1 Introduction

Research and research programmes have explicitly demanded to seek solutions to today’s societal challenges and have emphasised the importance of addressing societal needs and ethical questions in research and development (e.g. Owen and Pansera 2019). Since new sciences and emerging technologies are mostly embedded in fields of conflicting interests and are of high complexity, there is a need for multi-actor decision processes, including actors of the wider public (e.g. Chilvers and Kearnes 2016).

One effect of this trend was the new role assigned to societal actors within the Responsible Research and Innovation (RRI) concept which was also introduced in that period of time (see for example Jasanoff 2003; Owen et al. 2013). The paper aims to examine and evaluate Social Labs as an approach for implementing Public Engagement (PE) processes to foster RRI. The Social Lab (SL) approach as described by Zaid Hassan (Hassan 2014) enables a process of acting rather than planning by using experiments and involving different groups of people, with each lab focusing on a specific challenge. This paper is asking how the Social Lab approach is able to respond to frequent challenges of PE processes? We will answer this question by referring to the empirical findings of the applied Social Labs in the NewHoRRIzon project - an EC funded project which aimed at further integrating RRI in the European innovation system. The following section lays out a definition of PE and stakeholders, the relation to RRI as well as pitfalls and challenges of PE. Section 8.3 describes the operationalisation of PE in SLs in the context of the NewHoRRIzon project. The methodology section comprises the data sources as well as the qualitative analysis approach applied to the material. The results section comprises empirical results of the SL processes which are then contrasted against theoretical insights in the thereupon following section of discussion. The conclusion summarises the main insights and concludes with an assessment of the SL approach’s potential for solving challenges of PE in general as well as with regard to the specific context of RRI.

2 Theoretical Embedding

In the context of RRI, co-production of knowledge by scientists and societal actors is often part of the research process. Similarly, research on PE in research and innovation processes is no new endeavour and abundant in many different fields, such as Science and Technology Studies (STS) (Balázs et al. 2020).

The term ‘Public Engagement’ encompasses a wide range of meanings, with a variety of actors involved, while a standardised and clear definition is lacking (see Sect. 8.2.1). Accordingly, there is a “lack of shared language about engaged research” (Holliman et al. 2015, 15). Mostly, participation and PE are used interchangeably - with regard to the inclusion of societal actors in research and innovation processes. Within the framework of the new RRI paradigm, PE was introduced as an ambitious term that embraces the idea of “publicly engaged science” (Stilgoe et al. 2014). Public Engagement is therefore an integral part of open and inclusive research and innovation processes where societal actors can give relevant input.

In the context of Responsible Research and Innovation, PE introduces a circularity common to the concept of RRI as e.g. also identified by Timmermans et al. (2020); while PE is aspired and aimed at by RRI, RRI only prospers fully when societal stakeholders are already involved in the corresponding research process (Randles et al. 2016). Based on this understanding, we consider PE as an inclusive process in which we invite stakeholders to co-create innovative formats addressing RRI in their respective working environment.

In this section, we introduce selected challenges of PE in the specific context of RRI, serving as an exemplary systematisation of critical dimensions of PE in research that potentially arise in every form of participatory approach. These challenges represent important bottlenecks for successful PE in research and innovation processes. Listed challenges will later be used as a baseline for evaluating the Social Lab approach’s potential for realising Public Engagement in and for RRI.

2.1 Challenges of Public Engagement in Research and Innovation Processes

While the relevance of PE - in particular in the realm of RRI - is uncontested, the way this can be put in practice is far from being a clear and streamlined process. Potential challenges are repeatedly discussed and several attempts have already been made to create guidelines, to condense best practice examples on stakeholder processes and participatory formats. This section extends on challenges identified in Marschalek (2018) and elaborates on stakeholders to be engaged, their envisaged roles, involved feelings, group dynamics and impacts.

2.2 Selection of Participants

In many engagement processes “the public” to be involved remains unclear (Wickson, Delgado, and Kjolberg 2010). However, a clear identification of individuals or groups of people the process wants to engage with is crucial. Therefore, much effort needs to be invested for mapping and approaching potential participants. Usually, engagement processes start with decisions about who is to be included in the research and innovation actions. Depending on the research focus, the purpose of the inclusion or the design of the process, different societal actors might need to be included. As Fern Wickson, Ana Delgado and Kamilla Lein Kjølberg (2010, 757) emphasised, it is “essential to recognize the heterogeneous nature of ‘the public’ in engagement activities”. The RRI principle does not rule out any potential participants (Timmermans and Stahl 2014), and the challenge of defining principles for inclusion remains pertinent.

