Collaborative networks and an innovative educational design in an agroecology program: the PIES AGILES experience in Mexico

1 Introduction

This article analyzes the experience of the Interinstitutional Specialty Program in Food Sovereignties and Strategic Local Advocacy Management (PIES AGILES, for its Spanish acronym) in Mexico—a 2-year (2021–2023) nationwide interinstitutional educational program that brought together an educational community of close to 300 participants across 18 states. This initiative was driven by the National Council of Humanities, Sciences, and Technologies (Conahcyt; presently a Secretariat) through four of its public research centers—CIATEJ, ECOSUR, CIESAS, and COLMICH¹—located in various cities across the western and southern regions of the country. The program was designed, coordinated, and facilitated by an interdisciplinary team of researchers and professionals from these centers who shared a commitment to promoting agroecology and food sovereignty through alternative educational models.

PIES AGILES emerged within a national context of institutional changes and a shift of public policies toward an open science and humanistic approach, in order to address specific national problems, including those related to food systems and food sovereignty in Mexico (Consejo Nacional de Ciencia y Tecnología, 2023). Federal public policies promoted a phase in agroecological scaling (2018–2023), culminating in the 2020 presidential decree for the gradual phase-out of glyphosate, a widely used chemical herbicide in Mexico (Toledo and Argueta, 2024). PIES AGILES arose as a response to this decree, with the purpose of training professionals capable of understanding and transforming local and regional food systems, and a focus on promoting agroecological transitions across diverse territories.

The educational program was grounded in critical and constructivist pedagogy, and applied a project-based and hybrid or blended learning format that combined a place-based (territorial) and virtual learning strategy. In all, the program advanced a situated, collective, and plural learning model (Sánchez-Reyes et al., 2025). A learning management system was established supported by an online learning platform (Moodle), providing remote and asynchronous connectivity—a key factor in its success, given that the program took place during the later stages of the COVID 19 pandemic. Furthermore, the program took into account the rhythms and temporalities of the participants’ context, such as local agricultural cycles and the communities' political, organizational, and festive activities, thereby enabling students to remain in and reside within their communities while pursuing their studies.

Although many efforts are underway to reduce educational inequity, studies show that exclusion, inequality, and a sense of alienation persist in Mexican higher education (Alcántara Santuario and Navarrete Cazales, 2014; Bernabe, 2021). Thus, a program policy aimed at fostering access to education and inclusivity favored the recruitment of students with diverse profiles and backgrounds, paying close attention to those from rural areas and marginalized zones. Participants included young professionals (recently graduated), farmers and producers, agroecology promoters, teachers, members of social organizations, and local actors with varied life trajectories, experience and age. Notably, the program admitted students without academic degrees, recognizing the value of empirical knowledge, experience and non-formal education. There was also a high participation of women, which accounted for 54% of the student body. Finally, approximately one-third of the students identified themselves as part of an indigenous or ethnic group, reflecting Mexico's ethnic and linguistic richness, predominantly Nahua, Maya, P'urhépecha, and Tseltal among others, and spoke an indigenous language (Heinze and Sánchez Reyes, 2026).

The diversity of participants and cultures encouraged the visibility and recognition of different knowledge systems, especially of forms of knowledge, traditional practices and livelihoods that have sustained a diversity of agroecosystems over time—in one of the planet's richest biocultural regions (Toledo and Barrera-Bassols, 2017). We believe this created space for a dialogue of knowledge, as referred to in agroecology (Martínez-Torres and Rosset, 2014), fostering exchanges and complementarities, beyond academic disciplines, in which praxis and theory regained their significance, generating useful knowledge to support territorial transformations (Domené-Painenao et al., 2026). The opening of multiple dialogic spaces allowed for the convergence of similar and compatible experiences, creating and strengthening networks among students and communities, and also connections with other local actors and stakeholders, including schools, government bodies, social initiatives and research projects among others. This suggests that although this educational program was driven by public policy, it elicited a genuine and realistic response from participating communities, creating a sort of interface where dialogue, co-construction, and research with real-world impact took on meaning. Therefore, we identify expansive networks or connections, extending in different directions, among these territorial learning spaces that emerged from the program.

