Elsevier

Ecological Modelling

Volume 250, 10 February 2013, Pages 352-362
Ecological Modelling

Development of a broadened cognitive mapping approach for analysing systems of practices in social–ecological systems

https://doi.org/10.1016/j.ecolmodel.2012.11.023Get rights and content

Abstract

This paper presents a new cognitive mapping approach for analysing systems of practices in social–ecological systems. These systems are mapped from people's views collected during open-ended interviews. Cognitive maps are made up of diverse variables (e.g., operations, drivers, constraints) linked to each other by a range of relationships: cause–effect, fluxes of matter, information flows and sequence of two operations. Individual cognitive maps heuristically model the practices and decision-making processes expressed by interviewees. The mathematical formulation of cognitive maps allows individual cognitive maps to be aggregated into a social cognitive map. The latter can be used to model the system of practices used by a particular group of people. Using this approach, we analysed the practices and decision-making processes linked to grassland management in a Belgian grassland-based livestock farming system. Our work confirmed that a social cognitive map could be drawn up for multiple locations. The results showed how this inductive cognitive mapping approach overcame two limitations frequently highlighted in previous studies: the diverse interpretations of variables and relationships; and the difficulty in revealing the rationale in cognitive maps.

Highlights

► We developed an approach to model systems of practices in social–ecological systems. ► Open-ended interviews are coded into individual cognitive maps (ICMs). ► For a given set of actors, ICMs can be aggregated in social cognitive maps (SCMs). ► Analysis of SCMs highlights most important relationships within systems of practices. ► Our approach was applied to grass forage management in Belgian grasslands systems.

Introduction

In social–ecological systems (Holling, 2001, Walker et al., 2004), decision-making tends to be extremely complex because of the intricacy of these systems (Ascough et al., 2008). In the agricultural context, the scientific community has developed various models of these systems and used them as simulation tools to support managers’ decisions (Edwards-Jones, 2006, McCown et al., 2009). Farmers’ strategies are based on the interaction of their perceptions about their ecological, economic and social environments. These strategies are translated into practices through decision-making processes. As external factors change, strategies and practices are continuously adapted. The study of managers’ practices and their drivers is an important factor in modelling agricultural systems and highlights the need to model both system complexity at the farm level and system diversity at the regional level (Landais et al., 1988).

Studies of practices in agricultural systems can be grouped into two broad scientific approaches: one based on social issues (anthropological science) and the other on technical issues (engineering science). The social approaches are inductive, linked to anthropological and social sciences, and view practices as social constructs from a constructivist point of view (Darré, 1996, Darré et al., 2004). They focus on understanding managers’ perceptions and representations of social–ecological systems, either as a whole or divided into sub-systems, in terms of practices, knowledge, etc. (Darré et al., 2004). The outputs of such studies provide a good understanding of the studied situations, but they are not easy to incorporate into bio-economic simulation (Mathieu, 2004, Papy, 2004). Conversely, the technical approaches involve studying complex interactions among elements of the studied systems (Janssen and van Ittersum, 2007, Darnhofer et al., 2010). They use theories from artificial intelligence or management science in order to build farming systems models and decision-support systems (Aubry et al., 1998, Girard and Hubert, 1999, Dounias et al., 2002, Keating et al., 2003, Cros et al., 2004, Louhichi et al., 2004, Merot et al., 2008, Vayssieres et al., 2009). These models can be used to simulate and evaluate scenarios in order to support managers via decision support systems (DSS). In most of these bio-economic models, the involvement of actors (managers or non-scientific experts) is limited to their validating, enriching or specifying the structure of a model developed by scientists (Gouttenoire et al., 2010).

Historically, models of social–ecological systems have tended to ignore the social components (Dent et al., 1995) leading to the limited impact of DSS in rural resource management. This has led various authors to highlight the need for incorporating social aspects into DSS (Edwards-Jones, 2006, Gouttenoire et al., 2011). In order to improve decision-making in social–ecological systems, greater understanding is needed of the knowledge used by managers (such as farmers) in managing their systems (Girard and Hubert, 1999). A way to address this challenge is to develop models based on farmers’ perceptions that reflect the way they perceive their own agro-ecosystem in an inductive way. The structure of these models should be focused on the farmers’ practices. The objective of our study was to develop a socio-technical modelling tool to inductively identify and model farmers’ systems of practices.

Cristofini et al. (1978) were the first to use the term ‘system of practices’, referring to ‘a consistent combination of practices’ (Gras et al., 1989). In an organizational context, a ‘system of practices’ was later defined as the actions shaped by normative structures (Levitt, 1998) or as the complex network of structures, tasks and traditions that create and facilitate practice (Halverson, 2003). In our study, the definition of ‘systems of practices’ provided by Cristofini et al. (1978) was broadened thus: a farmers’ system of practices is (i) a particular combination of elementary practices, (ii) factors influencing practices, (iii) elements affected by these practices and (iv) the way in which all of them are linked to each other.

