Podcast: How Synthesis Shapes a Scientist

What experiences have shaped your perspective on science? 

Ecologists Elizabeth Borer and John Drake have a formative experience in common: they both were postdocs at NCEAS in the early 2000s, when synthesis was still blossoming as an approach to scientific discovery. Now professors - Elizabeth at the University of Minnesota, and John at the University of Georgia - they share how conducting collaborative synthesis science as early career researchers has influenced their work and views on science in our first-ever episode of NCEAS Portraits, The Podcast Edition. 

They both are also panelists at a symposium organized by NCEAS for next month's Ecological Society of America Annual Meeting, entitled Insights and Advances from Synthesis Science in Environmental Science.

Listen to their conversation with host Jenny Seifert.

*A tiny clarification: at 27:47, John mentions an organization that goes by NutNet, which is the Nutrient Network, a global experimental network focused on grassland health that Elizabeth co-leads.

The following transcript has been edited lightly for length and clarity.

Jenny: What would you say is the evidence of your NCEAS postdoctoral research and experience on your work and career today?

Elizabeth: I have a few arms of an answer for this. The first is I have developed a deep comfort with collaborations. I have a desire to work with others to form questions collaboratively, to generate a larger vision for what we should ask and answer and how we should approach that. The second is I have a deep appreciation for meta-analysis as a fantastic tool for hypothesis generation, more than testing. I think it’s a really wonderful way to quantitatively understand what we do know and what we don’t know, and if there’s bias in what we’ve ever studied. And the third is related to that. I have kind of a cottage industry. I’m deeply interested in generality, and my cottage industry is doing meta-analyses to compare evidence across ecosystems, but also using multiple tools. So, experiments, observations, models and meta-analysis to gain insights into focal questions. I’d say each of those are threads that really began during my times and interactions at NCEAS and have continued through in various forms since then.

Jenny: John, what are your threads?

John: You know, it’s interesting, they’re very much like Elizabeth’s. They are mostly not about the content, scientific theory, or any particular dataset. It really has to do with the way that I think about science as an enterprise in the first place. And you can say broadly methodology – the way I go about doing my science – but it’s not about any particular technique. For instance, I don’t think of research as something that a person does or even what a lab group is doing. I think it’s really a community endeavor. And one of the things about NCEAS and synthesis science in general is they taught me to think of our theories as evolving in a conversation among a large number of people. And I think that changes the way that you think about your ideas, it probably changes your attitude toward intellectual property and proprietary knowledge. And I think that it emphasizes the value of teams. And those teams might be explicit teams, or they might be implicit teams. Like, there are a lot of us that are involved in a conversation here, and I don’t even know who all of those people are, but I recognize their thoughts and ideas have somehow gotten wrapped up in what I’m doing and probably vice versa. One of the things that I think synthesis science does well, that it does for science in general, is that it helps us to quickly sift through all the ideas that are on the table, and let the real gems emerge. As a community, then, we can pivot around those ideas and work on them together.

Jenny: John, what of all of those things do you think was most valuable about either your experience with NCEAS specifically or with synthesis science in general that you still find yourself still promoting or doing today?

John: One of the things that I do at the University of Georgia is I involve a lot of students in team projects. And very often we bring in team members that are not even at our campus through teleconferencing or email. We use lots of distributed tools like GitHub and Overleaf in order to be able to work together in real time and share our results. I think that’s actually been huge for the efficiency of science and the extent to which we’re able to engage in collaborative projects that topically are much broader than any of us is able to manage on our own. The other really valuable thing has been position with respect to the field. One of the great things about NCEAS is it was this place where lots of people came from all over the world who were interested in approaching science in the same way. And the number of friends that I made there, and the extent to which I reach back to those people now – maybe not even has direct collaborators, but as people that are able to put me or, more importantly, the students in touch with other people who have some kind of resource. I think being able to connect the discipline of ecology from one end to the other is one of the real values that NCEAS has contributed to the field.

Jenny: Elizabeth, how about you? What do you find valuable about your experience with NCEAS that you still promote or do today?

Elizabeth: There are definitely threads that are going to run through what both John and I are saying. The first response that I have is the real value was meeting people who I was able to find as mentors. Not only well-established scientists who were coming through NCEAS, who I could seek out and bounce my brand-new ideas off of, but also not-yet-established scientists who have become very much my peers. It’s invaluable to have that network that persists and has grown. And the second thread is I discovered that I really love helping to facilitate scientific groups. I take great joy in thinking about what could a group create together that individuals couldn’t create alone. Who would the people need to be around that table? And then, after I brought them together, how can I maximize the vision and productivity and novelty, but also the fun of the science. I think partly that’s just who I am, but also it really began with my experience of putting together groups at NCEAS. I’ve discovered it helps me feel like, as a group we’re creating a much richer understanding of a topic. I also find myself really surprised when I’m in a group and it’s unusual for others to be operating this way. I’ve realized that I’ve so shifted my view of what science is and how it should function, that I almost feel like I’m on different sides of a continent when someone is confused by this sort of collaborative nature. For me it’s just joy.

