The procedural elements of a science–policy interface

Figure 14. Checklist of activities to ensure that knowledge is actionable and relevant for policy

Source: Authors’ own elaboration.

5.1.1 Scientific evidence production activities

To be useful for policy purposes, knowledge from many disciplines and epistemologies must be combined in a coherent format that summarizes known and unknown aspects, while recognizing knowledge gaps and conveying degrees of uncertainty and other limitations associated with the provided information (Gluckman, Bardsley and Kaiser, 2021). Evidence synthesis, i.e. the systematic identification, selection, appraisal and synthesis of relevant and robust studies on the same topic for overall understanding (including of divergences and controversies), how it varies by group or context,¹⁸ and translation of that knowledge to facilitate policy design and implementation is a key function of SPIs. The three main types of evidence synthesis are systematic reviews, systematic maps, and rapid evidence assessments. Systematic reviews focus on answering “What works?” by evaluating the effectiveness of interventions. Systematic maps have a broader scope, addressing “What is known?” Rapid evidence assessments can take either approach but use faster, streamlined methods.

Additionally, SPIs can generate new actionable knowledge. Various scientific evidence production methods have been proposed each contributing different levels of value (see Figure 15, condensed descriptions below) at different stages of problem-solving.

• Behavioural/action research: study of what people do, why and how this can be changed, as well as study of methods promoting uptake of new routine practices at various levels. These methods often (but not always) involve knowledge co-creation with non-academic actors (see next sub-section on knowledge co-creation and integration).
• Evaluation: monitoring the implementation and impact of initiatives, interventions or policies.
• Modelling: mathematically simulating real-world scenarios and options in a virtual environment.
• Foresight: identifying emerging trends and exploring different future scenarios based on various factors and assumptions, often involving different stakeholders to ensure the inclusion of different perspectives.
• Data analytics: systematic analysis of raw data to draw conclusions.
• Qualitative research and analysis: study of (typically non-numerical) data obtained through various methods to understand how problems are viewed and experienced, as well as options, implementation considerations, and metrics. This could include in-depth case study analysis, discourse analysis, systems and structural analysis, etc.
• Risk-based assessment: particularly amenable for the assessment of all relevant aspects of a technology/innovation, as well as comparison of the relative outcomes/effectiveness (also see Box 15). In the absence of scientific consensus, governments should act cautiously and diligently and should avoid steps that may damage human rights, human health or the environment.¹⁹
• Guidelines: Systematically developed recommendations based on evidence syntheses.

Figure 15. Examples of scientific evidence production methods and their possible contributions in problem-solving stages

Source: Adapted from Global Commission on Evidence to Address Societal Challenges. 2022. The Evidence Commission report: A wake-up call and path forward for decision-makers, evidence intermediaries, and impact-oriented evidence producers. Hamilton: McMaster Health Forum. https://www.mcmasterforum.org/docs/default-source/evidence-commission/evidence-commission-report.pdf

Box 15. Risk assessment policy requirements

5.1.2 Activities for knowledge co-creation and integration

Although scientists/researchers contribute significantly to the process of creating knowledge for designing public policies (e.g. Ojanen et al., 2021 in the case of forestry; Singh et al., 2021 for food systems; Iyalomhe et al., 2013 for climate change adaptation; Balvanera et al., 2020, for ecosystems and people), traditional, Indigenous Peoples’ and place-based knowledge and perceptions are as important as science for providing contextually relevant evidence (Figure 16). SPIs should involve integrating knowledge among different disciplines and social actors. This is particularly the case with agrifood systems which are complex and not only involve various sectors, scales, and temporal scope but also more fundamental disputes among social actors about values to be promoted for enhanced sustainability and disagreement on what the problem is, who is responsible, what solutions are appropriate, and what knowledge should be brought to bear on the problem (see Section 3).

In the absence of common vocabulary,²⁰ conceptual frameworks, indicators, information systems and research methods (all of which makes interdisciplinary systemic evidence synthesis more challenging) as well as high heterogeneity of social values, a more conventional science–policy advice approach bears the risk of perpetuating the promotion of sectoral research studies and partial perspectives. Such context warrants giving greater value to problem-focused transdisciplinary research and calls for co-creation and integration of diverse cultural understandings, requiring an approach that not only acknowledges differences but also fosters conditions for mutually beneficial collaboration. Intercultural co-creation of knowledge is the result of a process in which different ways of learning, creating, innovating, and transmitting knowledge co-exist in a safe environment, allowing conditions for new knowledge to emerge, for building bridges of understanding among cultures, for consensus building and conflict resolution (based on Rosado-May et al., 2020). A safe environment entails addressing power, respecting differences, ensuring space for disagreement, assigning equal value to diverse knowledge creation approaches, and fostering a spirit of mutual benefit among participants, where no one is exploited (see Box 16 for a successful example attributed to the coexistence of diverse ways of learning, constructing, innovating, and transmitting knowledge).

