Open Access
Issue
Nat. Sci. Soc.
Volume 32, Number 2, Avril/Juin 2024
Page(s) 216 - 225
Section Regards – Focus
DOI https://doi.org/10.1051/nss/2024048
Published online 28 November 2024

© J. Lombard Latune et al., Hosted by EDP Sciences, 2024

Licence Creative CommonsThis is an Open Access article distributed under the terms of the Creative Commons Attribution License CC-BY (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, except for commercial purposes, provided the original work is properly cited.

Introduction

Exploring the relationships between humans and nature: a need for new research tools

From the perspective of an ecologist examining an ecosystem, human activities are generally considered as an exogenous interfering force that do not align neatly with ecological theory (Collins et al., 2000). Historically, ecologists have investigated the biophysical, ecological, and evolutionary processes occurring in ecosystems unaffected by human influences, attributing any ecological and evolutionary changes to natural variations in energy and material flows or to natural selection (Alberti et al., 2009). However, in response to the growing global environmental crises, an increasing number of ecologists call for the inclusion of humans in the study of ecosystems by integrating social sciences into their own research field. For several years, they have asserted that effective conservation policy and management require both a knowledge of ecosystems and an understanding of human societies that interact with and depend on these ecosystems (Hooper et al., 2005). Consequently, the concept of social-ecological systems (SES) was introduced, generating a growing body of theoretical and empirical work relating to the ongoing changes and uncertainty in SES (Bretagnolle et al., 2019; Folke et al., 2010; Kates et al., 2001). This research combines transdisciplinary1, multiscale and multitemporal approaches, integrating physical, biological, and social sciences, and incorporating institutional and governance analyses. Such ambitious research calls for the development of new concepts and tools (Kaneshiro et al., 2005; Plummer and Fitzgibbon, 2004).

How serious games contribute to research on social-ecological systems so far

A growing body of literature on serious games (SG), that is, games used for non-entertainment purposes, has underscored the benefits of these tools for studying complex SES (Reckien and Eisenack, 2013; Zvoleff and An, 2014). Due to the multitude of names, the term SG encompasses various forms such as ‘serious games’, ‘role-playing games’, or ‘participatory simulations’ that are ‘experi(m)ent(i)al, rule-based, interactive environments, where players learn by exchanging information, by taking actions and by experiencing their effects through feedback mechanisms that are deliberately built into and around the game’ (Mayer, 2009).

SG use a variety of structures (e.g., with or without a digital interface) and have varying goals (Flood et al., 2018) depending on the target audience. For the general public and scholars, SG are valuable in providing interactive tools to educate the players and promote their behavioural change (Tsai et al., 2019; Wu and Lee, 2015). For environmental managers, SG offer innovative solutions for policy capacity building (Flood et al., 2018), particularly when coupled with computerized interfaces (Ruankaew et al., 2010). Finally, researchers interested in environmental governance have stressed the value of SG in fostering social learning, conflict resolution, collaboration, and dialogue (Den Haan and Van der Voort, 2018; Flood et al., 2018; Muro and Jeffrey, 2008). Given their potential, SG have been used to gain social or political insights into different SES, mostly in the management of agriculture, water systems, or climate change (Edwards et al., 2019).

As research tools that bridge various disciplines and foster exchanges between researchers and practitioners, SG are instrumental in understanding both the social and biophysical dynamics of SES. We, the authors of this article, have elaborated and/or used SG in diverse contexts, including skills training, teaching for students in agronomy and ecology, as well as implementation in transdisciplinary projects. Our exploratory study began with the intuition, based on our experience of SG as ecologists, social scientist, geographer and agronomist, that SG could provide unexpected benefits to researchers, and particularly by enabling ecologists to formulate new research questions on SES.

We first propose three key properties for research tools or concepts that can help raise new research questions: identifying knowledge gaps, recognizing mismatches between theoretical expectations and observations (‘anomalies’), and uncovering neglected interactions that can alter researchers’ representation of the systems under study. We then draw on a case study to investigate whether SG have these properties. Finally, we summarize our findings and conclude how SG may be used to generate new research questions on SES.

Identifying three key properties to renew research questions on social-ecological systems

Drawing on research in epistemology (see references below), we identified three key properties of research tools and concepts that may help researchers open new research fronts and raise new research questions.

Key Property 1: SG may allow researchers to detect knowledge gaps about the SES under investigation.