Jenny: That makes an interesting segue into my next question, which is related to working in the collaborative environment of synthesis. You both spent an early part of your career at NCEAS. What would you say are some of the common challenges that early career scientists might experience doing collaborative synthesis science, and how might have those challenges changed since you’ve been a postdoc?

Elizabeth: The first thing I think about when you ask that is fear. In your early career, there’s a lot of insecurity. There is fear that ideas spoken out loud in a group of more senior people may be perceived as dumb or uninteresting. There’s also the fear that someone who is more well-known may intentionally or unintentionally steal an idea or take it and claim it as their own. I mean, it requires trust to work in a collaborative group. I think one way to move around that is to have a really thoughtful and engaged facilitator who’s aware of the different roles that people play and the power structure in the room. I’d also say, early on – and this is back in 1998 when I first started interacting with people at NCEAS as a grad student – synthesis was not considered a science. There was a very large swath of ecologists who said, ‘You should not be going to a postdoc like that. This is not a place where you can do science. You’re just writing review papers. That’s not new science and you’ll never get a job that way.’ So, the early postdocs that went to NCEAS had to overcome that perception, that as an early career scientist you would never be able to make a career writing a bunch of review papers, which is the early perception of what has now become what I guess you’re calling synthesis science. I don’t know that I would name it that, but nonetheless, there’s now a recognition that this is a real way to test ideas, to frame hypotheses, to really push the field forward. But early on that was not the case. I think I’m about one click ahead of you, John, in when I was at NCEAS, and my sense was, even during the time that I was there, that had really changed. It really shifted from 1998 through to even half a decade later and had become a very desirable postdoc to land. What was your experience, John?

John: I think that’s spot on. I arrived in 2004, and it was about probably 2002 that I became involved in a distributed graduate seminar. And by 2003 my view was that this was the plum postdoc in the field. I think it already seemed to me that those postdocs were choice candidates for faculty jobs around the country. So, something certainly happened in that first generation of postdocs that was very impressive to the field. I would emphasize what Elizabeth said about the risks of doing science in that fashion. In science, we have had for a long time a kind of protégé model of scientific training, and at a place like NCEAS or just in synthesis working groups, there’s no senior person whose role it is to groom you for the next position, to provide you with that kind of career advice and professional development. I think what needs to happen then, is that becomes more of a community endeavor. That requires established people in the field to be generous with their time, ideas, and advice. It requires that, as junior people are trying to find their place in the field, they’re willing to take risks and reach out and cultivate relationships with people that they might not be working closely alongside. I think our field has done that to a large extent, and I’m proud of us for having done so. But it did seem to me something that was missing from the discipline as a whole fifteen years ago. Also, Elizabeth mentioned fear as a concern, a concern that my ideas might be appropriated. I think it's hard to identify where a particular idea came from. And so, I think it requires that we develop a different approach to recognition, and I think an intellectual tolerance – tolerance for ideas that seem conceptually or methodologically out of line with the way we think about things. And tolerance for the fact that I may not be able to trace exactly what my contribution to a particular project was. That the idea and research outcome really are more than just adding together a bunch of people’s different ideas. It’s the idea that emerged in a conversation among a bunch of people with different points of view.

Elizabeth: John, I have a question for you that relates to what you just were talking about. Our models of postdoctoral training are something I’ve been thinking a lot about here, as I’ve taken on new roles at the University of Minnesota. Do you think the synthesis center model of postdoc training is better than the more standard model of one or two postdocs in a single lab with a mentor in that lab, and do you feel there are aspects of either one that the other programs should be considering adopting or throwing out?

John: I think it is right now kind of an either-or, but my hope is that our field can move toward a blended model in which a postdoc can find themselves in both roles. One of the great things about the postdoc community at NCEAS is the mutual support and sharing of knowledge and contacts and experiences. If you’ve got twelve or fifteen people going on job interviews and bringing back the intel from all of that, that’s way more valuable than having a supervisor who interviewed for a job ten years ago and maybe only had a couple of interviews. I also think there’s some kind of energy that comes out of that cohort model for postdocs. On the other hand, what we need in science is not just people who are first-class researchers, but people who are facilitators of other people’s research, who are teachers and trainers, who are organizers, who are able to manage a broad range of responsibilities, some of which are not technical. You learn a lot of that in the lab environment. And all of those are skills we need to make sure the next generation of faculty are endowed with before they start the faculty position. My idea for a hybrid is that maybe there are places – they could be centers or institutes like NCEAS or they could be academic departments – where there’s both a cohort of postdocs, maybe with some support from the institution so they have real obligations to it, but they also have roles within labs, so they can benefit from both kinds of training.

