 157:75-89. https://doi.org/10.1016/j.landurbplan.2016.05.020
Anderson, N. M., K. J. H. Williams, and R. M. Ford. 2013. Community perceptions of plantation forestry: the association between place meanings and social representations of a contentious rural land use. Journal of Environmental Psychology 34:121-136. https://doi.org/10.1016/j.jenvp.2013.02.001
Audouit, C., V. Pasqualini, R. De Wit, H. Flanquart, P. Deboudt, and C. Rufin-Soler. 2019. Comparing social representation of water quality in coastal lagoons with normative use of ecological indicators. Marine Policy 101:137-146. https://doi.org/10.1016/j.marpol.2017.08.023
Barreteau, O., D. Giband, M. Schoon, J. Cerceau, F. DeClerck, S. Ghiotti, T. James, V. A. Masterson, R. Mathevet, S. Rode, F. Ricci, and C. Therville. 2016. Bringing together social-ecological system and territoire concepts to explore nature-society dynamics. Ecology and Society 21(4):42. https://doi.org/10.5751/ES-08834-210442
Barsalou, L. W. 2009. Simulation, situated conceptualization, and prediction. Philosophical Transactions of the Royal Society B: Biological Sciences 364:1281-1289. https://doi.org/10.1098/rstb.2008.0319
Baumgart-Getz, A., L. Stalker Prokopy, and K. Floress. 2012. Why farmers adopt best management practice in the United States: a meta-analysis of the adoption literature. Journal of Environmental Management 96(1):17-25. https://doi.org/10.1016/j.jenvman.2011.10.006
Bazin, P., and T. Schmutz. 1994. The rise and fall of hedgerow landscaping in Europe. Revue Forestière Française 46:115-118.
Bergamaschi, A. 2011. Attitudes et représentations sociales. Revue Européenne des Sciences Sociales 49(2):93-122. https://doi.org/10.4000/ress.996
Berque, A. 1995. Les raisons du paysage: de la Chine antique aux environnements de synthèse. Fernand Hazan, Paris, France.
Berthet, E. T. A., C. Barnaud, N. Girard, J. Labatut, and G. Martin. 2016. How to foster agroecological innovations? A comparison of participatory design methods. Journal of Environmental Planning and Management 59(2):280-301. https://doi.org/10.1080/09640568.2015.1009627
Bieling, C., and T. Plieninger. 2017. The science and practice of landscape stewardship. Cambridge University Press, Cambridge, UK. https://doi.org/10.1017/9781316499016
Blicharska, M., and A. Van Herzele. 2015. What a forest? Whose forest? Struggles over concepts and meanings in the debate about the conservation of the Białowieza Forest in Poland. Forest Policy and Economics 57:22-30. https://doi.org/10.1016/j.forpol.2015.04.003
Bouamrane, M., R. Mathevet, H. Levrel, H. Huntington, and A. Agrawal. 2017. Community participation and adaptation to change in biosphere reserves: a review and a Mediterranean European coastal wetland case study (Rhone Delta Biosphere Reserve, Southern France). Pages 120-138 in D. Armitage, A. Charles, and F. Berkes, editors. Governing the coastal commons: communities, resilience and transformation. Routledge, New York, New York, USA. https://doi.org/10.4324/9781315688480-7
Brown, K., W. N. Adger, P. Devine-Wright, J. M. Anderies, S. Barr, F. Bousquet, C. Butler, L. Evans, N. Marshall, and T. Quinn. 2019. Empathy, place and identity interactions for sustainability. Global Environmental Change 56:11-17. https://doi.org/10.1016/j.gloenvcha.2019.03.003
Buijs, A. E., B. J. M. Arts, B. H. M. Elands, and J. Lengkeek. 2011. Beyond environmental frames: the social representation and cultural resonance of nature in conflicts over a Dutch woodland. Geoforum 42:329-341. https://doi.org/10.1016/j.geoforum.2010.12.008
Buijs, A. E., B. Pedroli, and Y. Luginbühl. 2006. From hiking through farmland to farming in a leisure landscape: changing social perceptions of the European landscape. Landscape Ecology 21:375-389. https://doi.org/10.1007/s10980-005-5223-2
Burel, F., and J. Baudry. 1995. Social, aesthetic and ecological aspects of hedgerows in rural landscapes as a framework for greenways. Landscape and Urban Planning 33:327-340. https://doi.org/10.1016/0169-2046(94)02026-C
Calatayud, F., S. Ladet, G. Balent, and L. Fahrig. 2012. A mapping method of crop composition and configuration to study their effect on biodiversity of European agricultural landscapes. Pages 308-311 in J. Gensel, D. Josselin, and D. Vandenbroucke, editors. Proceedings of the AGILE'2012 International Conference on Geographic Information Science. Presented at the Multidisciplinary Research on Geographical Information in Europe and Beyond, Avignon, France.
