Introduction

Societies are facing unprecedented social, environmental, and economic sustainability challenges. These ‘grand challenges’—so named because their outcomes are felt globally (Molnar et al., 2023)—are difficult to address, given the complexity of the social and ecological systems from which they arise. Several scholars, scientific organizations, and funding bodies have argued that traditional disciplinary scientific approaches fail to deliver the multifaceted, systemic solutions needed to address these challenges and advocate a new kind of integrated approach called convergence (National Academies of Sciences, Engineering, and Medicine, 2021; National Science Foundation, 2024a; Roco, 2020). Convergence research demands innovation beyond disciplinary boundaries to create new, multidisciplinary solutions to vexing social and scientific challenges (National Science Foundation, 2024a; Sharp et al., 2011). Convergence research is gaining popularity, particularly in countries such as the United States. For example, the National Science Foundation (NSF) has created a portfolio of funding streams to support convergence research (National Science Foundation, 2024b, 2024a).

Despite the growing attention toward convergence research, how to operationalize convergence is not always clear. Societal challenges are often incredibly complex, and not all aspects are fully understood. Angeler and colleagues (2020, p. 98) reason that “It is therefore unclear what will converge to what, for what, and for whom; that is, convergence is not goal-oriented and cannot be a priori concretely operationalized.” In other words, there is no one-size-fits-all template for convergence, since the methods, approaches, and experts needed to explore different challenges will vary (Angeler et al., 2020; Sundstrom et al., 2023b). While the openness of the convergence approach is crucial to encourage exploration and innovation, the conceptual ambiguity of convergence has its limitations. Knowing how to start a convergence process—most crucially, developing effective convergent research questions—can be difficult without guidance (Sundstrom et al., 2023a, 2023b).

While recognizing that presenting a prescriptive roadmap of convergence is counter to the basic tenets of convergence, we explore how research groups can more effectively undertake convergence. We acknowledge that several high-level frameworks and principles of convergence provide the general outline of convergence or describe what convergence is (e.g., Gajary et al., 2024; Peek et al., 2020; Roco, 2020). Our work builds on these frameworks to provide guidance not on what convergence ‘is,’ but rather on how to ‘do’ convergence, by providing strategies to build convergence into the everyday interactions of a convergent network. The quotidian or day-to-day interactions have been a rich field of research and critique (Neal and Murji 2015, Lefebvre, 2014). Researchers of team science have raised the importance of everyday interactions (Dusdal and Powell, 2021; Hall et al., 2018). However, this literature has not connected the quotidian to the support of convergent research. By providing such strategies, we contribute further conceptual clarity of convergence in theory and offer actionable insights for convergent networks in practice.

Our case study is the Multiscale RECIPES (Resilient, Equitable, and Circular Innovations with Partnership and Education Synergies) for Sustainable Food Systems Network (henceforth the RECIPES Network). The Network targets solutions for food waste reduction using a convergence approach. Food waste is a complex societal, environmental, and economic challenge that occurs at every stage of supply chains, results from diverse drivers, and requires vast coordination to solve (Vittuari et al., 2019). Not only does the food waste challenge lend itself well to a convergence approach, but RECIPES was also required to adopt a convergence approach. RECIPES is supported through the NSF Sustainable Regional Systems Research Networks (SRS-RNs) program area. The SRS aims “to fund convergent research and education that will advance sustainable regional systems science, engineering, and education to facilitate the transformation of current regional systems to enhance sustainability” (National Science Foundation, 2024c). Fitting in that goal, RECIPES purports to gather a convergent team of researchers, students, and stakeholders and harness their combined expertise to generate novel research and solutions for food waste reduction (National Science Foundation, 2024d).

In this context, our paper is guided by the research question: How can research teams integrate disconnected disciplines and experts working on food waste to generate convergent research? This research question was driven by a practical need to assess convergence within the Network and help the Network improve its convergent processes. To answer this question, we interviewed RECIPES Network members. During semi-structured interviews, individuals shared their experiences of convergence in the RECIPES Network, and based on these experiences, identified barriers to convergence and strategies implemented to encourage convergence. Section (3) describes the method for conducting and analyzing interviews. Section (4) details findings from the interviews, and we reflect on strategies to embed convergence into the everyday activities of Network members in Section (5).

Background

Convergence research

Convergence was borne from the acknowledgement that many of today’s social and scientific challenges are complex, and “deep integration” across disciplines and sectors is necessary to tackle such challenges (National Science Foundation, 2024a). Organizations such as the National Research Council and the National Science Foundation in the US have traditionally focused mainly on the integration of basic research and “hard sciences” (life sciences, physics, mathematics, engineering, and medicine) to solve complex challenges (National Research Council, 2014; National Science Foundation, 2025). Other scholars note that tackling society’s most pressing challenges requires “…integration of knowledge about societal transformations, their underlying dynamics, their inherent opportunities, as well as the instruments and technologies to manage these transitions” (Pedersen, 2016, p. 2). In other words, social sciences and humanities will be key ingredients of integrative approaches. Regardless of the disciplines involved, the focus on “deep integration” emphasizes that experts must do more than come together or exchange information. Instead, true convergence happens when research practices or ways of thinking are transformed so that novel frameworks, paradigms, or disciplines emerge (Ivanov et al., 2024; National Science Foundation, 2024a).

While proponents argue that convergence is necessary to address complex social and scientific challenges, how convergence differs from other forms of disciplinary integration is not universally understood. Convergence shares its integrative research approach with many other concepts, including multi-, inter-, cross-, trans-, de-, anti-, and undisciplinarity, as well as concepts such as team science, fusion, and participatory science (Cannon, 2020; Pedersen, 2016; Roque et al., 2022; Shen, 2021). Indeed, some scholars use these concepts synonymously (see National Academies of Sciences, Engineering, and Medicine (2019); Morss et al., 2021; Thompson et al., 2023). Others argue that the level of conceptual integration differs across these concepts (Peek et al., 2020). For example, multidisciplinary projects produce knowledge from different disciplines about a common challenge; yet the disciplines remain separate. In contrast, interdisciplinary projects integrate information, data, and tools from multiple disciplines. Continuing up the ladder of disciplinary integration, transdisciplinary projects build new worldviews and paradigms that transcend disciplinary approaches, which are necessary but still insufficient approaches for convergence. Convergence approaches, Peek et al. (2020) contend, bring together a wide community of researchers and actors to address real-world challenges holistically.