Box 16. A shared vision for effective governance of forest management in Noh Bec, Mexico

Figure 16. A Siona woman registering growth of a medicinal plant in her chagra, a community forest garden, in Ecuador. SPIs should engage with different stakeholders and visions of sustainable development, including Indigenous Peoples and local communities.

Photo credit: Daris Piaguaje, a Siona Indigenous photographer, Alianza Ceibo

Encouraging a multidirectional process in the co-creation of knowledge involves small-scale producers, Indigenous Peoples,²¹ scientists and others mutually contributing to the exchange. While co-creation introduces increased complexity, requiring not just an interdisciplinary but a transdisciplinary approach, it builds trust, democratizes knowledge and lays the foundation for transformative changes, fostering equity and full participation in transforming agrifood systems. It should however be recognized that fostering epistemological effectiveness in knowledge co-creation and integration can be challenging, and there is often a lack of clarity on how to operationalize such knowledge co-production (Dolinska et al., 2023), specifically how an SPI could incorporate varied local perspectives and improve community agency to engage with the SPI (see Box 17).

Box 17. Improving community agency to engage with SPIs

Figure 17. Tovo the octopus gleaner, a community-produced video promoting octopus conservation efforts in Madagascar. Innovative visual communication methods help better engage local communities in policy formulation processes.

Photo credit: Louise Jasper/Blue Ventures

Establishing common ground and research methods, as well as building a mutual understanding of guiding concepts among all stakeholders involved in the research, can entail considerable investments of both time and resources. Such prior agreements are nonetheless crucial to develop trust among stakeholders before engaging in knowledge co-creation activities per se that require:

• Aligning stakeholders on research objectives and methodologies.
• Involving all stakeholders in the process of data collection, analysis and interpretation.
• Acknowledging all stakeholders as co-authors of the results and eventual publications arising from the co-creation process.
• Adhering to the principles of free, prior, and informed consent²² when involving Indigenous Peoples or local communities in the co-creation process.
• Sharing the results of the work with all interested members of the involved community or communities, ensuring accessibility and dissemination of knowledge.

Novel technologies, including AI, may offer significant potential in supporting and streamlining evidence-informed decision-making, including facilitating and synthesizing citizen engagement and feedback, providing policy simulations, and offering near-time insights and recommendations based on data analysis (see Box 18). An AI-driven study reviewing the distribution of global research over the last 13 years found that despite a 60 percent increase in agrifood systems publications, there are very low levels of research targeting the poorest, most food-insecure, climate-vulnerable countries (Porciello et al., 2024). Nonetheless, ensuring successful application of AI requires adequate human oversight and regulations and policies to mitigate issues such as lack of transparency and strong bias.

Box 18. AI to support inclusion of knowledge coming from public sector and citizen science approaches

5.1.3 Activities to ensure that decision-making processes are oriented towards the use of evidence

An SPI has the potential to contribute to creating an environment that values and utilizes evidence, leading to more informed, effective, and sustainable policy decisions. Mechanisms and tools to promote evidence-informed policy mainly rely on fostering continuous engagement with decision-makers. This also involves instilling a culture of learning within the SPI through the use of knowledge in all decision-making and implementation activities, including monitoring and evaluation and regular capacity building activities. Finally, more dedicated tools such as guidance documents, best practices and methodologies for using different types of evidence could also be envisaged (see Box 19).

Box 19. Coordinating the role of evidence producer organizations

5.2 Capacity development activities

Effective functioning of a national SPI depends on the ongoing and continuous development of individual, organizational and community capacities. Each of these capacity levels is a distinct area of focus that all SPIs should address. Additionally, the capacities are nested such that the three levels are mutually supportive and reinforcing. Given the aims and goals of a national SPI, capacity building efforts should ensure the even development of evidence, decision-making and brokering capacities (e.g. Gustafsson, Díaz-Reviriego and Turnhout, 2020). The following offers a brief outline of capacity components at all three levels.

