From External Intervention to Endogenous Growth: A CAS-Based Analysis of Poverty Alleviation Mechanism with University Participation in Rural Collective Entrepreneurship

Abstract

Rural collective entrepreneurship poverty alleviation within the university participation context is regarded as a “socio-technical-economic” hybrid system, which aims to generate long-term economic benefits and social well-being for rural collectives through the knowledge of universities and realize the effect of poverty alleviation. However, the existing research has largely overlooked the dynamic mechanisms involved, especially how rural collectives transition from a passive response to a proactive creation in the context of university participation. Thus, we employ Complex Adaptive Systems (CAS) theory’s “detectors-IF/THEN rules-effectors” framework through a longitudinal case study. These findings demonstrate that (1) detectors have transitioned from “specialized knowledge embedding” to “diverse knowledge embedding,” which enables broader information scanning; (2) IF/THEN rules undergo cognitive destructuring to cognitive restructuring, fostering adaptive knowledge orchestration strategies; and (3) effectors shift from exploiting vertically related opportunities to horizontally related opportunities. (4) Cross-phase evolution: The knowledge flow mechanism of “knowledge spillover-organizational learning-knowledge absorption” propels “detectors, IF/THEN rules, and effectors” from the passive response phase to the proactive creation phase. This study advances theoretical understanding of CAS and research on entrepreneurship for poverty alleviation.

1. Introduction

Poverty has become one of the most pressing challenges for society in the world today. Rural regions in developing countries remain vulnerable to land degradation, food insecurity, and malnutrition. At the same time, rural regions in developed countries also suffer from great structural barriers to growth. There is a significant urban–rural divide and a narrow range of key industries in the economy. In 2024, there were about 8.5% of people around the world living below the international poverty line and getting by on less than $2.15 per day. Moreover, the incidence of extreme poverty is anticipated to experience a slight increase, rising from 8.4% in 2019 to 8.5% in 2024. This escalation in poverty presents a substantial threat to social cohesion and stability [1]. Therefore, it is unsurprising that scholarly inquiry has predominantly concentrated on addressing issues related to poverty. Scholars widely contend that entrepreneurship has the potential to alleviate poverty by providing rural regions with access to resources [2], training [3], and support that would otherwise be unavailable [4,5].

In the national innovation system, universities are generally considered essential knowledge producers that facilitate economic development through knowledge transfer, knowledge flow, and the transformation of knowledge into technological innovation for enterprises [6]. However, the knowledge frequently utilized aligns with or reinforces the interests of the existing elite power class while neglecting the needs of rural regions. Thus, the well-established academic tradition acknowledges that universities represent the standard pathway to technological innovation, which provides local employment and prepares students for jobs to accelerate entrepreneurship in urban settings, and poverty is frequently concentrated in rural regions [7], where they confront more unique entrepreneurship challenges relating to geographical [8], institutional [9], and market access conditions than urban regions [10]. Although the participation of multiple actors is widely recognized as a key factor for the development of rural regions, extant research on the understanding of how universities function in rural regions is scarce [11]. Previously, universities participated in entrepreneurship that focused more on commercializing innovations, establishing incubators, and joint ventures with commercial companies [12]. However, in the context of advancing entrepreneurship for poverty alleviation, further investigation is required to elucidate the system’s mechanisms for universities to participate in promoting further rural development.

As the world’s largest developing country with the most extensive rural population, China presents a critical and significant case for examining rural development. In 2020, the Chinese government issued policies emphasizing that scientific and technological development should address the country’s major needs and promote the health and well-being of its people. This approach expands the regional mission of universities beyond metropolitan centers to support rural economic development. Rural regions are inherently complex systems composed of farmers, cooperatives, agribusinesses, and other entities [13]. These actors interact within specific natural, economic, and social environments, forming relatively stable structural relationships and operational logics. Universities, as external knowledge providers, drive multidimensional transformation within rural systems through systematic knowledge integration. This knowledge-driven systemic intervention not only alters the internal knowledge orchestration rules and collective cognitive schemas within the rural system but also breaks existing path dependencies. It propels rural systems from closed to open, from fragile to resilient, ultimately achieving a comprehensive leap in both systemic function and structure. Consequently, this study views rural areas engaged with universities as complex “socio-technical-economic” hybrid systems, reflecting the interactions of rural collectives within university participation [14]. Given this, we employ the “stimulus-response” model from Complex Adaptive Systems (CAS) theory to investigate how rural regions in Tongyu County, China, can achieve poverty alleviation through entrepreneurship with university participation. This study aims to answer two key questions:

RQ1: What are the detectors, IF/THEN rules, and effectors within the context of university participation in rural collective entrepreneurship for poverty alleviation?

RQ2: What are the dynamic evolutionary mechanisms within the context of university participation in rural collective entrepreneurship for poverty alleviation?

For methodical data coding and analysis, we used NVivo software (Version 12.0), which was based on the two research questions mentioned above. Through this process, we identified related patterns among factors. The research provides substantial theoretical and empirical contributions, summarized as follows: First, this study presents a novel theoretical viewpoint by examining how rural collective entrepreneurship could alleviate poverty within a context of university participation, a topic that has not been sufficiently explored. Second, this study broadens the research scope of university participation by transitioning the analytical emphasis from metropolitan-centric paradigms to rural development situations and provides practical guidance to industrialized economies aiming to execute poverty alleviation via rural collective entrepreneurship, thereby enhancing our understanding of rural revitalization efforts. Third, this research provides a more thorough comprehension of system intervention and adaptation by emphasizing the importance of knowledge flow as a catalyst for adaptive behavior, thereby contributing theoretically to the “stimulus-response” model in CAS theory.

This study is structured as follows: Section 2 provides a literature review of research on entrepreneurship and poverty alleviation, university and rural development, and research gaps. Section 3 describes the research design, Section 4 presents the case analysis, Section 5 discusses the case findings, and Section 6 summarizes the research conclusions and implications.

2. Literature Review and Analysis Framework

2.1. Entrepreneurship and Poverty Alleviation

Traditional development economics emphasizes capital accumulation, productivity gains, and economies of scale as macro-drivers of regional progress, but this perspective ignores entrepreneurial systemic consequences and hinders the sustainability of value creation. A burgeoning academic consensus acknowledges entrepreneurship as an effective solution to poverty alleviation; some typical literature is shown in Table 1.

Table 1. Recent literature on entrepreneurship poverty alleviation.