Given the ambiguity surrounding convergence research, unsurprisingly, no clear method for doing convergent research exists. Several frameworks and principles of convergence have been proposed (e.g., Gajary et al., 2024; Moran et al., 2022; Peek et al., 2020; Roco, 2020; Sixt et al., 2022), although their advice remains at a high level. For example, Roco (2020) identified seven principles to facilitate convergence research platforms, including adoption of a holistic view that finds “unity in diversity;” establishment of a common goal to address long-term system challenges; evolution of the research process through cycles of convergence, integration and divergence; a focus on system-centered actions and problem solving; awareness of concurrent causal pathways in complex systems; and concurrence of activities and collaborations towards a shared goal. While providing more direction than high-level definitions of convergence, such principles do not answer the question of ‘how’ to do convergence in practice. Some scholars describe the concept as “riddled with uncertainty” (Sundstrom et al., 2023b, p. 2), and the lack of both conceptual clarity and implementation guidance can impede convergent science (National Academies of Sciences, Engineering, and Medicine, 2019; Gajary et al., 2024; Sundstrom et al., 2023b).

Nevertheless, the literature reveals patterns about what facilitates convergence research. Scholars have found that teams are better equipped to converge when they have strong leadership, regularly interact, share information and resources, and build an atmosphere of trust, community, and willingness to try new things (Bukvic et al., 2022; Morss et al., 2021; Peek et al., 2020; Sundstrom et al., 2023b; Villablanca et al., 2024). At the National Academies of Science, conference participants noted “…having key leaders on board with sufficient leadership continuity over time is needed for convergence to become part of the fabric of an institution.” (National Academies of Science, 2019, p. 19).

A particularly critical element of successful convergence is the development of a shared vision or mission and shared goals to rally around (Bukvic et al., 2022; Ernakovich et al., 2021; Rigolot, 2020), along with a common language for diverse experts to effectively communicate about the challenge area (Cullen et al., 2023; Moran et al., 2022; Morrison-Smith et al., 2022). Additionally, institutional support and access to adequate resources (e.g., space, time, long-term funding) are needed for successful convergence (National Academies of Sciences, Engineering, and Medicine (2019); Bukvic et al., 2022; Cannon, 2020; Finn et al., 2022; Sixt et al., 2022).

On the other hand, convergence research projects often face common challenges, such as rigid funding structures and outdated institutional designs built to support or teach disciplinary research (Irwin et al., 2018; Sundstrom et al., 2023b; Wilson, 2019). Scholars have noted difficulties collaborating across disciplines with different cultures, conceptualizations of a problem, methods, data, or publishing criteria (Croog et al., 2023; Lombardozzi et al., 2023; Mercer and Ryan, 2016; Renwick, 2016). Securing buy-in from project partners or students pursuing an academic career can be challenging, given that convergence research is seen as riskier than traditional research approaches (Purvis et al., 2023; Wilson, 2019). Further, convergence is time-intensive yet rarely recognized in tenure and promotion criteria (Cannon, 2020; Mercer and Ryan, 2016; Morss et al., 2021). Even if buy-in is secured, knowing how to start a convergent process can be difficult, as some groups assume it will happen automatically (Sundstrom et al., 2023b; Westerhoff et al., 2021).

Despite unresolved challenges, certain funding bodies are increasingly developing programs and calls to foster convergence research. NSF, for example, identified convergence research as one of its ten “Big Ideas” in 2016, signaling the start of its convergence funding portfolio (National Science Foundation, 2018). Similarly, convergence research institutes and programs are emerging worldwide, focusing on challenge areas ranging from health sciences to environmental sustainability (Ernakovich et al., 2021; Wilson, 2019).

The RECIPES Network

The ideation of the RECIPES Network originated from two smaller NSF conference grants led by the Rochester Institute of Technology (RIT) and Johns Hopkins University (JHU) in 2019. These conferences brought together researchers from across the United States to focus on food waste. The two teams from the conference grants merged to submit a proposal in January 2021. Beginning in November 2021, the RECIPIES Network received $15 million from NSF for five years of support. The Network listed 40 faculty members in the original proposal across five research areas: engineering, computer science, and mathematics; social, behavioral, and economic sciences; physical, chemical, and biological sciences; public health; and design. The faculty are from 15 different universities across the United States. The Network began in 2021 during the COVID-19 pandemic. Thus, many team members had not met in person until 2022 at the first network-wide meeting.

The Network is structured around 10 thematic clusters and the Network Coordination Team, consisting of the Director, Co-Directors, and Project Manager. The clusters cover topics, not disciplines, such as Data, Valorization, Typologies, among others. As the Network evolved, members organized new clusters such as the Policy and Student clusters. Members organize themselves into one or more clusters, holding regular meetings (typically monthly). Cluster meetings are working sessions to progress projects originating from the clusters. The Network holds meetings every other month, focusing on communicating activities relevant to the Network, discussing network processes like convergence, hosting outside speakers, and sharing findings from clusters.

RECIPES recognized early in its inception that the team was large and diverse. Thus, through a nearly year-long process, the team worked toward what convergence meant within the Network. To facilitate this work, design team colleagues led Network-Centered Design Studios, which include network-centered, human-centered, and life-centered approaches to shift Network researchers’ mindsets and work toward convergence and to provide opportunities and offerings that enable such convergent thinking and action to occur (Ashton et al., 2024). The Design Team also led a series of listening sessions and interviews on convergence. Additionally, the Network hosted a Convergence Café at both annual meetings to create space for discussions about convergence. Together, these discussions resulted in the Guiding Principles and Community Norms (Agarwalla et al., 2024) (Appendix A) and the Tenets and Tensions document (Espat et al., 2024) (Appendix B), both developed to aid collaboration and convergence. Both documents were published after our interviews, meaning neither participants nor interviewers knew the final form of the Principles and Norms or Tenets and Tensions documents at the time of the interviews.

Methods

A mixed-methods research design consisting of qualitative semi-structured interviews and a quantitative ranking exercise was used to understand how members of the RECIPES project perceived convergence to date in the project. This study was approved by Duke University Campus Institutional Review Board (protocol ID# 2024-0149).

Recruitment

As of January 2024, all personnel of the RECIPES project were invited for interviews. An initial recruitment email and two follow-up emails were sent via the RECIPES listserv, and the interviews were announced in a RECIPES full Network meeting. Individuals were given information about the purpose of the interviews and how to schedule an interview. Interested individuals contacted the research team, who scheduled an interview and shared the consent form.

Interviews

Virtual interviews took place between January and March 2024. Twenty-five interviews were conducted, with one interview involving two participants, for a total of 26 participants. Four co-authors either led or observed the interviews. Of these four, one co-author was deeply engaged with the RECIPES Network; two co-authors were formal Network members but were either new to the Network or did not frequently engage with the Network; and one co-author was not a formal Network Member. Participants were aware before the interviews that the interviewing team was responsible for assessing convergence within the Network. Interviews lasted approximately 60 min and were recorded via Zoom. Upon entering the Zoom room, the research team read the consent form and received a verbal agreement to participate from individuals. After the interview, the interviewer stopped the recording and asked several demographic questions, which were entered into an Excel file. The recording of one interview failed, but hand-written notes were typed and included in the analysis. One person, upon reading the consent, decided not to proceed with the interview.

