Cultivating Sustainability: Quebec’s Living Labs as Ecological Catalysts

Abstract

Agriculture is often considered a major factor in environmental degradation. This case study delves into the use of sociotechnical experiments—and, more specifically, agroecosystem living labs (ALLs)—to facilitate the transition of conventional agricultural practices toward heightened sustainability. Our research indicates that achieving successful collaboration, such as an experiment, necessitates the alignment of expectations, the establishment of trust, the cultivation of patience, and the allocation of substantial resources. This investigation into agroecosystem living labs contributes to our comprehension of the actors’ networks, their interactions with experimental sites, and the dynamics of open innovation.

1. Introduction

Our environment is currently experiencing ecological disruptions that are putting biodiversity and ecosystems at risk. In 2015, the United Nations responded to these concerns by adopting the 2030 Agenda for Sustainable Development, which includes 17 Sustainable Development Goals (SDGs) [1], which all directly or indirectly aim to support sustainable food production and consumption, thus making agri-food central to the goals’ attainment [2]. Many authors affirmed that standard agriculture significantly contributes to environmental degradation [1,3,4]. Current food production methods pose major challenges, including intensive land and water use, erosion, resource scarcity, ocean acidification, increasing greenhouse gas emissions, climate change, the disruption of the nitrogen and phosphorus cycles, and biodiversity loss [2,3,4,5,6,7,8].

Agriculture plays a key role in the quest for environmentally, economically, and socially sustainable solutions [9,10]. Nevertheless, meaningful changes to our agricultural production methods are required to meet current challenges. Technological innovation is critical to promoting alternative forms of agriculture and contributing to sustainable development, resilience, and food security [11,12]. Even so, while these advances are essential, they will not solve environmental sustainability issues. To effectively meet the challenges, the focus on product and process improvement must shift to a broader approach to bring about structural changes to the organization of production and consumption systems [11,12].

It, therefore, appears necessary to adopt disruptive innovations, which represent new configurations of actors, institutions, and practices, to completely transform production chains and systems [12]. These innovations require a major shift in dominant sociotechnical regimes through technological evolution, changes in user practices, regulatory adaptations, transformations within industrial networks and infrastructure, and adaptations to the symbolic meaning of culture [13].

Agroecosystem living labs (ALLs) stand out as the ideal framework to realize meaningful change in the agricultural sector. ALLs have the potential to broaden and accelerate the adoption of current good practices and foster the implementation of new innovative practices and techniques in agriculture [14]. ALLs are a transdisciplinary approach involving various stakeholders, such as farmers, researchers, and other actors who wish to contribute to the design, monitoring, and evaluation of new agricultural technologies and practices in working landscapes to enhance their effectiveness and early adoption [14]. They represent a type of open innovation focused on technology, practices, and knowledge that emphasizes the sustainability and resilience of agricultural and agri-food systems [15].

This study extends the scholarly discourse through an exploration of the Living Lab–Québec (LLQ). It elucidates previously unexplored dimensions of ALLs by examining the consequential value of such experimental platforms in facilitating the sustainable utilization of natural resources. This research inquires the following questions:

How is the actors’ network of an ALL orchestrated, and what are the anticipations, insights acquired, and hurdles encountered during the implementation of such experiments?

Our study pinpoints specific critical aspects to consider when creating an ALL, making it a valuable resource for practitioners, actors, agricultural enterprises, managers, and facilitators such as governments, who can use ALLs as reference models to manage and develop new ALLs.

The paper is structured as follows: the next sections explore the study’s conceptual foundations, detail the research methodology, and present the empirical results. In conclusion, future ALL research directions are suggested.

2. Literature Review

Sociotechnical experiments refer to initiatives conducted, often, in real-world settings and that aim to explore, test, or implement innovations in both social and technical aspects of a system simultaneously. These experiments typically involve interdisciplinary collaborations between social scientists, engineers, policymakers, and other stakeholders to address complex societal challenges by integrating technological innovations with social and organizational practices. The goal of sociotechnical experiments is to understand how technological advancements interact with human behavior, organizational structures, cultural norms, and institutional frameworks, thereby shaping the outcomes and impacts of the intervention [11].

The actors’ network structure and their expectations and learning ability and processes constitute the pillars of sociotechnical experiments. They represent the three internal processes of strategic niche management (SNM), which is an approach mobilized by Transition Management aiming to proactively manage the development of innovation niches to facilitate the emergence, development, and diffusion of innovations. Aligning the actors’ expectations, guiding learning processes, and legitimizing stakeholder protection and education are crucial activities for successful experiments [16]. Articulating the stakeholders’ expectations is essential for attracting attention, resources, and new actors to innovation projects [17,18,19]. Learning in sociotechnical experiments focuses on technical specifications and underlying assumptions [16,17,18,19,20,21]. The first type of learning corresponds to first-order learning, and the second is second-order learning on social practices [18]. Learning, whether of the first or second order, allows for challenging the dominant regime and promoting large-scale innovation [18]. User involvement, idea communication, and actor interactions are essential for learning [22,23]. Large and diversified actor networks promote second-order learning by determining the depth and breadth of learning processes [16]. Learning is, thus, linked to actor networks composed of various stakeholders, such as businesses, researchers, governments, users, and decision-makers, who collaborate, share knowledge, and support innovation [11]. However, to achieve success, these three internal processes must be complemented by considering the external context of such experiments [16,18,19,24,25].

Sociotechnical experiments are critical for fostering innovation. They articulate and accept the potential of innovation while learning about local changes in culture, practices, and structures [26]. They also articulate the stakeholders’ expectations, learn and refine innovations in real environments, and identify obstacles to their implementation [11]. Many innovations fail when this step is neglected or when the focus is solely on technological aspects [21]. Sociotechnical experiments promote innovation on multiple fronts and remain open to adjustments and improvements [11]. They help identify obstacles to implementation and the diffusion of innovation, thus increasing its acceptance and legitimacy [11,19,24]. User involvement is decisive in this process [24]. Experiments serve as laboratories, windows, and agents of change, facilitating the transition to sustainable innovations [11,19,21]. They are conducted with groundbreaking innovations, involve a broad actor network, and are protected within technological niches [11]. Various types of sociotechnical experiments exist, including LLs [27,28,29], distinguished by their realism and direct user involvement in the innovation process. In other words, they integrate real users into authentic contexts [28,30,31,32].