The main objective of science is to advance knowledge production, which is guided by the identification of knowledge gaps, typically revealed through establishing a state of the art. Interdisciplinary or even transdisciplinary approaches can renew the process of establishing states of the art. The confrontation of various types of knowledge may outline unsolved research questions (Girard, 2013; Zvoleff and An, 2014). SG, developed by game designers from multiple disciplines and involving stakeholders with empirical knowledge, are at the forefront of identifying these knowledge gaps (Agogué et al., 2015). SG may also lead participants to explore innovative and collective solutions, further identifying knowledge gaps. Such a relationship between exploration of ideas and search for missing knowledge is explained by Hatchuel and Weil (2008) through design theory.

Key Property 2: SG may enable researchers to detect anomalies and thus discuss or enrich existing theories.

Scientific advancement occurs through the continuous discussion of theories (Cariou, 2019): theories are enriched, refined, or even contested when confronted with empirical data. As Hatchuel et al., (2018) points out: ‘the unknown lies in the anomalies detected between a state of the art and a state of the facts’. Detecting anomalies leads to a revision of theories and a fortiori knowledge. SG provide a unique setting where actors in the same SES, who do not necessarily interact in real life, are invited to explore solutions together. Unlike theoretical predictions, this interaction process may allow researchers to detect anomalies and subsequently revise existing theories.

Key Property 3: SG may cause researchers to change their representations of the SES under study and take previously neglected interactions into account.

Scientific knowledge builds on the representations of reality through models, concepts, and formalisms. Identifying new relations or dimensions may allow researchers to renew their representations of reality and thus identify new research questions (Toffolini et al., 2020). SG are specifically designed to unveil poorly known interaction processes between humans and between humans and nature. They may lead ecologists to broaden their research focus from ecosystems to SES, revealing interaction patterns of greater complexity that encompass ecological, social, economic and political processes (e.g., Rakotonarivo et al., 2021, Lardon and Piveteau, 2005). Interestingly, renewed representations of SES may result from both the design phase and the implementation phase of SG.

Methodology

Selecting a set of serious games for a comparative case study

To explore the potential benefits of SG for ecological research, we analysed the design and implementation of six SG. Three of these SG were developed with contributions from the authors of this paper, while the other three were selected based on a preliminary survey regarding the use of SG by French ecologists. All six SG address the issues of biodiversity, natural resource management, and/or climate change (see Tab. 1).

Inspired by the companion modelling approach (ComMod) (Étienne, 2014), they were all designed in the context of French public participatory research, with partial or complete co-construction with local stakeholders. Indeed, the ComMod approach relies on the involvement of stakeholders to define and develop a model of the SES of interest. This conceptual model is then translated in a SG in accordance with the practitioners’ community ethical guidelines (Barreteau, 2003). When relevant, the process begins with the development of a SG, which is later converted into a computerized model. Here, the case studies cover a range of contrasting SES: Foster Forest aims to better understand the implementation of climate change adaptation in French forestry; BiOffset investigates the emergence of a governance system for offsetting biodiversity loss due to land development; BotNidVeau assesses the potential of new collective agri-environmental schemes for protecting a wetland region; AdaptaMeije examines how local actors may enhance the resilience of mountain grassland landscapes to climate change; Secoloz focuses on the integrated management of ecosystem services (biodiversity, cattle breeding, water availability, tourism, etc.) in an agropastoral landscape; and CapBiomasse explores the exchanges and cooperation surrounding forest biomass mobilization for the local energy transition. For each game, we sought to identify whether it enabled researchers to detect knowledge gaps, anomalies, and/or neglected interactions or processes.

Tab. 1

Presentation of the six case studies.

Data collection and analysis

We conducted eight semi-structured interviews (Newcomer et al., 2015) with SG designers2. The designers were informed about the research context of the study, and gave their consent to participate by completing a form in accordance with the General Data Protection Regulation (EU 2016/679). The interviews3 addressed the reasons for implementing the game, how the researchers obtained the knowledge needed to build the game, the discussions that occurred during (simulation) and after (debriefing) the game, and what researchers could draw from the game for their own research. We based our interviews on the classical steps described to implement a companion modelling approach, focusing on: (1) methodological approaches (reading literature, conducting interviews, testing); and (2) choices concerning the creation and implementation of the SG.

We transcribed the audio-recorded interviews and coded them according to the three key properties mentioned above. We triangulated our results using various sources of data: interviews with different researchers for two SG and documents (articles, reports, PhD theses). Additionnally, at least two researchers analysed the material for each case study.