Elizabeth: I agree, I think there are elements that are really beneficial in both, and I think that some hybrid model is fantastic. I think there are many ways we can achieve that. I think we could take pieces of each and create something that will give our more junior colleagues the tools they’ll need to not just stay in academia, but to take these ideas and translate them elsewhere.

Jenny: Let’s switch gears a bit and look to the topic of the symposium that you are participating in at the Ecological Society of America’s meeting this August. The title of that symposium is “Insights and Advances from Synthesis Science in Environmental Science.” In what ways have you seen synthesis science unleash new insights or make major advancements in your own fields, and what about synthesis science do you think enables that?

John: We’re taught as ecologists to think about complex systems. The experimentalist approaches a complex system by dissecting it down to its individual pieces. The synthesis scientist approaches a complex system by looking for all of the different kinds of information that can constrain your picture of how that system works. And so rather than going down to a smaller scale, it reaches out to a broader scale. Synthesis science has given us ways of thinking about the understanding that many of the problems we face in our society have this multi-scale, multi-causal characteristic. I also think it’s one of the areas that synthesis needs to continue to improve on. A lot of what synthesis scientists do is discursive and narrative, and I think that’s important and valuable. But I think it could be made more rigorous with new methodologies for quantitatively representing what these complex systems are and assimilating the ways we communicate about this complex system to the data that presumably are informing what our knowledge is. So, I’m particularly excited about new quantitative methodologies that people are developing along those lines.

Jenny: Elizabeth, what do you think?

Elizabeth: One of the things that I’ve seen is this change in how we ask questions. I feel this thing we’re now calling synthesis science – which I’m not even certain I know exactly what that is – as a field began with case studies. We began with reviews of the literature, and NCEAS really moved it into a quantitative framework. In terms of pushing ideas forward through synthesis science, one is this concept of the leaf economic spectrum. This is the idea that, with photosynthetic organisms, which are all of our energy captured on earth that we use, if we understand the relationship between the different leaf traits, we may be able to understand how environmental change will then modify what species will be in an environment in the future. Another one is this idea of multiple nutrient limitations. In ecology, we’ve historically added one nutrient – nitrogen, phosphorus, or potassium – into ecosystems and said, what happens. And it’s clear that’s not what’s happening. We’re changing where carbon is, and we’re changing phosphorus through agriculture, and understanding how those come together to influence terrestrial systems and aquatic systems is a huge question that we need to address to understand global change. And I would say one big synthesis effort at NCEAS led to a pretty substantial advance because of using that quantitative meta-analytical framework. But I agree that we need more tools. Using synthesis centers to really identify and pursue the development of new quantitative techniques would be one really great way forward for synthesis centers.

Jenny: So, would you say that this quantitative methodological expansion might be the next big frontier for environmental synthesis science?

John: I do think that new quantitative methodologies are needed and there’s lots of different directions to go. I think they are bound to happen and I hope that environmental science leads the way in that. But I think there’s something deeper that synthesis science, that NCEAS-type science has already offered to the community, and that is we can take methods that we think of as being comparative and apply them to lots of new things. Biologists have done comparative biology since the origins of zoology – what does the development of two different animal species look like; how are they the same and how are they different? We’ve taken the idea of comparison and applied it to all kinds of things. For decades, population ecologists would study the dynamics of populations – do they have cycles or how dense are they – as individual case studies. But what synthesis science said is, well you could actually take an idea like cycling, and come up with a quantity to index the intensity, and then plot that all over the globe. And then ask whether or not there are large-scale determinants of that particular pattern. I think what the distributed experiment like Nutrient Network has done is to take that idea of comparative methods applied to large-scale problems in an experimental direction. I think integrating more distributed experiments like that is one of the key ways synthesis science could move forward in the future. I think that has to be done by connecting field biologists to conceptual biologists, and then connecting them with statisticians, modelers, and computer scientists who are able to represent such complex ideas in a tractable way.

Jenny: Elizabeth, do you have anything to add?