Living Labs (LLs)

LLs stand out as privileged spaces for sociotechnical experimentation [27]. In these spaces, collaborations among public, private, and citizen stakeholders are established to develop and evaluate new technologies, services, and products in real-world contexts [32,33]. These laboratories emphasize a user-centered approach [27,34] and are characterized by their ability to address complex issues in diverse environments [33,35,36]. LLs are often defined by their distinctive attributes. Thus, the state of the art on LLs reveals several of these characteristics. Table 1 provides a concise overview of the various characteristics of LLs.

The literature review on LLs highlights their exploration of various fields, demonstrating their transdisciplinary nature and adaptability [29,30,33,35,37,38,39,40,41,42]. Among the application areas, we find health and well-being, education, environment and sustainability, public services, urban planning, energy, mobility, tourism, industry and manufacturing, Information and Communication Technologies (ICT), food, as well as social and community contexts. This diversity illustrates LLs’ ability to tackle a wide range of societal and technological challenges through their open and collaborative approach. However, despite this broad coverage, the fields of agriculture and, more specifically, agroecosystems remain relatively underexplored [15].

Studies on the operational mechanism of ALLs are still uncommon. It is from this perspective that our study specifically focuses on ALLs. These represent a domain that is both new and distinct, crucial for sustainable development, and distinguished by its challenges and uncertainties, as well as by its immersion in an open environment, where the close connection with the territory and natural resources, such as lakes, brings a dimension that cannot be replicated in a conventional framework. They also involve experiments with their own innovation cycle, thus highlighting the complexity and richness of these learning environments. It is in this context that our research aims to decipher the functioning mechanisms of this particular type of LL, the ALL, by examining its learning processes, the transferability of its practices, its challenges, and its key actors.

Table 1. The characteristics of LLs.

Characteristics	Previous Research
User-Centered Approach	LLs are powerful tools that involve users at every stage of innovation, thereby enhancing success rates through a user-centered approach [27,31,34,43]. In this approach, users are not just sources of information but also testers, developers, and contributors, distinguishing LLs from other innovation methods [37,44]. LLs employ two key mechanisms related to user involvement, namely ideation and evaluation, to capture new ideas and mitigate business risks [38,45,46,47,48].
Actor network	LLs gather various actors, such as governments, academics, developers, researchers, industry, citizens, users, and public and private organizations to collaborate in seeking innovative solutions to a problem [29,45,49,50,51,52,53]. These actors contribute to creating new services, fostering mutual understanding, and sharing knowledge, all of which are essential for addressing local issues [31,54,55,56].
Real context and activities	LLs involve key activities, such as co-creation, exploration, experimentation, and evaluation [29,36,50,53,57,58,59,60,61,62,63]. Co-creation entails a joint design by users and producers, while exploration discovers emerging uses and market opportunities. Experimentation implements real-world scenarios with user communities, and evaluation assesses concepts, products, and services in real environments. LLs are distinguished by their use of real environments to test new artifacts, thereby increasing their value and adoption upon market launch [36,39,40,64]. Real environments encompass both physical and virtual locations [29,44,61].
Challenges	LLs face challenges, which include managing temporality and actor stability, governance amidst diverse skills and interests, unforeseen results, efficiency dependent on knowledge sharing, user engagement, cognitive and motivational barriers, ethical issues related to the value created by users, as well as the financial sustainability and scalability of innovations [31,34,35,38,42,47,53,55,58,59,61,65,66].
Results of innovation	The outcomes of LLs can range from incremental to radical, resulting in tangible (designs, products) and intangible (services, ideas, knowledge) outcomes [33,44,54,55,58,67,68,69,70]. They can lead to systemic, social, and technological innovations [57,68,71]. The LL process can generate economic (tangible), business (including health and well-being), and utility value (user experience with a product or service) [65,72].
Sustainability	Sustainable innovation and LLs are closely linked [55,60,69]. LLs focus on sustainable products and services, emphasizing sustainability through continuous learning and development [73]. In an LL, sustainability refers to its viability and responsibility to the community in which it operates [65]. To address sustainability, LLs must take responsibility for their ecological, social, and economic impacts, addressing sustainability issues in their innovation processes [35,43,65,72].
Methods and tools	LLs adopt a multi-method approach to cover the innovation process at three levels: technological, social, and economic, using a variety of methods and tools from diverse disciplines such as ethnography, psychology, sociology, strategic management, and engineering [30,39,40,47,67]. Methods and tools in LLs vary widely across laboratories, with a “harmonization cube” proposed to facilitate exchanges and widespread adoption of methodologies [36,41,67,74]. FormIT is also a tool used in LLs focusing on user involvement through an iterative design cycle [47].
Approaches	Besides the user-centered approach, LLs mobilize several other approaches. They are at the core of open innovation, facilitating knowledge exchange to drive innovation [28,37,75,76,77]. Within LLs, openness and co-creation with broad participation [28,35,43,65,72] are crucial, considering users as co-creators [57]. In terms of participation, LLs can be open or closed [35,69]. Coordination in LLs follows both bottom-up and top-down principles, incorporating local needs and external initiatives [34,56,78,79]. Cooperation and stakeholder participation are crucial for the success of LLs [28,31,61,74].
Communication, trust and motivation	Communication [35], trust [31], and motivation [32,34,42] are key elements of LL approaches. Social proximity promotes trust and facilitates the exchange of tacit knowledge [80]. Understanding and motivating actors to participate in LL activities, considering their individual and mutual motivations, is essential [34,42,65]. Intrinsic and extrinsic rewards are necessary to encourage actors’ contributions to the project [42,81,82,83].

Table 1. The characteristics of LLs.