It is within the community that these connections and relations we call collaborative networks arise, which are here defined as alliances that serve to coordinate different social groups; given today’s highly specialized world, these networks act as intermediaries enabling social organizations, governments, and citizens to collaborate. Their goal is to link ideas and solutions across local, regional and global levels (Zimmerman, 2004). In the case of the educational program here examined, these collaborative networks are co-created by students and their learning communities, motivated to promote and advance agroecology: The student–learning community–institution triad generates a new mode of coordination for agroecological transitions and community wellbeing. Relations are forged based on shared interests and common challenges, linking to other community projects and institutional programs—scaling both out and up, i.e., expanding throughout and across territories and articulating from the bottom up (articulating, as used in Spanish, meaning to join, link and interconnect). In the process, participants in the program assume a dual role: that of a student and trainee and that of a networker, a sort of mediator between the community and institutional structures.

Despite their emerging relevance, a gap remains in the literature regarding the functioning of these types of situated, collective and plural learning programs, as well as how collaborative networks relate to agroecological scaling. This is partly because such educational models are rare. While we have identified online initiatives in Mexico, Chile, Colombia, Spain, and other countries, none combine this learning-by-doing approach to project-based and blended learning. We believe there is a critical need for further research into these educational models that foster collaborative networks, particularly those rooted in peasant, indigenous, and plural agroecologies. The recognition of these forms of connectivity is essential to understanding the mechanisms behind the expansion and territorialization of agroecology. Collaborative networks not only bring visibility to the practices and knowledge associated with food systems, but also propose frameworks of autonomy and cooperation that contest and challenge the global agro-industrial model. We believe these networks constitute the engine and hold the potential to transform current food systems.

In this article, we describe the PIES AGILES experience with a focus on networks, both some theoretical foundations and the networks established during the program (2021–2023). The objective is to demonstrate how this educational program contributes to building collaborative networks and in turn local and community capacities, scaling agroecological processes by the transition of students themselves into subjects of territorial transformation. The research question for this study is therefore: How does this educational program contribute to the visibility, emergence, and strengthening of networks aimed at building community capacities and scaling agroecological processes?

2 Methodological approach

This study is based on a mixed-methods design with an exploratory-descriptive scope (Creswell and Creswell, 2017), enabling us to address the research question through the analysis of collaborative networks. We aim to explore how learning communities interconnect, and subsequently relate these findings to our own (the authors) experience in the program. To achieve this, we employed techniques such as network graphing with a specialized software. Furthermore, our interest lies in analyzing how the program's design fosters, proposes, and gives rise to specific forms of collaborative networks established among students, learning communities, and institutions, specifically institutional research programs.

2.1 Positionality and ethical considerations

We (the authors) participated in the program as core faculty, with various activities including curriculum design, teaching and tutoring (both online and in person), and program operation and evaluation. Thus, this article builds on our own direct experience in the program.

Academic output, research results and information generated during the program (2021–2023) and used in this study, was securely stored and protected in an institutional repository managed by the Center for Research and Assistance in Technology and Design of the State of Jalisco (CIATEJ), the academic institution responsible for the program. Information was gathered from evaluation and monitoring reports carried out by CIATEJ and the coordinating team—which are frequently requested for public government programs. Data used to structure the networks were taken from students’ research protocols (their final project) along with other documents and evidence (all likewise stored in the institutional repository). Information and data obtained and used for this study was authorized by CIATEJ.

All personal information related to students and participants, including identifiable individual data, is treated as strictly confidential and is not publicly available. Repository materials are not openly accessible; however, access to specific research outputs may be considered upon a reasonable request to CIATEJ’s Graduate Studies Office, subject to institutional authorization and applicable ethical, confidentiality, and data-protection requirements.

2.2 Study context: the PIES AGILES program and educational experience

PIES AGILES was designed based on specific epistemic, pedagogical, and educational principles. First, it positioned itself from a constructivist perspective, that is, placing the student at the center of knowledge creation and assuming a leading role in the learning process, which occurs through cooperation and social development (Vygotsky and Cole, 2018). To this end, Project Based Learning was the core teaching method, which allowed students to engage with real, collective and personally meaningful projects. This method was adopted after evaluating the first trimester, which was considered too conventional and with disarticulated disciplines, as well loaded with excessive content and activities. Drawing on the experience of an agroecology degree from a Venezuelan university (Domené-Painenao and Herrera, 2019), the program was restructured on Project Based Learning and the following trimesters were all focused on the development of a Territorial Action Plan as the student’s central project. Hence, course topics, assigned tasks and practical activities all contributed to the gradual development of this community-centered action plan/project, all the while addressing a real and significant problem in the food system through agroecology. Thus, learning and doing are merged, turning into intentional action, a concept supported by Piaget (Castorina et al., 1996), who proposed that all learning activity and intellectual processes are rooted in action. Knowing is only possible through the interaction between the conscious mind and the world, resulting in a capacity for understanding that stems from an interest that encompasses affectivity.