Research Contents	Research Method
Entrepreneurship can alleviate poverty through remediation (accessing key resources), reform (institutional change), revolutionary (changing the capitalist system and society) [3].	Literature Review
Regional development through collective agency (entrepreneurial exaptation) in rural Sweden [15].	Ethnographic
Social cooperation and Self-Help-Groups in rural regions stimulate entrepreneurship and economic development [16].	Modeling Simulation
Collective entrepreneurship can alleviate poverty by helping poor individuals integrate into societies [2].	Case Study
Microentrepreneurs in rural regions can promote cluster formation and economic development through collective entrepreneurial action in weak institutional environment [17].	Case Study
Digital financial inclusion in rural regions reduces income volatility by promoting household entrepreneurial activities [18].	Empirical Study
BOP entrepreneurs can build social capital to escape poverty [19].	Case Study

At its core, entrepreneurial poverty alleviation relies on the identification, evaluation, and exploitation of entrepreneurial opportunities through multi-stakeholder participation within rural systems; it transcends mere individual venture creation, as addressing the needs of vulnerable regions requires collective entrepreneurial action rather than isolated efforts. Collective entrepreneurship integrates various actors into unified entities, delineating participant roles and managing shared objectives [20]. Such an approach allows resource-constrained entrepreneurs to obtain financing and resources that would otherwise be unattainable, while collaborative knowledge enhances the recognition and exploitation of opportunities beyond the individual [2].

Entrepreneurial opportunity is defined as the integration of new resources by entrepreneurs to introduce new goods, services, and raw materials for value creation and is at the heart of entrepreneurship research [21]. Prevailing academic research predominantly adopts a “single-opportunity” perspective, overlooking the inherent relatedness of rural entrepreneurial opportunities (i.e., how initial opportunities shape sequential ones). Rural entrepreneurship opportunities demonstrate two key dimensions of relatedness characteristics:

(1) Structural-related characteristics, where they involve vertical industrial chain integration (e.g., synergistic coordination between crop valorization, production, processing, and distribution) and horizontal resource sharing (e.g., cross-sectoral technology applications);

(2) Temporal-related characteristics, where initial phase opportunity exploitation generates knowledge spillovers that influence subsequent phase opportunities (e.g., agricultural technology innovations cascading into improved production, processing, and marketing process techniques).

Rural entrepreneurial activities are characterized by collective collaboration among agribusinesses, farmers, and cooperatives [22], and exploiting collective entrepreneurial opportunities in rural regions confronts systemic constraints such as weak resistance to risk, low return on investment, natural constraints, and a monolithic industrial structure, which require the intervention of external actors to mitigate structural barriers [23].

2.2. University and Rural Development

Universities are perceived as “engines of economic growth” and “anchors” of regional innovation processes owing to their deep societal embeddedness [24,25]. Therefore, universities are required to address the socio-economic demands of marginalized populations while additionally advancing regional development alongside their traditional core educational and research missions [26,27].

Rural regions suffer from an abundance of challenges via their infrastructure and social trust [28], which affect their economic activities. However, rural contexts that have been historically neglected in entrepreneurship research are now receiving more attention regarding the application of scientific knowledge (public goods) to benefit these regions [29]. Although local government policies highlight the dual role of universities as both knowledge producers and educators in rural settings, it is clear that rural systems function through distinctive logics. Opportunities for collective entrepreneurship and poverty alleviation remain constrained by rural geographical isolation, divergent stakeholder interests, cognitive barriers, and limited prior collaborative experience, all of which factors impede university participation [30]. Nevertheless, the dual mandate of fostering regional economic development and addressing persistent social challenges in rural areas empowers universities [31]. The knowledge-driven regional development is achieved through complex, nonlinear pathways that directly link scientific research outcomes to societal impact. However, the inherent cross-level complexities between universities and rural systems, such as cognitive gaps between rural farmers and university scientists at the micro level and the misalignment between universities” “knowledge supply” (scientific knowledge) and rural collectives’ “knowledge demand” (practical knowledge) at the meso level, result in paradoxical tensions that often lead to resource misallocation. Thus, the implementation of these initiatives involves intricate processes of knowledge co-creation, requiring both top-down government policy incentives for universities to participate in rural region development and bottom-up collaborative knowledge creation [32].

In sum, knowledge asymmetries between expertise producers (especially universities) and target rural regions impede crucial knowledge flows between urban and rural areas. Current research has overlooked universities’ pivotal role as knowledge agents and the initiative of local rural actors in this process [33]. Considering that government promotes university participate in collective entrepreneurship, they can incorporate their specialized and multidisciplinary knowledge to utilize local resources [34], which meets the demands of the rural region’s development and stimulates rural collective endogenous development [35]. It is therefore required to investigate the system mechanism through which government promotes university participation in collective entrepreneurship drives poverty alleviation (see Figure 1).

Figure 1. Key actors within the rural system and their interconnections.

2.3. Research Gaps

Current literature on the relationship between university knowledge and rural endogenous development exhibits three limitations, as outlined below:

First, the linear conception of knowledge constrains existing research’s understanding of endogenous growth. Existing research predominantly adopts the neoclassical and endogenous growth perspective, treating knowledge as a generic factor of production within a linear “research-development-dissemination” model [27]. This approach typically views universities as knowledge providers and rural regions as passive recipients, thereby overlooking the agency of local actors-such as key opinion leaders-in the knowledge absorption process. Therefore, this study fails to reveal how the embedding of university knowledge reshapes cognitive patterns and schemas in rural regions, thereby influencing knowledge orchestration behavioral shifts. The essence of endogenous growth lies in local actors’ localized adaptation, absorption, and creative application of external knowledge [36]. The linear perspective of existing research obscures this nonlinear feedback process driven by cognitive activation and knowledge-based behavior [37].

Second, the static “structure-function” perspective struggles to capture the evolution of the system. Much of the current research employs a static lens [38], focusing on identifying participating actors, such as universities, agribusinesses, and their predefined roles [39]. However, it generally fails to draw on Complex Adaptive Systems (CAS) theory to interpret the processes of system self-organization and evolution. While this static lens can describe the state of a system at a specific point (e.g., “who is doing what”), it cannot explain how internal structural relationships and functions change over time [40]. For instance, it does not elucidate how a university might transition from an external “leader” to a “collaborator” within the rural system, or how agribusinesses evolve from a passive “technology adopter” into an “opportunity co-developer” with endogenous capacities. Such fundamental shifts in roles and relationships are a central symbol of how external knowledge input from universities catalyzes endogenous growth in rural regions. Consequently, the static lens in existing research obscures the evolutionary mechanisms of key variables along the pathway from passive response to proactive creation by rural actors. This limitation ultimately prevents the answer to the fundamental question: how can a rural region initially driven by external university knowledge ultimately achieve endogenous development?

Third, the failure to unpack the micro-level “mechanisms of action” has resulted in the endogenous growth process remaining largely unexamined [41]. Much of the literature has mainly focused on establishing a correlation between the input of university knowledge and rural economic outcomes [42], without revealing the micro-level mechanisms that transform external intervention into endogenous growth. While researchers have identified the “input”-the embeddedness of university knowledge-and the “output” -endogenous growth, they have not unraveled the intricate mechanisms that link external interventions to the formation of endogenous growth. In particular, the literature fails to clarify the causal chain and nonlinear interactions through which university knowledge reshapes local actors’ cognitive models, thereby stimulating their knowledge orchestration behaviors and ultimately driving the exploitation of entrepreneurial opportunities. A deeper analysis of these micro-level cognitive models and knowledge-related actions is necessary to explain how externally embedded resources are internally transformed to generate sustainable, endogenous motivation for development within rural regions.