Characteristics	Previous Research
User-Centered Approach	LLs are powerful tools that involve users at every stage of innovation, thereby enhancing success rates through a user-centered approach [27,31,34,43]. In this approach, users are not just sources of information but also testers, developers, and contributors, distinguishing LLs from other innovation methods [37,44]. LLs employ two key mechanisms related to user involvement, namely ideation and evaluation, to capture new ideas and mitigate business risks [38,45,46,47,48].
Actor network	LLs gather various actors, such as governments, academics, developers, researchers, industry, citizens, users, and public and private organizations to collaborate in seeking innovative solutions to a problem [29,45,49,50,51,52,53]. These actors contribute to creating new services, fostering mutual understanding, and sharing knowledge, all of which are essential for addressing local issues [31,54,55,56].
Real context and activities	LLs involve key activities, such as co-creation, exploration, experimentation, and evaluation [29,36,50,53,57,58,59,60,61,62,63]. Co-creation entails a joint design by users and producers, while exploration discovers emerging uses and market opportunities. Experimentation implements real-world scenarios with user communities, and evaluation assesses concepts, products, and services in real environments. LLs are distinguished by their use of real environments to test new artifacts, thereby increasing their value and adoption upon market launch [36,39,40,64]. Real environments encompass both physical and virtual locations [29,44,61].
Challenges	LLs face challenges, which include managing temporality and actor stability, governance amidst diverse skills and interests, unforeseen results, efficiency dependent on knowledge sharing, user engagement, cognitive and motivational barriers, ethical issues related to the value created by users, as well as the financial sustainability and scalability of innovations [31,34,35,38,42,47,53,55,58,59,61,65,66].
Results of innovation	The outcomes of LLs can range from incremental to radical, resulting in tangible (designs, products) and intangible (services, ideas, knowledge) outcomes [33,44,54,55,58,67,68,69,70]. They can lead to systemic, social, and technological innovations [57,68,71]. The LL process can generate economic (tangible), business (including health and well-being), and utility value (user experience with a product or service) [65,72].
Sustainability	Sustainable innovation and LLs are closely linked [55,60,69]. LLs focus on sustainable products and services, emphasizing sustainability through continuous learning and development [73]. In an LL, sustainability refers to its viability and responsibility to the community in which it operates [65]. To address sustainability, LLs must take responsibility for their ecological, social, and economic impacts, addressing sustainability issues in their innovation processes [35,43,65,72].
Methods and tools	LLs adopt a multi-method approach to cover the innovation process at three levels: technological, social, and economic, using a variety of methods and tools from diverse disciplines such as ethnography, psychology, sociology, strategic management, and engineering [30,39,40,47,67]. Methods and tools in LLs vary widely across laboratories, with a “harmonization cube” proposed to facilitate exchanges and widespread adoption of methodologies [36,41,67,74]. FormIT is also a tool used in LLs focusing on user involvement through an iterative design cycle [47].
Approaches	Besides the user-centered approach, LLs mobilize several other approaches. They are at the core of open innovation, facilitating knowledge exchange to drive innovation [28,37,75,76,77]. Within LLs, openness and co-creation with broad participation [28,35,43,65,72] are crucial, considering users as co-creators [57]. In terms of participation, LLs can be open or closed [35,69]. Coordination in LLs follows both bottom-up and top-down principles, incorporating local needs and external initiatives [34,56,78,79]. Cooperation and stakeholder participation are crucial for the success of LLs [28,31,61,74].
Communication, trust and motivation	Communication [35], trust [31], and motivation [32,34,42] are key elements of LL approaches. Social proximity promotes trust and facilitates the exchange of tacit knowledge [80]. Understanding and motivating actors to participate in LL activities, considering their individual and mutual motivations, is essential [34,42,65]. Intrinsic and extrinsic rewards are necessary to encourage actors’ contributions to the project [42,81,82,83].

3. Methodology

This study builds upon a qualitative research methodology centered around a specific case study. The adoption of a qualitative methodological approach for data collection and analysis is justified by its relevance in studying phenomena that are still relatively unexplored [84]. LLs have been around for about two decades and have been deployed in various contexts. However, the application of the LL concept within agroecosystems is a relatively recent innovation. Until 2017, this form of sociotechnical experimentation primarily focused on technological innovation, was rare in an agricultural context. Since then, Europe and North America, particularly Canada, have seen the emergence of LLs, including AcadieLab. This evolution marks a turning point in the approach to sociotechnical experimentation, aligning it more closely with the specificities and needs of the agricultural sector.

The use of case study as an investigative method is motivated by its potential to provide a deep understanding of the subject under study by examining its functioning and interactions within its real contextual environment [85]. This approach allows specific events to be linked to theoretical knowledge, thus enriching practical understanding [86]. For the selection of the case study, three main criteria were established: (i) firstly, the case must illustrate an open innovation initiative adopting the LL approach, engaging diverse actors in innovation activities, and placing users at the center of the development process; (ii) secondly, the case must involve an agroecosystem; and (iii) thirdly, the case must be situated within the Canadian context. These criteria were defined based on the research objectives and a prior review of the literature. Based on this, the Living Lab–Québec (LLQ), located in the Lake Saint-Pierre watershed and part of Agriculture and Agri-Food Canada’s (AAFC) Living Laboratories initiative, was chosen as the case study. This choice is detailed in Section 4.

3.1. Data Collection

The data were collected through semi-structured interviews conducted between April and May 2023 without a predefined sample definition. Our study is based on four interviews with LLQ participants. The interviews aimed to capture the functioning of the LLQ by exploring its organization, actors, and results and gathering data on the expectations, learnings, and challenges encountered.

To achieve this, semi-open questions were developed to explore relevant aspects of the LLQ and divided into five main themes. Each theme aimed to shed light on a specific aspect of the initiative, from its introduction and context to its outcomes and impact. The first theme focused on LLQ’s introduction and context, seeking to understand the roles, responsibilities of participants, and motivations behind LLQ implementation. The second theme examined LLQ’s internal organization, analyzing its activities, objectives, organizational structure, functioning, life cycle, and management challenges. The third theme explored the network of involved actors, their composition, respective roles, level of participation, and communication and decision-making mechanisms. The fourth theme aimed to analyze LLQ’s outcomes, evaluating the types of results produced, beneficiaries, transferability, and environmental, social, and economic impact. The final theme addressed learning and knowledge transmission, studying technical learning mechanisms, joint learning, and knowledge sharing. Final questions were asked about trust within LLQ, key points, suggestions for further interviews, and opportunities for additional feedback. The questions were oriented toward all these dynamics. However, the discussion could sometimes focus on a particular aspect, depending on the participant’s experience and collaboration background.

The interviews were conducted virtually via Zoom and Teams, with audio recording for transcription and analysis, lasting an average of 50 min. The main participants, who held various roles, such as scientific coordinator, innovation management specialist, and agronomist, were anonymized to protect their privacy.

3.2. Data Analysis

The empirical data collected were transcribed verbatim. The transcriptions were coded using Nvivo (NVivo 12.7.0 (3873)) to identify and categorize the characteristics of the LLQ according to the emerging themes from the preliminary research. These themes encompass user roles, actor networks, context, activities, challenges, outcomes, sustainability, approaches, communication, trust, and motivation. This study then aligned these characteristics with the three key components of the SNM: expectations, learning, and actor networks. Thus, data related to activities, context, approaches, methods and tools, and challenges were categorized under lessons learned. For instance, challenges often emerge only after the completion of LL activities, thereby turning each challenge into a valuable learning experience. Data regarding user roles, trust, communication, and motivation were categorized under actor networks as they relate to stakeholders. Finally, data on outcomes and sustainability were considered expectations. Figure 1 provides an overview of the outcome of this analytical process, with learning, expectations, and actor network development as the main units of analysis. This inductive approach allowed us to highlight the peculiarities of ALLs, particularly through a comparison between LLs and LLQ, and explore learning in a broader sense while discussing its transferability. The approach supports the goal of generalization and transferability. The results are presented in the following sections.