Celio, E., C. G. Flint, P. Schoch, and A. Grêt-Regamey. 2014. Farmers’ perception of their decision-making in relation to policy schemes: a comparison of case studies from Switzerland and the United States. Land Use Policy 41:163-171. https://doi.org/10.1016/j.landusepol.2014.04.005
Chapin III, F. S., G. P. Kofinas, and C. Folke. 2009. Principles of ecosystem stewardship. Resilience-based natural resource management in a changing world. Springer, New York, New York, USA. https://doi.org/10.1007/978-0-387-73033-2
Conrad, E., M. Christie, and I. Fazey. 2011. Understanding public perceptions of landscape: a case study from Gozo, Malta. Applied Geography 31:159-170. https://doi.org/10.1016/j.apgeog.2010.03.009
Dany, L., I. Urdapilleta, and G. L. Monaco. 2015. Free associations and social representations: some reflections on rank-frequency and importance-frequency methods. Quality & Quantity 49:489-507. https://doi.org/10.1007/s11135-014-0005-z
de Sainte Marie, C. 2014. Rethinking agri-environmental schemes. A result-oriented approach to the management of species-rich grasslands in France. Journal of Environmental Planning and Management 57:704-719. https://doi.org/10.1080/09640568.2013.763772
De Vreese, R., A. V. Herzele, N. Dendoncker, C. M. Fontaine, and M. Leys. 2019. Are stakeholders’ social representations of nature and landscape compatible with the ecosystem service concept? Ecosystem Services 37:100911. https://doi.org/10.1016/j.ecoser.2019.100911
Deconchy, J.-P. 2003 (1984). Croyances et idéologies. Pages 331-355 in S. Moscovici, editor. Psychologie sociale. Quadrige, Presses Universitaires de France, Paris, France.
Doise, W. 1985. Les représentations sociales: définition d’un concept. Connexions 45:243-253.
Doise, W. 1992. L’ancrage dans les études sur les représentations sociales. Bulletin de Psychologie 45:189-195.
Doise, W., A. Clemence, and F. Lorenzi-Cioldi. 1993. The quantitative analysis of social representations. Harvester Wheatsheaf, London, UK.
Durkheim, E. 1968 (1912). Les formes élémentaires de la vie religieuse. Presses Universitaires de France, Paris, France.
Elands, B. H. M., and K. F. Wiersum. 2001. Forestry and rural development in Europe: an exploration of socio-political discourses. Forest Policy and Economics 3:5-16. https://doi.org/10.1016/S1389-9341(00)00027-7
European Council. 2000. European landscape convention. Council of Europe, Strasbourg, France.
Farr, R. M. 1991. Les représentations sociales: la théorie et ses critiques. Bulletin de Psychologie 45:183-188.
Farr, M. R. 2003 (1984). Les représentations sociales. Pages 379-389 in S. Moscovici, editor. Psychologie sociale. Quadrige, Presses Universitaires de France, Paris, France.
Flament, C., and M.-L. Rouquette. 2003. Anatomie des idées ordinaires: comment étudier les représentations sociales. Armand Colin, Paris, France.
Galand, C., and E. Salès-Wuillemin. 2009. La représentation des drogues chez les étudiants en psychologie: effets des pratiques de consommation et influence de l’entourage. Les Cahiers Internationaux de Pyschologie Sociale 84:125-152. https://doi.org/10.3917/cips.084.0125
Gibon, A. 2005. Managing grassland for production, the environment and the landscape. Challenges at the farm and the landscape level. Livestock Production Science 96:11-31. https://doi.org/10.1016/j.livprodsci.2005.05.009
Gilg, A. 2009. Perceptions about land use. Land Use Policy 26:S76-S82. https://doi.org/10.1016/j.landusepol.2009.08.018
Gruev-Vintila, A., and M. L. Rouquette. 2007. Social thinking about collective risk: how do risk-related practice and personal involvement impact its social representations? Journal of Risk Research 10:555-581. https://doi.org/10.1080/13669870701338064
Guimelli, C., and M.-L. Rouquette. 1992. Contribution du modèle associatif des schèmes cognitifs de base à l’analyse structurale des représentations sociales. Bulletin de Psychologie 405:196-202.