Studying systems of practices implies a degree of complexity: systems of practices are not only constrained by their environment (e.g., market, climate, seasons, consumer choices), but are also highly influenced by human factors (actors’ preferences and perceptions). The importance of these human factors underlines the need to analyse actors’ local knowledge. In this context, knowledge-driven modelling techniques, such as cognitive mapping approaches, seem to be promising alternatives for implementing DSS in terms of taking account of social aspects (Fairweather, 2010).

Cognitive mapping approaches have been used to identify people's perceptions of complex social systems (Özesmi and Özesmi, 2004). In this field of study, the work of Axelrod (1976) was seminal. He was the first to use directed graphs (i.e., a network of nodes and directed edges) to show causal relationships based on actors’ descriptions, and he called these representations ‘cognitive maps’. Kosko (1986) applied fuzzy causal function (i.e., weighting the edges, from −1 to 1) to the relationships, creating ‘Fuzzy Cognitive Maps’ (FCM). Recent scientific studies have used cognitive mapping techniques in various domains, such as management studies (Pinch et al., 2010), finance (Koulouriotis et al., 2005) and medical sciences (Stylios et al., 2008, Papageorgiou, 2011). In an organizational setting, the Strategic Options Development and Analysis (SODA) technique has been developed by Colin Eden and Fran Ackermann (Ackermann and Eden, 2010). This technique is used to represent problematic situations in individual or collective cognitive maps. Taking account of a complex system of goals and objectives, it allows participants to explore options and find negotiated solutions to resolve problematic situations.

In environmental sciences, cognitive mapping techniques have been used mainly in environmental conflict management (Özesmi and Özesmi, 2003, Özesmi and Özesmi, 2004) and forest management (Mendoza and Martins, 2006, Tikkanen et al., 2006, Isaac et al., 2009, Kok, 2009, Wolfslehner and Vacik, 2011). Ten studies have applied FCM to agricultural systems analysis (Table 1) in order to: (i) understand farmer perceptions about pesticides (Popper et al., 1996) on their own farms (Fairweather, 2010) or about environmental management measures (Ortolani et al., 2010); (ii) describe practices in agro-ecosystems (Isaac et al., 2009); (iii) assess the impact of agricultural systems on the environment (Özesmi and Özesmi, 2003) and crop yield (Papageorgiou et al., 2009, Papageorgiou, 2011) and the impact of policies on agricultural systems (Hukkinen, 1993, Newig et al., 2008); and (iv) evaluate the sustainability of agro-ecosystems (Rajaram and Das, 2010, Fairweather and Hunt, 2011).

Cognitive mapping approaches are flexible tools that can model people's diverse drivers and motivations without excluding the technical dimensions linked to the studied system of practices. Özesmi and Özesmi (2004) developed a multi-step FCM approach for analysing how people perceive an ecosystem and for comparing and contrasting the perceptions of different people or stakeholder groups. The authors looked at particular examples of environmental conflicts, each one linked to one ecosystem, such as the creation of a national park or the erection of a hydroelectric dam. Fairweather (2010) has applied Özesmi and Özesmi (2004)'s approach to the study of identical ecosystems in different places. He has shown how maps from several farmers, each describing his/her own farm, can be used to build a group map that represents how a group of farmers think their farm ecosystem works. These maps created with farmers focus on the farm system as whole, overlooking details about how parts of the system work. Considering the complex nature of social–ecological systems, he suggested that further work on building causal maps for particular parts of the farm system would be needed to describe fully how the system works (Fairweather, 2010). Based on FCM, we have developed a new approach for examining a particular part of the farm system – the system of farmer practices, as defined earlier.

In this article, we initially describe a Cognitive Mapping Approach for Analysing Actors’ Systems Of Practices (CMASOP) in social–ecological systems. We then apply this approach to the analysis of forage management in a grassland-based livestock farming system, as a case study. One original aspect of the CMASOP approach is its application of cognitive mapping for gaining a detailed understanding of an important part of social–ecological systems – people's practices.

Section snippets

CMASOP approach

The CMASOP approach is based on using open-ended interviews to create individual cognitive maps (ICMs). These ICMs are then used to build a social cognitive map (SCM). The four steps are illustrated in Fig. 1 and are described below.

Case study: grass forage management in a grassland-based livestock farming system

We used the CMASOP approach to analyse grass forage management (harvest, preservation and conditioning of grass forage) in the livestock farming systems of Ardennes and Famenne, two grassland-based systems in Belgium (Fig. 2). The study carried out among the farmers in this area is described here.