4. Results

4.1. Living Laboratory–Québec

Launched in 2018, Agriculture and Agri-Food Canada’s (AAFC) Living Laboratories Initiative established several LLs across the provinces to support agricultural research in Canada. The network’s development focused on two components: a theoretical component and a practical one. The theoretical (or conceptual) component involved a meeting with the G20 [14] to study, develop, and define the concept of ALLs. The practical component involved implementing LLs in several Canadian provinces. Introduced in 2019, Living Lab–Atlantic on Prince Edward Island was the first, followed by Living Lab–Eastern Prairies in Manitoba, Living Lab–Ontario, and Living Lab–Québec (LLQ) in 2020. Those four LLs were the first to be funded by AAFC, which drew on the success of the Acadie Lab LL to implement the initiative.

In this study, we focused on LLQ, whose main collaborator is the Union des producteurs agricoles (UPA), an agricultural trade union. At the outset, the UPA first identified water quality as the environmental issue to be studied and, in a second step, targeted three territories around Lac Saint-Pierre based on certain criteria. The first criterion related to the choice of watershed as the location for the activities and the watershed’s nature, which had to be agricultural. The second criterion was that the water quality in the watershed in relation to agricultural practices had to have already been documented along with any prior efforts by local organizations and stakeholders to mobilize agricultural producers to adopt good agricultural practices. The third and final criterion was that the watershed had to drain directly into Lac Saint-Pierre. The three watersheds bordering Lac Saint-Pierre that meet these criteria are the Rivière Pot au Beurre, Rivière Bois Blanc and South Shore watersheds.

The project was implemented in 2020 and officially ended in March 2023. The LL focused on the issue of water quality related to agricultural practices in Lac St-Pierre. The general objectives were, therefore, focused on the sustainability and resilience of the agricultural system: soil health and cover, watercourse health, climate change, and biodiversity in agroecosystems. They were set out to identify ways of working with agricultural producers to ensure uninterrupted year-round coverage and minimize water and wind erosion for watercourse health and management. However, one of the specific aims of the LL was to involve the agricultural community from the earliest stages, develop the project in cooperation with the farming community, and thus facilitate and accelerate the adoption of good farming practices and on-farm research protocols. In addition, LLQ also sought to generate knowledge that could be mobilized to launch similar projects.

4.2. Characteristics

According to a previous study [15], ALLs are distinguished by their connection to nature and living systems requiring experimentation in real-life situations, their territorial roots, the significance placed on the final product due to its high social value, as well as the consideration of a unique and integrated constellation of institutions, actors, and knowledge in the agri-food domain. This study also highlights the central role of the end-user in collaboration and the implementation of a co-creative approach.

4.2.1. Farmers-Led Approach

In the context of the LLQ, farm producers and businesses were the main users and were at the core of the innovation development process. A key principle of the LLQ was to involve them in every decision, research activity, and related project (such as information days and coaching/mentoring activities). The LLQ initiatives were, therefore, focused on the needs, ideas, and issues of producers and businesses, aiming to provide scientific support to address their concerns. Additionally, agricultural producers were also involved in testing new developments.

User needs were identified through co-development meetings, with scientific actors providing resources and knowledge support. The main objectives established by the UPA were presented to agricultural producers and businesses, encouraging them to share their ideas, suggestions, and preferences for refinement. After initial meetings with farmers and agricultural businesses, it became apparent that community dynamics and willingness to innovate varied significantly from one watershed to another and from one farmer to another, with some experimenting with cover crops for years while others were novices. As one research participant stated:

“Take cover crops, for example, we had producers who had been growing them for 10 years, others for 5 years, and some not at all.”

Additionally, some producers were seen as stakeholders whose opinions were valued, while others were active participants in meetings. Thus, only a few had research protocols on their farms, reflecting a diversity of engagement in the innovation process.

In both ALLs and LLs, users play a central role in collaboration. The roles of users also vary from one producer to another in ALLs, as in LLs (see, for example, [44]). However, a major distinction lies in the fact that, while in LLs, the users are individuals bringing their ideas and testing innovation, ALLs expand this definition to include agricultural businesses, also considered as users, in addition to producers (see also [15]). Furthermore, ALLs differ from general LLs by classifying users according to their level of knowledge of the agricultural practices being tested, unlike traditional LLs, where the users’ level of knowledge is not considered.

4.2.2. Actor Network

As part of AAFC’s LL initiative, the main partners vary depending on the laboratories and can be indigenous or non-profit organizations. The LLQ is a collaboration between three main actors: the UPA, agricultural producers, and AAFC. It includes the participation of scientific coordinators and experts such as agronomists and agricultural advisors. The UPA, as the project initiator, plays a central role, while scientific coordinators facilitate the liaison between AAFC and partners to ensure the alignment of scientific activities with project objectives. Agricultural advisors, working closely with farmers, play a crucial role in daily monitoring, gathering information on new practices, producing reports on the results, and disseminating these findings. They also contribute to establishing research protocols and ensuring the proper conduct of research activities, even conducting ad hoc data collection activities when access to farms is logistically challenging. This structure enables effective cooperation among all actors, strengthens ties with producers, and facilitates the identification of opportunities for activities with farmers while ensuring that research remains relevant and aligned with the producers’ needs.

The LLQ involved innovation management specialists tasked with supporting the innovation process, clarifying the LL concept, and ensuring better integration of social aspects. These specialists also coordinated several international collaborations. The UPA played a key role in creating a consortium of experts to support farmers, involving partners engaged in environmental conservation such as Ducks Unlimited and Nature Conservancy of Canada, as well as researchers from the Université du Québec à Trois-Rivières, McGill University, and Laval University. The project brought together diverse stakeholders, including four regional UPA federations around Lake Saint-Pierre, economists, regional county municipalities (MRCs), social scientists, and other actors supporting innovation and research.