Another central element guiding PIES AGILES—a program that educated and motivated students to participate in and with the community based on their own experience—was popular education, understood as a political-pedagogical practice of resistance and transformation that brings to the fore indigenous, ancestral, and earth-based knowledge in order to build a different reality (Walsh, 2014), and in particular social change in Latin America (Jara, 2010). The program encouraged critical and political approaches in the assigned activities, and drew inspiration from the agroecological “formación” of social movements that not only train but also aim to construct a better human being through critical reflections and actions (McCune et al., 2014). Therefore, activities based on participatory action research were pivotal in advancing the logic of collective learning, as was the creation of learning communities to establish the program's educational and territorial structure. Learning communities are defined as organizational structures directed toward developing collaborative processes of knowledge co-creation to achieve shared goals (Merçon, 2021).

An additional distinctive characteristic of the program was its extensive geographic coverage: students and their learning communities were located in close to 300 different communities throughout the country, grouped and organized into regional nodes mainly in central and southern Mexico (Figure 1). Locations included villages, towns, and cities, ranging from coasts to valleys and mountains, and spanning different climatic zones, all of which reinforced the program's territorial and biocultural diversity.

Students took four subjects in each academic term: Political Agroecology, Agroecosystem Redesign, Community Communication, and Participatory Action Research Methodology, the core subject that served as an unifying thread to the entire academic program and guided the development of the student’s central project. To begin with, students conformed and/or structured a learning community, which worked alongside and accompanied them throughout the program, starting with a problematization process and the framing of a research question, thereby identifying a main shared problem. Together, a community action plan was co-designed, after which proposed activities were carried out and evaluated, adjustments were made and even the research question was adjusted and reformulated. Reflection and analysis were constant throughout the process. In the second year, the project scaled up, especially through connections with other projects, such as linking to other learning communities that shared similar action-research topics, thus forming thematic learning communities. This led to joint activities, both in person and online, including courses and workshops, webinars and virtual discussion forums, as well as the creation of collaborative networks among others.

The establishment and use of an online learning platform defined a learning pathway for designers, facilitators, and students alike. It enabled the mapping of each learning community’s journey, documenting evidence and findings, and systematizing results from both course-specific activities and the development of their community action plan—on a digital platform and repository. These activities took on various forms, including written documents, audio and video recordings, photographs and graphics, among others.

The educational program was structured into five trimesters, with the last one dedicated to the final project, consisting of a research protocol that included a critical and reflective systematization of the student’s learning experience, as well as a revised and updated version of the community action plan, now formally named the Territorial Action Plan. This plan had a broader territorial scope and required establishing connections with partners whose work and experience aligned with the research question, with the aim of outlining a roadmap for continued action in the long term, even beyond the program. It is worth mentioning that after completing the program, students could decide to apply to a following professional internship, and thus continue to implement and develop their Territorial Action Plan for nine additional months (with an equivalent stipend).

2.3 Network map: connections between action plans at the nation level

Information for this study included student research protocols, as well as institutional evaluation and monitoring reports that in many instances collected, analyzed and systematized data from the program. In order to construct and graph the network, a census sampling of all 287 research protocols was used (the units of analysis), from which data was obtained regarding: (i) the project's main research topic, (ii) the regional node it belonged to, and (iii) its alignment with Conahcyt’s National Strategic Programs (Pronaces, in Spanish), a series of ten priority issues and national challenges identified by the federal government (Secretaría de Ciencia, Humanidades, Tecnología e Innovación (Secihti), 2026): Toxic agents and polluting processes, Water, Culture, Education, Energy, and Climate change, Health, Human security, Socio-ecological systems, Food sovereignty, and Housing. Data was reviewed and curated to eliminate duplicates, standardize category nomenclature, and verify consistency across data fields. Based on this coding, a relational database of nodes and edges was created and subsequently exported to Gephi (vers. 1.0), a software application used to visualize, explore, and analyze complex networks. A multilevel network approach was chosen because the objective was not merely to represent relationships among students, but to visualize the simultaneous articulation of actors, territories, thematic agendas, and national research priorities.