In order to prevent limited participation, Alexander Bogner (2012) coined the term “invited participation” as a form of steered engagement where stakeholders are deliberately selected to overcome potential self-selection biases. As Ulrike Felt and Maximilian Fochler (2008) argue, uninvited forms of civic engagement equally need being considered not to overburden invitees as sole representatives of the public. Doing so, however, provokes new questions, namely what kind of knowledge and expertise of the subject in question is required to join the research process; which public, in fact, is welcome to participate? (Delgado, Lein Kjolberg, and Wickson 2011).

Having passed all the hurdles of stakeholder identification, new challenges arise, since PE processes see themselves increasingly confronted with the problem of “stakeholder fatigue” (Delgado, Lein Kjolberg, and Wickson 2011, 834). Chosen stakeholders might not automatically be willing or able to join the research and innovation processes. Public interest in highly technical topics cannot be taken for granted, awareness for even having a stake in the research topic might need to be created (Bogner 2012).

While the inclusion of neglected knowledge in innovation processes is an explicit goal of PE in the realm of RRI, not having a specific expertise in the area of research might also lead respondents to feeling unconfident to accept an invitation (Marschalek 2018; Valkenburg 2020). When invitees doubt the relevance of their perspectives or that their own interests will not be respected, they might distrust the process as a whole. Trustful interactions are important stepping stones for setting off collaboration processes. Fears that engagement might ultimately be instrumentalised by more powerful actors for opposed motives in contrast are lethal for successful multi-stakeholder processes (Selsky and Parker 2005; Asveld, Ganzevles, and Osseweijer 2015).

2.3 The Role of the Engaged and the Timing of Public Engagement in Research Processes

According to some researchers of participatory processes, such as Sheila Jasanoff (2003), it is impossible to include all members of society directly in techno-scientific research processes. From an RRI perspective, research and innovation is to be responsive, i.e. responding to new knowledge as it emerges, to imminent perspectives, views and norms (Von Schomberg 2011; Owen et al. 2013). There seems to be consensus that involvement at an early stage of the innovation process is important to enable societal embeddedness in the research design, good governance and reflexive approaches (Delgado, Lein Kjolberg, and Wickson 2011; Von Schomberg 2019).

Whether PE processes are implemented at early or late research stages also correlates with the role(s) attributed to the stakeholders. While engagement might happen as one-way communication, attributing a passive and listening role to participants, societal stakeholders might also be engaged in two-way communication, a still passive process of consultation to more active involvement. Following Tina Nabatchi (2012), deliberative processes only start if societal actors are actively involved in problem definitions with the highest possible degree of participation once final decision making is in the control of the public (Nelimarkka et al. 2014). This model also suggests that these different levels of participation are no closed containers, rather engagement might shift along the continuum of the research process (see Fig. 8.1).

Fig. 8.1
figure 1

Spectrum of participation. (Nelimarkka et al. 2014; based on Nabatchi 2012; and IAP2 2007)

Full size image

2.4 Management of Expectations

When involving societal stakeholders, not only researchers, but also invited stakeholders have expectations of the process ahead. The management of expectations hence becomes a crucial and difficult part throughout the engagement process. In case created expectations of the engaged are not fulfilled, stakeholders might well get frustrated (Marschalek 2018). Therefore, it seems to be particularly important to agree upon a certain level of participation upfront, aligning levels of participation (see Fig. 8.1) with clear promises made, in order to prevent unrealistic expectations of the process and a mismatch of announced actions. The International Association for Public Participation (IAP2 2007), which created the original version of Fig. 8.1, put forward ‘promises’ communicated to prospective participants for every level of participation employed (see Fig. 8.2).

Fig. 8.2
figure 2

- Promises to the public based on the Spectrum of Participation (based on IAP2 2007, own illustration)

Full size image

Also, every engagement process needs to come to an end. This raises the question when participatory processes should be terminated. Recurrent critique has been expressed on the premature closing of engagement processes, leaving participants uninformed about both the way their inputs will be used and the outcomes of the research processes in general (Marschalek 2018). Participants often are no longer involved in the evaluation phase of the processes (Burgess and Chilvers 2006) and can therefore no longer contribute with their perspectives. In order to guarantee that outcomes of engagement processes are also fed back into R & I actions and policies, Rosina Malagrida (2015) recommends planning every engagement process with follow ups and iterative feedback loops.