Elizabeth:  Yea, I actually think there are a couple. One is questions and one is process. I still see us not having a strong process for accessing and compiling data from multiple studies. We have made various attempts across the field for close to the 20 years now, since the start of NCEAS, when we were measuring with a ruler on graphs to extract data. I mean, those were the first years, and then there were some computer programs. But we still don’t have the ability to query and receive data, for example, from multiple studies in a single format. So, we’re still limited by access to data, which is disappointing 20 years in. My dream, and I think a place that synthesis centers could really push us forward, is to have the limitation be the articulation of exciting questions and not have the limitation be extracting and finding those data. Discoverability is a word that gets used quite a bit, but I think it’s even more than that. It’s also just simply, once you’ve discovered it, accessing it and having it in usable form. In terms of questions, interconnectivity across disciplines is something that we could still stand to really benefit from. We could use synthesis centers to better understand what earth may look like by synthesizing across the disciplines. For example, disease ecology and ecosystem ecology, boy we all study microbes, but man we speak different languages, right? I see those as areas where we don’t necessarily have strong conversations. And synthesis centers could be extremely important hubs where that could happen. But I would say that they stand alongside empiricists, for example, and theoreticians, who are out in the field and creating those new data. Because a synthesis center doesn’t have anything to stand on if there aren’t really strong field biologists generating data and theoreticians generating news insights into the interrelatedness among variables. And so, I see it as one place, but not a panacea to our future understanding and synthesis across fields or subdisciplines.

John: Yea I think the data problem is a really tricky one. The other thing, I think, is workforce enhancement. The kinds of technical tools that scientists are expected to have and that contribute the most to science are changing quickly. And I think there are not enough scientists being trained right now to tackle the problems our agencies and funding organizations have set out and that I think society expects from us. And I think that synthesis centers, probably more than any other scientific institution, have that birds-eye-view of the discipline and are able to advocate for the direction and buildup of a much stronger scientific workforce.

Jenny: In the last few minutes, I wanted to give you an opportunity to ask each other another question.

EB: Oh, totally, I could sit and have a beer with you John. So, both of us have taken on some pretty substantial administrative leadership roles. Has your NCEAS experience given you tools that you have used in these administrative leadership roles?

John: I haven’t thought about it quite from that point of view. It certainly has affected my idea about how scientific institutions can and should function. The Center for the Ecology of Infectious Diseases is modeled on some of my experiences at NCEAS. We don’t have funding for a postdoc program, but we believe that that would be a great contribution to science if we were able to do that. We do have a working group model that enables us to draw on the collective expertise of all of the people at University of Georgia who work in the ecology of infectious diseases. I think one thing that comes out of NCEAS is thinking of science broadly, generously, and as a human-driven enterprise. That probably plays into my thinking about leadership and the management of scientific organizations. What I mean by that is, in technical disciplines, I think it’s very easy for people to be trained up and, to borrow a concept from genetics, canalized in their way of thinking. And one of the things that an organization like NCEAS does is it teaches you to question the conceptual framework that you’re using to frame a problem or to ask a question, or to acknowledge what kinds of data might be relevant to that question. And that kind of pluralism about science is certainly important to running a scientific organization. That’s a good question, Elizabeth, I’ll have to think about it more. How about you?

Elizabeth: I would say along the lines of my answer to an earlier question, I have discovered that I just truly love that co-creation of knowledge. I come into the room with an idea, but I hope I don’t leave the room with that same idea. What that requires at the base though is good listening and creating an environment of trust and flexibility, and an egalitarian approach to ideas generation. I feel all of those are threads that came from my experience at NCEAS. I do remember walking by at least one working group where everyone was sitting around and there was a single person at the front of the room talking, and I thought, oh my god, what a horrible model for creation of knowledge. In contrast, I participated in some groups that were fabulous – rapid moving, lots of opinions, and good scales of conflict. I feel a lot of that has translated over into how I think about administration – creating openness, creating trust, listening. I’m director of grad studies, and so I spend a lot of time asking the grad students what do you want, what do you need? Trying to create an environment of trust and identifying some common vision and trying to enact that. I think a lot of those are threads that began in my experience working in larger groups at NCEAS.

John: You know that resonates with me, too. The experience in working groups has caused me to be attentive to the social dynamics of collaboration. Being attentive to those social dynamics means that you can identify how different participants are interacting – which of those interactions are positive, and which ones might be damaging to the esprit de corps or the intellectual objectives of the working group. And trying out different tactics for facilitating the group to be able to achieve the best outcomes it can achieve. Sometimes that involves redirecting. It involves promoting and advocating. It involves leveling different playing fields that have become unlevel in the course of conversation. It involves enforcing social norms, whether that has to do with intellectual ownership or access to resources and data, or you know, all of the things that happen when people get together. It was certainly by participating in working groups at NCEAS that I started to tune into what those very human individual interactions are.