Before the interviews, the interview guide was pre-tested with members of the Design Team who, as described earlier, have been integral to the Network’s convergent process. Questions were adjusted to enhance clarity or removed to sharpen focus. The final interview questions asked how participants perceived convergence in the RECIPES Network. First, participants were asked to describe an important convergence challenge in the Network, including how the challenge appears in the Network and what has been or could be done to overcome the challenge. Then, participants were asked to rank five common convergence challenges from most to least challenging or indicate that the challenge was not applicable within the RECIPES Network. To develop the list of challenges, one team member led a literature review of convergence in 2023 (Ernakovich et al. 2021, National Academies 2019, National Research Council 2015, and Petersen et al. 2021, inter alia). We also reviewed comments from “convergence conversations” led by the Co-Design cluster in 2023. Assessing this literature and internal network conversations, we developed a list of challenges. Participants were then asked to describe their top-ranked challenges in more detail before discussing what had facilitated convergence. Participants were also asked about the purpose of convergence in RECIPES, whether they perceived convergence to be a goal or a means to an end. Footnote 1 whether they felt convergence had occurred within the Network, and how to know if convergence has happened. Several additional questions were asked but not included in this analysis. Interviews were professionally transcribed using the service Rev.com. Transcriptions were then reviewed, corrected to match the recordings, and anonymized.

Analysis of ranking

We conducted graphical and regression analyses to examine the ranking of challenges. As described above, participants were first asked to state a convergence challenge. Then, they ranked five challenges identified in the literature, allowing participants to incorporate their challenges into the ranking or eliminate challenges they deemed irrelevant. Participants could incorporate additional challenges into the ranking, which provided valuable insights into the qualitative analysis. However, the additional challenges were unique from the five original challenges. Initially, we classified these additional challenges as ‘Other’ challenges (Other), totaling six challenges. In the quantitative ranking analysis, we did not interpret the Other challenges or categorize them into new challenges. In the thematic analysis, however, we used inductive reasoning to code the Other into themes that expanded upon and reshaped the original five, predetermined challenges.

One participant reported and ranked two Other challenges in addition to the five predetermined challenges. Thus, we would have had seven challenges, but only one participant ranked that additional challenge. Because that participant ranked their second Other challenge fourth out of the seven, we dropped this fourth-ranked Other challenge and readjusted their lower-ranked items upwards, so the rankings were from six to one for all participants; otherwise, we would have had a ranking from 6 to 0. Further, two other participants did not rank one of the challenges. We did not alter their rankings and allowed for missing values. To preserve the underlying distribution of the rankings, we developed histograms of the rankings for six challenges using Stata (SE 18.0 64-bit x86-64 Revision June 7, 2023). We calculated the mean rank for each for comparison. We used the Kruskal–Wallis equality-of-populations to test if challenges derive from the same underlying distributions. We also estimated a rank-ordered logistic regression model by maximum likelihood, clustering standard errors by participant identification number. From that regression, we estimated the difference between coefficients to understand the relative importance of each challenge.

Analysis of interviews

De-identified interviews were coded using the NVivo software (release 14.23.4). A two-step coding process was employed. First, structural coding was used to identify the segments of text related to each interview question (Saldana, 2016, p. 98). Thus, our initial codebook mirrored the interview guide (Guest et al., 2011, p. 56). Second, descriptive coding was used to further distinguish the responses within each parent code (Saldana, 2016, p. 102). One co-author, who was not a Network member nor an interviewer, analyzed all interviews. A co-author with extensive qualitative research experience, who conducted the interviews and was deeply familiar with the material, closely reviewed the entirety of the coding after both rounds with the coder to identify additional codes and ensure a consistent interpretation of the data across authors. Following Halpin (2024), another team member double-coded one interview and achieved a Kappa score of 0.89. After coding was complete, similar codes were merged into broad categories, which are presented in the results.

Positionality statement

Only one co-author is a fully integrated Network member who has been with the Network since the beginning. The co-authors who conducted most of the analysis and interpretation of results were not formally part of the RECIPES Network (n = 2) or were only loosely/newly involved with the Network (n = 2). For example, the coder was not a Network member, and the supervisor who closely reviewed the codes with the coder had only recently joined the Network to do this work. A level of objectivity was brought to the analysis by taking this approach. However, we also acknowledge that no analysis, regardless of the co-authors’ connection to the RECIPES Network, is ever fully objective.

Results

Participant demographics

Twenty-six individuals agreed to be interviewed (21% response rate). One person declined to participate after we shared the informed consent information. Table 1 presents participant demographic characteristics.

Table 1 Participant demographics.
Full size table

Participant perceptions of convergence in RECIPES

Below, we summarize how participants perceived convergence in the RECIPES Network. Anonymized participant identification numbers are presented after quoted material, denoted (IN#).

Participants largely agreed on the purpose of convergence in the RECIPES Network. Echoing the two pillars of convergence defined by NSF, participants saw convergence as necessary to (1) bring individuals with diverse expertise together and (2) tackle a complex, societal challenge—in this case, wasted food. One participant explained, “you can’t solve a problem like wasted food—a large complex societal issue like wasted food—without convergence…there are just too many moving parts, too many influences and impacts on the entire system that, without working together, we don’t get anywhere” (IN12). There was a sense that convergence was needed to “create something that is more than just the sum of its parts” (IN22), which included new ways of thinking, new approaches to solve challenges of food waste and new solutions for food waste management. From a more practical perspective, one participant suggested that convergence was adopted because “it’s what NSF is funding” (IN7).

Despite agreement on the necessity of convergence in RECIPES, participants documented challenges along the path of convergence, described below.

Ranked challenges: a quantitative analysis

To understand which challenges posed the largest barriers to convergence in the Network, participants were asked to report their top convergence challenge and rank our predetermined list of five common convergence challenges. As discussed above, due to inconsistencies with the additional challenge category, the data were adjusted to develop a consistent list of the five challenges identified in the literature: Common Goals (lack of shared common goals), Discipline Differences, Geographic Distance, Team Dynamics, and Team Turnover. Including the additional, ranked challenges as Other, the ranking ranged from 6, the highest rank, to 1, the lowest. Figure 1 below illustrates the mean ranks of the six challenges.

Fig. 1
figure 1

Histogram of ranking challenges to convergence—scores ranged 1 to 6 (n = 26).