2.5 Group Dynamics

Stakeholder processes often include diverse actors who bring their own and specific perspectives, knowledge and value systems with them. Whilst heterogeneity might be a guiding principle in the process in order for the “complexity of public problems [being] represented by the infinity of other stakeholders involved in partnership and collaboration, each with their own and legitimate value frames, and even ideologies, with regard to such public problems” (Blok 2019, 252), more homogeneous groups are more likely to reach agreement on problem definitions and the required solutions (Selsky and Parker 2005). Since RRI aspires to solve the grand societal challenges of our times, multi-stakeholder processes are pertinent (Blok et al. 2019).

Diverging perspectives, values, needs and consequently problem definitions pose a challenge for fruitful collaborations. In this regard, Vincent Blok (2019, 245) problematises the emphasis on consensus, harmony and alignment, predominantly discussed in literature of partnership formation. Chantal Mouffe’s (2013) conceptions of agony, which is not seen as constraint but as open contestation of diverging knowledge, resonate with the context of RRI. Diverging perspectives can hence be interpreted as the result of a democratisation process, whereas consensus must be taken with care, possibly resulting in an “oppression of marginalised voices” (Valkenburg 2020, 354). Conflict might hence be productive in fostering processes of learning and reconfiguration, however, not all perspectives can be reconciled in an inclusive and respectful way (Van Bouwel and Van Oudheusden 2017).

This plurality-consensus dilemma is not solvable by maintaining an emphasis on consensus processes. Balancing these tensions at the micro-level might also lead to intransparency, i.e. ambiguity in order to achieve a “balance between being understood, maintaining a specific-self-image, and not offending others” (Christensen and Cheney 2015). In this sense, transparency is a normative concept playing out in engagement processes which is strongly related to trust and power differentials within the group of engaged societal actors (Selsky and Parker 2005). While transparency is a core element of RRI processes (Owen et al. 2013; Wickson and Carew 2014), it can also be an “outcome of responsible innovation” (Blok 2019, 226).

2.6 Effects and Impact

Since PE activities are cost and time intensive for all engaged parties, it is legitimate to ask for their results. As Martina Nitsch and colleagues emphasise, the strength of participatory approaches lies in their “contribution to empowerment and social change “(Nitsch et al. 2013, 44). Repeatedly, however, there is no identifiable relationship between PE and resulting processes in research and innovation (Scholl et al. 2012). In many processes “the how trumps the why” (Stilgoe et al. 2014, 5), and the implementation of engagement processes is deemed more important than achieving a certain result. Accordingly, the usefulness of PE processes might be questioned.

Engagement processes often only involve a handful of people, but even attempts on larger scale such as the VOICES project (Broerse et al. 2014) are criticized not to reach far enough or deeply enough to counterbalance hegemonic research and innovation processes, being “microscopic against the backdrop of global science and its governance” (Stilgoe et al. 2014, 11). Again, the issue of promising too much is at stake. The scope of engagement processes is hence finite, their ability to create large impact seems to be limited. Small scale real life applications, which directly relate to engagement processes - as for example introduced by the concept of transdisciplinarity (Jahn et al. 2012) are still missing.

2.7 Requirements in Public Engagement Processes

While PE processes are difficult to handle, a considerable number of engagement activities have already taken place. A proliferation of participatory processes, however, stands in contrast to the lack of standards for PE activities and their expected results (Stilgoe et al. 2013; Emery et al. 2015; Rip 2018). While there have been several attempts to create common standards and understandings, up to date, they have not been successful. In a recent review of the uptake of RRI among EU Member States’ research and innovation programmes, John Pearson (2019), 109) emphasises the problem of void concepts of ‘engagement’, which are downsized to “selling science” to the public as “potential customers of innovations”. More participatory approaches, hence, might not necessarily imply more responsible and responsive research and innovation processes, but might rather be stripped of their transformative potential (Stilgoe et al. 2014; Gianni and Goujon 2019).

This review of well-known challenges of Public Engagement processes in the context of R & I underlines the difficulty of implementing participatory R & I processes successfully. To sum up, in the following a list of challenges as discussed in the literature translated into basic requirements to be addressed in PE processes, is presented.

2.8 List of Requirements

  • Targeted selection and invitation of participants including reluctance to participate

  • Clear roles for participants

  • Management of expectations

  • Addressing of Group dynamics

  • Visible effects and measurable impact

  • Outlines and criteria for engagement process

In the following, the Social Lab approach as a tool to operationalise PE in the context of RRI will be evaluated against this background in later sections. At first, emphasis is put on the definitions of participation and engagement.