Halfacree, K. H. 1993. Locality and social representation: space, discourse and alternative definitions of the rural. Journal of Rural Studies 9:23-37. https://doi.org/10.1016/0743-0167(93)90003-3
Herzlich, C. 1972. La représentation sociale. Pages 303-325 in S. Moscovici, editor. Introduction à la psychologie sociale. Larousse, Paris, France.
Höijer, B. 2011. Social representations theory. A new theory for media research. Nordicom Review 32(2):3-16. https://doi.org/10.1515/nor-2017-0109
Holm, S. 1979. A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics 6(2):65-70.
Howarth, C. 2006. A social representation is not a quiet thing: exploring the critical potential of social representations theory. British Journal of Social Psychology 45:65-86. https://doi.org/10.1348/014466605X43777
Howarth, C., E. Andreouli, and S. Kesi. 2014. Social representations and the politics of participation. Pages 19-38 in P. Nesbitt-Larking, C. Kinnvall, T. Capelos, and H. Dekker, editors. The Palgrave handbook of global political psychology. Palgrave Macmillan, London, UK. https://doi.org/10.1007/978-1-137-29118-9_2
Jeziorski, A. 2014. Étude des représentations sociales du développement durable dans une perspective didactique: une contribution à la formation des enseignants à l’éducation au développement durable (thèse en didactique). Université Laval Québec, Canada et Aix-Marseille Université, Marseille, France.
Jeziorski, A., and A. Ludwig-Legardez. 2013. Éducation au développement durable: la difficulté de concevoir une action éducative interdisciplinaire. Revue Francophone du Développement Durable 1:30-59.
Jodelet, D. 1989. Folies et représentations sociales. Presses Universitaires de France, Paris, France.
Jodelet, D. 1991. L’idéologie dans l’étude des représentations sociales. Pages 15-33 in V. Aebischer, J.-P. Deconchy, and E. M. Lipianski, editors. Idéologies et représentations sociales. Les Éditions Delval, Cousset, Switzerland.
Jodelet, D. 2003 (1994). Les représentations sociales. Presses Universitaires de France, Paris, France.
Jones, N. A., H. Ross, T. Lynam, P. Perez, and A. Leitch. 2011. Mental models: an interdisciplinary synthesis of theory and methods. Ecology and Society 16(1):46. https://doi.org/10.5751/ES-03802-160146
Krien, N., and É. Michel-Guillou. 2014. Place des risques côtiers dans les représentations sociales du cadre de vie d’habitants de communes littorales. Les Cahiers Internationaux de Psychologie Sociale 101(1):101-122. https://doi.org/10.3917/cips.101.0101
Lai, P.-H., Y.-C. Hsu, and S. Wearing. 2016. A social representation approach to facilitating adaptive co-management in mountain destinations managed for conservation and recreation. Journal of Sustainable Tourism 24(2):227-244. https://doi.org/10.1080/09669582.2015.1062018
Lebart, L., and A. Salem. 1988. Analyse statistique des données textuelles. Dunod. Paris. France.
Lemée, C., M. Guillard, G. Fleury-Bahi, N. Krien, C. Chadenas, E. Chauveau, M. Desse, M. Coquet, M. Lamarre, and O. Navarro. 2019. What meaning do individuals give to coastal risks? Contribution of the social representation theory. Marine Policy 108:103629. https://doi.org/10.1016/j.marpol.2019.103629
Lo Monaco, G., A. Piermattéo, P. Rateau, and J. L. Tavani. 2017. Methods for studying the structure of social representations: a critical review and agenda for future research. Journal for the Theory of Social Behaviour 47:306-331. https://doi.org/10.1111/jtsb.12124
Luginbühl, Y., P. Howard, and D. Terrasson, editors. 2015. Landscape and sustainable development: the French perspective. Routledge, London, UK. https://doi.org/10.4324/9781315591360
Lynam, T., and K. Brown. 2011. Mental models in human-environment interactions: theory, policy implications, and methodological explorations. Ecology and Society 17(3):24. https://doi.org/10.5751/ES-04257-170324
Lynam, T., R. Mathevet, M. Etienne, S. Stone-Jovicich, A. Leitch, N. Jones, H. Ross, D. Du Toit, S. Pollard, H. Biggs, and P. Perez. 2012. Waypoints on a journey of discovery: mental models in human-environment interactions. Ecology and Society 17(3):23. https://doi.org/10.5751/ES-05118-170323
Mathevet, R. 2004. Camargue incertaine: sciences, usages et natures. Buchet-Chastel, Paris, France.
