A Peek into the Future of our Science: We are all in this Together
by Pam Davis-Kean, Luke Hyde, John Schulenberg
In attending various SRCD sessions, we were struck by the number of advertised and especially unadvertised interdisciplinary sessions that included scholars examining phenomena and mechanisms at multiple levels. In some ways, the unadvertised interdisciplinary sessions were the most compelling, those that were problem focused and included without fanfare, for example, neuroscientists and epidemiologists together. This suggests that the sometimes loud and sometimes quiet interdisciplinary evolution that our science has been experiencing for years is gaining some real traction.
The promise and challenge of interdisciplinary research for the future of our science is that we need to work outside our normal way of thinking – or better yet, that we need to find a “new normal” way of thinking. Imagine if it was normal to actively seek how other disciplines approach the understanding of adolescence, to start conversations with those whose research seeks to understand adolescence at different levels of analysis and/or with different methods than our own research. It is in these conversations that we can deal with our own disciplinary strengths, weaknesses, and biases and explore new thoughts and ideas of how the multiple proximal and distal mechanisms work together to shape development and behavior across adolescence. No one discipline can write the full story of how youth develop across time. By adding the richness of ideas, methods, and findings from other disciplines, our science benefits and the community benefits when our interventions can be more targeted and successful by being problem-focused rather than disciplinary-focused.
There are many signs that the future of our science will involve integrating mechanisms from multiple levels. Many of us are moving our own research programs in this direction, and more of our new scholars are being trained from the beginning in multiple disciplines.
- Research consortiums abound as we learn about the drawbacks of individual small-scale, convenience-based studies limited to a single level of explanation (Falk et al., 2013; Simmons, Nelson & Simonsohn, 2011).
- James Coyne has a recent fascinating blog on this general topic, about how single studies of convenience samples that consider only one level of explanation can lead us astray and hurt our scientific credibility (http://blogs.plos.org/mindthebrain/2015/01/21/nimh-biomarker-porn-depression-daughters-telomeres-part-1/).
- Our science is moving toward seeking how, for example, biological (e.g., neural, genetic) mechanisms work together with social mechanisms to shape development. As we begin to understand more nuanced interactions and effects across levels of analysis, such efforts require larger studies with multiple perspectives to find these effects and to link them across domains of study.
- We are learning that in order to have an impact on social policy, our research needs to consider multiple perspectives and disciplines with converging evidence. Moreover, we are finding that the public and social policy makers are often more convinced when research shows the biological embedding of experience and behavior. Thus, to significantly impact social policy in today’s real world, interdisciplinary approaches may be the most convincing.
- More and more funders are impressed with and are actively seeking truly interdisciplinary collaborations. For example:
- NIH has been giving more attention to innovation in its funding priorities, and many of us have learned that being actively interdisciplinary is a smart way to be innovative in our grant applications. With collaboration across disciplines, we can find new approaches that innovate beyond only one discipline, which raises the overall “innovation” and impact on science.
- NIMH emphasis on the Research Domain Criteria (RDoC) approach has changed the way in which NIMH science is funded and the way in which many studying psychopathology are thinking about understanding behavior. This approach pushes us to understand mechanisms across multiple levels of analysis from cells to symptoms to behaviors. Ultimately, it pushes for collaborations with experts across these levels of analysis for competitive grant applications.
- The new request for applications from the Collaborative Research on Addiction at NIH (CRAN) for the Adolescent Brain and Cognitive Development (ABCD) study, which will be a large scale longitudinal study of 10,000 youth to examine the multi-level mechanisms and consequences (from neural to contextual) of teen substance use, requires neuroimagers and epidemiologists, among others, to work together.
- NSF for years has been advocating interdisciplinary approaches, as illustrated by Interdisciplinary Behavioral and Social Science Research (IBSS), and Research Coordination Networks (RCN).
- The Institute of Medicine/National Research Council, whose consensus studies typically pave the way for increased funding priorities for government and private funders, recently put forth a consensus study on health and well-being during the transition to adulthood (http://iom.edu/Reports/2014/Investing-in-the-Health-and-Well-Being-of-Young-Adults.aspx). Among the major recommendations regarding needed research and social policies is more interdisciplinary research and collaboration on the topic.
- Professional research societies, including SRA and SRCD, are giving more serious attention to breaking down barriers/silos between disciplines. Indeed, in SRA’s mission statement is this: “We value diversity in scholarship, including the study of diverse populations; disciplinary perspectives; and methodological approaches.” This helps define the new normal that is interdisciplinary scholarship.
So there is a need for interdisciplinary research, and clearly many of us are already conducting such research. But need and opportunity are not enough. We also need to get great scientists who want to collaborate with others working at different levels of explanation. But this can be exceedingly difficult.
- Scholars from different disciplines may have trouble valuing the approaches of another discipline or may simply not realize the strengths of another discipline. For example, as we have written previously about (Falk et al., 2013), neuroscientists may think more about “recruitment” of people for their studies and less about “sampling”, that is: who do these research participants represent more broadly?