Given the complexity of managing the LL, facilitators and animators were integrated to support co-design activities and communication between stakeholders, ensuring the absence of bias and fostering idea generation (e.g., through prioritization exercises). As one participant stated:

“What we will have, which we didn’t have from the beginning, is what we call a facilitator, that is to say a person who understands the mechanisms of co-development, innovation, discourse, open innovation, whether it is a neutral person who accompanies us in our co-development meetings to ensure that we are always on the right track, and that we do not deviate towards biases that we, as researchers or producers, might pull the cover to our side, and then oops at some point, we derail a co-development meeting because we discussed a bunch of things that, ultimately, did not bring us closer, but rather separated us.”

The expertise of facilitators and animators in co-design, open innovation, and communication management was crucial for maintaining stakeholder engagement and group cohesion. This approach ensured a high level of interest and engagement despite slow activity cycles, highlighting the importance of animation in the project’s success. The openness and integration of voluntary actors characterized the LLQ. Figure 2 provides an overview of the project ecosystem. Research participants emphasized the importance of integrating facilitators, animators, and local actors from the outset of the project to ensure continuous field monitoring and promptly address logistical issues.

In all types of LLs, the presence of multiple heterogeneous actors is a constant, requiring a multi-actor approach [15], particularly in ALLs, where agricultural advisors and facilitators play a crucial role. Agricultural advisors, who play a similar role to coordinators [44] in general LLs, are essential in gathering and organizing information about user needs within ALLs. As one of the research participants asserted:

“The participation of agricultural advisors in all activities requiring farmer involvement enabled them to stay informed and strengthen messages and communications during field follow-ups, thus enhancing their connections and easily identifying opportunities for activities with farmers,”

Facilitation, though costly [44], is essential in ALLs, unlike in general LLs, where it may be less necessary. Facilitators in ALLs perform a similar role to orchestrators [44] in general LLs, which is crucial for building trust and fostering collaboration. Animators, comparable to messengers [44] in general LLs, facilitate the innovation process by gathering and sharing ideas, underscoring the importance of animation in all types of LLs. The involvement of local actors is also crucial in ALLs and necessary for LLs related to specific territories, such as urban and rural LLs. This study reveals that local actor engagement is not required in other types of LLs, which are not directly tied to a specific innovation location.

4.2.3. Real Context and Activities

This case study considers the real context through experimental activities conducted directly in the field. LLQ activities focus on innovation and scientific research, following a three-phase operational cycle. The first phase, corresponding to the co-development phase, involves the co-creation with farmers to identify their needs and problems, plan future activities, and develop ideas to support the search for solutions, including best management practices beneficial at the farm level. Initial co-development meetings within the LLQ aimed to define research activities, protocols, and farms participating in experimental activities while ensuring that stakeholders fully understand the research project to ensure its viability and validity. The second phase involves testing and experimentation by researchers in the field to test ideas developed during the co-development phase. Finally, the evaluation phase analyzes and synthesizes the collected data, including user feedback. The results feed into a new co-development phase, thus fostering continuous improvement of innovation on farms. This innovation cycle promotes an iterative and participatory approach, crucial for aligning innovations with practical field requirements and the real needs of farmers, thereby effectively contributing to the advancement of research and innovation in the agricultural domain.

The LLQ also emphasized other activities to support agricultural producers, funded by AAFC but separate from formal research activities. These activities were initiated at the request of agricultural producers willing to test new practices aligned with the objectives established at the beginning of the project. The UPA, the producers involved in experimentation, and agricultural advisors were responsible for defining and implementing these activities. Experiments, previously planned by farmers and their advisors, were to be conducted in a specified area, with a specific volume of data collected by an advisor, providing a structured framework for experimentation.

The LLQ also conducted a series of experiments called “trial networks”, documenting the impact of cover crops on nitrogen fertilization for corn over a two-year period. These experiments allowed farmers to quantify reductions in mineral nitrogen and fertilizer, as well as cost savings on material purchases while improving soil quality. In parallel, an in-depth study on riparian buffer strips was carried out by a research team to compare different vegetation management models, including herbaceous, shrubby, arboreal, and mixed varieties. This study concluded that there were no significant differences between models of varying widths and types of vegetation. Researchers also monitored biodiversity, including pollinating insects and birds, as well as soil microbiomes, enabling farmers to better understand biodiversity in their riparian strips. Another series of experiments was conducted on the feeding of cows and pigs with trace elements such as copper, zinc, and manganese. The aim was to determine if these nutrients were being overfed, which could lead to their accumulation in the soil and waterways due to their presence in manure. The results revealed instances of excessive administration on some farms, highlighting the need for more precise management of trace elements to achieve cost savings and prevent environmental pollution while optimizing nutrient absorption by the crops. These experiments have not only contributed to improving agricultural practices in terms of fertilization management, biodiversity, and animal feeding but have also provided advantageous economic prospects for farmers by reducing costs and promoting more sustainable agriculture.

The LLQ also focused on other activities, such as knowledge sharing within the agricultural community, by collaborating with other producers and external organizations as part of the project’s information dissemination component. This was achieved through information days, newsletters, and articles. Researchers also organized activities aimed at designing experiments tailored to each type of farmer and ensuring regular follow-up.

The activities of co-creation, experimentation, and evaluation are similar in both types of LLs, but within the LLQ, the exploration phase was not undertaken or was implicitly included in the co-development phase. The absence of the exploration stage raises uncertainty about the exact composition of the innovation cycle, questioning whether it is limited to the three aforementioned stages regardless of the type of LL or if this configuration is specific to the analyzed case. Moreover, the activities of the ALLs are characterized by a higher number of activities compared with general LLs, including integration days, support activities, information dissemination, and a trial network, aiming to ensure the reproducibility of experiments for scientifically reliable probable results conducive to acceptable statistical analyses. Unlike general LLs, where the context can be physical or virtual, an ALL is inherently linked to a specific territory (see [15]). Also, unlike general LLs, activities within an ALL align with seasonal cycles, with co-development in winter, experimentation in summer, and evaluation in autumn, reflecting and integrating with the agricultural work cycle. This seasonal timing dictates that activities within an ALL fit within seasonal work cycles [15], extending the time required for their completion compared with other types of LLs. Thus, while the innovation development cycle is iterative in both general LLs and ALLs, it is notably slower in the latter.

4.2.4. Challenges

Challenges are at the heart of ALLs, which aim to find solutions to major issues, including environmental ones. As such, the challenges encountered as part of an ALL relate to aspects such as stakeholder expectations, understanding the LL approach, the health context, weather conditions, and even communications.

One of the main challenges encountered within the LLQ was the adaptation of researchers and farmers to the LL approach, marked by a limited understanding of their roles and how LLs operate. Scientific researchers, unfamiliar with open innovation, struggled to grasp their roles, LL functioning, and the integration of farmers into research activities. As one research participant stated:

“They are more used to conducting experiments in experimental stations with controlled conditions nearby.”