3 Social Labs to Operationalise Public Engagement for RRI

Zaid Hassan brought the Social Lab (SL) approach forward with his book “The Social Labs Revolution “(Hassan 2014). As he describes, complex challenges of today’s world, such as food security or climate change, cannot be solved with business as usual solutions. Instead of relying on complex planning processes to seek solutions, Hassan argues for an approach of doing rather than planning by using experiments and prototypes and involving the most diverse groups of people. A Social Lab termed by Zaid Hassan in analogy to other types of labs, is not a method but rather a paradigm or an approach which is social, experimental and systemic. SLs are social in the sense that they bring together stakeholders from different fields who actively work together beyond pure consultation as increasingly demanded in PE processes. SLs are experimental as the involved team continuously tries out innovative solutions through an iterative approach and prototyping interventions. The interventions are systemic since they do not address symptoms but the root cause of why things do not work in the first place. Thus, inductively the team moves forward evaluating what has worked out and what has not, adapting actions to new information. As described in the Social Labs Fieldbook (Hassan et al. 2015), a lab always focuses on a specific challenge. It represents a stable space supporting the required practice.

The European funded New HoRRIzon (NH) project chose the SL approach for particularly fostering PE processes in the context of RRI and modified the approach for its needs (Timmermans et al. 2020), i.e. integrating RRI in research and innovation systems on national and international level. NH targets the 19 programme lines of the European funding programme, i.e. Horizon 2020.

For each programme line, a SL was organised, bringing together stakeholders of the programme line to work on RRI practices and uptake. Each lab aimed to better include RRI in the field. So-called pilot actions were to be developed to address challenges, which were identified for each programme line based on a previous phase of document analysis and expert interviews. The participants leading these pilot actions are henceforth called pilot action hosts.

Each SL had predefined roles based on common project outlines, which supported the SL as a whole and the pilot action development and teams:

  1. 1.

    The SL manager was responsible for setting up the Social Lab process and team. In addition, the manager connected the individual SL with other labs and the project as a whole.

  2. 2.

    The SL facilitator was responsible for designing and facilitating all three face-to-face workshops of a SL putting an emphasis on co-creative workshop techniques.

  3. 3.

    SL assistants supported SL managers and facilitators.

As shown in Fig. 8.3, each SL consists of three main pillars: face-to-face workshops, pilot actions (or short: pilots) and the setup of communities of practices. Each lab team met three times for two-days workshops, but communicated also in between within a period of almost two years.

Fig. 8.3
figure 3

Social Lab structure in NH project

Full size image

The first workshop served to start and form the SL, scoping the challenge statement and starting the prototyping. Further, it served as selecting pilot hosts and forming sub-teams who supported the hosts. The second workshop aimed at evaluating the ongoing pilots, reiterating and in some cases creating additional pilots. The third and last workshop focused on the sustainability of the pilot actions and put a conclusive assessment area at its core.

4 Methodology

The list of requirements in public engagement resulting from literature analysis in the paper is applied to a specific case, i.e. on the public engagement process as operationalised by SL in the NewHoRRIzon project. A case study allows for an in-depth examination of a particular case (Yin 2009). One of the central aims in qualitative research is to develop hypotheses and to build theories (Mayring 2000) and to classify (Lazarsfeld and Barton 1957). Classifying means to systematically order the material following the needed classification rules that are theory driven and empirically solid.

The coding system to develop allows for a systematic analysis of the data, while the rules for applying the codes have to be fixed in a coding plan with the purpose to increase reliability and validity across different researchers working with the same material. While there is a variety of techniques for interpretation available in qualitative research classification and structuration are the most appropriate forms of interpretation for our case, i.e. the SL approach in NewHoRRIzon. The aim here is to explore the data material according to the defined rules and categories (codes) and describe the transverse section in the data; thus to apply the predefined codes from literature (list of requirements in public engagement) to the data material.

Next to the literature review (Sect. 8.2), we analysed the following materials. The analysis is based on different materials and data sources that documented the 19 SLs and two cross-sectional workshops shortly described in the following section:

  • SL workshop templates (pre and post)

  • Documentation of the two cross-sectional workshop

In order to document the three workshops, SL managers were asked to fill in a template before and after each workshop, i.e. at six different moments within the process.

The documentation of the cross-sectional workshops describes the process and outputs of both two-day workshops, where all SL managers and facilitators as well as members from the Advisory Board (and participants of the SL in the second cross-sectional workshop) came together to fertilise each other and to share insights and lessons learnt.

The material and data sources described above were used as the empirical basis to distil common challenges and solutions, by applying qualitative analysis, combining deductive and inductive coding (see for example Flick 2014).

5 Results

The focus of this section is not the output of the SLs but the experiences from the SL process. The following follows the structure of requirements of Sect. 8.2, which also served as codes in the qualitative analysis.

