The challenge of leading interdisciplinary research projects

Consensus is not always possible.

17 August 2021

Kyodo News/Getty Images

Adelie penguins in a nesting area near Japan's Syowa Station research facility in Antarctica. Multidisciplinary teams enable researchers to study wild animal behaviour using AI-powered technologies.

Working at the interface of multiple disciplines on critical societal problems is a test for experienced researchers and novices alike. Nature Index spoke to three researchers about how to make it work.

Team building tools to cultivate trust

Katherine Daniell’s role at the Australian National University in Canberra straddles three schools and institutes: the 3A Institute in the School of Cybernetics, the Fenner School of Environment and Society and the Institute for Water Futures, where among her colleagues are engineers, ecologists, anthropologists, geophysicists, economists and statisticians.

Katherine Daniell

Daniell, who investigates collaborative approaches to decision-making and governance for sustainable development including the transformation of societies and environments driven by artificial intelligence (AI), says success in any inter-disciplinary project begins with researchers respecting the views of other team members which could be very divergent across multiple disciplines and sectors.

“Without trust and respect, projects can easily break down into unproductive conflict, leading to poor inter-disciplinary team morale and project outcomes," says Daniell. "Once people feel they are valued and respected, trust can be more easily built and sustained”.

Her experience of interdisciplinary working started as an undergraduate with a double Bachelor’s degree in civil engineering and arts at the University of Adelaide in South Australia. She encourages young researchers to start their inter-disciplinary research journey early.

“It won’t impact your career prospects, it will just set your career on a different exciting path,” she says.

She has experimented with different strategies to establish personal bonds and trust within teams. “This has included lots of play, questioning, listening, shared experiential work, team-based challenges, social activities and different formulations of introductions, from only first names to providing more personal and professional information,” she says.

Inter-personal skills such as leadership and team management may be overlooked in training for early career researchers that emphasises building research skills. But early exposure to these facets of research, as Daniell experienced, can help future leaders learn to draw insights from multiple disciplines.

“Researchers often need to work as convenors and facilitators to bring people together, and where relevant, to provide additional inspiration and knowledge that may not be available, not impose our own views and values on others,” says Daniell.

Broaden your horizons while searching for collaborators and mentors

Koutarou D. Kimura combines the disciplines of animal behaviour, neuroscience and data science in his role at Nagoya City University and Osaka University in Japan, using AI and machine learning to study animal behaviour and neural activities.

Koutarou D. Kimura

The professor, together with his student Shuhei Yamazaki, and collaborators working in AI, developed STEFTR software, which automatically detects behaviours characteristic to a particular animal group.

“The project unexpectedly popped out from conversations with a long-collaborating robotics professor, who was interested in the emerging AI techniques. I realized that the behaviour of animals could be analysed using AI,” says Kimura.

The method allows researchers to find behavioural features in multiple groups of animals in a very short time. This saves on the very time-consuming task of manually extracting behavioural features from big data generated by high-resolution cameras for lab animals or remote sensing technologies in the wild. Kimura says it will be “the game changer”.

For example, STEFTR can distinguish between normal and mutant animals. Testing this method on a variety of species, such as worms, insects, mice, penguins and seabirds, has identified previously unobserved forms of movement.

Due to his pure biological background, Kimura started looking for collaborators with the necessary skills in AI. His advice to kick-start an interdisciplinary project is to find partners with deep knowledge in each discipline the project covers and communicate with them frequently.

“[This includes] not just exchanging the emails when necessary, but visiting their lab, having dinner together, and spending time to discuss unnecessary things,” as a way of building trust, he says. Unfortunately, adds Kimura, these social activities are not possible now because of the pandemic, “but hopefully will be in the near future.”

Kimura emphasizes that it is important that not just the lead researcher but all members of the collaborating team and their supervisors value inter-disciplinary research.

“The advisors should [try to] understand at least the basic ideas of the other field and have good relationships with other supervisors,” he says.

Ensure equal representation and embrace diversity

An interdisciplinary champion in her field of marine socio-ecology, Rachel Kelly provides guidance on cross-disciplinary research relating to marine conservation, ocean infrastructures and ocean literacy, for stakeholders and early-career researchers particularly.

Rachel Kelly

Her recent paper in Socio-Ecological Practice Research highlights barriers to research that spans disciplines, and 10 best practices to overcome them.

Kelly, who works at the Centre for Marine Socioecology in Tasmania, Australia, and with the Future Ocean and Coastal Infrastructures Consortium in Canada, says that discipline-specific scientific terminologies and acronyms can be a major hurdle for interdisciplinary work.

“In our projects, we’ve found that it’s really important to carve out time at the beginning to address this in a safe space, where everyone can ask questions and not worry about being teased or chastised for not understanding,” she says.

Kelly says it is not always possible to find consensus across disciplines or people, but that is a fact researchers should embrace.

“The aims of marine research projects I have been a part of demanded an appreciation and inclusion of diverse and complex knowledge to co-create solutions to complex ocean challenges," she says. "Oversimplifying this knowledge would render the outputs less valuable.”

According to Kelly, good mediation and equal representation of the team members are key to successful interdisciplinary research projects.

“When leading teams, I have appreciated the importance and value of ensuring that everyone has the chance to speak and contribute – and also seen how dominating voices can derail discussions and team-building,” she says. “It’s eye-opening to realise that the ‘superstars’ of one field are virtually unheard of in another.”