In response to the three limitations, this study uses the “stimulus-response” model of CAS theory as theoretical foundation, which was chosen for the following reasons:

First, in existing rural studies, the term “system” is frequently mentioned (see Table 2). Here, an agricultural system is defined as encompassing the knowledge, practices, and contextual factors of rural stakeholders. With the evolution of innovation theory, this systemic perspective has developed into the more structured concept of the Agricultural Innovation System (AIS), defined as “a network of organizations and individuals focused on bringing new products and processes into economic use, together with the institutions and policies that shape their interactions and knowledge flows” [43]. This perspective explicitly considers innovation as the result of interactions among various actors. Recently, the research focus has broadened from agriculture to rural regions more generally, giving rise to the concept of the Rural Innovation Ecosystem (RIE). This is understood as a complex socio-economic system composed of diverse organizations-such as agribusinesses, cooperatives, universities, and government agencies-and individuals, including farmers, interconnected through specific relationships, collaborations, and network patterns [44,45]. The RIE emphasizes understanding patterns of actor interaction in innovation, the role of innovation policy, and the functions of innovation support, such as research and extension services [46]. The conceptual progression from AIS to RIE indicates a scholarly consensus that views these as CAS networks of multiple agents exhibiting self-organization and co-evolution through nonlinear interactions. According to CAS theory, such systems evolve through rule interactions between actors and their environments [47]. In these systems, multiple interdependent components engage in nonlinear interactions that generate emergent properties; this means that understanding individual elements enhances comprehension of systemic behavior [9]. When universities have embedded knowledge in rural systems, the rural region demonstrates essential characteristics of CAS. Diverse stakeholders, such as farmers, cooperatives, agribusinesses, government agencies, and research institutions, participate in complex, nonlinear interactions. These interactions promote top-down collaborations and bottom-up opportunity exploitation. Consequently, the rural system goes through ongoing learning processes and knowledge flows among actors [48], generating reinforcing feedback loops that cumulatively drive its structural and functional evolution. While existing theories of university–enterprise cooperation and endogenous development acknowledge the importance of knowledge embeddedness and multi-stakeholder structures, they largely rely on a static “structure-function” analysis. Such a perspective describes “who does what” but fails to capture the dynamic and evolutionary essence of CAS. As a result, current research cannot adequately explain the micro-level mechanisms-such as how embedded university knowledge reshapes local cognitive models and triggers behavioral shifts in knowledge orchestration, nor can it trace the macro-level pathways through which these interactions foster the exploitation of entrepreneurial opportunities, thereby coalescing into sustained endogenous developmental motivation. Thus, our research addresses this gap by integrating CAS theory to reveal how universities’ knowledge embeddedness is transformed into endogenous development motivation within rural regions.

Table 2. Recent literature on rural system.

Research Contents	Research Method
Smallholder systems as local integrated human-natural systems comprising smallholder actors [49].	Case Study
The rural system is a coupled social-ecological system composed of a resources subsystem and actor system [50].	Empirical Study
The rural innovation system is a complex open socio-economic ecosystem for rural revitalization and development [28].	Qualitative Analysis
The rural system is an open system defined by distinct structures and functions that arise from the interaction between its natural environment, resource endowment, and other factors [51].	Case Study
The construction of a system includes system arrangement (behavioral rules) and system structure (particular scope) [52].	Empirical Study
The circular agricultural production system seeks to ensure the economic, environmental, and social sustainability of agriculture via efficient resource utilization throughout the value chain [53].	Empirical Study
The agricultural innovation system is a collaborative network of actors and organizations that co-create agricultural innovations (products, processes) [54].	Qualitative Analysis
A rural innovation ecosystem is a network of interdependent actors co-evolving through collaboration to achieve shared value propositions for rural development [44].	Qualitative Analysis
A Rural Entrepreneurial Ecosystem refers to the synergistic network of actors and institutions that facilitates the discovery, evaluation, and exploitation of entrepreneurial opportunities [55].	Qualitative Analysis

Second, the “stimulus-response” model of CAS theory can offer insights for analyzing mechanisms for alleviating poverty through collective entrepreneurship in the contexts of university participation, and the model includes three fundamental components:

(1)

Detectors receive environmental stimuli, processing inputs to facilitate information perception;

(2)

IF/THEN rules regulate actors’ behavioral responses subsequent to information processing;

(3)

Effectors execute actions that alter actor states and environmental conditions.

Extending this logic to rural collective entrepreneurship, we establish an integrative framework to clarify how knowledge is both a strategic asset and a catalytic driver in systemic adaptive evolution processes. Specifically, the government’s bottom-up policy promotion for university participation functions as a detector that categorizes and interprets knowledge signals. This cognitive schema transformation reshapes rural actors’ perceptions of academic knowledge through dynamic evolution, which systematically guides their knowledge orchestration rules. The resultant behavioral (effectors) manifest as related opportunity exploitation, generating feedback loops wherein (1) top-down implementation: government-driven universities participate in disseminating scientific knowledge and shaping rural actors’ adaptation cognition, knowledge orchestration strategies, and related opportunity exploitation; and (2) bottom-up feedback: rural collectives’ bottom-up knowledge flow enables dynamic assessment of knowledge orchestration efficacy through detector adaptation. This integrated perspective reveals the complex linkages between cognition scheme, knowledge orchestration, and entrepreneurial opportunities within both bottom-up and top-down multilevel adaptation networks involving government, universities, and agriculture stakeholders, ultimately reflecting the CAS principle of adaptation and complexity.

3. Research Design

3.1. Case Selection: Tongyu County, Jilin Province, China

Tongyu County is located in Baicheng City, Jilin Province, on the eastern border of the Horqin Grassland. The county covers an area of 8496 square kilometers; the county comprises 16 administrative townships with a registered population exceeding 360,000 (see Figure 2).

Figure 2. Tongyu County administrative region map.

The region is located in the hinterland of the Songnen Plain, with flat and open terrain, but more than 65% of the land is saline. Such challenging natural conditions, combined with systemic constraints in agricultural production (lack of scientific and rational farming systems, delayed technological implementation, and a rudimentary industrial chain), have resulted in the long-term development paradox of “planting a wide range of crops for a thin income.”

Confronted with dual challenges, China’s Ministry of Education promotes China Agricultural University (CAU), Jilin University (JLU), and other institutions to help Tongyu County, leveraging scientific and technological support to advance high-yield crop technology and saline-alkali soil remediation. This initiative aims to facilitate the start of collective entrepreneurship among agribusinesses, cooperatives, and farmers, thereby establishing a sustainable development path characterized by “saline governance-high-yield crop cultivation-deep processing of agricultural products” and constructing a new paradigm of rural revitalization with the trinity of “environmental restoration-agricultural efficiency enhancement-industrial upgrading.” By 2024, the GDP is reach 11.03 billion yuan; this case serves as a typical example of the rapid development of ecologically fragile regions in the Northeast within the university participation context.

The two reasons why Tongyu County was chosen as the research case are that:

(1) Typicality. The case illustrates how collective entrepreneurship opportunities are exploited within the context of university participation. This poverty alleviation initiative was selected as the “Seventh Model Project of Targeted Poverty Alleviation Innovation” by China’s Ministry of Education and has gained social recognition.