5.1 Selection of Participants and Persuasive Efforts of Invitation

Forming a SL team, steering motivation, ownership and trust from the very beginning, i.e. already at Workshop 1, in order for team structures to evolve within the SL, was reported as quite a challenge.

It all started with the recruitment procedure to achieve a good balance between different stakeholder groups and different levels of hierarchies. A SL manual (Braun et al. 2020) clearly outlined the numbers of participants in each lab, which ranged from 15 to 20 persons. Also, the lab teams should be diverse representing different stakeholders as well as gender identities, age groups, and regions. According to these outlines, a stakeholder mapping was undertaken by the SL managers to identify institutions and persons relevant in the thematic field of the lab. The following recruitment process was based on this mapping, combined with a combination of targeted e-mails to selections through stratified random sampling and snowballs.

However, many of the designated participants could not see any ‘value’ in participating, or declined because they did not see themselves up to the task. In this case, clear communication was necessary in order to convince SL participants to take part in the process. They had to be informed about the exact timing and dates of the workshops, and also the extent of the process in general. Time still remained an issue. Although lab managers had sent out save-the-date announcements, it was difficult for participants to find the time for the three two-days-workshops and additionally for the required activities in between the workshops. Also, the need to commit over such a long period of time had a deterrent effect.

The path taken - albeit not inherent to the SL approach - fitted the approach well and resulted in 19 diverse SL teams. However, the challenge of reaching less represented groups remained pertinent and partly unsolvable; despite targeted invitations, civil society representatives remained underrepresented across many of the 19 SLs.

Furthermore, due to the drop-out of participants, after workshop 1 further participants had to be recruited. Labs tried to keep the team structure and replace participants, or they tried to get new participants because of their particular relevance for the selected pilot actions. Although lab teams tried their best to re-recruit after workshop 1 and, the general numbers of participants declined up to workshop 3 across all labs.

5.2 The Roles of the Lab Participants

During the first workshop, the topic of the SL, the roles within and the intentions of the process and goals of each workshop were explained. Also, the idea of the pilot activities (including the budget frame) were communicated.

Signing up for the role of a pilot host and to be clothed with tasks respectively proved to be a prerequisite for a successful pilot action as mentioned in many reporting templates. No pilot host was coerced to take on this role, rather, the pilot action co-design and selection processes were set up steering ownership of the ones collaborating, with pilot hosts and teams opting in themselves. In the cases where hosts remained unclear or were not able to dedicate as much time as required, pilot actions were either dropped or worked out poorly as reported by the lab management team. In cases where hosts fully took responsibility and dedicated sufficient time and effort, pilot actions prospered better. However, not all stakeholders were eager or able to take part in the research process.

5.3 Management of Expectations – Transparency

Quite a high level of commitment was expected from participants; they were asked to attend all three workshops within a process that lasted up to two years and further required to be active in between the workshops which could not easily be achieved. Whilst travel and accommodation costs were reimbursed participation was voluntary and unpaid. Not everyone was able to combine this task with their everyday working life depending on the employment situation and the type of institution they were affiliated with. Also, not every stakeholder group seemed to be equally accessible and open to the topic of RRI and the process of the SL.

While a clear outline of the process ahead was important and included in every invitation, the demands of the SL in case of time and resources needed partly represented insurmountable barriers to participation for those, who could not use their working hours for the SL. Participants invested much of their time for the lab process and the pilot action implementation. In case they did not become aware of the value of this endeavour, they dropped out. However, across all SLs about 27% of the participants dropped out; those who stayed were highly committed.

This sense of ownership among the participants is crucial not only for the implementation of the pilots, but also for the overall and voluntary engagement over this long period of time between the workshops. Ownership needed to be continually nourished by the lab management teams. They not only addressed the different values and interests of their diverse lab participants, and applied many group forming activities, but also helped the pilot teams to define their next steps and organised support for these tasks. As the time of the face-to-face workshops was not sufficient for implementing the pilot activities, lab managers stayed in permanent communication with the pilot teams throughout the whole SL period of one and a half years. They organised online meetings and calls with the pilot teams, telephone conferences to discuss pending issues or contents and to interlink the pilot activity with other SLs of the project.

In this way, while still supported by the lab teams and managers, participants could enact their own responsibility for the action and create their own tasks at their own pace rather than being called upon. This is fundamental for involving the participants and aligning the outcomes with promises – the participants themselves define what to expect.

At the level of content, visioning exercises such as future sentences (which have been collected and discussed in preparatory workshops and provided to the lab teams) were used as a tool to bring everyone on the same page. The steered reflection of the participants’ own attitudes, roles, potential barriers and enablers worked well in order to foster a sense of agency. The same holds for explicitly identifying best practice examples of RRI on a systemic, institutional and personal level.