- Similarly, some social scientists may worry about reductionism in the brain or about the need to examine questions using neuroscience as a potential passing “fad” and may focus more on the weaknesses of current approaches, rather than the strengths of these approaches when applied to their discipline.
- Moreover, many scientists across disciplines have seen this type of collaboration as not feasible based on time or resources.
To get past these and other barriers, the following guideposts are offered to help encourage our interdisciplinary future.
- Working together. If we want to be a part of interdisciplinary research, regardless of what level of explanation we emphasize, we are all “social” scientists. That is, to work together scientifically, we have to be social and relate with one another.
- The last thing we want to hear when working with someone from a different discipline/level of explanation is that they are right and we are wrong. We can all appreciate the scientific ideal that we have to be open to be proven wrong and also that there are often multiple “right” answers to the same question. And at the same time, we can all appreciate that social intelligence, in terms of seeing things from another’s perspective, is essential to social integration and influence.
- We are better able to influence others from different disciplines if we can see things from their given perspective. By understanding their expertise, advances, concerns, and tools, we set the stage for better seeing their blind spots and helping them see the blind spots, and vice-versa. This open approach takes individuals willing to come to a table to leverage each other’s strengths to purpose a common scientific goal.
- Training. Training can take many forms.
- For more senior scholars, training could mean being open to new collaborators or it could mean gaining enough training in another discipline to appreciate the discipline and distinguish “good” science and collaborators versus those doing less stringent research. This type of training can be a big challenge for senior researchers – with whom to collaborate and how to know that the added methods are well respected in the other discipline.
- For more junior scholars, training may mean a post-doc, a longer period in graduate school with multiple mentors, or a later career development award. This type of training can make junior scholars more attractive to job searches and help give more options (i.e., multiple disciplines) when looking for a job/career.
- Communicating. Beyond gaining training, these types of collaborations often need one or more “bridge” people and those that can help to translate across disciplines and know enough to value core strengths of each discipline.
- This can be a colleague with cross-disciplinary training and/or one that has had enough experience in the other discipline to understand the different language.
- Ultimately, many potentially successful collaborations across disciplines likely fail because collaborators don’t speak each other’s language or are unable to articulate how their approach can be commensurable with the other approach.
- Structure. To really change to more interdisciplinary science, we need larger projects with more collaborating investigators.
- But that change can be risky early in a career (i.e., resulting in real or perceived tenure difficulties) or later in a career (e.g., not wanting to give up credit or resources). Thus, to nurture this direction, we will need a change in publication, promotion, and grant processes to give “credit” to larger teams (i.e., as done in Physics and other “big data” sciences).
- More open science (e.g., data sharing) can give many more researchers access to data to analyze in new and exciting ways. Again, though open access data is likely best for science broadly, this can undermine the work that original investigators have done, and thus we need to find ways to give credit for making data open and potentially losing some of the leverage that data may have given investigators were they private (Gorgolewski, Margulies, & Milham, 2013).
- Funding. The above guideposts are predicated on appropriate funding opportunities.
- Without sustainable science funding, it is difficult to really see disciplines as complimentary and not “us versus them”. When funding is very tight, it makes researchers more likely to become entrenched and over-advocate for their approach to the exclusion of other approaches.
- Lean budget times are not likely to be interdisciplinary-friendly. Yet even during such times, when funders encourage interdisciplinary collaboration, we will see increased interdisciplinary emphasis in our science.
There is no doubt in our minds that interdisciplinary research is the future of our science, and we would like to get there sooner rather than later.
- We need more shining examples of convincing interdisciplinary research regarding adolescence. When these collaborations work, they can be models for others.
- We need to see the clear advantages for our science and for social policy, the larger gains we can make by working together.
- As more early career scholars become trained from the beginning in interdisciplinary research, then it becomes the new normal and can help to change the overall ethos of our field.
Your turn: What other disciplines/levels of explanation do you include in your research? What have been the costs and benefits of your own interdisciplinary experiences?
Falk, E. B., Hyde, L. W., Mitchell, C., Faul, J., Gonzalez, R., Heitzeg, M., Keating, D., Langa, K., Martz, M., Maslowsky, J., Morrison, F.J., Noll, D.C., Patrick, M., Pfeffer, F.T., Reuter-Lorenz, P.A., Thomason, M., Davis-Kean, P., Monk, C.S., and Schulenberg, J. E. (2013). Neuroscience meets population science: What is a representative brain? Proceedings of the National Academy of Sciences (PNAS), 110 (44), 17615-17622.
Gorgolewski, K. J., Margulies, D. S., & Milham, M. P. (2013). Making data sharing count: a publication-based solution. Frontiers in Neuroscience, 7, 9.
Simmons, J.P., Nelson, L.D., & Simonsohn, U. (2011). False-positive psychology: Undisclosed flexibility in data collection and analysis allows presenting anything as significant. Psychological Science, 22, 1359-1366.