Structural indicators for social networks were estimated and analyzed. Connectivity degree (degree centrality) accounted for the total number of direct connections or edges a specific node has; modularity facilitated the recognition of relatively cohesive thematic communities; and betweenness centrality identified nodes acting as bridges between different groups within the network. Density estimated the proportion of observed connections relative to the total possible; while diameter, average path length, and the number of connected components made it possible to assess the network's level of cohesion and integration. Visualization was carried out using the ForceAtlas2 algorithm with its standard settings to distribute nodes based on connection intensity, which allowed to identify high grade nodes (most connections), thematic clusters, and intermediary (bridging) nodes. This analysis simulates a physical system in which nodes behave like particles subject to opposing forces: elements with more connections tend to cluster together, whereas less connected groups are spatially apart, facilitating the identification of structural patterns that would be otherwise difficult to discern. Subsequently, the modularity algorithm was applied to identify structural communities, and descriptive metrics were calculated using Gephi's integrated tools. Qualitative data and information was used, particularly research protocols as well as activity and synthesis reports describing collaborations, to further interpret connections, structural indicators and network patterns.

3 Results: weaving connections and territorial networks

Within each regional node a diverse array of allies collaborated, including local actors, groups and institutions, as well as government programs, universities, research centers, and non-governmental organizations. Various learning communities were established, usually one per student but in a few cases more, as well as thematic learning communities among students that shared interests in action-research topics like soil fertility, maize, water management, and agrobiodiversity. Outcomes included collaborations for capacity building, as well as knowledge and experience exchanges during student gatherings, workshops and community events. Challenges ranged from overcoming geographical distances, individualism and personal conflicts, to navigating differing cultures and worldviews, competition and burnout; in urban contexts, a lack of land was reported as a significant obstacle.

Strategies proposed by students—especially collective strategies developed within learning communities—seemed to succeed in strengthening collaborative networks. This reinforcement had a positive impact on social organizations and educational institutions participating in these processes. The students' continuous engagement and support enriched these networks. Through an organized, iterative "back-and-forth" process, students promoted a constant state of action-reflection that favored learning by praxis. Thus, the collaborative networks in which they participated, combined with their understanding of local culture, enabled and strengthened their formative and educational process.

A remarkable dynamic process of territorial articulation and collaboration was apparently developed throughout the program, here represented as an ascending spiral of networking intensity structured in four levels (Figure 2). The first level focused on identifying and connecting with pre-existing networks in the territory. Subsequently, the second level links theoretical understanding with practice through situated learning, and the third level consolidates the construction of new networks and active collaboration among colleagues, partners and stakeholders. Finally, the linking process matures and ascends to the fourth level, where scaling, autonomy, and long-term sustainability of territorial alliances are achieved.

3.1 Network analysis

The network helped to visualize and understand the interplay between the program's educational and training processes, territorial dynamics, and the national research agenda. This led to a program and national-level thematic classification, with the following research topics: water, food and health, biocultural diversity, agrobiodiversity, communication, social and solidarity economy, education, feminist approaches, community gardens, agricultural pests, agroecological practices, productivity, seeds, soils, and the defense and care of the territory, all of which constituted an intermediate level that linked student action-research projects with on-the-ground issues.

The performed analysis yielded a multilevel network composed of 324 nodes and 1065 edges. The network graph (Figure 3) presents the four kinds of nodes: student projects, research topics, regional nodes, and the national strategic programs of Pronaces. The size of nodes is proportional to their degree of connection (centrality), while the colors correspond to recognized cohesive thematic communities (modularity). Food Sovereignty constitutes the largest thematic community, with 69 nodes (21.3% of the network), followed by Soils/Socio-ecological Systems (39 nodes; 12.0%), Education (34; 10.5%), Food and Health (24; 7.4%), and Biocultural Diversity/Culture (23; 7.1%). Edges (connections) were defined based on relationships of affiliation (student-topic), thematic linkage (topic-Pronaces), and territorial anchoring (student-regional node). Given that these connections represent associations and not information flows, the modeled network was non-directed and non-weighted, assigning the same weight to all connections in order to focus on links and not on their intensity (Table 1, includes other structural indicators).