Individual level capacity

Individual level capacity recognizes the need for diverse and complementary skills and abilities of the SPI participants (e.g. employees, partner representatives) to undertake the work of the SPI. At a minimum, a national SPI should consider the following questions as guides to ensure sufficient capacity is in place. If some capacities are limited or absent, the SPI could devise means and strategies to build or acquire them internally or externally (see Box 20).

• Does the SPI possess the human capacity to provide the requisite input? Does the SPI have the diversity of skills and abilities (e.g. understanding of different forms of evidence, policy knowledge, brokering ability, etc.) to carry out its programme of work? Is there a diversity of participants and representation of different social positions and epistemic beliefs? Do SPI staff have sufficient field experience and familiarity with the range of relevant agrifood systems dimensions to enable the SPI to absorb and understand the range of contexts and issues related to national challenges?
• Are SPI staff sufficiently skilled in effectively translating and communicating evidence? Do SPI staff exhibit interpersonal skills to focus on relationships and interaction, and engagement skills to include citizens and stakeholders? Are there skilled individuals proficient in effective internal as well as external communication?
• Are incentives in place to ensure participation by evidence holders? Some SPI evidence contributors (e.g. scientists, researchers, traditional knowledge holders, practitioners) are fully employed in positions or have major responsibilities outside of the SPI (e.g. researchers in academia). In these cases, the incentives for their participation in the SPI may be low, due to competing requirements of their employer or profession. The SPI should consider ways in which it can encourage such knowledge holders to contribute effectively to the SPI. Official recognition and visibility of SPI contributors may be one mechanism. SPIs can also work with employers to recognize and encourage SPI participation.
• Do SPI staff have sufficient skills and ability to understand political processes, regional policy priorities, geopolitical sensitivities, and policymaking in general?
• Are SPI staff well-versed in evidence-informed decision-making tools and techniques? Are they able to integrate evidence in an inclusive way, and can they handle evidence controversies?
• Do SPI staff have leadership, facilitation and negotiation skills, in order to be able to balance different interests, and political sensitivity?
• What skill gaps or needs exist and how does the SPI plan to address them, either through formal training or on-the-job training?
• How will the SPI assess the success of its approaches to building and maintaining human capacity?²³

Box 20. The FSNet-Africa early career researcher development model

Organization level capacity comprises the abilities of the organization to enable the key activities of collaboration, learning and coherence (Section 4.2). Primary organizational capacities include governance, content, reflexive, relational and managerial capacities. Hiring professionals (including for time-bound tasks) could be considered if the existing individuals in the SPI do not represent all the required skills but these are needed at short notice.

• Governance capacity pertains to the ability of the SPI to undertake governance functions identified in Section 4.3. The SPI should have governance experts who understand how power and influence are wielded and have the authority to set rules, guidelines and structures to enable the desired forms of deliberation and decision-making. It is crucial to empower Indigenous Peoples to remain the primary agents in sharing and utilizing their knowledge. They must have the authority to determine the creation and implementation of policies that may directly or indirectly impact them, their territories, or their natural resources. Governance skills also play a pivotal role in designing joint work with other SPIs and across different scales (local–regional–global) necessary for effective collaboration, learning and alignment.
• Content capacity refers to the range of skills, abilities and backgrounds within the SPI, primarily covered by criteria for expert/member nomination. At the organizational level, the SPI team should ideally also include individuals who can effectively navigate different knowledge domains (e.g. Indigenous Peoples’ and traditional knowledge, western academic science; see Box 21), diverse disciplines (e.g. natural and physical sciences, policy and social sciences), and knowledge gained from practice. Appreciation and integration of diverse content capacity can be enhanced through cross-training.
• Reflective capacity is essential for an SPI with a diverse staff, operating at the boundary of evidence and decision-making to address critical agrifood system challenges. This requires providing space and time to process the complexity of the tensions, trade-offs, controversies and implications inherent in multifaceted knowledge-policy interfaces. Does the organization create and maintain opportunities and incentives that encourage reflection and learning?
• Relational capacity involves expertise in setting up organizational systems that cultivate social capital and trust among staff, as well as between the national SPI and other partners and stakeholders. Relational capacity ensures that formal and informal ties are valued and maintained, both to ensure a healthy workplace and to enable effective knowledge exchange, collaborative inquiry, effective reflexivity, bridging knowledge divides and decision-making. Scientists should have their independency guaranteed and they should also be protected from undue pressure to act contrary to their scientific responsibilities and from any possible intimidation or retaliation. They should be provided with a safe and effective mechanism to raise concerns about issues they encounter, whether in the private or public sector, that may jeopardize human rights. Safe and efficient recourse mechanisms should also be put in place in case scientists face threats and harassment.
• Managerial capacity concerns the skills and ability of the SPI to design, implement and sustain work processes, as well as to assess performance and identify opportunities for growth, self-correction and improvement.
• Capacity as a boundary organization includes the ability of the organization to: (1) take stock in a complex landscape of topics and processes; (2) identify gaps in content and processes while anticipating future needs; (3) effectively communicate the SPI composition, processes and structure to multiple stakeholders; (4) translate and communicate SPI deliberations, controversies, and advice/decisions to multiple stakeholders.