Results: analysing the interest of serious games to foster a SES approach

Serious games may help researchers identify knowledge gaps

In all six cases, we found elements confirming the value of SG in identifying knowledge gaps. This was particularly salient for Foster Forest, CapBiomasse, and Secoloz. For instance, discussions during the Foster Forest sessions revealed a lack of knowledge about the behaviour of tree species under drought conditions. These discussions also highlighted scientific controversies, such as whether carbon storage is higher in old or young trees, whether mixed forest are more resilient than even-aged forests during extreme climatic events, and how these factors can be quantified. Furthermore, new research paths emerged for which knowledge is missing, such as using carbon storage as a source of income.

After several playing sessions of CapBiomasse, the game designers realized the importance of elaborating new indicators to assess the economic impact of the energy transition on the territories in question. They also pointed out the need for new biodiversity indicators to assess the impact of agricultural biomass-based energy transition based on soil biodiversity.

The Secoloz game highlighted knowledge gaps in the ecological impact of management actions (e.g., land clearing or pebble removal) on agropastoral scrublands, especially in terms of thresholds and tipping points on lichens or soil erosion, for instance. These knowledge gaps make it difficult to formulate definitions shared by heterogeneous stakeholders, as seen with the term ‘natural open grasslands’.

Serious games highlight surprises rather than anomalies

Our analysis indicates that SG reveal ‘surprises’ rather than anomalies, leading to discussions of ecological theories. For instance, the AdaptaMeije game revealed that despite the close proximity of mountain landscapes, the local inhabitants were quite disconnected from their environment. Furthermore, some stakeholders in the ski industry viewed climate change as an opportunity rather than a threat, especially those in higher-altitude mountains with a competitive advantage over lower-altitude locations. Lastly, while scientists expected that enhancing solidarity, local development, and autonomy would increase the resilience of the area, the game sessions revealed that a strategy based on a strong tourism industry, which would increase incomes, might better enable the inhabitants to attain their goals.

In the Secoloz game, players unexpectedly prioritized water quality over biodiversity issues as the mayor had no means of coercion to preserve water quality, leading to spontaneous self-organization among players.

Serious games may change researchers’ representations of the system under study

Bioffset revealed unexpected strategies among the members of environmental protection groups. These actors actively engaged in biodiversity offsetting systems contrary to the game designers’ expectations of them remaining as ‘environmental fire keepers’ to keep all their power of alert. This raises questions about the consequences of including environmental protection groups as providers of offset measures on biodiversity conservation.

In AdaptaMeije, the game sessions enriched the understanding of collective action and decision-making processes. A few stakeholders, expected to play an active role in the collective dynamics, were instead withdrawn. The sessions also unveiled strong power games with an important impact on the economy and innovation dynamics. Main barriers to resilience were more related to sociological and political aspects than to ecology.

During BotNidVeau sessions, the farmers changed their perception of the birds featured in the game. The appropriation took place during the game and led to the emergence of better crop exploitation as well as a conservative bird management through new rules.

Regarding Foster Forest, participatory game construction playing sessions revealed two overlooked issues considered to be crucial by forest managers: (1) hunt game dynamics, which impacts tree growth and the direct income of forest owners; and (2) the reluctance towards clearcutting due to its impact on landscape value and the strong reactions it triggers from local inhabitants.

Lessons and perspectives about the use of serious games for research on social-ecological systems

Our comparative analysis suggests that SG are relevant for helping researchers identify knowledge gaps and change their representations of SES. Regarding the identification of knowledge gaps, the findings highlight that researchers involved in SG necessarily change their position (Hazard et al., 2020) from external observers to part of the system under study, raising new questions about their own role and impact. To respond to non-academic questions and expectations, researchers need to take a new perspective on their research, which can reveal knowledge limitations. In addition, SG, as tools generating design processes, lead to the exploration of new concepts or ideas, which may call for innovative research (Hatchuel and Weil, 2008; Vourc’h et al., 2018).

SG may trigger changes in representations among researchers for two reasons. First, taking into account the social component of SES, SG highlight the importance of considering a wider range of variables as opposed to a strict use of variables coming from a single discipline as used in traditional ecological approaches for instance. In so doing, they lead to a change in representations not only in the social models but also in the ecological models. Furthermore, SG are interesting tools to collect original data from unprecedented interactions between actors. Second, SG specifically aim to identify collective solutions for the management of ecosystems. As a result, they often call for new representations of these ecosystems: when designing a SG, researchers generally require a more systemic understanding of the SES under study, thoroughly considering the interactions between people and nature (Rakotonarivo et al., 2021). SG can lead researchers to change their focus and consider processes that were formerly disregarded. They can also be used to explore and test innovative paths for further investigation with standard research methodologies (Lardon and Piveteau, 2005).