Table 1. Network structural indicators derived from the Gephi software analysis.

The network exhibits a structure organized around a clearly prominent thematic community—food sovereignty—which constitutes the largest modularity-based community in the network (see Figure 4). This pattern corresponds to the program's original purpose, which as noted before, emerged in a political context promoting agroecological transitions and the strengthening of local food systems. In this regard, the centrality of this node reflects both the program's thematic orientation and its structural alignment with Pronaces. Organized around this core are several densely interconnected thematic clusters, particularly those relating to agroecology, seeds, soils, community practices, and social and solidarity economy. These clusters stem from the program's pedagogical approach, which is grounded in situated learning and collaborative work within learning communities. Since the projects originate from real-world issues within specific territories, it is to be expected that they would converge on structural aspects of the food system, thereby generating coherent thematic groupings. In terms of network analysis, these clusters exhibit a differentiated community structure, as reflected in the modularity analysis (Q = 0.445; Table 1). This suggests that student projects tend to group around related thematic areas while remaining connected through shared thematic, territorial, and programmatic links.

A key element explaining the graph's morphology is the role of thematic areas as bridging nodes. These connect multiple student nodes with different Pronaces, creating a hierarchical network structure. This pattern is directly linked to the program's curricular design, which promotes problem-based learning and interdisciplinarity. Thus, a single student can be linked to multiple themes and, through them, to different strategic axes. This helps explain the observed pattern of connections and the absence of fragmentation within the network.

The distribution of regional nodes is also noteworthy. In the network graph (Figure 3), these nodes appear integrated to the thematic clusters, which could indicate that the territory is not an external element but a constitutive part of the educational process: knowledge is produced within and from the territory. Furthermore, the variation in connection density among territorial nodes suggests processes of regional specialization, wherein certain territories concentrate a higher number of projects on specific themes. This is consistent with findings from the systematization process, which revealed differences in local capacities, actors, and issues.

From the territorial perspective, the network displays a clear differentiation in thematic focus, revealing specialization processes linked to specific regional and local contexts. For instance, the Jalisco-Nayarit regional node stands out for its strong orientation toward agroecology, social and solidarity economy, alternative food networks, and marketing processes. This results from its collaboration with universities, research centers, and local market actors, which promote both production as well as economic and organizational impacts. In contrast, the Veracruz regional node focuses on themes such as biocultural diversity, agroecology, community gardens, water issues, and traditional practices. These are closely tied to community processes, cultural diversity, and the recovery of local knowledge, all reinforced by the participation of indigenous communities and intergenerational dynamics.

On the other hand, the Puebla-Tlaxcala regional node is characterized by a focus on soils, biodiversity, agroecological practices, and community organization, linking production with rural educational processes and the strengthening of grassroots organizations. In the case of the Chiapas node, the network focuses on agroforestry systems, soils, and agroecological practices, grounded in strong community and family foundations where kinship ties play a central role in the continuity of these processes. Finally, the Mexico City-Morelos regional node presents a distinct configuration, centering on themes such as education, community communication, culture, and urban agroecology, reflecting its hybrid urban-rural character and its connection to cultural, educational, and social organization spaces.

The network (Figure 4) shows a lower edge density for certain Pronaces, such as Energy and Climate change, Housing and Human security, which appear more at the periphery. This uneven distribution can be interpreted at two levels. On one hand, it reflects the program's thematic priorities, which are more focused on food systems and the agroecological sphere. On the other hand, it reveals a potential disconnect between specific national strategic programs and the identified issues emerging from the territories where the students operate. This suggests a need to broaden the program's thematic scope to achieve greater coverage across the variety of Pronaces.

Finally, the observed pattern of connections together with the existence of intermediary nodes, provides structural evidence that suggests consistency with the study’s premise that the program contributes to the formation and articulation of collaborative networks. The network is characterized by multiple inter-node connections and the absence of completely disconnected components, indicating structural integration across local, regional, thematic, and national levels. Nonetheless, these structural properties should not be interpreted as evidence of uniformly strong or direct collaboration among all participants. Rather, results suggest that local learning-community networks are connected through thematic, territorial, and programmatic links, seeking to contribute to a broader network with national scope. To conclude, it is important to highlight that the program’s discourse has intentionally fostered—from its inception—the connections and types of bonds that students seek as subjects, i.e., active community and territorial participants, and ultimately promoted the scaling of agroecologies.

