Table 4 Quotes illustrating opportunities for reflexivity in a layering of learning domains (e.g. cognitive, epistemological and normative).

Team A

“I’m not sure if our role really should be to improve how the actors work with each other…is it creating a tool together with the stakeholders, that the stakeholders ideally would adopt? Or using information for data decision making? Or if our role is to kind of change that network and how the actors work with each other? I don’t think that is our role as scientists, but I might be wrong.” […] “that’s probably linked more to my role […] where we usually generate information. We are not generating tools, we’re generating information that policymakers can use for their decision making. But someone else within the team may be generating tools that can help to link up or develop the network between stakeholders a bit better. So that person may have a slightly different view.” (researcher, workshop #1).

Epistemological, normative, relational domains: Researcher recognising onto-epistemological pluralism and multiple potential researcher roles (positionality) within the group which shapes co-production purpose. Relational domain: in building trust, acknowledging different views within the team

Team B

“I’m not entirely sure how a researcher would actively broker power or, to be honest, if that should be part of our role. I can see how research and research outputs can influence and change power dynamics, and I understand in terms of the Chambers modes that that is an important aspect. But in terms of deliberately and actively doing that, I guess I’m not totally clear on what that would even look like, to be honest.” (researcher, workshop #1)

Cognitive and normative domains: Opening up cognitive learning as to what constitutes ‘power’ in sustainability science contexts. Reflexive opportunity as to the role of scientists (positionality, normative learning domain)

Team C

On brokering power: “[…institution] was meant to be independent, right, and so that’s a bias that probably doesn’t fit with social scientists as far as truth, but [institution] are meant to be independent. And so, I think brokering power amongst different stakeholders, and I think if you asked most scientists across [institution], they wouldn’t be comfortable saying that because they wouldn’t acknowledge their power. But also, they feel like they’re meant to be trusted advisors that are independent.” (researcher, workshop #1)

Cognitive and normative domains: Opening up cognitive learning as to what constitutes ‘brokering power’ in sustainability science contexts. Reflexive opportunity as to the role of scientists (positionality, normative learning domain) and science institutions

Team D

“For example … on ‘predicting trends and patterns’ and ‘explaining processes’ I think it is useful to have team members who do still think that that’s very important because it is a primary way in which scientists are held to account, right? And it’s also a core expectation of scientists because we are trying to reframe things a bit about one way of dealing with complexity and uncertainty is not to say, “We as scientists can come in and reduce it,” but we can help co-develop methods to navigate that uncertainty and complexity without reducing it. But at the same time, there is a strong demand for us to be able to explain, predict, reduce uncertainty wherever we can, and I think it is important to hold both those things.” (researcher, workshop #1)

Epistemological, normative and relational domains: Acknowledging and validating diverse researcher positionalities and skill sets, by directly linking to purpose (reframing discussion while retaining legitimacy) within sustainability science contexts.