Finally, while SG reveal surprises that can be useful for researchers, they are less effective at detecting anomalies that could enrich ecological theories. To the best of our knowledge, this is because SG have not to date been specifically used to question ecological theories. Indeed, the underlying ecological models are often oversimplified to focus on social processes. However, SG, especially those co-designed by scientists and other actors, could be used from this perspective, aligning with Toffolini et al. (2020), who show that participatory design approaches can renew agronomy models.

In conclusion, although this analysis is exploratory and would require further investigation, it suggests that SG have the potential to generate original research questions that would consider both people and nature in social-ecological systems. To enhance this potential, we suggest five courses of action.

  1. Begin with a collective clarification of researchers’ ‘hidden assumptions’ about the functioning of the SES and their expectations from the SG. This initial knowledge building is greatly valuable for SG design and analysis, especially if there is a non-academic commissioner.

  2. Involve heterogeneous stakeholders in the SG design team, in particular for developing the conceptual model and simulation tool. This may help revealing unexpected knowledge gaps.

  3. Advocate for long term SG-based research projects to provide feedback opportunities to the participants and develop reflexivity on the entire process — sometimes even on a loop ending up on a lightened version of the SG. This can facilitate collective discussions on changes in SES representations.

  4. Encourage ecologists to participate in the SG workshops, and remain open to surprises during the game/simulation to gain new research insights.

  5. Implement a monitoring strategy (during the design and implementation of the game) to identify moments when participants identify knowledge gaps, neglected interactions, or when researchers change their representations of the system under study.

Acknowledgments

This research was funded by a grant overseen by the French National Research Agency (ANR) as part of the ‘Investissements d’Avenir’ Programme (LabEx BASC; ANR-11-LABX-0034), in the framework of the INDISS project. The authors are grateful to Lorène Prost for her pertinent feedback on the paper. This research benefited from interactions with members of IDEAS (Initiative for DEsign in Agri-food Systems at INRAE).

Authors’ contributions

NFL, EB, and JLL conceived the ideas and designed methodology; JLL and EB collected the data; TF and VS conducted the literature review; JLL, EB, TF, VS, and NFL analysed the data; EB and JLL led the writing of the manuscript; JLL, TF, and VS were interviewed on the SG they contributed to design. All authors wrote parts of the article, contributed critically to the drafts and gave final approval for publication.

References


1

Interdisciplinary approaches call on a wide range of disciplines, whereas transdisciplinary research additionally includes knowledge from non-academic sources (see Nowotny, 2003).

2

See Annex A for the characteristics of the respondents.

3

See Annex B Interview guides for SG analysis case studies.

Cite this article as: Lombard Latune J., Berthet E.T., Fouqueray T., Souchère V., Frascaria-Lacoste N. 2024. Analysing the potential of serious games to raise new research questions on social-ecological systems. Nat. Sci. Soc. 32, 2, 216-225.

Annex A Characteristics of the interviewees

Serious game Interviewee Discipline Professional position (at the time of the interview) Research institution
Secoloz A Geography Post-doctoral fellow IRD
  B Geography Senior researcher INRAE
CapBiomasse C Sociology Lecturer IMT Mines Alès
AdaptaMeije D Ecology PhD student CNRS
  E Interdisciplinary land planning Post-doctoral fellow ETH-Zurich
Foster Forest F Ecology Post-doctoral fellow UQO
(QC, Canada)        
BiOffset G Environmental geography Post-doctoral fellow AgroParisTech
BotNidVeau H Agronomy Research engineer INRAE

Annex B Interview guide for SG case study analysis

Introduction of the interview

Reminder of the research project context: INDISS (Innovation and Design In Sociotechnical Systems) is a research project that organises exchanges and collaborations between design specialists (design agronomists, researchers in design sciences) and researchers focused on knowledge production (ecologists, agronomists, geneticists, ecophysiologists, sociologists). These collaborations are articulated around thirty ‘innovation projects’, which concern four main areas: ideotyping and crop variety selection; agroecological cropping system design; water quality collective management; and biodiversity collective management. In this last field, we are interested in the potential interest of role-playing games for research in ecology.

We invite you to participate in this research project because of your involvement in the design, development and/or implementation of role-playing games for your research on socio-ecosystem management.

The aim of our enquiry is to find out to what extent serious game devices dealing with natural resource management can be used to formulate new questions or directions for research in ecology. In order to identify this phenomenon, we propose to retrace with you the entire game creation process. The aim is to identify whether (i) knowledge gaps, (ii) mismatches between theoretical expectations and observations, and (iii) neglected interactions emerged through this process.