(2) Data Availability. Tongyu County XYF has achieved rapid growth through the exploitation of entrepreneurial opportunities within collaborative participation with Jilin University (JLU) and China Agricultural University (CAU). Data for this study can be drawn from two key sources: (1) comprehensive secondary data available through relevant institutional platforms and (2) primary data collected via in-depth interviews with key stakeholders-including university researchers, Tongyu County firm managers, participating farmers, and agricultural cooperatives. Furthermore, since 2016, one co-author has conducted regular field research in the Tongyu County, providing an extensive firsthand understanding of this case.

3.2. Research Methods and Data Collection

Considering that the case study is appropriate for exploring the process-focused and mechanism-driven questions. Thus, this study employs a case study methodology to examine what the poverty alleviation mechanisms are with university participation in rural collective entrepreneurship (“what” and “how”). This study provides a detailed analysis of particular processes, acknowledging that the complexity of rural collective entrepreneurship mechanisms for poverty alleviation is inseparable from the reality of the situation.

Data collection employed a combination of semi-structured interviews and document analysis, which was conducted in three phases (October 2022, March 2024, and July 2025). The sampling strategy employs a theoretical sampling approach, which focuses on individuals who have participated in university-rural cooperation projects for a minimum of two years and have been pivotal in collective entrepreneurial activities. The final participant group comprised university technical staff, rural revitalization office managers, agribusiness representatives, and key farmers who acted as opinion leaders-all of whom were pivotal to understanding the poverty alleviation mechanisms with university participation in rural collective entrepreneurship. Through this methodological design, the study accumulated 469 min of interview recordings, resulting in over 640,000 words of transcript data. These primary sources, along with secondary materials such as university websites, publications, and academic papers indexed in CNKI (see Table 3), enabled comprehensive data triangulation. This methodology builds a solid data foundation for investigating the endogenous transformation mechanism.

Table 3. Case data.

Data Type	Interview Information			Code
Primary Data	Interviews	University technical support staffs	151 min	FT1; FT2; FT3
		Rural revitalization office Manager	97 min	FT4
		Tongyu County farmers and firm managers	84 min	FT5
	Video interviews		137 min	FT6
	Official website		23 articles	A1
Secondary data	Books		3 documents	A2
	CNKI paper		3 articles	A3
	Third-party mainstream media		13 articles	A4
	Other materials (e.g., WeChat, small programs, etc.)		17 articles	A5
	Total number of words		64,035 words

3.3. Data Coding Procedure

This paper’s data coding procedure is implemented as follows (see Figure 3):

Figure 3. Data coding procedure.

First, a coding team was formed: To minimize subjective bias in coding, the two authors jointly formed this team. During the coding process, the two coders carefully reviewed materials to eliminate items unrelated to the research topic and explained the problematic components through discussion, thus reducing potential errors.

Next, data presentation was conducted: We employed the first-order, second-order structured data analysis methodology [56], providing a systematic analytical framework for building a dynamic theory model with qualitative rigor. Adhering to the principle of “letting the data speak for itself,” two coders continuously compared raw data with emerging concepts, ultimately aggregating core concepts. To coordinate different interpretations, the research team used two methods: internal discussions focused on achieving interpretive consensus through iterative data analysis and debate, and external expert interaction to analyze and strengthen conceptual connections.

Then, conclusion and validation: Using initial codes and categories, employing inductive reasoning and constant comparison to evaluate statements, identifying common elements and universal characteristics. Continuously analyze existing theories, codes, and new data to clarify the relationships between categories [57].

Finally, to achieve theoretical saturation, four entrepreneurship scholars and four management PhDs reviewed the validated theoretical model until they reached theoretical saturation. The integrity of data coding adheres to qualitative indicators:

(1)

Identifiability: Employ transparent analytical procedures (including graphics) to ensure the identifiability of research conclusions.

(2)

Reliability: Guaranteed by standardized coding techniques and inter-coder consistency (>95%).

(3)

Trustworthiness: Enhanced by using three different types of data-interview data, archival records, and field observations-to make the research more valid.

(4)

Transferability: Contextual descriptions spanning temporal and geographical aspects provide a comparative framework and applicability for similar rural settings.

This study adopted a sequential coding process consisting of open coding, axial coding, and selective coding, followed by a cycle including source material, labels, concepts, themes, and dimensions, resulting in a three-level data structure (see Figure 4).

Figure 4. Schematic representation of data analysis and coding.

4. Case Analysis

This study divides the knowledge embedding role of universities in the agribusiness entrepreneurial opportunity exploitation process into a passive response phase and a proactive creation phase based on critical events within the context of university participation, and this periodization was endorsed by four entrepreneurship scholars and four management PhD candidates, as seen in Figure 5.

Figure 5. The key events of Tongyu collective entrepreneurship poverty alleviation within the university participation context.

The passive response phase (2016–2017) indicates that Tongyu rural collectives were impeded by ingrained cognitive imprints and fear of failure and exhibited reluctance to engage in technological innovation or modify market strategy. As a result, they were only forced to passively accept the Science and Technology Courtyard model due to survival pressure after witnessing its success at CAU in Quzhou, Hebei Province.

This stage is followed by the proactive creation phase (2018–2024) in Tongyu. In this phase, the behavioral dynamics of the rural collectives shifted from external force to internally spontaneous actions. They actively pursued transformation, took the initiative to seek saline land restoration and deep processing technical support, and cooperated with Jilin University to create an engineering research center and jointly invest in the development of saline land and saline-tolerant soybeans to create a brand of agricultural products.

4.1. The Passive Response Phase

4.1.1. Detectors: Specialized Knowledge Embedding

Initially, CAU established an STC in Quzhou County, Hebei Province, to gradually promote the transformation of the traditional outcome-oriented development model of agriculture to a problem-oriented one. The XYF firm’s managers in Tongyu observed the STC’s “Four Zeros” in Quzhou:

(1)

Zero-distance contact (technicians permanently stationed at agricultural production sites);

(2)

Zero-threshold access (unrestricted access to services);

(3)

Zero delayed responsiveness (real-time problem solving);

(4)

Zero-cost delivery (full-service delivery provided at no cost).

The effectiveness of the implementation of this model led to the expansion of STC to other townships in Tongyu, where it primarily achieved two objectives: (1) identifying appropriate crop needs based on saline texture conditions, and (2) deconstructing entrenched cognitive schema through university-embedded technology and market knowledge.

4.1.2. IF/THEN Rules: Cognitive Deconstruction-Dependent Knowledge Orchestration

Engaging with university expertise as an effector begins to deconstruct the cognitive schema of local firm managers. On one hand, risk-averse managers tend to focus on adapting existing markets for salinity-tolerant agricultural products to fit with the university’s specialized technical knowledge. On the other hand, Tongyu’s agroecological constraints, which include a persistent drought cycle and distinct soil conditions of “eastern alkali and western sand,” have long formed a risk-averse, intuition-based cognitive model of agriculture. This manifests in a deeply ingrained cognitive model of “extensive cultivation with low yields” among agricultural stakeholders. While rural firm owners recognized village-enterprise linkage as a potentially effective way for STC development, persistent skepticism toward scientific knowledge among local farmers necessitated empirical validation through field demonstrations. Consequently, Tongyu’s farmers, cooperatives, agribusinesses, and CAU decided to do a pilot test in Wujingzi Village, Wulanhua Township. In summary, the government’s top-down policy ultimately deconstructed the content and model of rural collective cognition in Tongyu by supporting university participation in rural regions and providing access to knowledge resources for collective entrepreneurship.