In general, clarity in respect to the SL methodology, its objectives and its process was a major challenge and an issue recurrently reported by the lab managers. A certain level of standardisation was necessary to allow for comparable processes in all 19 SLs, but it was also important to regard each SL as a place for experimentation with much room for ideas and options for individual reaction. While the lab process was conceived as an open process, still guiding questions, visions and social challenges needed to be kept in mind. This struggle for balance between structure and standardisation on the one hand and openness on the other hand, sometimes resulted in clashing positions: “The presentations broke the flow of the workshops: The idea of the workshop to be an exchange of ideas, to get active, to engage and to debate was strongly contrasted by the presentations, where participants were required to be rather passive listeners” (SL10, WS1). In contrast, for other participants the rather open design of the workshops also created moments of frustration and it did not lead to their intended effect of bringing them out of “their own nuclear bubble” (SL19, WS1).

Having an open and clear communication will help in several aspects. Statements and observations by SL management teams show that it is very important to clearly explain the SL method, the entire process and the specific objectives right at the beginning. This “may help to prevent lacking clarity and understanding” (Marschalek et al. 2021, 49). Secondly, clear communication and transparency on goals, requests, limitations and documentation fosters the building of trust towards the organisation as well as its SL team. Thirdly, the whole process itself has to be transparent to allow all actors to understand why certain steps were undertaken including the follow up.

Given the time and resource efforts that are requested from participants, it played out well to distinctly outline the value of participation in the SL. For some participants it was sufficient to raise awareness that their own reflections and ideas might impact the R & I world and that their input is inspiring for other RRI stakeholders. Outlining the consequences of undesirable effects of irresponsible R & I might trigger motivation to change them. Further, this method might support the involvement of different stakeholder groups, who are not interested in taking part in the first place. During the workshop participants were informed about activities of other labs and the project website was increasingly filled with information on workshops and pilot activities of all 19 labs. Cross-sectional learning and dissemination activities show results of lab processes of all programme lines addressed. Tangible pilot outcomes, such as printed guidelines developed by pilot teams, were distributed across the labs or can be found in the project repository. Informing participants about the big picture helped them to look beyond their own nose and reflect on the wider impact. For instance, participants appreciated the chance to participate in “something innovative and being able to shape it” (SL7, WS3). Any individual appreciation, such as “awarding” participants (SL 4, 9, WS 3) for their engagement were welcome. However, from practical experience we emphasise that the more an incentive or issue is directly connected to the (professional) environment of the participant, the higher the commitment and engagement will be. In case the professional interest is lacking, intrinsic and extrinsic motivation become even more important. Expense allowances to especially support civil society actors to participate, and activities developed in the lab which help bring forward already existing ideas or initiatives will positively impact the participation and engagement in the lab.

5.4 Group Dynamics and Settings

According to the ‘SL design Workshop’ (November 8–9, 2017) participatory workshop activities and techniques were applied in the labs to create a sense of ownership and commitment to the lab and its activities. Certain exercises which should enhance collaboration and co-creation, such as group forming activities or workshop rules and explicitly applied attitudes such as ‘active listening’ would help lab teams to get to know each other and “cultivate mutual trust among SL participants” (Braun et al. 2020, p 12). The engagement and motivation of participants requires a sound basis of trust and the feeling of being valued as a person. Only if participants trust that information and ideas presented are valued by others, they will share information openly. Consequently, much emphasis has been put into trust building from the very beginning in order to allow the participants to grow as a team.

Within SLs also friction was noted at different occasions, for instance different ideas on what lab participants intended to achieve with their Pilot Action led to controversies and tensions within the pilot teams could also lead to redesigns of the action or re-arrangements of pilot teams (Marschalek et al. 2021, 27). However, disagreement is somewhat intended by composing such heterogeneous lab teams as it emphasises the diversity of engaged perspectives since diverse stakeholders might also bring in diverging perspectives. Often, lab participants could recognize the diversity of perspectives, values, and different points of view and appreciated its added value.