“If we go back a bit, we had a researcher who had an idea, he would set up a research protocol, a research proposal, test his hypotheses, come up with conclusions or results that could be conclusive. And then he would try to sell either the innovation, the new practice, or the results to the farming community.”

Unlike their usual methods focused on controlled experiments on private farms, the LL approach requires co-creation with farmers. Farmers wanted to be involved in the annual evaluation of results, necessitating a reorganization to ensure the sharing of preliminary results each winter and farmer involvement in result evaluation. Lack of understanding and insufficient communication also led to farmers inadvertently using experimental plots, resulting in data loss. Variability in expectations among stakeholders exacerbated this challenge. As one participant highlighted:

“I would say that among all the producers who participated, there were really expectations at varying speeds.”

Some were well-informed about the LL approach and adjusted their expectations accordingly, while others struggled to understand their role and research objectives. The main obstacle to adaptation and expectation alignment stemmed from the lack of initial initiatives to clarify LL roles and operations, underscoring the importance of better guidance and participant integration from the project’s outset.

The COVID-19 pandemic also posed a significant challenge for the project. The first season with farmers was lost due to a complete halt in activities. Pandemic-related restrictions hindered exchanges, relationships with partners, and co-creation meetings, thus impeding the progress of the LLQ. Out of the three series of planned co-creation workshops for each watershed, only two were able to take place in person. Restrictions also severely limited field activities, including the usual support for farmers, and interrupted research and sample analysis activities for an extended period. As explained by one participant:

“Agriculture Canada researchers were at home, and then the laboratory closed. It took about a year before the laboratory reopened for them to analyze all the samples they had collected.”

Despite farmers continuing their agricultural activities, the lack of research results due to the suspension of activities has led to frustrations. These challenges were compounded by the difficulties posed by unpredictable and variable weather conditions, exacerbating the obstacles faced by the ALL.

Geographical location also hindered collaboration among the LLQ stakeholders. During the pandemic, they had to regroup and adopt virtual communication tools, leading to the relocation of some meetings to online platforms. Also, the LLQ activities, organized between 2020 and 2023, spanned a relatively short period for experimentation in an agroecosystem, where work cycles are annual. This time constraint limited the process to only three experimental cycles, reducing the LLQ’s flexibility. A longer duration is essential for increased flexibility in such projects, adding an additional challenge to this initiative.

In LLs in general, and more specifically in ALLs, the challenges are varied. However, certain obstacles such as weather conditions, time constraints, and geographical location impact ALLs more significantly. These challenges have little or no effect on traditional LLs. Consistent with the SNM, aligning stakeholders’ expectations, especially those of farmers, from the outset is crucial. It is essential that each participant understands the role of the different actors and how an LL operates. Investing time and resources to clearly explain the objectives and expected outcomes to farmers is fundamental to avoid the formation of unrealistic expectations regarding the project.

In this case, governance was not an obstacle, thanks to a flexible approach that involved stakeholders in meetings and decision-making based on specific research needs and topics. Not all stakeholders were required to attend every meeting or be involved in every decision, depending on the specific needs of the LL and the farmers. As explained by one of the participants:

“What we did was target them on research themes so that we didn’t have everyone online at the same time, so we targeted, for example, cover crops, and then we invited all the farmers who were interested in conducting research at that level of that theme.”

A schedule and participant list were thus established for each meeting, and only the input of stakeholders directly affected by the decisions was sought. However, it remains uncertain whether this governance model can be generalized to all ALLs, although the literature indicates variability in governance approaches from one LL to another.

4.2.5. LLQ and Sustainability

The LLQ aims to actively engage the agricultural community in the project, accelerate the adoption of environmentally friendly agricultural practices, generate useful and transferable knowledge, improve water quality, and preserve soil health and biodiversity while minimizing water and wind erosion. The expected outcomes, both tangible and intangible, impact social, economic, and environmental spheres. However, evaluating the outcomes remains premature, as the project data are still being analyzed. Early findings from the experimentation reveal that the main obstacles to the adoption of sustainable agricultural practices by farmers include a lack of financial, informational, and technical support. Despite some challenges, the collaboration between farmers and researchers has been beneficial, fostering the democratization [87] of innovation development and acknowledging farmers as experts. This dynamic has facilitated a mutual understanding of roles and a more open and collaborative research approach.

The implementation of the LLQ is part of an accelerated adoption approach to beneficial agro-environmental practices. By favoring the ALL approach, it aims to stimulate the adoption of innovations and the co-construction of solutions to agro-environmental challenges, with a focus on climate change, biodiversity, as well as soil and water quality. Thus, practices within the LLQ are deeply rooted in an environmental perspective. While general and agroecological LLs focus on sustainability, the general LL model may prioritize environmental, social, or economic dimensions separately. In contrast, the ALL embraces a holistic approach, integrating all three pillars of sustainability.

4.2.6. Methods, Tools and Approaches

The LLQ implemented various methods and tools, such as prioritization exercises and brainstorming.

Other approaches have been identified for the LLQ in addition to the user-centered approach. Considering LLs as related to the open innovation paradigm [28,37,75], it is possible to observe the application of the open innovation approach within the studied LL. Co-creative and cooperative approaches have also been explored through co-creation with agricultural producers. The collaboration between scientific and agricultural communities to conduct research activities identifies the collaborative approach. Similarly, the participative approach has been implemented through the involvement of various actors in the LL activities. In terms of participation, the LLQ presents itself as an open LL, welcoming and including all actors interested in the project. Regarding the coordination approach, two distinct approaches emerge in this context: the top-down approach (see [56]) through the activities of the ALL and the bottom-up (see [56]) approach by considering the needs and ideas of agricultural producers.

Although details on the specific methods and tools used in the case study are limited, and considering that they may vary depending on the LL and its specific needs, it is complex to determine whether the methods and tools used in general LLs are directly transferable to ALLs or vice versa. Nevertheless, there is evidence to support the existence of a diversity of methods and tools applicable in these contexts. The alignment between the approaches identified in the literature review and those observed in the case study suggests a potential for the generalization of these approaches. Regarding the participation of actors, the LLQ is distinguished by its openness. However, despite the literature asserting the impossibility of generalizing the nature of participation within LLs, the question remains open as to whether the characteristic openness of the LLQ can be seen as representative of all ALLs.