Full size image

Of the six challenges, the Other challenges had the highest mean rank. Nine of the 26 participants noted an Other Challenge, which was not on our list. Of these Other challenges, eight participants listed their Other challenge as either six or five (the highest ranks), with the ninth participant ranking their Other challenge fourth. Thus, over one-third of the sample ranked a heterogeneous mix of other challenges as the most important challenge that the Network faced. Of the predetermined challenges, the order of importance was Common Goals, Discipline Differences, Team Dynamics, Geographic Distance, and Team Turnover (Fig. 1). With the pairwise Kruskal–Wallis equality-of-populations rank test of challenges (See Table 2), Team Turnover is the lowest ranked challenge and is statistically different from the remaining challenges, except for Geographic Distance, which is the second lowest scored challenge. Geographic Distance is statistically different from Other, Discipline Differences, and Common Goals. Other, as the highest rank challenge, is statistically different from the remaining challenges, except for Common Goals.

Table 2 Pairwise Kruskal–Wallis equality-of-populations rank test of challenges with ties (\({\chi }^{2}\) (1) and (p-value)).
Full size table

Extending this analysis, a rank-ordered logistic regression model of the six challenges was estimated (See Fig. 2). With Team Turnover as the base value, the other challenges are ranked higher, except for Geographic Differences. Testing the coefficient values yields similar results to those of the Kruskal-Wallis test, except that the coefficients for Geographic Distance and Common Goals are not statistically different. In total, the four top challenges (i.e., the greatest challenges) are statistically different from the challenge with the lowest mean rank (Team Member Turnover).

Fig. 2: Regression coefficients from the rank-order logit model of challenges to convergence.
figure 2

Note: The dots are the point estimates. The lines represent the 95% Confidence Interval. Detailed presentation of the rank order logit is in the appendix.

Full size image

Table 3 shows the estimated difference in the rank of challenges. Relative to Other Challenges, the predetermined challenges have an estimated rank that is 1.5 to nearly 3 points lower, suggesting the strength of Other Challenges for those who offered them. While Common Goals is not statistically different from Disciplinary Differences and Team Dynamics, it has an estimated rank of one point over Geographic Distance and Team Turnover. Collectively, Other Challenges is the most pressing of the challenges, followed by Common Goals and Discipline Differences. The least important challenges are Team Dynamics, Geographic Distance, and Team Turnover, although Team Dynamics is ranked higher than Team Turnover.

Table 3 Pairwise contrast of estimated coefficients.
Full size table

As discussed later, the low rank of Team Turnover is consistent with qualitative interview responses, indicating that most participants did not see this as a challenge for RECIPES. Geographic Distance was seen as both a convergence challenge and a facilitator, contributing to its lower mean rank as a challenge.

Challenge areas and progress to overcome challenges: a qualitative analysis

The ranking exercise provides evidence of the relative importance of the five original challenges. It also provides a broad category that we initially called Other. With the qualitative analysis of the entire interview, we were able to analyze the Other and recast the predetermined challenges into six challenge areas for convergence (Table 4). According to the quantitative analysis, we present the challenges below in order from highest to lowest rank. The Other challenges represent two emergent challenges discussed by participants: Understanding of Convergence and Structural Challenges. We present Understanding of Convergence and Structural Challenges first, as they represent the highest-ranked challenges and are rooted in the Other challenges and subsequent sections of the interviews. Even participants who did not offer an Other challenge mentioned challenges that connect with these two emergent challenges. We discuss Understanding Convergence directly before Common Goals, since there is an overlap in the descriptions of these two challenge areas. Four challenges align with those in our ranking exercise: Common Goals; Discipline Differences; Team Dynamics; and Geographic Distance. We renamed Team Dynamics to Team Dynamics and Logistics to address the more expansive ideas surfaced in the interviews about the challenges the Network found in collaborating. Each challenge area is discussed below, along with an indication of whether the challenge was resolved or is ongoing and any strategies used to mitigate or overcome the challenge. Although the challenges are presented as six distinct categories, many links between the categories exist, as noted throughout the text.

Table 4 Reported challenges.
Full size table

Structural challenges

Falling into the Other category, Structural Challenges appeared to be a significant challenge for the Network, influencing its day-to-day work. Budgetary issues were the most discussed structural challenge affecting convergence. Participants felt that rigid budget structures limit flexibility in pursuing emergent convergent ideas. One participant expressed, “the dollars are doled out at the beginning…and so that’s the true bedrock structure, is the funding conduits” (IN11). One felt that “NSF needs to reimagine its requirements and its policies in how these types of budgets are scoped to begin with” (IN18). Several participants found the allocated budget to be limiting, noting that additional funding would allow for more frequent in-person meetings, provide seed funding to pursue ideas that emerged beyond the scope of the original proposal, and allow RECIPES to invite additional partners where there were “missing perspectives” (IN19). Additionally, several participants noted the challenge of funding timelines. For example, “…the funding timeline for supporting, it might not be long enough in a way because it just takes so much time to build up the relationships, especially if you want it to be such a big group of people” (IN7).

Academic structures present other structural challenges, such as a lack of reward for the effort of convergence work, or the requirement to deliver on pre-existing projects (i.e., time allocation can often be inflexible). Others mentioned academic budgetary structures as a barrier to convergence. For example, part-time funding on a project may not leave sufficient time for convergence. Additionally, “you can’t take summer funding away from a faculty member, you don’t want to take salary away from someone who’s on this project if their part isn’t working out so much” (IN4).

Furthermore, participants felt that cross-institutional convergence was challenging due to factors such as differing semester/term schedules or difficulty coordinating across multiple players. Others felt that traditional grant structures – where one lead institution is awarded the funding and the “prestige of having the grant” (IN23) – can lead to power dynamics and “logistics that get in the way” (IN23) of doing convergent work.

Most structural challenges are ongoing, given that these issues are largely beyond the control of the Network to solve. One participant explained, “no one of us individual researchers has the power to fully change the problematic dynamics of academia, the concept of research, the concept of grants and funding, and all those things” (IN25). Budgetary issues were seen as something that cannot be entirely resolved. Yet, as one participant described, “at least I feel more clarity now around what our flexibility is to change things [in the budget]” (IN18).

Understanding of convergence

Another commonly identified challenge that emerged from the Other challenge was the lack of shared understanding about convergence in practice. For example, participants disagreed whether convergence was a means to an end (i.e., a method or process) or an end goal. Most participants viewed convergence as a means to an end in the RECIPES Network – the end goal being food waste reduction. For example, “It’s wonderful to get everyone working together and out of our silos, but ultimately, there is an end goal, that we want to address something in our society that is a major issue” (IN20). While some stakeholders saw convergence as a means to an end, they understood NSF and/or the RECIPES Network to define convergence as a goal. For example, one participant stated, “I feel like convergence is the end goal, because for what I’ve been communicated about this whole project, that is what we’re trying to achieve” (IN9). Finally, others felt that convergence was both a goal and a means to an end in RECIPES.