Nevertheless, tensions have to be dealt with carefully in order not to block the entire SL process or to have a biased discussion throughout, with pilot actions being supported only by a few team members. Throughout the face-to-face encounters it was challenging to make sure that all voices are heard and to manage that participants had equal opportunities to speak up their mind. Everybody must feel represented and empowered to talk and to let others talk too, with each input being valued equally. This process needs support from SL managers and facilitators and hence their roles are essential to manage the diversity of personalities (aimed for in SLs) and usual group dynamics. For the facilitators it was a challenging task to work with the group diversity: “The main challenge thus was to take everybody with us from the basics to a practically usable result at an immense pace” (SL 11, WS 1) (Marschalek et al. 2021, 26). The idea creation and development process of the pilot actions mostly happened during the face to face workshops in which all participants were enabled to contribute. The selection of actions was often organised in transparent voting procedures (e.g. voting with ones’ feet or sticky dots), however, given the heterogeneity of the groups and diverging priorities, not all participants were satisfied with the results.

Charming localities as well as offering an interesting programme to work with a remarkable team expresses appreciation for the participants and their engagement. Therefore, several SL workshops were organised in pleasant surroundings, providing for a calm ambiance and both spaces for recreation and interaction. Often, bright rooms with lots of windows, nice views or furniture were chosen. Also, a flexible setting with chairs and tables that can be moved to provide space for different workshop methodologies was emphasised as key for choosing workshop rooms (D 7.4, 30). Team building initiatives such as ice skating together as a group after the workshop or an RRI inspired activity game were organised by many labs. By bringing participants closer to each other in different activities and environments e.g. through walkshops, i.e. guided discussions in groups while walking (see Wickson et al. 2015) - a good working atmosphere can be created. The analysis shows that the spatial and timely setting can also support trust building among team members, requiring sufficient time for interaction between SL team members.

5.5 Effects and Impact of SL Participation

The analysis shows direct and indirect effects on the participants and their wider networks resulting from their participation in the lab process. The direct effect and impact relates to participants as this quote shows: “The greatest transformative effect was probably on the SL participants themselves because the SL gave them the opportunity to dedicate time to learn about and engage with RRI” (SL7, WS3).

A direct output of the SLs are the pilot actions. Overall, the 19 SLs each developed between one to five pilot actions, dedicated to implementing RRI in practice at different levels, resulting in a total of 57 pilot actions. These pilot actions were co-designed by teams of SL participants, from an early prototyping to their implementation. All of these pilot actions, however, are at their very core a product of stakeholder engagement processes. The pilot actions comprised tangible outputs (such as RRI Career matrix or an open web repository for sustainable energy), RRI training modules for different stakeholders, awareness raising activities, etc. Thus, the indirect effect and impact is associated with the pilot actions which addressed and engaged further stakeholders.

SL participants and in particular the pilot hosts mostly doing the lion’s share of the pilot actions do have the clear potential of becoming change agents, as they have a clear goal in mind for improving or changing their working environment. Often they were inspired by the RRI concept which only a few of them had explicitly heard of before. In this sense, as RRI was the umbrella topic of each lab, the lab approach helped to spread the idea of the RRI concept and allowed for personal experiences with the notion, which could contribute to a personal uptake of RRI (which was clearly stated in workshop 3 by most of our participants). While early signs of change in some of the participants’ institutions was noted, a real impact on the institutional level takes time and needs actions along several different angle points step-by-step. Also, only a few signs of impact on the policy level could be detected, i.e. RRI taken up at national level in funding instruments.

6 Discussion

The subsequent discussion section interlinks the results gained from our SL process with challenges and requirements as identified in the literature (see Sect. 8.2.1). It critically reflects on the way and the extent the SL approach meets the requirements for PE processes in the context of RRI and whether it offers appropriate solutions to overcome frequently discussed challenges. Findings from the literature, i.e. requirements in public engagement processes, have been contrasted with process elements in the SLs.

The following table plots common requirements identified against the way they are or can be addressed using the SL approach. The later section chronologically elaborates on the way the SL approach deals with these requirements (Table 8.1).

Table 8.1 Evaluation of the Social Lab Approach
Full size table

As indicated in the table below (c.f. section 1 in the table), targeted selection and invitation on the basis of a commonly agreed and pre-established set of criteria was applied in the SL. Following the approach of invited participation (Bogner 2012), much attention has been paid to stakeholder mapping and to individual selection and invitations of potential candidates (1a). Accordingly, the recruitment processes were time demanding. However, although reasonable numbers of dropouts were reported across the labs, still a sufficient number of participants could be committed to the labs (1b). The problem that CSOs are difficult to attract remained an issue also in the SLs (1c). How to best incentivise participation of CSO representatives is still an open question.