4.2.7. Communication, Trust, and Motivation

Within the LLQ, communication was diverse, encompassing both in-person and virtual meetings. Initially, the first co-development meetings with farmers took place in person in each watershed. However, this communication had to be adjusted based on the varying dynamics of farmers and their level of practice adoption. With the onset of the pandemic, meetings shifted online to maintain momentum. Virtual gatherings allowed for the discussion of specific research topics, bringing together stakeholders interested in particular themes such as intercropping. Furthermore, online meetings were favored when geographical distance made in-person meetings challenging. This flexibility in communication methods ensured the regular and effective dissemination of information on research progress, strengthening the collaboration between researchers and agricultural producers.

Trust played an important role in the LLQ. Good communication within the LL was based on trust between the actors. For several years, the federal scientific community in the agricultural sector was totally off the farming community’s radar. Good communication between the federal scientific community and Québec’s agricultural community helped establish trust between the two groups, enabling them to cooperate and work together effectively. As a result, a high level of trust was established between the researchers and farmers.

Within the LLQ, actors were not remunerated for their participation, and farmers contributed on their own initiative. They did, however, receive compensation for any yield or soil losses associated with the research project. Farmers were motivated by the fact that they were considered experts and members of the project development team. Even so, their motivation stemmed from being at the heart of the activities. They could voice their concerns and opinions, and their ideas and suggestions were considered. The lack of remuneration for agricultural producers in the LLQ is explained by the fact that the project sought to engage their personal will and initiative to adopt new, more sustainable practices. Besides the agricultural producers, the entire UPA team and the agricultural advisors who worked on the project received financial assistance. The agricultural advisors, for example, received funding and compensation for support activities and some of the research and data-gathering activities.

In the literature on LLs, communication and trust are unanimously recognized as fundamental pillars. There is a close relationship between these two aspects. Within the LLQ, stakeholders have expressed the need to integrate communication specialists, thereby facilitating collaboration among actors. In the literature on traditional LLs, communication specialists are identified as orchestrators [44] who can motivate actors to contribute to the project. Motivating these actors to engage in the project is crucial, and this can be achieved through intrinsic or extrinsic rewards. This dynamic also applies to the LLQ, where agricultural producers are motivated by extrinsic incentives, while other actors, such as agricultural advisors, benefit from financial compensation. Thus, it can be asserted that communication, trust, and motivation remain essential aspects in all LLs.

The

Table 2. Characteristics of LLQ as sociotechnical experiment.

LEARNINGS	Activities and context	Real context refers only to a physical territory.
		The activities include co-development and fine-tuning meetings, experimentation, evaluation, support activities, the test network, information dissemination activities, innovation, and research.
	Challenges	Challenges often relate to actor expectations, communication, health context, weather conditions, geographic location, timeframe, and the comprehension and adaption of the LL approach.
	Approaches, methods, and tools	User-driven approach, co-creative and cooperative approach, collaborative approach, participatory approach, coordination approach, and open innovation.
		Methods and tools include prioritization exercises and brainstorming.
ACTOR NETWORKS	Users	They represent agricultural producers and enterprises.
		They are at the center of ALL activities.
		Their role is to express their needs, issues, and ideas and test new developments.
		There may be several user groups based on their knowledge level.
	Stakeholders	Multiple heterogeneous actors
		Involvement of agricultural advisors, facilitators, animators, local actors, communication specialists, and other stakeholders.
		Governance involving actors based on needs.
	Communication, trust, and motivation	Communication (in person and virtual) requires ongoing adjustments to meet certain challenges.
		There must be trust between researchers and farmers.
		Provide intrinsic incentives for producers and monetary compensation for yield losses. Provide compensation for other actors who give up their time.
EXPECTATIONS	Results	The results are focused on the adoption of good agro-environmental practices, the generation of a transferable knowledge base, user involvement, the minimization of water and wind erosion, and biodiversity conservation.
		Varying expectations (in terms of the results and actor roles) that required readjustment.
	Sustainability	Generally ecological, economic, and social.

summarizes and categorizes the different characteristics of LLQ according to the components of the internal SNM process.

4.3. Transfer of Learning

A single experiment cannot change a sociotechnical regime. For real regime change and a sustainable transition, single experiments are insufficient. That is why experiments require repetition in a range of contexts and ongoing learning based on feedback from interactions [11,24]. Van Den Bosch refers to the process of scaling up: reproducing experiments in different contexts and linking them with other projects, fields, and initiatives. Scaling up does not refer to repetition without variation but rather to the development of something new in a variety of contexts [26]. That implies that experiences that are successful in one context are repeated intelligently in other contexts, incorporating lessons learned and always adapting to the new context [19]. Learning within experimentation is, therefore, limited, requiring that experiments be repeated in other contexts to learn more about different concepts in different contexts [26]. This repetition supports the idea that different experiments in different contexts can gradually complement and strengthen each other [26,88] and thus replace the existing regime.

Along with AAFC’s LL initiative, the new Agricultural Climate Solutions (ACS) program has been under development since 2022. The program covers a longer timeframe, with a possible extension until 2031. It is also focused on the implementation of ALLs across Canada but in a slightly different context. ACS target new territories and new issues more focused on carbon sequestration and greenhouse gas reduction by creating ALLs based on the same approach for a longer period and building on learnings from the first initiative to enable open innovation or co-development with the farming community. The program is still under development, and it is difficult to say at this stage whether the same actors from AAFC’s first LL initiative will participate in ACS. In Québec, the UPA is the main partner in the new project, which will include many of the actors from the former program within a wider partner network and will likely take place across the province. It will be important for this new program to link the new LLs with those initially developed to share the knowledge, ideas, learnings, and methods that are already developed.

5. Discussion

Through field research, this study identified the components of the internal SNM process (expectations, learnings, and actor networks) through the ALLs. ALLs emerge as a form of sociotechnical experimentation that can make an effective contribution to the sustainability and resilience of agricultural systems, which have specific features that often differentiate them from traditional LLs. Initiating and managing an ALL requires the development of a broad actor network and the alignment of expectations and trust among these actors.

5.1. The Actors’ Network

With regard to the creation of actor networks, this research shows that an ALL brings together multiple heterogeneous actors. In addition to the project’s main actors, this type of LL requires the involvement of other actors, specifically agricultural advisors, facilitators, animators, local stakeholders, and communication specialists. Unlike general LLs, where the involvement of these actors is optional, the actors’ lack of involvement in an ALL can affect the project’s success. It is very important for the actors to be involved from the earliest project stages. It is also important to compensate them to keep them motivated. This research revealed that it is not necessary to have all actors attend every meeting or the decision-making process. The ALL, therefore, adopts a particular type of governance, targeting actors according to the topics that are being addressed. However, the type of governance within the ALL requires further discussion to determine whether it is similar in all ALLs and whether there is potential for transfers to other contexts.