Participants also held differing views on how to spot convergence in practice. Several participants noted that convergence happens when a diverse group jointly produces outputs such as journal articles, white papers, or other materials, particularly when the group can demonstrate that various disciplines, theories, and methods were used in that work. Others reasoned that if convergence is seen as a means to an end, then assessing whether the group addressed issues related to food waste reduction would prove whether convergence had happened. Several participants noted indicators of convergence that were not output-oriented but represent day-to-day activities such as talking to, learning from, or sparking new ideas with disciplinary experts and organizations with which an individual would not have otherwise collaborated. Other participants noted that personal exchanges between different institutions could be considered convergence. Several participants identified procedural indicators of convergence, such as including historically excluded fields in the research or giving a variety of perspectives equal weight in answering a research question. One participant felt that concrete indicators of convergence were difficult to identify, but “…you know it when you see it, or you know it when you’re a part of it” (IN8). Finally, one participant noted that convergence is not a “universal process” and that “it can be really unique to each project and each collaboration” (IN22).

Given the lack of shared understanding about convergence, participants found it challenging to convince the team that convergence is important and achievable. For example, “We are encouraged to do convergence research, but I think there [are] things that should be done on why we have to do convergence research in a way to let us understand more the topic” (IN13). Another expressed, “I think one challenge…is that again, related to perception - it’s not impossible to do convergence. It’s maybe impossible to do it in a way that everyone 100% agrees with or buys in with or likes, but it’s actually happening” (IN14).

Although participants felt that different understandings of convergence persisted, one participant said, “I would definitely say I think people are more comfortable and conversant in convergence now than at the beginning of the RECIPES Network” (IN18). Regarding the importance of convergence, one participant described initial steps taken to resolve the confusion: “The biggest thing, thus far, has been professors and faculty members and just people on the project that aren’t students or postdocs actually then really communicating the importance of all of this to us, just because it’s so easy for us to get lost in the stress of all our work…that we lose that drive” (IN21).

Common goals

Participants expressed a shared understanding that the broad goal of the Network was to find solutions for food waste reduction. Despite this, the lack of clarity on shared common goals was the challenge with the highest mean rank of the five predetermined challenges (third overall). Several factors contributed to this. Beyond the broad food waste goal, participants expressed less agreement when “drilling down to slightly narrower goals” (IN11). This was due in part to the regular tension between project and individual goals, where some Network members were pursuing goals and deliverables needed for career progression, in turn making them less open to converging. For example, “we have different researchers who have different thought processes on what the outcome and what the overall direction should be because what different people need to get out of the project, need and or want, depending on the stage in their career…is so different depending on what your background [is] and your department associates with success” (IN5). Participants noted strategies that partially resolved this challenge, such as dedicated time at in-person meetings to discuss the Network’s shared goals and discussing more specific goals within the clusters. However, several participants saw this as an ongoing challenge, since individuals will always need to balance project goals with career demands.

Differing perspectives about convergence as a main goal of the Network were also apparent (i.e., was convergence of similar or lesser importance than the goal of finding solutions for food waste reduction). One participant acknowledged the tension of having dual goals: “One thing that came up in the in-person meeting [was that] some people really wanted to talk about food waste and some people found that we’re talking too much about convergence and not really getting anywhere about it. So, I think that was an interesting tension in having that dual goal in the Network and how we are marrying the two” (IN2). Another participant felt that convergence should not be a main goal of the Network: “And we’ve been hearing a lot where folks are saying, ‘Enough, we get that convergence is important, but we got to return to what the important ultimate end game here is.’ So yeah, so I think there’s got to be a balance, not overemphasizing convergence to be equal to the goal around food waste” (IN18). As these quotes suggest, balancing the focus between food waste and convergence is an ongoing challenge in the Network.

Disciplinary differences

Participants noted that each discipline came to the project with a distinct way of working (e.g., methods, epistemologies, languages, publishing practices) and disciplinary cultures (e.g., promotion criteria, familiarity with collaboration, desired research outputs). Given the diverse backgrounds of Network members, individuals were not always familiar with each other’s fields or how they relate to food waste, which shaped the Network’s routine interactions. As a result, each team member’s contribution to a convergent process was not always obvious. One participant explained, “Obviously, everyone has their different areas of expertise, which is actually a real benefit to this project, but at the same time, it does take some time, I think, to really understand what each other does, whatever one’s expertise is” (IN20).

Another challenge is related to team composition. One participant felt that the Network composition was not ideal, explaining, “We’re trying to make all of this stuff fit and converge when there’s missing pieces and there’s people who we’re trying to force into converge when their research expertise is…on the periphery” (IN19). Similarly, one participant noted the challenge when an individual lacks content knowledge: “I see a lot of designers just building solutions for sustainable outcomes without ever being able to frame that sustainability in a more scientific way” (IN2).

Disciplinary differences had the second-highest mean rank of the predetermined challenges (fourth overall), and participants indicated that this was a persistent challenge for the group. One participant said, “It’s never like one’s job is ever done on that front, and there are always places you can improve” (IN7). Another stressed the time needed to overcome disciplinary differences: “We can’t learn what people do in one year or two years, it still takes a while” (IN15). However, participants also noted the significant strides made to bridge disciplinary differences. Again, the in-person meetings were mentioned as crucial time points when people got to know each other and their respective areas of expertise. Similarly, repeated contact through monthly cluster meetings helped people become more comfortable with one another. The cluster structure was mentioned: “And then the different clusters that they’re not labeled by discipline supports working together across disciplines” (IN12). One participant noted that “having great leadership at the top” (IN25) and a coordination team can help connect people and ideas. Finally, efforts to create a shared language were credited as bridges across disciplinary differences. For example, “I think they worked really hard in the beginning to put together a lingo document and trying to get a shared language for all of us that are working on food waste, even if we have different academic or industrial backgrounds. I think there was a lot of power in trying to work on that to target the disciplinary differences” (IN10).

Team dynamics and logistics

Participants noted that, despite the focus on convergence, silos still formed. Participants mentioned silos at the individual, cluster, and university levels. The formation of silos is related to a lack of effective communication across groups. For example, one participant noted, “when I think about opportunities for convergence with other clusters, I just wouldn’t really know enough about what the other clusters are actively working on…I don’t know how to find out that information or if it is available” (IN24). Effective integration of students was also noted as a challenge. That is, students seemed to have their cluster silo or were limited to engaging with the projects of their advisors. One participant explained, “I think we have been mindful of academic hierarchies and really trying to create space for students to feel empowered, to take leadership roles within the Network, but I would say that that power hierarchy remains” (IN26).