Mathevet, R., F. Bousquet, C. Larrère, and R. Larrère. 2018a. Environmental stewardships and ecological solidarity: rethinking social-ecological interdependency and responsibility. Journal of Agricultural and Environmental Ethics 31(5):605-623. https://doi.org/10.1007/s10806-018-9749-0
Mathevet, R., F. Bousquet, and C. M. Raymond. 2018b. The concept of stewardship in sustainability science and conservation biology. Biological Conservation 217:363-370. https://doi.org/10.1016/j.biocon.2017.10.015
Mathevet, R., M. Etienne, T. Lynam, and C. Calvet. 2011. Water management in the Camargue Biosphere Reserve: insights from comparative mental models analysis. Ecology and Society 16(1):43. https://doi.org/10.5751/ES-04007-160143
Mathevet, R., J. D. Thompson, C. Folke, and F. S. Chapin III. 2016. Protected areas and their surrounding territory: socioecological systems in the context of ecological solidarity. Ecological Applications 26(1):5-16. https://doi.org/10.1890/14-0421
Mathevet, R., C. Vuillot, and C. Sirami. 2014. Effective nature conservation on farmland: Can we change our own models, not just the farmers? Conservation Letters 7:575-576. https://doi.org/10.1111/conl.12064
Matless, D. 2003. Introduction: The Properties of Landscape. Pages 227-232 in K. Anderson, M. Domosh, S. Pile, and N. Thrift, editors. Handbook of cultural geography. Sage, London, UK. https://doi.org/10.4135/9781848608252
Michel-Guillou, E., and G. Moser. 2006. Commitment of farmers to environmental protection: from social pressure to environmental conscience. Journal of Environmental Psychology 26:227-235. https://doi.org/10.1016/j.jenvp.2006.07.004
Moliner, P. 1993. Cinq questions à propos de réprésentations sociales. Cahiers Internationaux de Psychologie Sociale 20(12):5-14.
Moliner, P., and J.-C. Abric. 2015. Central core theory. Pages 83-95 in G. Sammut, E. Andreouli, G. Gaskell, and J. Valsiner, editors. The Cambridge handbook of social representations. Cambridge University Press, Cambridge, UK. https://doi.org/10.1017/CBO9781107323650.009
Moliner, P., P. Rateau, and V. Cohen-Scali. 2002. Les représentations sociales: pratique des études de terrain. Presses Universitaires de France, Rennes, France.
Moloney, G., Z. Leviston, T. Lynam, J. Price, S. Stone-Jovicich, and D. Blair. 2014. Using social representations theory to make sense of climate change: what scientists and nonscientists in Australia think. Ecology and Society 19(3):19. https://doi.org/10.5751/ES-06592-190319
Moloney, G., and I. Walker. 2002. Talking about transplants: social representations and the dialectical, dilemmatic nature of organ donation and transplantation. British Journal of Social Psychology 41(2):299-320. https://doi.org/10.1348/014466602760060264
Morlot, R., and E. Salès-Wuillemin. 2009. Effet des pratiques et des connaissances sur la représentation sociale d’un objet: application à l’hygiène hospitalière. Revue Internationale de Pyschologie Sociale 21:89-114.
Moscovici, S. 1972. Society and theory in social psychology. Pages 66-91 in J. Israel and H. Tajfel, editors. The context of social psychology: a critical assessment. Academic, London, UK.
Moscovici, S. 1976 (1961). La psychanalyse, son image et son public. Presses Universitaires de France, Paris, France.
Moscovici, S. 1988. Notes towards a description of social representations. European Journal of Social Psychology 18:211-250. https://doi.org/10.1002/ejsp.2420180303
Moscovici, S., and I. Marková. 1998. Presenting social representations: a conversation. Culture and Psychology 4:371-410. https://doi.org/10.1177/1354067X9800400305
Moscovici, S., and I. Marková. 2000. Ideas and their development: a dialogue between Serge Moscovici and Ivana Markova. Pages 224-286 in G. Duveen, editor. Social representations: explorations in social psychology. Polity Press, London, UK.