4 Discussion: the territory as a space for learning and connecting

The structure of this educational initiative transcends mere transmission of contents, to provide a space for reflection and action through dialogue among participants and local actors in the territory. The Participatory Action Research approach allowed theory to be linked to praxis, in a context of socio-environmental crises at different scales, and in which family and community play leading roles. This model foments critical inquiry and co-production of situated knowledge and is aligned with agroecological pedagogies that prioritize participatory praxis (Nicklay et al., 2023; Domené-Painenao et al., 2023).

Here, network connections are conceived not only as functional structures, but rather as spaces for collective learning, knowledge exchange, reciprocity, and building trust, all of which strengthen the social fabric and food sovereignty of territories (Rosset et al., 2021). One characteristic of this network structure is the presence of relatively short paths between nodes (maximum distance of five steps and mean path length of 2.6). These structural properties indicate that different components of the network can be reached through a limited number of intermediary relationships. However, path length should be understood as a structural property of the modeled network and does not, by itself, indicate the actual transfer of information, knowledge, or learning across territories. This agrees with observations by Granovetter (1983), who points to distant contacts (weak links) that bring external innovations to small communities. This theory could explain how distant learning communities such as some organizations manage to motivate social coordination, particularly in daily activities that seek to resolve problems and transform reality. Not being exempt from conflict and tension, these organizations nonetheless manage to encourage collective action (Zimmerman, 2004).

Learning communities were no doubt essential to weaving an initial network on a territorial level through their regional nodes, and subsequently to connect to each other around a common theme, which facilitated the development of specific thematic networks (thematic learning communities). The network’s structural relationships can be interpreted as opportunities for knowledge sharing and the exchange of experiences among territories, noting that network metrics describe the structure of these connections rather than directly measuring the actual exchange of knowledge and experiences. When prioritizing endogenous topics such as seeds, access to water, and care—which are common issues across different agricultural landscapes—dialogue is established among communities and communal groups. This approach is similar to that promoted by the Farmer-to-Farmer movement (Holt-Giménez, 2006), and allows for revitalizing indigenous and/or local knowledge and re-signifying daily life as a space for scientific and cultural production. A fundamental pillar of this approach is to yield the floor and listen to others, which grants power and agency to territorial subjects (Domené-Painenao et al., 2020). Following this premise, agroecological education becomes pluralistic and political, serving as a strategy of resistance by social movements (Meek and Tarlau, 2016).

Scalability and interconnection are foundational aspects to the network. Learning communities do not operate in an isolated manner, but rather possess a modular structure that allows linkages from local learning cells to regional and national networks. Articulation among learning communities goes beyond the mere management of a project-oriented educational process to address local food issues; it manifests the principle of knowledge co-creation, which constitutes an ontological pillar of agroecology (Utter et al., 2021). The transition of isolated experiences toward reticular structures reveal the configuration of localized food systems, where diverse social actors, collectives, and institutions converge in nodes of horizontal governance.

The program’s methodology, which involved participatory approaches, the dialogue of knowledge, and the democratization of voices, were also important. This experience defies the pedagogical design to move toward critical interculturality that may be one of the reasons for its success. As López Rivas and Limón Aguirre (2016) suggest, there is a need to promote an education that does not generalize, but rather raises awareness of gender and generational diversity (women, youth, children). This approach values “that which is our own” over “that which belongs to others,” seeking to reduce intergenerational gaps and recognizing knowledge systems at risk (of disappearing) that sustain relations of interdependence between humans and non-humans, and that promote alternative food systems.