Some mentioned a lack of willingness to converge, either due to pre-existing projects or because researchers can become set in their ways (i.e., closely linked to Structural Challenges). For example, “It feels like there are some people who maybe…They’ve got their set of tools. They’re open to how to improve their set of tools, but it doesn’t feel like they’re necessarily there to think broadly about these issues and really expand the applications” (IN7). Another suggested that willingness is not the core issue: “everyone’s really conceptually bought into convergence, which is really cool. But I’m not sure we quite have the structures in place to know what to do with all the different opinions” (IN25). One offered a pragmatic perspective, explaining that as researchers, “we’re not necessarily trained to be good team players” (IN14).

Several participants mentioned the challenges inherent to remote work, mostly discussed in the context of the COVID-19 pandemic. Some felt that RECIPES got off to a slow start due to its remote start. One participant explained that the remote nature of the Network also made it harder to get clarity about shared common goals or bridge disciplinary differences: “It’s way easier to do that when you’re sitting in a room together and have snacks versus if you’re sitting on a computer screen and you’re like, ‘Okay, get ready for this three-hour meeting where we’re going to hash all of this out’” (IN22).

Team dynamics had the third-highest mean rank among the predetermined challenges (fifth overall), likely due to the diverse responses given when asked if this was a challenge and, if so, whether it was ongoing. Perspectives about this challenge could vary given that team dynamics spans a wide range of issues; people’s positions within the Network varied (e.g., student versus faculty); and perceptions about this challenge reflect the specific interactions of each participant. For example, one participant expressed, “I think a project like this attracts good people, empathetic people who care and are willing to learn and are willing to work hard…And that’s why I think it’s not really a challenge because everyone that I’ve encountered on this project has been really great” (IN21). In contrast, others mentioned specific clusters with team dynamic challenges, and one participant felt that, “everyone’s really nice, but it’s still the power differential is still there” (IN19).

Geographic distance

The geographical spread of the RECIPES Network was identified as a distinct challenge, as discussed above, from working remotely. However, many of the impacts (e.g., difficulty building relationships, feeling connected, or engaging in convergence) were similar. However, one participant noted a distinct challenge of geographic distance—the challenge of effective communication when different team members have highly localized research foci.

While geographic distance was mentioned as a challenge, some participants also recognized that a geographically spread team has benefits for convergence. For example, “I think the nature of this project too requires that we have folks who are spread across the country. We want both national and regional effects. So really, you have to have people who are spread apart seeing these things happen and how they happen in different contexts, different locations, all of that” (IN4).

The advantages of geographic distance are likely one reason this challenge ranked fourth among the five original challenges (sixth overall). Some participants felt that geographic distance had been overcome with tools such as Zoom, which gained popularity during the pandemic. One participant explained, “When you’re on Zoom, it doesn’t really make a difference where people are. So yeah, it’s not a perfect tool, but it is probably the best option we have to communicate amongst the different groups, and I think it’s been pretty okay so far” (IN3). Finally, participants noted that even if most of their interactions were on Zoom, the in-person annual meetings helped to bring the researchers together.

Occurrence of convergence in RECIPES: in the “little moments”

In terms of convergent outputs, one participant described convergent work on a research paper: “Even just in the literature review that we were doing, I think there was quite a bit of exchange about how different disciplines, how they define resilience and sustainability and equity…I think there was definitely some new exchange on that front” (IN7). When considering convergence as a means to an end, one participant explained that the outcomes achieved by the Network (regarding food waste) suggest that convergence has happened. From a different perspective, convergence as a means to an end could be described as small behaviors that help the team reach its goal. For example, “Convergent behavior, the behaviors that you have to do in order to get convergence, lots of collaboration, lots of emails back and forth, lots of boards and talks and all those things, are necessary for what we’ve been asked to do in this grant, what we’ve told our funders we are going to do” (IN25).

Some provided tangible examples of convergent collaboration, such as the collaboration between RECIPES Network members and a grocery store chain to implement real-world food waste reduction strategies, as well as the process used to develop the Wasted Food course. Others suggested that less tangible or output-driven connections were valuable. For example, “The parts that I think have been convergent, the parts where I’ve learned about other people’s research, the parts where we’ve made cross-university cross-discipline connections, the things we have been able to incorporate I think have been at least really interesting” (IN25). As indicated in the previous quote, examples of convergence often took the form of learning from or talking with other disciplinary experts. This form of convergence seemed particularly relevant for the student group: “I feel like we see a lot of it [convergence] also in the student and postdoc group where we have students from different clusters but also different disciplines. And I feel like there we have been able to go a little bit beyond what our expertise are and just try to learn from each other” (IN9). In terms of inclusion as a procedural sign of convergence, one participant noted, “the Network has done a pretty good job at empowering students, postdocs, and our non-academic partners to really be part of that [convergence]” (IN11).

Finally, reflecting the ambiguous nature of convergence, one participant said, “I really do think that people are doing convergence, they just don’t always know that they’re doing it” (IN14). The lack of awareness that convergence in practice could be because convergence was seen by some not “like the big bang, it’s not like there is singular convergence” (IN5) but rather “those little moments, sparks of doing something differently, moments of getting out of their comfort zone, pretty much translating the goals of convergence into [these] more tangible practices in people’s lives that they can experience, that they can try to cultivate” (IN2).

Discussion—developing an ethos of everyday convergence

Participants described the numerous challenges encountered by the RECIPES Network on their convergence journey. Yet, as demonstrated by the participant reflections above, these challenges did not stop the Network from seeking strategies to bridge disciplinary differences and foster a convergent team dynamic, ultimately finding diverse ways to operationalize convergence. Notably, participants’ descriptions of convergence within the Network often focused on small, everyday events and interactions, such as learning from colleagues, writing a paper, or making new connections, that had been layered with a convergence perspective. This highlights the importance of convergence thinking permeating the everyday activities of Network members. Below, we outline five strategies used in the RECIPES Network to develop an ethos of everyday convergence.

Before presenting the strategies, we situate our contribution to the existing literature. A critical theoretical framing that informs our work is the sociological consideration of the quotidian. Neal and Murji (2015, p. 813) argue.

…an understanding of the everyday as moments of translation and synthesis in which the “big” folds into, shapes and is concretized in, but also co-constituted by the “small.” For sociologists, the micro, the slight, the most mundane, and the banally ordinary practices, emotions, social relationships, and interactions also reflect convergences with and manifestations of wider social factors, forces, structures, and divisions.

Back (2015, p. 834) argues “…the everyday matters because it offers the opportunity to link the smallest story to the largest social transformation.” Thus, the everyday practices that follow reflect the “folding in” of the big moves of convergence into the small practical moves of collaboration, communication, and deconstructing inhibiting structures, wherever possible. These small moves point to ways for any team of researchers to address large, complex problems that NSF has suggested are appropriate for convergent research.