To validate the effectiveness of university expertise and address initial skepticism, the Tongyu rural collective strategically selected saline-tolerant maize as the primary crop for verification. A clear market demand motivated this decision, prompting the search for high-yield cultivation techniques. Thus, the Tongyu rural collective obtained soil testing and customized fertilization techniques from CAU. Subsequently, through targeted demonstration trials involving model farmers (e.g., Han Xiaoxian) in Wujingzi Village and guided by STC experts, the effects of collective knowledge recombination creation in rural regions with university participation were empirically verified. Meanwhile, Tongyu’s rural collective also reused agrometeorological knowledge, which encompasses the 24 solar terms and the proverb “Planting Corn at Qingming Festival, Flowering at Grain Rain Festival,” to determine optimal sowing schedules for salt-tolerant maize varieties, thereby meeting their specific growth requirements and identifying initial opportunities to cultivate salinity-tolerant maize.

Overall, the main characteristic of knowledge orchestration in this phase was Tongyu’s reliance on CAU-embedded specialized expertise in high-yield cultivation technologies for maize markets. This dependency knowledge orchestration strengthened the depth and breadth of subsequent collaborative efforts between CAU and Tongyu’s rural collective in exploring technological and market knowledge. Motivated by current maize market demand for technological supply, they identified initial opportunities to cultivate salinity-tolerant maize by combining recombinant high-yield cultivation technologies with reused indigenous agrometeorological knowledge. This strategy challenged the long-standing rule of knowledge orchestration, which stated that saline land should only be used for extensive cultivation with low yields. After these technologies were disseminated, the strategy expanded to other regions through guanxi-based networks that utilized interpersonal trust and local institutional ties.

4.1.3. Effectors: Vertical-Related Opportunities Exploitation

Building on the initial exploitation of opportunity (salinity-tolerant maize), the university and the Tongyu Rural Collective systematically identified related opportunities along the maize value chain. Upstream efforts focused on the introduction of improved seed varieties and the establishment of the Wujingzi Farmers’ Specialized Planting Cooperative. These vertical-related opportunities activated the multiplier effect of knowledge diffusion, and “word-of-mouth” demonstration among cooperative members facilitated wider adoption so as to enable more farmers to benefit from “rice with technology.” Downstream, corn was utilized as a raw material to develop new products such as “Wujingzi” corn ballast and green storage feed, further expanding vertical-related opportunities and enhancing the entire corn value chain.

During the passive response phase, the relational dynamics were characterized by government-driven institutional interventions, which facilitated the top-down embedding of university-specialized technologies and market knowledge (serving as “detectors”) and disrupted the Tongyu rural collective’s traditional “market-pull” cognitive model and its market-oriented cognitive content. This cognitive deconstruction enabled dependent knowledge orchestration, mediated by STC experts. Functioning through market-responsive IF/THEN rules, the model generates symbiotic dynamics by combining knowledge recombination and reuse mechanisms, ultimately establishing novel knowledge orchestration rules for agricultural modernization. As collaboration deepens, universities and rural collectives are shifting from exploiting initial opportunities to expanding into related activities upstream and downstream (acting as “effectors”). This transformation generates a new “points-to-lines” integration effect, linking separate opportunities into interconnected value chains. The resulting knowledge flows are reinforced through bottom-up feedback mechanisms, which facilitate the embedding of university-driven knowledge in the proactive active creation phase.

4.2. The Proactive Creation Phase

4.2.1. Detectors: Diversified Knowledge Embedding

Triggered by vertical-related opportunities, Tongyu’s STC has aggregated a diverse range of knowledge. After demonstration experiments with technology scenarios, the government has promoted the top-down participation of universities such as JLU in the development of the Tongyu region. This initiative has also encouraged the participation of farmers, cooperatives, and enterprises in Zhanyu Township, adjacent to the Wulanhua area. By integrating interdisciplinary knowledge, the universities, along with the STC and Tongyu rural collective, are planning to develop appropriate new crop categories. This process creates a large-scale “knowledge aggregation box” that not only catalyzes qualitative knowledge transitions through quantitative accumulation but also simultaneously induces emergent cross-border market behaviors.

4.2.2. IF/THEN Rules: Cognitive Reconstruction-Interactive Knowledge Orchestration

University-embedded knowledge has reconstructed the cognitive scheme of rural collectives in Tongyu. Through collaborative entrepreneurship, they have realized the benefits of technology, which prompted them to take advantage of STC’s multidisciplinary technologies to explore new market opportunities. Meanwhile, their cognition scheme has undergone a restructuring: intuitive thinking evolved into a dual mode combining intuition with analytical reasoning. Following the learning curve and cost principle, this new model enables them to evaluate novel salt-tolerant crops more systematically. Moreover, they have transitioned from a technology supply model driven by market demand to a dual mode that matches the technology and market. In collaboration with experts from JLU, Tongyu’s rural collectives critically assessed the feasibility of developing a market for improving saline soil texture through scientific technology.

The restructured cognition of Tongyu Rural collectives enabled them to assess the effectiveness of interactive knowledge recombination with JLU and CAU. The team’s goal was to determine the relationship between market demand and technological supply. Specifically, whether market demand stimulates technological supply or whether technological supply propels market demand, thereby creating synergies between technology and markets. In cases where market demand stimulated technological supply, Professor Gao’s team from JLU partnered with Tongyu’s rural collectives to experimentally validate solutions for saline soil and straw waste treatment. Through iterative near-critical water experiments and continuous knowledge recombination, creation, and reuse, they established a shared consensus that near-critical water technology meets the needs of the waste management market.

In terms of technological supply propelling market demand, Professor Zhang’s team from JLU and Tongyu’s rural collectives jointly advanced the market for deep processing of agricultural products. Such an outcome was achieved through the recombination creation of technical knowledge related to deep processing technology and food science, as well as the reuse of high-yield cultivation techniques. Thus, within the logic of co-evolution between technological supply and market demand, Tongyu’s rural collectives, JLU, and CAU transformed the inherent constraints of saline-alkali land into structural market barriers (e.g., specific pedo-climatic requirements for premium crops). They identified two initial opportunities: the remediation of saline-alkali soil using near-critical water technology and the extension of the industrial chain through the application of deep processing technologies. These initiatives aim to reconfigure the principles of knowledge orchestration, converting production constraints into a differentiated competitive advantage that establishes a defensible market position.