Pasher, J., S. W. Mitchell, D. J. King, L. Fahrig, A. C. Smith, and K. E. Lindsay. 2013. Optimizing landscape selection for estimating relative effects of landscape variables on ecological responses. Landscape Ecology 28:371-383. https://doi.org/10.1007/s10980-013-9852-6
Pe'er, G., L. V. Dicks, P. Visconti, R. Arlettaz, A. Báldi, T. G. Benton, S. Collins, M. Dieterich, R. D. Gregory, F. Hartig, K. Henle, P. R. Hobson, D. Kleijn, R. K. Neumann, T. Robijns, J. Schmidt, A. Shwartz, W. J. Sutherland, A. Turbé, F. Wulf, and A. V. Scott. 2014. EU agricultural reform fails on biodiversity. Science 344:1090-1092. https://doi.org/10.1126/science.1253425
Perichon, S. 2004. L’impossible reconstruction des bocages détruits. Espace Géographique 33:175-187. https://doi.org/10.3917/eg.332.0175
Perichon, S. 2005. Les représentations sociales associées au bocage. Analyse des relations agriculteurs/territoire en haute-bretagne. Bulletin de la Société Géographique de Liège 46:71-80.
Plieninger, T., T. Kizos, C. Bieling, L. Le Dû-Blayo, M.-A. Budniok, M. Bürgi, C. L. Crumley, G. Girod, P. Howard, J. Kolen, T. Kuemmerle, G. Milcinski, H. Palang, K. Trommler, and P. H. Verburg. 2015. Exploring ecosystem-change and society through a landscape lens: recent progress in European landscape research. Ecology and Society 20(2):5. https://doi.org/10.5751/ES-07443-200205
Quétier, F., F. Rivoal, P. Marty, J. de Chazal, W. Thuiller, S. Lavorel. 2010. Social representations of an alpine grassland landscape and socio-political discourses on rural development. Regional Environmental Change 10:119–130. https://doi.org/10.1007/s10113-009-0099-3
R Core Team. 2014. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.
Rateau, P. 1998. Variable d’implication et évolution structurale d’une représentation sociale. 4ème Conférence Internationale sur les Représentations Sociales, Mexico City, Mexico.
Raymond, C. M., J. O. Kenter, T. Plieninger, N. J. Turner, and K. A. Alexander. 2014. Comparing instrumental and deliberative paradigms underpinning the assessment of social values for cultural ecosystem services. Ecological Economics 107:145-156. https://doi.org/10.1016/j.ecolecon.2014.07.033
Roussiau, N., and C. Bonardi. 2001. Les représentations sociales: état des lieux et perspectives. Editions Mardaga, Sprimont, Belgium.
Salès-Wuillemin, E., R. Morlot, A. Fontaine, W. Pullin, C. Galand, D. Talon, and P. Minary-Dohen. 2011. Evolution of nurses’ social representations of hospital hygiene: from training to practice. European Review of Applied Psychology 61:51-63. https://doi.org/10.1016/j.erap.2010.06.001
Sammut, G. 2015. Attitudes, social representations and points of view. Pages 96-112 in G. Sammut, E. Andreouli, G. Gaskell, and J. Valsiner, editors. The Cambridge handbook of social representations. Cambridge handbooks in psychology series. Cambridge University Press, Cambridge, UK. https://doi.org/10.1017/CBO9781107323650.010
Sammut, G., E. Andreouli, G. Gaskell, and J. Valsiner. 2015a. Social representations: a revolutionary paradigm? Pages 3-11 in G. Sammut, E. Andreouli, G. Gaskell, and J. Valsiner, editors. The Cambridge handbook of social representations. Cambridge University Press, Cambridge, UK. https://doi.org/10.1017/cbo9781107323650.003
Sammut, G., E. Andreouli, G. Gaskell, and J. Valsiner. 2015b. The Cambridge handbook of social representations. Cambridge University Press, Cambridge, UK. https://doi.org/10.1017/CBO9781107323650
Sammut, G., and C. Howarth. 2014. Social representations. Pages 1799-1802 in T. Teo, editor. Encyclopedia of critical psychology. Springer, New York, New York, USA. https://doi.org/10.1007/978-1-4614-5583-7_292
Scherr, S. J., and J. A. McNeely. 2007. Biodiversity conservation and agricultural sustainability: towards a new paradigm of ‘ecoagriculture’ landscapes. Philosophical Transactions of the Royal Society B: Biological Sciences 363:477-494. https://doi.org/10.1098/rstb.2007.2165