Box 21. Knowledge Makers Programme

Community level capacity

The community of national, regional and global SPIs is made up of individuals and organizations that together comprise a profession with the expertise essential for undertaking effective boundary activities. A national SPI should consider itself to be an important and integral member of this broader SPI community. As a result, the national SPI bears the responsibility to enable national level staff to work across scales (both top down as well as from the bottom up), exchange training and tools important for sustaining the profession, access training opportunities, and take on leadership, teaching or other positions of visibility in the profession. An important goal for national level SPIs is to actively participate in the SPI community while simultaneously cultivating nation-specific identity focused on agrifood systems transformation at the national level.

5.3 Learning and reflexivity for the SPI in relation to desired impact

Learning, in a wide sense, refers to the actual or prospective ability of the SPI to assimilate new knowledge and experiences, with a view to improving its overall effectiveness, relevance, credibility and legitimacy. Learning can take many forms and can be categorized in many ways, e.g. from experiential at the individual level to learning at the level of organizations (“learning organizations”).

The academic literature distinguishes formal and informal learning practices in SPIs (e.g. Obermeister, 2020 on science advisors), where the informal is more difficult to ascertain as it occurs in shadow spaces that can be difficult to observe. An important distinction could be made between the following two types of learning:

• First loop learning, which aims to improve the achievement of an existing objective.
• Double loop learning, which involves questioning the objective (Borie et al., 2020).

Similarly, another important distinction could be made between instrumental learning and transformational learning (Pallett and Chilvers, 2013):

• Instrumental learning is goal-oriented and empowers actors to work in and/or engage with the SPI as it currently exists.
• Transformational learning refers to the SPI gaining insights about itself (e.g. its intended or unintended practices, impacts, etc.) and adapting and evolving accordingly.

The understanding of the impact of SPIs is a dynamic and evolving field. No consensus exists on a set of standardized indicators for evaluating impact across SPIs to enable comparison and draw lessons on what a successful SPI looks like. This is mainly because policy processes are complex and SPIs involve incremental, diffuse and slow-to-manifest impacts over a long period of time. In this context, conventional evaluation methods based on assessing clearly identified effects of specific interventions on pre-defined outcomes are ill-suited and even spurious.

The most common approach is informed by a simple and linear conceptualization in which knowledge informs and shapes policy (see Section 3.3). The implication of such a simplified version of SPI for impact assessment is a focus on accounting for specific policy impacts of SPI activities based on narrow indicators such as citation in policy documents. Theory-informed discussion in the literature on SPI has cautioned against such a simplified version, which does not do justice to the range of impacts according to different actors participating in SPIs (see Box 22).

Measuring such impacts requires the mobilization of a range of methods summarized in Figure 18.

Box 22. Range of SPI impacts

Figure 18. Methods employed to assess the impacts of boundary-spanning activities

Source: Adapted from Posner, S.M. & Cvitanovic, C. 2019. Evaluating the impacts of boundary-spanning activities at the interface of environmental science and policy: A review of progress and future research needs. Environmental Science & Policy, 92: 141–151. https://doi.org/10.1016/j.envsci.2018.11.006

Finally, understanding the diversity and complexity of SPIs is crucial for strengthening their effectiveness across various policy domains (see Box 23 that highlights the significant variations in SPIs across different policy areas in Japan). By examining these differences, common challenges and opportunities for enhancing communication and best practice-sharing among sectors can be identified, ultimately improving the impact of SPIs.

Box 23. Cross-sectoral perspectives on SPIs in Japan