Discussion

Systems of practices and the way people talk about them are complex (Landais et al., 1988, Darré et al., 2004). Elements taken into account for decision-making stem from highly diverse fields – socio-economic, ecological and psychological (Cerf, 1996) – and decision-making processes are themselves affected by elements of these diverse fields. In addition, approaches used to study the practices are multidisciplinary: agronomy, mathematics, management and socio-anthropology. In this context and

Conclusion

In this study we illustrated how cognitive mapping approaches could be used for analysing farmers’ systems of practices. The twofold nature of these approaches, qualitative and quantitative, allows the studied objects to be considered in terms of their whole complexity and a model to be built based on actors’ perceptions of social–ecological systems. Another key point of the analysis of systems of practices is the diversity among farms. The automation of our analysis approach takes into account

Acknowledgements

The Walloon Agricultural Research Centre provided financial support for this work through the MIMOSA project. We wish to thank the farmers for the time they devoted to interviews and Aude Bernes for her participation in interviewing farmers. We would also like to thank two anonymous reviewers for their helpful comments.

References (63)

  • T. Kamada et al.

    An algorithm for drawing general undirected graphs

    Information Processing Letters

    (1989)
  • B.A. Keating et al.

    An overview of APSIM, a model designed for farming systems simulation

    European Journal of Agronomy

    (2003)
  • K. Kok

    The potential of fuzzy cognitive maps for semi-quantitative scenario development, with an example from Brazil

    Global Environmental Change

    (2009)
  • B. Kosko

    Fuzzy cognitive maps

    International Journal of Man-Machine Studies

    (1986)
  • D.E. Koulouriotis et al.

    Development of dynamic cognitive networks as complex systems approximators: Validation in financial time series

    Applied Soft Computing

    (2005)
  • S. Madelrieux et al.

    Patterns of work organisation in livestock farms: the ATELAGE approach

    Livestock Science

    (2009)
  • G.A. Mendoza et al.

    Multi-criteria decision analysis in natural resource management: a critical review of methods and new modelling paradigms

    Forest Ecology and Management

    (2006)
  • A. Merot et al.

    Analysing farming practices to develop a numerical, operational model of farmers’ decision-making processes: an irrigated hay cropping system in France

    Agricultural Systems

    (2008)
  • U. Özesmi et al.

    Ecological models based on people's knowledge: a multi-step fuzzy cognitive mapping approach

    Ecological Modelling

    (2004)
  • E.I. Papageorgiou

    A new methodology for decisions in medical informatics using fuzzy cognitive maps based on fuzzy rule-extraction techniques

    Applied Soft Computing

    (2011)
  • E.I. Papageorgiou et al.

    Application of fuzzy cognitive maps for cotton yield management in precision farming

    Expert Systems with Applications

    (2009)
  • S. Pinch et al.

    Cognitive mapping of creative practice: a case study of three English design agencies

    Geoforum

    (2010)
  • T. Rajaram et al.

    Modeling of interactions among sustainability components of an agro-ecosystem using local knowledge through cognitive mapping and fuzzy inference system

    Expert Systems with Applications

    (2010)
  • C.D. Stylios et al.

    Fuzzy cognitive map architectures for medical decision support systems

    Applied Soft Computing

    (2008)
  • J. Tikkanen et al.

    Applying cognitive mapping approach to explore the objective-structure of forest owners in a Northern Finnish case area

    Forest Policy and Economics

    (2006)
  • J. Vayssieres et al.

    GAMEDE: a global activity model for evaluating the sustainability of dairy enterprises. Part II – Interactive simulation of various management strategies with diverse stakeholders

    Agricultural Systems

    (2009)
  • J. Vayssieres et al.

    Modelling farmers’ action: decision rules capture methodology and formalisation structure: a case of biomass flow operations in dairy farms of a tropical island

    Animal

    (2007)
  • J. Vayssieres et al.

    Integrated participatory modelling of actual farms to support policy making on sustainable intensification

    Agricultural Systems

    (2011)
  • B. Wolfslehner et al.

    Mapping indicator models: from intuitive problem structuring to quantified decision-making in sustainable forest management

    Ecological Indicators

    (2011)
  • F. Ackermann et al.

    Strategic options development and analysis

    Systems Approaches to Managing Change: A Practical Guide. The Open University in Association with Springer-Verlag

    (2010)
  • R. Axelrod

    Structure of Decision: The Cognitive Maps of Political Elites

    (1976)
  • Cited by (0)

    1

    Tel.: +32 61 231010; fax: +32 61 231028.

    2

    Tel.: +32 10 473723; fax: +32 10 472428.

    View full text