4.2.3. Effectors: Horizontal-Related Opportunities Exploitation

Within the context of university participation, rural collectives have identified the shared technical demands between initial opportunities and other markets or domains, thus enabling the exploitation of horizontally related opportunities. On one hand, they have applied near-critical water technologies to remediate saline-alkali soils and converted straw waste into nutrient-rich materials, benefiting industries such as solid waste management, the fruit cultivation industry, and the tea cultivation industry, while also facilitating the establishment of a biomass-improved soil and modern green agriculture engineering research center. This center promotes the agricultural value chain from the “field” to expand to the “dinner table.” On the other hand, corn was deep-processed into grain pancakes and cookies, contributing to the grain processing industry. Similarly, rice was deep-processed into weakly alkaline rice, supporting the rice processing sector. Soybeans were deeply processed into many products, including protein powder, meal replacement powder, mung bean vermicelli, soybean sauce, and mixed grain porridge, thereby fostering the soybean product manufacturing and grain processing industries.

Overall, upstream agricultural product follows a unified model involving companies, cooperatives, and farmers, characterized by standardized seeds, technologies, services, and centralized procurement. Downstream, agricultural products are marketed through digital platforms supported by the Ministry of Education, which help rural collectives meet the customer demand. Eventually, around the Tongyu corn production project, upstream and downstream industries will be further developed, gradually introducing soybean processing, miscellaneous grain processing, fertilizer manufacturing, logistics, and other related sectors. Through exploiting these related opportunities and the bottom-up knowledge flow model, a poverty alleviation effect from collective entrepreneurship has been achieved.

During the proactive creation phase, the relational dynamics were characterized by the government-driven top-down embedding of university-diversified technology and market knowledge (detectors), which prompts rural collectives to update their cognitive content toward a dual intuitive-analytical model. This shift enabled them to interactive knowledge recombination and reuse based on technology-market synergies, where change the market-driven knowledge orchestration rules to exploit horizontally related opportunities (effectors) through technology-market synergistic new knowledge orchestration rules, which facilitate the exploitation of horizontally related opportunities (effectors). This progression culminated in the “Lines Move into Surface” effect (a geometric metaphor denoting progressive system expansion) and ultimately feeds back to the detector through the bottom-up knowledge flow to realize the exploitation of new related opportunities.

4.3. Cross- Phase Evolution Mechanism

A further analysis of the cases reveals that the key mechanism to facilitate the transition from the passive response phase to the proactive creation phase is the knowledge flow of “knowledge spillover-organizational learning-knowledge absorption.” Knowledge spillover refers to the unintentional, non-contractual dissemination of scientific technology and practical knowledge from universities to agribusinesses. Organizational learning refers to agribusinesses and universities continuously enhancing their knowledge absorption effects through cognitive and action learning. Knowledge absorption refers to the internalization and utilization of both explicit and tacit knowledge by actors [58]. The circular mechanism of “knowledge spillover, organizational learning, and knowledge absorption” exhibits significant characteristics at various phases:

4.3.1. Surface Knowledge Spillover-Cognitive Learning-Explicit Knowledge Absorption

During the initial implementation of Tongyu agricultural extension programs in which the university participates, technical materials and the maize crop market analysis encountered significant barriers to knowledge transfer. These documents were filled with specialized academic jargon and allowed only superficial, top-down knowledge to flow to local rural regions. However, this initial exposure provided the basis for further observation and imitative learning through actual demonstration plots featuring salinity-tolerant maize varieties. Direct observation of these demonstrations is critical, as it allows participants to progressively understand the technical agronomic principles and market implications. Through the cognitive learning process, the initial passive reception of information evolved into proactive knowledge assimilation, enabling rural collectives to absorb explicit knowledge.

4.3.2. Deep Knowledge Spillover-Action Learning-Tacit Knowledge Absorption

Absorbing explicit knowledge can have profound tacit knowledge spillover effects. In collaboration with JLU, Tongyu’s rural collectives have established engineering research centers to facilitate multidisciplinary, bottom-up cooperation. These initiatives enable scientific and market-oriented knowledge to be converted into generalized technological applications. At the same time, tacit knowledge spillovers promote the participation of universities and rural collectives in joint research to optimize deep-processing technological solutions and systematically summarize market insights on saline soil remediation through action learning. These collaborative efforts employ retrospective logic to overcome the constraints imposed by established knowledge frameworks, enabling the absorption of tacit knowledge about advanced processing techniques and market expertise. Eventually, the absorbed tacit knowledge begins a cyclical process that triggers a new round of knowledge spillover.

In summary, the university and Tongyu’s rural collective have utilized the knowledge flow mechanism of “knowledge spillover-organizational learning-knowledge absorption” to achieve an integration of understanding and practice (“the unity of inner knowledge and action”).

5. Discussion

5.1. The Poverty Alleviation Mechanisms with University Participation in Rural Collective Entrepreneurship

To summarize, guided by the “detectors-IF/THEN rules-effectors” framework, this study constructs a model to elucidate how university-participated rural collective entrepreneurship transitions for poverty alleviation from passive response to proactive creation.

During the passive response phase, the system operates under a linear, input-driven logic. Empirically, rural collectives undergo cognitive deconstruction of their traditional frameworks as they receive specialized knowledge from universities, leading to dependent knowledge orchestration and opportunity exploitation that is confined to vertical deepening within existing industry chains. Theoretically, the system represents a CAS in its initial configuration, where university specialized knowledge serves as the primary tag guiding cognitive deconstruction, while knowledge flows unidirectionally as the key resource flow. Rural detectors are calibrated to identify, capture, and process relevant knowledge-particularly technological knowledge (both specialized and cross-disciplinary expertise) and market knowledge (both existing and emerging market intelligence) within the university participation context. Once viable signals are detected, rural collectives translate the scientific knowledge of the university into locally interpreted language, thus activating systematic response mechanisms. These responses are governed by predefined IF/THEN rules (e.g., “IF the university provides technique X, THEN apply it to crop Y”), and these rules, which serve as the critical cognitive-behavioral nexus that bridges detectors and effectors in rural systems, facilitate intervention strategies that expand cognitive beliefs about potential opportunities while maintaining external dependency. The effectors consequently produce predictable, linear outputs through vertically related opportunities emerging from knowledge flows within the same industry chain. Vertical-related opportunities emerge from knowledge flows within the same industry chain, where shared resources (e.g., raw materials) link upstream and downstream segments. This stage reflects a system’s significant resistance—understood as the ecosystem’s capacity to withstand external interventions while preserving existing structures and functions [59]. This resistance is particularly evident in the cognitive schema of rural systems, which filters and selectively adapts university-embedded knowledge while maintaining core operational patterns. Such resistance, rather than representing system failure, constitutes an important ecological property that shapes the rule change [60]. This perspective extends the “ecosystem resilience” framework [61] by highlighting how resistance serves as a crucial precursor to subsequent adaptive transformations.