Convergence frameworks, case studies, and strategies for convergence/cross-disciplinary engagement have been proposed, each providing crucially important insights. Frameworks often provide high-level overviews of the phases of convergence (e.g., Gajary et al., 2024; Peek et al., 2020; Sixt et al., 2022) or the components of convergent processes (Moran et al., 2022). Case studies, on the other hand, often detail specific procedures used, for example, to engage non-academic stakeholders (Burt et al., 2023; Cullen et al., 2023), develop convergent research questions (Ivanov et al., 2024), or break down silos and maintain forward momentum guided by common goals (Ernakovich et al., 2021). Reviews compile crucial elements of a convergent approach, such as flexibility, complexity, and good communication (Cannon, 2020; Drusdal and Powell, 2021; Hall et al., 2018), or practical actions for fostering convergence (Ding et al., 2020). Our strategies focus on cultivating a mindset or ethos of convergence, aiming to complement both abstract frameworks and project-specific strategies for convergence by providing actionable yet adaptable strategies to show how convergence can be integrated into the everyday interactions of a convergent network. In doing so, we answer the call from Lakhina et al. (2021, p. 300), who suggest that we not only view convergence as a method or outcome, but also as an “ethic that motivates a higher order alignment on why we come together.”

Further, we acknowledge that an ethos of convergence is not always sufficient for successful convergence. Many scholars have noted the necessity of institutional change to facilitate convergence research and provide suggestions for systemic changes within academia and funding agencies (e.g., Bukvic et al., 2022; Croog et al., 2023; Gajary et al., 2024; Irwin et al., 2018; Pedersen, 2016; Peek et al., 2020). In line with other published literature, our participants indicated that funders could develop more flexible and long-term funding structures, and universities can reassess promotion criteria and integrate disciplinary silos (Gaziulusoy et al., 2016; Morton, Eigenbrode, and Martin, 2015; Nyboer et al., 2021; Slade et al., 2024). While we should not lose sight of systemic shifts needed to facilitate convergence or the discussions needed between research networks and systemic actors (e.g., funders, universities), we chose to focus on strategies within the control of the research network.

Community or cohort building is the first essential ingredient of convergence (Espat et al., 2024). Bringing together diverse individuals to work towards a common goal requires trust-building, respect, and building personal ties (e.g., Bukvic et al., 2022; Ernakovich et al., 2021). The necessity of community building is not unique to convergent teams and is well documented in literature on collaborative efforts, such as community collaboration (Franco et al., 2007; Lansing et al., 2023), team dynamics (e.g., Hakanen et al., 2015), and, of course, cross- or multi-disciplinary collaboration (Ding et al., 2020; Harris and Lyon, 2013). Although the RECIPES Network faced challenges in community building, not least due to the remote start during the COVID-19 pandemic, participants noted several ways individuals were encouraged to build community and address structural challenges. The in-person annual meetings or simple ice-breaker activities at the start of remote meetings were often mentioned as a time when Network members could informally connect and get to know each other beyond their research interests. For example, one stakeholder shared, “I think just seeing people in-person and just being able to have a chat that’s not targeted at a specific meeting goal is helpful to actually get interested in people. Because I think a lot of times that convergence really gets framed as combination of skills and expertise, but I doubt that that’s how people actually start a collaboration” (IN2). These activities encouraged Network members to invest in the collective work, a necessary step to mitigate Structural Challenges and develop an Understanding of Convergence. Furthermore, such activities helped overcome challenges related to Disciplinary Differences and Team Dynamics. Collectively developing goals and norms also instilled a sense of community, thereby addressing the Common Goals challenge, supporting similar reports in the literature (Bukvic et al., 2022; Dusdal and Powell, 2021; Hall et al., 2018). In summary, for a research network to move beyond its disciplinary boundaries and comfort levels, individuals first need to feel part of a community that collectively is ready to leap into the unknown of convergence.

Second, the Network dedicated time to discussing what convergence meant to the Network. As highlighted throughout this paper, no single definition or roadmap for convergence exists (Sundstrom et al., 2023b), and information on what convergence means, what enables it, and how to foster convergence effectively is lacking (National Academies of Sciences, Engineering, and Medicine, 2019). While groups might have a broad understanding of the approach, participants shared that it was useful for the Network to dedicate significant time and effort to understanding what convergence meant in the context of the RECIPES project and then institutionalizing practices that encourage convergence. In this way, dedicated convergence discussions addressed both the Common Goals and the Understanding of Convergence challenges. As previously described, the RECIPES Network built this collective understanding by hosting convergence-centered discussions at annual network-wide meetings, network-centered design studios, listening sessions, and interviews. This focused exploration and collective sense-making led to the publication of the Guiding Principles and Community Norms document (Agarwalla et al., 2024), which was released after our interviews. The power of community norms lies in their specific and actionable nature, providing a guide for everyday interactions, rather than working towards a “big bang” of convergence (IN5). For other groups engaging in convergence, the Guiding Principles and Community Norms can serve as a template that can be adapted to suit their needs.

Third, although convergence is an inherently collaborative endeavor, top-down guidance from enthusiastic leadership is needed (Morss et al., 2021). Specifically, the RECIPES leadership was intentional about facilitating convergence. Furthermore, participants in our interviews emphasized that leadership sets the ‘tone’ of convergence. That is, the leadership can either view convergence as a checkbox or prioritize it so that it permeates the Network’s everyday work. One participant drove home the importance of regular reminders to converge when explaining, “I feel like there used to be little communication about the importance of convergence in the project, and only in the past few months have I actually really seen that word being thrown around consistently in dialog when working with members on the project. And I do think that’s the most important step in it all is actually reminding people that this is one of the overall goals and we have to work towards it” (IN20). Similarly, convergence is about experimentation, and leaders have a role in encouraging the testing of nascent ideas. In this way, leadership can help address challenges of Team Dynamics, creation of Common Goals, and facilitate a shared Understanding of Convergence. In the RECIPES Network, this leadership came from the Principal and Co-Principal Investigators. However, it is possible that this leadership could come from another ‘leader’, such as coordinators responsible for facilitating convergence. The importance of leadership aligns with findings from Ernakovick et al. (2021, p. 5), who wrote, “we found that we could successfully formulate convergence research objectives using bottom-up approaches, but progress towards achieving tangible research results was difficult without top-down organization, momentum, and goal setting.”