Furthermore, this approach to learning based on the territory grants a central role to students, whose search for knowledge transcends technical-productive matters, also addressing political, social, and cultural issues. Thus, students were asked to play a double role: as actors with defined tasks within the institutionality, and as thinking territorial, collective, and historical subjects (Zemelman Merino, 2010), that alongside their organization adopted a political stance that allows for visualizing complex food systems (often marginalized by conventional academia). Thus, students were able to recover, for example, native varieties of maize and ancestral practices, through active listening and dialogic inquiry with elders; able to relate the communal logic with institutional interests like those represented by Pronaces. Clearly, the flow of information was facilitated when there were common themes and interests among stakeholders. Many of these students were experienced and committed to their organization, a key aspect considered in their application to the program. They resembled “‘militant researchers’, in reference to scholar-activism in knowledge politics (Borras and Franco, 2025), as professionals that generate positive knowledge through emancipatory processes.

Based on the leading role of students and their communities, we identify four distinctive properties in the program’s design that we believe favored the development of networks:

• Methodological transversality and integrative praxis: The design is composed of transversal axes, such as the Participatory Action Research seminar, which function as cross-cutting themes and nodes of curricular confluence. In contrast to rigid structures, this approach serves as a toolbox for different critical processes, like problematization, the development of action plans, and the systematization of experiences of community work. The integration of disciplines allows for activities in fields such as agroecology and community communication to produce tangible results with social impact, such as collective recipe books based on local knowledge and online platforms for territorial mapping and visualization.
• Flexibility and synchronicity with the time and own rhythms of communities: The academic structure is subordinated to climatic, agricultural, and festive cycles of culturally diverse territories. Such flexibility requires planning to coordinate the institutional calendar with the biocultural rhythms and cycles of communities, e.g., the periods to collect wild edible plants, harvest crops, and celebrate agro-religious festivals. The effectiveness of the program depends on its adaptability to the dynamics and lives of students and their learning communities as central local actors.
• Intergenerational inclusion and epistemic justice: the design promotes an educational approach which values people’s heritage and personal identity (López and Limón, 2016), taking into account gender, generational, and other differences. Focusing the dialogue of knowledge on the inclusion of elders and relationships with non-humans (Toledo, 2022), knowledge systems at risk are recovered and mutual awareness between diverse ways of understanding agriculture and life are fostered; this design demands the equitable redistribution of pedagogical power (those who learn and those who teach), as well as commitment to epistemic justice.
• Place-based education and the territory as an open classroom: The virtual space is grounded, territorialized, through real-world experience. The online learning platform does not function as a mere repository of information, but rather as a vehicle for decentralization that turns, e.g., maize fields, backyard gardens, and apiaries into legitimate classrooms. This model promotes learning based on the experiences of those who inhabit each place, like the way the Zapatista movement approaches community education through their “Educational Milpa” model (Sartorello, 2021). It is this “place” which allows subverting conventional academic hierarchies, thereby enabling remote communities (those distant to educational institutions) to take ownership and an active role in their own learning process, through their political and productive praxis, which also decentralizes science and its power structures.

In general, this study finds that the program’s characteristic design and structure not only facilitates students’ technical and methodological capacities, but can also contribute to the construction of collaborative networks—reinforcing the agroecological web that sustains processes of local advocacy and impact. Thus, networks, networking capacity and alliances are key educational outcomes, as well as guiding pedagogical principles and central tenets of the program. Collaborative networks, as shown to be promoted by the program, can support local decision making and the creation of collective strategies for scaling agroecology and food sovereignty initiatives.

5 Study limitations

This experience also posed challenges and demanded processes of reinvention. The management of an online distance degree with an extensive geographic coverage brought a wide range of internet connectivity issues, as well as difficulties for the faculty to be close and support different action-research projects. Furthermore, the researchers involved in the educational program were mainly agroecologists and experts in related fields, and not well trained in curricular development, teaching and online education. The team’s hyperactivity was a product of the multiple roles they needed to take on, resulting in exhaustion and little time and space for reflection processes. Despite these limitations, we were motivated by this learning experience—characterized by a combination of online and in-person interactions—that we found stimulated the co-creation of knowledge and strengthened tangible processes of agroecological grassroots organizations.

Finally, It is important to take into account the context and impacts of structural violence in Mexico (Salazar and Álvarez, 2018), including forced displacements promoted by corporate interests, organized crime, paramilitary groups, community defense forces, and domestic violence among others. These have impacted formative processes in many rural territories, and in some cases led to student drop-out, as well as to a lack of trust in the (institutional) program and its handling of sensitive information regarding their organizations. In a country where disappearances and murder rates of social leaders are among the highest in North and South America, it is essential to develop care and safety protocols. Therefore, carrying out this type of educational program required the recognition of multiple uncontrollable factors and uncertainties that affected the territorial educational processes.