Communication is very important in this type of collaboration between the actors and, more specifically, between agricultural producers and other stakeholders, especially when the project is complex in terms of the number of stakeholders, researchers, and territories involved. Communication must be open, ongoing, and continually readjusted to meet challenges. For example, when the geographic location of the study area or the health context poses challenges that are not conducive to in-person meetings, virtual communication tools may be mobilized to facilitate exchanges. Some meetings may require a physical presence, and these tools are not always a solution, but they can help in many situations. As soon as the ALL is launched, clear instructions must be drawn up regarding communications, content, and frequency. Research has shown that communication and trust are closely linked. As a result, good communication necessarily requires trust among actors.

In terms of actor participation, the ALL is an open living laboratory [35,69]. From the outset of the project, it is important to have the openness to consider that all the actors involved in the initiative have the same value in terms of contribution, expertise, knowledge, and competence. It is also important to promote openness on the part of all actors and of researchers in particular so they can integrate and work directly with farmers, taking their realities into account when drawing up protocols and including them in data analysis to keep them informed on the advancement of this research. In addition to exchanging information, doing so could support an effective analysis of the results. In other words, involving farmers when determining tasks and analyzing the results can help resolve problems more quickly. Openness is, therefore, important in this type of project. However, the open nature of the LL discussed here cannot be generalized to all ALLs. Future studies could, for example, seek to identify the best type of study in terms of actor participation. AAFC’s LL initiative brings together a variety of actors since their inclusion can lead to opportunities that are often unforeseen. With the launch of nine new LLs in 2021, the network is currently running 13 LLs across the different provinces and territories with about 1000 direct participants.

The notion of a community of practice is also much more significant in an ALL than in a traditional LL, and work tools must constantly be adapted. Users are central to the ALL. They represent agricultural producers and businesses whose role is to express their needs, problems, and ideas and test new developments.

5.2. Learning

As far as learning is concerned, ALLs have demonstrated that this type of LL raises challenges that are often linked to stakeholder expectations, communication, the health context, weather conditions, the spatial and temporal frameworks, and actors’ understanding and adaptation to the LL approach. This research also confirmed the findings of previous studies [15], revealing that LLs are rooted in a territory and adopt a multi-actor approach. Indeed, user-driven, co-creative, cooperative, collaborative, participative, coordination, and open innovation approaches were all identified. Also, the implementation of LLQ sheds light on the different activities that are possible. This research identified a cycle of co-development, experimentation, and assessment activities similar to those in general LLs. However, the cycle within the latter is made up of four activities, including exploration. It is not possible to affirm whether the exploration activity is not considered at the ALL level or whether it is incorporated in the co-development activity. Further research is required to clarify this. In addition to the first three activities identified, the ALL also undertakes support, test network, information dissemination, innovation, and research activities. In this sense, this study points to the need for further research to determine whether the coaching, trial networking, and information dissemination activities may be mobilized in other contexts.

As far as the timeframe, this research did not confirm the existence of other activities besides the eight that were identified. Future studies may investigate this. This research also pinpointed several methods and tools, such as prioritization exercises and brainstorming, that were mobilized as part of this project, but further research is required to find more methods and tools and clarify whether they may be transferred to other LLs. This research also showed that the timeframe is significant in this type of LL. It must be long enough for experiments to mature since iterative cycles are annual and follow seasonal agricultural production cycles. It is, therefore, worth pointing out that a different work cycle requiring a long timeframe is needed to cover all the activities undertaken as part of an ALL. Activities within an ALL are time-consuming and call for patience, understanding, and resilience on the part of everyone involved and of the actors from the scientific and farming communities in particular. Time and budget are essential to bring actors together at the start of the project. LLQ was a very flexible project that made adjustments based on the needs of producers and researchers. This flexibility enabled LLQ to experiment with other new activities, such as documenting the impacts of a sedimentation basin. Flexibility offers a huge advantage and opportunities, provided the LL takes place over a long period of time.

5.3. The Actors’ Expectations

The actors’ expectations and, more specifically, those of agricultural producers varied and thus required some adjustments. For example, farmers had very high expectations regarding the research results, the speed of their dissemination, and the researchers’ roles. That meant that the researchers had to adapt their approaches to the farmers to ensure they fully understood the research process. Sufficient resources must be mobilized from the outset to train all the actors involved in the LL approach, so they better understand the approaches’ process, principles, and rigor and to better guide expectations. It is important to clearly indicate the exact roles of the specialists and researchers to the farmers and confirm they all clearly understand the research requirements. Researchers must also take the time to understand producers’ expectations so they can better respond to them. Finally, it is essential to make certain that farmers are aware and remain mindful of the plots dedicated to experiments to avoid their accidental use and the consequent loss of research data.

6. Conclusions

This study investigates the factors for accelerating the adoption of good agri-environmental practices toward sustainability as well as for establishing a transferable knowledge base while actively involving users, minimizing water and wind erosion, and preserving biodiversity. We focus on understanding the LL approach and, more specifically, the role of each actor in the context of an open innovation experiment.

The AAFC’s LL projects officially ended in 2023, but research and assessments are ongoing. Also, the dynamics of the four ALL projects were very different. In Manitoba, for example, the ALL took place over a very large geographic area, making it difficult for actors to share information and communicate. That was not the case in Ontario, where only eight large agricultural businesses took part, and discussions and information sharing were much easier to organize. In Québec, some 40 farms contributed to the scientific activities. The issue there was ensuring that when research was carried out on a farm, there was some similarity or point of comparison with what was being done on other farms. In the case of the small province of Prince Edward Island, the project targeted a specific production (potatoes), whereas more crops were targeted in Québec and Ontario. The different dynamics from one project to the next make it difficult to establish a model or framework specific to the Canadian or Québec context at this stage, though future research could address that issue, as well as establish a comparative case between the four LLs to highlight the specificities of each context.

Future in-depth studies of the ALL at hand could adopt an ethnographic approach to determine exactly how the project was set up. Following the literature review, funding was recognized as a challenge, though no funding issues were identified within the LLQ. A significant aspect of experimentation is the protection of innovation, and funding constitutes a form of protection. Future studies could, therefore, look at the different forms of protection within ALLs to determine whether they are a factor of success or a hindrance to a project’s development.

Finally, we hope that the recommendations will stimulate further research in the field and that the results will help fill the gap in the literature required to lead and manage an ALL.