During the proactive creation phase, the system evolves into a self-organizing ecosystem. Empirically, universities embed diversified knowledge across chemistry, agriculture, and other fields, leading to a transformation in rural collective cognitive schema. This transition entails moving from intuition-dominant to dual intuitive-analytic models, while simultaneously restructuring cognitive content through a market-technology fit. This enables a transition to interactive knowledge orchestration and catalyzes horizontal-related opportunities. Theoretically, such change signifies fundamental CAS transformation where detectors become attuned to misalignments between technological supply and market demand, identifying both underutilized technologies and unmet market needs. New tags that change the focus of the system are local resources and related opportunities. This evolution marks a crucial shift in the internal rules of rural actors, from executing externally prescribed directives (e.g., “IF the university provides technique X, THEN apply it to crop Y”) to co-constructing collaborative frameworks (e.g., “IF a joint assessment identifies technique X as suitable for local market Y, THEN co-develop its application”). This transition enables a move from passive implementation to the joint exploitation of new related opportunities. The effectors consequently produce nonlinear, emergent outcomes through horizontal-related opportunities that apply technologies across different markets and sectors [62]. These dynamics demonstrate the system’s capacity for adaptive self-renewal through knowledge-orchestrated fit between technology and markets [63], ultimately driving coevolution from localized imbalance toward dynamic equilibrium [64], as conceptualized in ecosystem literature [65,66].

The cross-phase emergence arises by driving the “detectors-IF/THEN rules-effectors” changes through the knowledge flow mechanism of “knowledge spillover-organizational learning-knowledge absorption.” The effectors of the passive response phase drive critical inputs to detectors in the proactive creation phase, thus creating a feedback mechanism that promotes a shift in rural systems from passive response to proactive creation within university participation. By enabling a mechanism of knowledge flow (knowledge spillovers, organizational learning, and knowledge absorption), which promotes optimization of rules for knowledge orchestration in systems and catalyzes collective entrepreneurship contributes to poverty reduction (see Figure 6).

Figure 6. The poverty alleviation mechanisms with university participation in rural collective entrepreneurship.

5.2. Knowledge as a Core Resource Throughout the Collective Entrepreneurship Pathway Model for Poverty Alleviation

Further analysis reveals that knowledge is a core resource that runs the poverty alleviation mechanism with university participation in rural collective entrepreneurship

During the passive response phase, empirical evidence indicates specific functional characteristics among the system’s components. Regarding detectors, universities function as “knowledge filters,” selectively specializing knowledge for rural collectives based on local actual demands, thereby addressing initial knowledge asymmetry by enabling recognition of its usefulness. Concerning IF/THEN rules, universities function as “knowledge anchors” that deconstruct rural cognitive schemas through deconstruction of intuition-dominant models and rectifying miscognitive content where collectives knew something worked without understanding why, specifically by deconstructing the “markets for technological adaptation” cognitive content where technology simply needed market application. On the effector side, this configuration generates vertically related opportunities that improve rural regions’ autonomy over supply chains through industrial chain resource integration. Theoretically, this design represents a CAS configuration in which knowledge flows unidirectionally through filters controlled by universities. Rural actors adhere to dependent behavioral rules aimed at reducing decision-making uncertainty, while effectors generate predictable outcomes within established industry boundaries. This configuration corresponds with the increasing acknowledgment of universities as complex adaptive systems that must rapidly transform to address contemporary challenges [67]. Particularly in the context of transitions, universities adopt ecosystem models to promote responses to external pressures. The passive response phase identified in our study exemplifies how rural ecosystems initially preserve functionality by resisting external interventions, aligning with the conceptualization of Higher Education Institutions (HEIs) as CAS that can adapt to societal feedback and demands [68]. This path underscores the essential balance between maintaining system stability during the initial phase of knowledge embedding and laying the foundation for subsequent adaptive transformation.

During the proactive creation phase, empirical evidence demonstrates systemic transformation across all components. Regarding detectors, universities evolve into “knowledge-gathering boxes” that aggregate multidisciplinary expertise, accelerating technological iteration and stimulating cross-sector innovations while improving knowledge usability to address innovation complexity. The co-learning mechanism facilitates “theory-practice-re-theory” cycles that continuously generate and resolve new knowledge asymmetry challenges. Concerning IF/THEN rules, universities serve as “knowledge wallbreakers,” embedding rural collectives within multidisciplinary networks, enabling cognitive reconstruction toward dual intuitive-analytic models where the original “markets for technological adaptation” schema transforms into a “market and technology fit” cognitive content, with both elements mutually reinforcing through interactive knowledge orchestration. On the effector side, this configuration generates horizontally related opportunities that depend on multi-market objective coordination, enabling cross-market resource complementarity and risk diversification. Theoretically, this phase represents a self-organizing CAS where detectors dynamically identify latent knowledge gaps through iterative learning, rules transform into mechanisms for collaborative knowledge orchestration [69], and effectors create emergent results through opportunity relatedness-demonstrating the innovation ecosystem dynamics characterized by knowledge alignment drives the system toward sustainable adaptation [70].

On the cross-phase side: With the participation of universities, rural collectives have acquired and utilized knowledge via a “knowledge spillover-organizational learning-knowledge absorption” knowledge flow mechanism. Specifically, the superficial and deep knowledge spillover from the university enhances the progress of cognitive learning and action learning, facilitating the absorption of explicit and tacit knowledge [71]. Through this continuous flow of knowledge, the rural collectives have evolved from the IF/THEN rules and effectors in the passive technology adaptation phase to the proactive creation phase, which ultimately enables them to solve problems in a more adaptive and innovative way.

6. Conclusions

6.1. Research Conclusion

In order to realize the rural regions from “blood transfusion” to “systematic blood production,” this study attempts to explore how rural systems achieve collective entrepreneurial poverty alleviation with university participation. Applying the CAS theory to an in-depth analysis of a Chinese university’s assistance in Tongyu County, the research reveals the following findings:

First, during the passive response phase, detectors operate as “knowledge filters” through which universities embed specialized knowledge into rural systems, enabling rural collectives to identify and adopt solutions compatible with local resources. This detection process deconstructs established cognitive schemes. Correspondingly, these responses are governed by predefined IF/THEN rules (e.g., “IF the university provides technique X, THEN apply it to crop Y”), which serve as the critical cognitive-behavioral nexus that bridges detectors and effectors in rural systems. Correspondingly, these responses are governed by predefined IF/THEN rules (e.g., “IF the university provides technique X, THEN apply it to crop Y”), which serve as the critical cognitive-behavioral nexus that bridges detectors and effectors in rural systems. These rules facilitate intervention strategies that lead to dependent knowledge orchestration. On the effector side, this phase generates vertical-related opportunities that deepen existing value chains through knowledge resource orchestration.

Second, during the proactive creation phase, the detector evolves into a multidisciplinary “knowledge collection box.” This transformation enables rural collectives to transcend the constraints of individual domains and identify intricate innovation potentials across various domains. Regarding IF/THEN rules, universities function as “knowledge wallbreakers” by incorporating rural collectives into multidisciplinary knowledge networks through IF/THEN rules. This evolution signifies a significant change in the rule system towards principles focused on collaboration (e.g., “IF a joint assessment identifies technique X as suitable for local market Y, THEN co-develop its application”). These rules drive cognitive restructuring while strengthening interactive knowledge orchestration among all participants. On the effector side, this transition produces horizontal-related opportunities that enable cross-sector resource recombination and risk diversification. Theoretically, this transition reflects CAS evolving from an externally guided adaptation toward a self-organizing ecosystem capable of co-creation.