Overall considerations

  • Why did you decide to build a serious game? What was the context (research project…)?

  • What were the objectives of the game (awareness raising, empowerment, generating innovative functioning, producing data, decision support…)?

  • Which research consortium developed the game? (discipline of origin of the people involved in the construction and animation of the game).

Phases of game creation Sub-phases Questions
Knowledge inventory Field investigation Who did you interview? Are there actors who were met when this was not initially planned? Who are these actors (membership)? What kind of knowledge did they bring? i.e. Why were these additional actors met? Did you learn anything new during this stage?
‘Grey’ literature review What disciplines does the ‘grey’ literature review relate to?
Did you learn anything new during this stage?
Scientific literature review What disciplines does the scientific literature review relate to?
Did you learn anything new during this stage?
Game elaboration Conceptual model Please specify:
– whether a formalism was explicitly used, or whether it was a methodological tinkering with pre-existing data/models?
– what type of formalism was used: ARDI, PARDI1, UML2?
What types of interactions did you seek to represent, and how? What in particular did you want to capture? Were you surprised by any unknown or new interactions to be included in the model?
How was the conceptual model developed: in the lab (autonomously by the modeller or the consortium of researchers), during participatory workshops (co-construction of the conceptual model), alternating lab and participatory processes?
For each option mentioned above (i,ii,iii), please try to identify at what moments and to what extent one or more of the hypotheses (see heading) have taken place and/or if other phenomena in the field of ‘innovative design’ took place, particularly in relation to questions of scientific ecology.
Focusing the model on a few aspects, choice of question, refocusing on specific interactions Please explain:
– how the choice of the issue to be addressed was made, from the initial broad representation.
– how the model was simplified (choice of such actors and or such interactions and or such dynamics).
Same as above: during this ‘refocusing’, please try to see at what moments and to what extent one or more hypotheses (see introduction [i,ii,iii]) took place.
Choice of the tool Did you use agent based model, or board game with or without excel file allowing to follow some economic and/or ecological dynamics?
What were the questions raised when choosing the tool? (Do some of our hypotheses come true during this choice? Does the complexity of the representation enabled you to advance on a particular theme, and if so which one?)
What game mechanics were envisaged and why? What were the discussions around this?
Type of representation What was the kind of representation of reality: fictive, stylised, realistic, etc.?
Did the choice of representation raise questions during which our hypotheses could be verified?
What was the emphasis? The representation of biological phenomena? The interactions between nature and humans?
What discussions were linked to this?
Calibration What were the issues at stake in the calibration? How was the parameterisation carried out (literature review, additional interviews…)? What were the most complex aspects to parameterise and why? Do our hypotheses hold true during this stage?
Game simulation sessions Game introduction What information was provided? What questions were asked by the participants?
Design of the entertainment of the game simulation What game facilitation device was chosen and implemented during the game?
What were the issues, discussions and reflections at the origin of the animation device? Identify the debates and challenges, what was debated and why?
Collective highlights What collective highlights took place during the game, and as a result of what (e.g., unsatisfactory indicators, agreements between players)? On whose initiative? (one player, all players, the facilitator?)
Identify what debates took place at that moment and what was the content of these debates? Did they bring out controversies? Do our hypotheses (i,ii,iii) hold true?
Game session debriefing Feedback of the game What were the negotiations, the levers of action, the trade-offs?
Identify what stand out at this stage, what doubts were raised by the participants, what debates? Do our assumptions hold true?
Link to reality: how would you act in reality? Would you implement things tested in the game? What are the differences between how you act in a fictional situation in the game and what you would do in reality?
Why are there differences (or not)? Do our assumptions hold true?
Shortcomings and improvement of the game Did the participants mention proposals for improving the game? Such as new functions, interactions, land use, etc. If so, please specify.
Identify then what debates these proposals provoked? Did it highlight knowledge gaps?
Do any of our hypotheses (i,ii,iii) hold true in this discussion?
Game session impacts Analysis of the results from the game sessions by researchers Did the game analysis reveal new things? To what extent are the results new? Do our hypotheses hold true at this stage? Have you launched any new research programme on an ecological issue that emerged during the game?

1Actors, Ressources, Dynamics, Interactions – Problem, Actors, Ressources, Dynamics, Interactions (Étienne et al., 2014).

2United Modeling Language (Étienne et al., 2014).

All Tables

Tab. 1

Presentation of the six case studies.

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.