6 Conclusion

This situated educational process aimed to contribute to strengthening collaborative networks and scaling transformative agroecological processes. One of the defining characteristics of experiences such as PIES AGILES, is their capacity to visibilize (make visible and aware) the territory’s potential—that which resists and exists, manifesting and shaping itself in the landscapes, knowledge and practices of its participants. More than opening spaces for horizontal dialogue, this allows to interweave various life experiences that gather strength within the diverse and complex areas of the Mexican territory.

We highlight the importance of students as collective historical subjects within this web of encounters and gatherings promoted by the program`s activities. This role seeks to transform the student into a mediator, a sort of bridge-builder and facilitator that connects local communities, institutions and organizations. Furthermore, being recognized and represented by a learning community can place students in a position of both vulnerability and advantage, mediated by power relations that, in turn, influence territorial transformations.

These collective forms of learning also have a social character: learning is conceived not as an exclusively individual process, but rather as a sociocultural interaction. Each individual simultaneously transforms their internalized knowledge in a back-and-forth communication of ideas—listening to others and coming to understand the benefits of addressing tasks and problems as a part of a team committed to collective efforts. In this context, creating and organizing learning communities is considered a key stage in the development of the student’s collective project. The community members or organizations that are part of these learning communities work alongside students, as companions, seeking to enliven the exchange of ideas and experiences, while contributing with formative and educational elements in an intercultural context. This is central to the program as it enables transdisciplinary collaborations, to provide significant, situated, and mutual learning experiences that can encourage the co-creation of knowledge and practices. Thus, the learning community creates a prominent action-research space that moves beyond individual goals, to focus on territories as collective scenarios where problems can be solved, decisions are made, and diverse forms of knowledge interact—through dialogues that bridge theory and local agroecological knowledge.

The structural and quantitative analysis of the multilevel network (324 nodes and 1065 edges) suggests that PIES AGILES operated as an effective networking mechanism for scaling agroecology and food sovereignty on a national level. The topology of the network—characterized by the configuration of a single connected component, a low diameter, and a short median distance among nodes—evidences the absence of isolated nodes and indicates an elevated structural proximity among participants, territories, and the national strategic agenda of Pronaces.

As a whole, the existence of thematic areas as bridging nodes and the structural integration observed across the network suggest that this educational program can not only build individual capacities, but also contribute to the articulation of local community networks into broader regional and national relational structures. This process involved the participation and engagement of students, learning communities, researchers, and institutions, particularly through decentralized interinstitutional collaboration involving a team of researchers and facilitators based at and supported by different academic institutions across the country.

PIES AGILES serves as an example of an educational proposal in Mexico in favor of agroecological transitions, open science, and universal access to knowledge. Through community engagement toward transformations, and the collective effort involved, it raises visibility and aims to strengthen and promote the emergence of networks on a national level.

The pedagogical design of this program seeks to alter conventional academic hierarchies by encouraging the dialogue of knowledge, active listening and yielding, and epistemic justice. In addition, the combination of flexible transversal axes and synchronicity with the community’s local time and rhythms, shows that virtuality may be territorialized in order to validate legitimate learning spaces, like maize fields, backyard gardens, and apiaries, for knowledge co-creation and political resistance in the face of the global crisis.

This study suggests that the construction of networks in the program transcends its instrumental function in education and local advocacy management, establishing and consolidating a pedagogical principle and a key outcome to promote the horizontality of sustainable food systems. The observed continuity in local community processes—evidenced in the persistence of alliances and facilitation processes in some regional nodes beyond the program—indicates that collaborative networks build permanent community capacities that are essential to the social fabric and for food sovereignty in the long run.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Karen L. Nordstrom, UVM Institute for Agroecology, United States

Reviewed by: Patricio Silva-Avila, Universidad de Concepción, Chile

Carlos Alberto Echeverría Mayorga, Universidad Alberto Masferrer, El Salvador

Footnotes

Center for Research and Assistance in Technology and Design of the State of Jalisco (CIATEJ), College of the Southern Border (ECOSUR), Center for Research and Higher Studies in Social Anthropology (CIESAS), and College of Michoacán (COLMICH).