Third, the knowledge flow mechanisms of “knowledge spillover-organizational learning-knowledge absorption” underpin the evolution of detectors, IF/THEN rules, and effectors, enabling, within the context of university participation in rural collective entrepreneurship poverty alleviation, a shift from the passive response phase to the proactive creation phase.

6.2. Contribution

The theoretical contributions of this paper include the following three aspects:

First, this study contributes to poverty alleviation research by developing a path model for university-participated rural collective entrepreneurship based on the CAS theory’s “stimulus-response” framework. More specifically, university knowledge embedding acts as the detector, cognition-driven knowledge orchestration forms the IF/THEN rules, and related opportunity exploitation acts as the effector. This framework enables us to bridge a crucial gap between external intervention and endogenous development. While previous research has primarily conceptualized “capability” as the core of endogenous development [72,73,74], our model reveals that endogenous development extends beyond capability building to include entrepreneurial behaviors of “related opportunity exploitation” by rural actors. Thus, this research provides a novel perspective on the relationship between “intellectual resource injection” and “local resource activation,” illustrating how knowledge embedding transforms the cognitive schemas of rural collectives and restructures their knowledge orchestration rules. This transformation facilitates the exploitation of related opportunities by internalizing external rules, thereby enhancing research on poverty alleviation via rural entrepreneurship.

Second, this study enriches the research on university-rural interactions by revealing the knowledge orchestration process. While prior research has acknowledged that universities’ scientific knowledge benefits rural regions’ development [75], it has yet to reveal the dynamic mechanisms, particularly across different field contexts. Existing literature predominantly focuses on university-urban enterprise collaborations that operate within fields characterized by shared logics and compatible knowledge bases [76]. In contrast, our research reveals two modes of university participation that demonstrate an evolutionary path from specialized to diversified knowledge embeddedness, which bridges the academic field and rural field-each with its own distinctive logic. More importantly, our evidence indicates that universities function as adaptive agents, transcend traditional linear knowledge transfer models through interactive learning, and adjust with rural actors [77]. This paradigm reveals the mutual adaptation processes required for successful collaboration across field boundaries, wherein universities learn to navigate the distinct rule logics of rural contexts rather than merely transferring urban-developed models.

Third, this study expands research on the CAS theory. While prior studies have predominantly focused on the macro level [67,78], with insufficient attention given to the micro-cognitive level. This paper follows the “stimulus-response” model within CAS theory to investigate changes in the cognitive schemas of agribusinesses involving university participation. It examines how university participation is a key stimulus to drive the evolution of agribusinesses’ cognitive schemas. This exploration not only bridges a research gap in the cognitive dimension of CAS theory, providing a theoretical foundation for understanding how organizational knowledge orchestration emerged from the micro-cognitive level, but also significantly expands and deepens the application depth of CAS theory in entrepreneurship research, particularly within the context of agribusiness. Thus, this study incorporates knowledge elements into the CAS analytical framework, which deepens our understanding of adaptive evolution mechanisms in complex systems by elucidating how knowledge flows drive the functions of “detectors-IF/THEN rules-effectors.”

6.3. Management Implication

The management implication of this paper includes the following three main aspects:

First, this study highlights the knowledge embedded within universities (detectors) that is necessary to deconstruct and reconfigure the cognitive schema of agribusiness. This task requires agribusiness managers to adeptly interpret new technologies and market information introduced by external environments and enables them to recognize and address critical bottlenecks that hinder their development, thereby promoting effective knowledge orchestration strategies for participation in university contexts. Furthermore, it is recommended that agribusiness managers develop absorptive capacities to effectively interpret and contextualize external knowledge.

Second, our findings indicate that agribusiness must efficiently orchestrate knowledge to achieve the transformation of IF/THEN rules within the system. At the same time, universities should adopt a step-by-step participation strategy to progressively deepen the empowerment process. Effective university-rural collaboration requires universities to move beyond their traditional role as knowledge transmitters and instead establish specialized “knowledge filter” mechanisms to identify and translate disciplinary knowledge into contextually appropriate solutions. Additionally, it will be necessary to create flexible institutional structures that support multidisciplinary team formation and iterative learning cycles with rural partners. Concurrently, an evaluation metric system should be developed that prioritizes adaptive participation and long-term endogenous capacity building over the traditional outputs of technology transfer. Only through the integration of internal and external knowledge can structured knowledge chains be achieved, transforming disordered knowledge elements into structured knowledge chains. This approach amplifies the value of knowledge orchestration between universities and agribusiness while fully unlocking their potential in opportunity exploitation through systematic cognitive and rule transformation.

Third, this research suggests that in diversified university knowledge-embedding contexts, the exploitation of related opportunities (effectors) can function as the “resilience engine” for high-quality agribusiness development. Exploiting these related opportunities allows agribusiness to ensure alignment between technological supply and market demand. Specifically, vertical-related opportunities to enhance their core industries by utilizing existing resource endowments. This process can help increase autonomous control over the industrial chain and facilitate vertical resource integration. In addition, horizontal-related opportunities can promote multi-market synergies and significantly improve the resource orchestration efficiency. This strategic approach trans-forms initial opportunities into interconnected growth pathways that create synergistic value, thereby enhancing agricultural supply chain resilience.

6.4. Limitation and Future Direction

This study has several limitations and provides insights for future research.

First, the identified paths are based exclusively on successful cases, which may introduce selection bias. The exclusion of unsuccessful cases constrains our comprehension of the conditions that lead to the failure of university-rural collaborations in transitioning from passive response to proactive creation. Future research should incorporate both successful and failed cases. Through systematic comparative case analysis, it should reveal the underlying mechanisms determining the success or failure of such transformations, thereby constructing a more comprehensive path model.

Second, while this study reveals important relationships between knowledge embedding, cognitive schema transformation, and opportunity exploitation, the causal relationships among these variables remain exploratory. Empirical research faces significant challenges due to the lack of mature measurement scales for core concepts such as “cognitive deconstruction/reconstruction” and “knowledge orchestration strategies,” which presents challenges for empirical research. Subsequent studies should focus on creating validated measurement instruments and utilizing quantitative methods, including machine learning techniques to construct relevant keyword lexicons for extracting measurements of these variables from annual reports of agriculturally listed companies-to verify the causal mechanisms proposed in our model and systematically examine the causal relationships among variables.

Third, as a single case study, the findings of this research require further validation regarding their applicability. The contextual factors in the Tongyu case-its distinctive agricultural structure, level of economic development, and university participation model-may limit the applicability of this research model to other rural regions with varying socioeconomic conditions. Future research may explore, through cross-regional comparative studies, the impact of varying institutional contexts, resource endowments, and cultural factors on the evolution of university-rural partnerships. We specifically recommend investigating the functionality of the proposed CAS framework in markets characterized by diverse levels of development and infrastructure. These comparative investigations should utilize both qualitative and quantitative methodologies to ascertain the model’s boundary conditions and determine which components are fundamental and which are contingent upon the context. This approach enhances the robustness of the path and provides practical guidance for adapting university participation strategies to fit different rural contexts.