Fourth, in addition to enthusiastic leaders, funded roles that support convergence efforts are crucial (Pollock et al., 2019). One such role in RECIPES was that of a Network Coordinator, who oversaw the logistics of coordination and helped facilitate communication channels within and across clusters. For example, one participant described that scheduling challenges were eased, despite Network members being located across the country, due to the work of the Network Coordinator: “A lot of it is [the Network Coordinator] being able to work out these schedules and having us in these different meetings that happen really frequently, so I feel like we can keep up with each other pretty well” (IN9). The Network Coordinator also ensured that everyone was informed. One participant shared, “I think [the Network Coordinator] does an amazing job of making sure that everybody’s on the same page from a logistic standpoint. If you don’t have somebody like that to keep us all in the know, then things can fall apart quickly. So, that’s probably the best thing RECIPES is done is hire her” (IN10). Thus, dedicated convergence staff can help facilitate effective information sharing, a core aspect of the Team Dynamics challenge. Additionally, several participants mentioned the benefit of having a design team to facilitate the Network’s convergence efforts. Whereas disciplinary researchers are not often trained as ‘good team players’, the design team excelled in prioritizing ‘whole person’ engagement among Network participants, fostering critical reflections about convergence, and making space for challenging conversations about emergent tensions (Ashton et al., 2024). In effect, these roles counter the structural challenges researchers face and create the necessary infrastructure to facilitate convergence, allowing researchers to focus their (often limited) time on ‘doing’ convergence.

Finally, the RECIPES Network accepted convergence in all its forms. As we could see from the interviews, participants had varying ideas about what convergence was or when it happened. Rather than seeking consensus on these issues, the Convergent Tenets and Tensions documents acknowledge that individuals can think differently about convergence and recognize that “divergence is part of convergence” (Espat et al., 2024). This strategy helps address diverse perspectives arising from Disciplinary Differences, create a shared Understanding of Convergence, and mitigate Structural Challenges. Given the divergent thoughts and feelings about convergence, we caution convergent Networks or funding agencies against developing overly rigid or narrow metrics for convergence. Narrow metrics would limit the flexible, emergent, and pluralistic nature of convergence, reducing it to a tick-box activity or leading to unwanted consequences (Petersen et al., 2021).

These recommendations connect to and disrupt the frameworks of other researchers, as mentioned above. For example, Peek et al. (2020) present CONVERGE as a cyclical process moving from identifying researchers, educating and training them, agenda setting, connecting researchers, and funding convergence (Table 5). While a powerful framework to structure a team to engage in convergent research, our findings of everyday convergence complement and counter this framework. The challenges we uncover and the strategies we offer connect with the nodes of the CONVERGE framework. See Table 5 for direct comparisons. While we find compatibility between our findings and those of Peek et al. (2020), the interconnectedness of our challenges and strategies with the CONVERGE nodes suggests that everyday convergence recognizes that converging research involves more complex and multidirectional linkages between nodes. Thus, we find that challenges in convergence can lead to unstructured and iterative processes. Beyond the challenges, convergence is like ordinary life, which is more than mundane or a routine; convergence is “dynamic, surprising and even enchanting; characterized by ambivalences, perils, puzzles, contradictions, accommodations and transformative possibilities (Neal and Murji, 2015, p. 812).

Table 5 Connection of everyday convergence to Peek et al. (2020).
Full size table

To navigate the complexities of convergence and stretch toward the transformative possibilities, we offer everyday convergence. With these small moves, we argue that the Network and potentially other teams can work across the nodes of CONVERGE to bring about convergence. Based on Neal and Murji (2015), we suggest that everyday convergence involves researchers working on seemingly mundane and banal tasks and activities, which helps them connect with the shifts in social factors and forces that can move convergence forward. It is the everyday acts of researchers that support the movement between nodes in the CONVERGE framework.

The RECIPES Network brought together a diverse range of disciplines (Table 1), and many of the challenges and opportunities align with those discussed in the existing literature. For these reasons, while the five strategies presented above emerged from analysis of one convergence research network, they can be adapted to any convergent network.

Everyday convergence extends the more structured framework of Peek et al. (2020); however, an overly narrow focus on the everyday could lead to a myopia that prevents engagement in the large-scale moves necessary for convergence. Similarly, too much focus on “more structure, goal-oriented collaboration” can overlook the everyday practices that drive the entire enterprise. Thus, researchers must strike a balance between more structured frameworks of convergence and everyday convergence. Our presentation of everyday convergence serves as a corrective to the structured models that obfuscate the smaller actions necessary for convergence.

The strategies can be applied similarly to the convergence frameworks and principles cited throughout this text, serving as guidelines rather than prescriptive actions, since every convergent endeavor is unique. We see the strategies above as a useful complement to the high-level frameworks for convergence available in the literature. Further research investigating how convergence can be integrated into the everyday actions and interactions of convergent research teams would complement this analysis.

Limitations

We acknowledge the relatively low response rate. However, we collected responses from a diverse range of Network members. For example, Table 1 illustrates the breadth of disciplines represented, ranging from the social sciences and design to the natural sciences. Further, every cluster had at least one member who participated in an interview. That is, we did not speak only with individuals from clusters focused on convergence, such as the Co-design cluster; we spoke to all clusters. Finally, many of the challenges and opportunities expressed by participants resonated with the Tenets and Tensions of Convergence document created by the collective RECIPES Network, giving us confidence that the interviews reflected broader perceptions.

The interviewees were self-selected, meaning that those who were not motivated to discuss or engage in convergence may have held different views about whether convergence had occurred in the Network. Nevertheless, analysis of the interviews reveals that convergence has occurred in some form within the Network.

Conclusion

The RECIPES project is a large, discipline-spanning project that aims to create knowledge that transforms the current wasteful food system into one that promotes sustainability, equity, and resilience. As part of the NSF SRS-RNs grant scheme, the Network was required to take a convergence approach. It was clear from our interviews that this was no easy, straightforward task. Participants described numerous challenges and uncertainties surrounding convergence, yet also noted how, despite these challenges, convergence emerged in various forms throughout the Network. Our analysis underscores what many other scholars have recognized–no template or roadmap for convergence exists. Groups that engage in convergence must dedicate time and effort to develop a shared understanding of convergence and collectively decide how convergence will permeate the group’s work. Broader systemic changes to academic structures (e.g., promotion criteria) and funding mechanisms (e.g., more flexible, long-term funding) are necessary to facilitate convergence, which will necessitate adjustments to funder and university policies. However, in this paper, we focused on strategies within the control of research groups to encourage convergence. Based on the experience of the RECIPES Network, we outline five strategies for creating an ethos of everyday convergence: 1) focus on community building, 2) dedicate time to explore what convergence means to the group, 3) ensure the leadership is enthusiastic about convergence 4) fund specific roles responsible for facilitating convergence, and 5) embrace convergence in all its forms. The strategies can be applied by convergent networks in a similar manner to existing convergence frameworks and principles, serving as guidelines to be adapted rather than prescriptive actions, since every convergent endeavor is distinct. Further research investigating how convergence can be integrated into the everyday actions and interactions of convergent research teams would complement this analysis.