Sirami, C., N. Gross, A. Bosem Baillod, C. Bertrand, R. Carrié, A. Hass, L. Henckel, P. Miguet, C. Vuillot, A. Alignier, J. Girard, P. Batáry, Y. Clough, C. Violle, D. Giralt, G. Bota, I. Badenhausser, G. Lefebvre, B. Gauffre, A. Vialatte, F. Calatayud, A. Gil-Tena, L. Tischendorf, S. Mitchell, K. Lindsay, R. Georges, S. Hilaire, J. Recasens, X. Oriol Solé-Senan, I. Robleño, J. Bosch, J. A. Barrientos, A. Ricarte, M. A. Marcos-Garcia, J. Miñano, R. Mathevet, A. Gibon, J. Baudry, G. Balent, B. Poulin, F. Burel, T. Tscharntke, V. Bretagnolle, G. Siriwardena, A. Ouin, L. Brotons, J. L. Martin, and L. Fahrig. 2019. Increasing crop heterogeneity enhances multitrophic diversity across agricultural regions. Proceedings of the National Academy of Sciences 116(33):16442-16447. https://doi.org/10.1073/pnas.1906419116
Swaffield, S. R. 1991. Roles and meanings of ‘landscape.’ Dissertation. Lincoln University, UK.
Thenail, C., and J. Baudry. 2004. Variation of farm spatial land use pattern according to the structure of the hedgerow network (bocage) landscape: a case study in northeast Brittany. Agriculture, Ecosystems & Environment 101:53-72. https://doi.org/10.1016/S0167-8809(03)00199-3
Tscharntke, T., A. M. Klein, A. Kruess, I. Steffan-Dewenter, and C. Thies. 2005. Landscape perspectives on agricultural intensification and biodiversity - ecosystem service management. Ecology Letters 8:857-874. https://doi.org/10.1111/j.1461-0248.2005.00782.x
Tsirogianni, S. 2014. Social values. Pages 1809-1812 in T. Teo, editor. Encyclopedia of critical psychology. Springer, New York, New York, USA. https://doi.org/10.1007/978-1-4614-5583-7_294
Valence, A. 2010. Les représentations sociales. De Boeck, Bruxelles, Belgium. https://doi.org/10.3917/dbu.valen.2010.01
Van Dijk, T. A. 2006. Politique, idéologies et discours. SEMEN 21:73-102.
Vergès, P. 1992. L’evocation de l’argent: une méthode pour la définition du noyau central d’une représentation. Bulletin de Psycholology 45:203-209.
Vergès, P., Y. Bonnec, and N. Roussiau. 2002. Categorical and prototypical analysis: a study on the quality-process in hospital institutions. European Review of Applied Psychology 52:213-220.
Voelklein, C., and C. Howarth. 2005. A review of controversies about social representations theory: a British debate. Culture and Psychology 11:431-454. https://doi.org/10.1177/1354067X05058586
Vuillot, C. 2015. Paysages agricoles et biodiversité dans les représentations sociales et les modèles mentaux des agriculteurs: une approche comparée de quatre régions françaises. Dissertation. Human Geography, AgroParisTech and Université de Montpellier, Montpellier, France.
Vuillot, C., N. Coron, F. Calatayud, C. Sirami, R. Mathevet, and A. Gibon. 2016. Ways of farming and ways of thinking: do farmers’ mental models of the landscape relate to their land management practices? Ecology and Society 21(1):35. https://doi.org/10.5751/es-08281-210135
Wachelke, J. 2011. Social representations: a review of theory and research from the structural approach. Universitas Psychologica 11(3):729-741. https://doi.org/10.11144/Javeriana.upsy11-3.srrt
Wachelke, J., and R. Wolter. 2011. Critérios de construção e relato da análise prototípica para representações sociais. Psicologia: Teoria e Pesquisa 27(4):521-526. https://doi.org/10.1590/S0102-37722011000400017
Wagner, W., G. Duveen, R. Farr, S. Jovchelovitch, F. Lorenzi-Cioldi, I. Marková, and D. Rose. 1999. Theory and method of social representations. Asian Journal of Social Psychology 2:95-125. https://doi.org/10.1111/1467-839X.00028
Wagner, W., and N. Hayes. 2005. Everyday discourse and common sense: the theory of social representations. Red Globe, London, UK.