As outlined in the theoretical framework, roles (c.f. section 2 in the table) in engagement processes have to be made clear and put in line with expectations placed on the participants (see Fig. 8.3). The roles present in the SL approach were explicitly defined prior to the start of the process and communicated right from the beginning, thereby making it easier for participants to understand their responsibilities or demands connected to their roles. Placing the final decision making power in the hands of the SL team, starting with the first workshop the SL approach corresponded to the last spectrum of participation: empowerment (see Fig. 8.1). Concerning the timing (c.f. 2a) of the engagement processes, SLs as they are meant for applying deliberative processes from the problem definition via idea creation until full implementation of co-created ideas, they enable participation throughout the whole life span of the activity.

Much effort has been made to communicate intentions understandably. Having the requested effort in mind, it is essential to communicate expectations (c.f. section 3 in the table), goals and aims but also limitations of the SL (project outlines) and the common understanding of the (RRI) topic clearly from the very beginning. Reflections and feedback from the participants emphasised that at times they did not feel sufficiently informed and thus transparency could have been enhanced. SLs have a clear end once (c.f. 3a) the pilots have been fully implemented or finished. Reflection exercises within the workshops and the possibility for feedback and cross-checking at the second cross-sectional workshop (see Fig. 8.3) help to validate the communicated results before they are widely disseminated.

Pilot hosts did not work on their own but needed the support of the lab management and participants’ team. Therefore, much attention has to be paid to team collaboration and group dynamics (c.f. section 4 of the table). Lab teams were often invited to stimulating environments which offered many opportunities for working together, but also for informal encounters and team activities other than sole working. Lab teams could retreat from their daily business and thus better focus on the workshop activities. As already noticed in similar settings (Wickson et al. 2015), this approach also softened hierarchical or other imbalances between participants. Instead of always looking for consensus (c.f. 4a), the SL offered room to disagree. The pilot teams could organise their team processes independently from the lab teams, equally team members could refuse to support a pilot action and remain part of the SL process as a whole.

As indicated in the table, effects and impact (c.f. section 5 in the Table 8.1) have been achieved only partially. While direct and indirect impact on the participants themselves and their wider networks has been well documented, only a few signs indicate a broader uptake on e.g. policy level. Thus the impact requirement of PE has not been fully addressed as the link to the research and innovation system on a (EU) policy level has only partially been established. At the level of the participants, however, the SLs have contributed to empowerment of participants in the sense of Nitsch and colleagues (2013).

The pilot actions and their outputs represent small scale real life applications (c.f. 5a). These actions represent tangible outcomes of an engagement process and are as such a product of a stakeholder engagement process. Pilot actions hence underpin endeavours of the pilot host and thus strengthen their role as change agents in their institutional background. Allowing for a positive experience in practical implementation, pilot actions counteract the participation fatigue, as they provide remaining results. While their results are subject to ongoing analysis, already now they provide evidence of representing innovative activities and new materials for how to implement RRI in different contexts and achieve institutional change.

7 Conclusion

With this article we have described SL processes within a case study and investigated whether SLs represent a suitable approach to operationalise PE for RRI.

In general, we have shown that the SL approach provides a potential strategy to cope with some of the challenges discussed with regard to implementing PE processes in and for RRI such as providing a clear definition of roles to all involved actors, providing for transparent yet dynamic and co-creative processes, producing clear outputs (pilot actions), while allowing for space of disagreement or agony. Other challenges, however, remain obstacles for implementing engagement processes when using the SL approach, notably, these concern the underrepresentation of stakeholder groups, intense time demands, and resource requirements.

The SL approach further proved to work well with the circular character of PE as a central part of the lab approach and a core concept of RRI. Therefore, we argue that the SL approach contributes to solving some of the challenges of PE and is a fertile ground for supporting PE in and for the context of RRI.

To date, no common guidelines on how to implement SLs exist, especially not in the context of RRI. This is often the case within participatory processes – although signposted with terms, such as “Social Lab”, processes lack clear instructions or quality criteria (Rip 2018). The project team therefore had to start with some basic understandings on how to implement SLs. In this case, therefore, it took a while and much effort to come up with agreed guidelines which could be applied across all labs. Group exercises, such as a visioning conference, or a co-creation workshop carried out the different roles, lab methods and processes to be applied (see Braun et al. 2020).

Accordingly, looking back at 19 lab processes we can observe that although the labs followed the same rough structure, still the labs have been carried out individually. With regard to this fact further research and implementations are needed to investigate and define common criteria on how to successfully run a SL.

19 SLs offer(ed) a unique possibility to re-evaluate opportunities and limits of PE processes for RRI, however, more research and experimentations are necessary in this field - how can SLs be implemented in a sustainable manner? How can missing stakeholder groups like civil society organisations and policy makers be reached and included? How to cope with the intense resource demands of the SL process? With this chapter we hope to steer further interest, discussion and research in these matters.