If you're a young academic, science researcher, or research trainee, you are sure to have heard about the importance of setting up successful collaborations. You may also have read a line or two in your university's strategic mission statement about the importance of interdisciplinary research. There's a well-established notion in much of the scientific community that big breakthroughs occur when people work together and when previously separate disciplines are fused. Articles reporting such breakthroughs are likely to find their way onto the covers of top science journals and can be the foundation of a successful career.

But the collaborative track, especially across different disciplines, is full of hurdles. Many collaborations--most in fact--die a quick death in bottom drawers. Sometimes they don't even make it back from the conference coffee bar. There is simply not enough time in a busy researcher's schedule to bother with pie-in-the-sky plans; so much emphasis is placed on getting results--by tenure committees, grant-review boards, and so on--that the pressure to take the safe option wherever possible is strong .

But sometimes you realize you must reach out to others in order to move your project forward. Or maybe you're especially excited by the apparent potential of a collaborative approach. If you've thought it through and the plan makes sense, it's probably a good idea. So how do you make it happen?

In the paragraphs that follow, we discuss some of the problems we have experienced first hand or heard about in our conversations with other researchers. We hope that by showing you where the holes are, we can help you steer around them and get your collaborative effort--whether it's with someone in your own lab, just down the hall, at a different department or university, or on the other side of the world--off to the best possible start.

Where do you start?

It's late in the evening at an interdepartmental meeting or a conference, and you're engrossed in a heated science discussion that turns into a "what if?" Before you know it, you're planning a follow-up meeting to get the ball rolling on a new joint project.

Most collaborations start with serendipity, a chance encounter with another scientist, a scientific lecture, or a broadly focused article. Give yourself the opportunity to be inspired by attending networking events and other people's seminars, and by reading widely. When you've decided there may be scope for further development, nail down a time to meet to discuss it further. Put it on your calendar now--at the least make contact within a few days--and work out a way to communicate (phone, meeting, videoconference). You've got an idea, now you need an action plan. Don't let it slip away.

One of the first and most serious problems when people decide to work together is overcoming the communication barrier, because disciplines and individuals have different vocabularies, work ethics, and standards. The clearer the mutual expectations are at the outset, the easier it will be to avoid friction later. But it's not just about formal agreements; it's equally important to take time out to build mutual rapport and respect though meetings and informal discussions. You want the group to start thinking like a team instead of as individuals. You'll probably have to invest some time in learning another subject, whether it's by reading textbooks, attending lectures, or watching your collaborators do their work. One engineering fluid dynamicist, starting to work on modelling blood flow in the heart, watched heart surgery.

Typically, the most "interdisciplinary" team member(s)--those with a hand in more than one field--will help ensure communications between subject specialists. Certain team members--typically the more "social" scientists on the team, sometimes referred to as "glue" of a group--help bind collaborations together. Their positive, energetic, and interpersonal approach--and their enthusiasm--helps to keep the collaboration vibrant. The ability to see the other person's viewpoint and communicate it clearly is a vital skill.

Credibility is a major hurdle to interdisciplinary research. Often, in trying to convey the core ideas of a subject--and its value--to an outsider, a discipline or subject area can be simplified almost beyond recognition. This can cause disciplinary peers and others who dwell on narrow details to find the characterization (i.e., in a grant proposal) unacceptably simplistic. But this is characteristic of the kind of fundamental thinking that comes when an area of science is viewed in alien light. There is a necessary--and valuable--widening of perspective. But this will hinder support from funding bodies, journal reviewers--and departmental colleagues and chairs who write letters for tenure and promotion decisions.

Break through this resistance by emphasising the novelty and the potential impact of your new, interdisciplinary line of thinking. Have the subject specialists articulate how the new work will sharpen their own cutting edge.

It is possible to be a successful researcher in a cocoon of individualism, but collaborations require good management and team players, and these can be absent in academic settings. There's no secret about this stuff; promoting and training team workers is a huge priority in the research departments of most industries. It means regular, open communications, mutual commitment, and emphasis on the communal goal. Things like credit, authorship on papers, and the share of the work should be agreed on informally by everyone at the start of the project, and revisited periodically along with tentative timelines and benchmarks for progress.

It should be no surprise that money plays a major role in the success of a project, but often time is just as important. The need to shepherd these precious resources means that collaborative projects require better-than-usual project-management skills. Are the financial resources in place? Can the people making key contributions be counted on? Is the work plan realistic and proceeding on schedule? These are more difficult questions when the team is scattered and each sub-unit has its own priorities.

In the excitement of things, it is easy to forget that Joe--a fourth-year Ph.D. student who is expected to do all the sample preparation--is planning to finish his degree in less than a year. It is really disappointing and unfair to the other participants when the project has to be put on hold because the hand-power--and expertise--just isn't there to get the work done. All the time spent up to that point can quickly be wasted when the competition moves ahead as new workers are trained, or as team members and priorities change. In a collaborative project, people need to trust each other and candidly discuss strategies and contingencies.

Cutting-edge science tends to be very specialized. It is important that your collaborative project and its results do not end up so far ahead of their time--or out of a subject's usual comfort zone--that journals have a difficult time finding reviewers. A good mental exercise is to think about people in related fields who could be asked to comment on the work, or who would find the work exciting. At some stage, it might be worth inviting those scientists to discuss the work and get some informal feedback.

It is equally important that some of the team members bring high levels of credibility from their own disciplines; an upstart's wild idea may be dismissed, but an established scientist's radical proposals will at least be taken seriously. Get the support of some big guns, even if they don't do much of the real work. Also, promote your ideas and results through departmental lectures and by visiting other universities. Naturally, you'll want to present your results at scientific meetings and discuss them with other experts afterwards. Interdisciplinary research is often very interesting and can be exciting to a wider audience

A few final thoughts

Collaborations are fun! They help create supercharged groups where impossible things seem doable, and they offer a larger safety net of ideas when things get stuck. Interdisciplinary projects break boundaries and elucidate fundamental problems by combining skills and disciplines. As such, they let us experience a deeper and more passionate drive towards scientific discovery, creating new experts and even new branches of science along the way.

Magdalena Bak-Maier spent 10 years as a researcher working in neuroscience, a multidisciplinary research area, and is presently working as a training and development manager for research staff at the University of Bristol (U.K.). Her insights on interdisciplinary collaborations are based on training and development events she has organized for researchers and postgraduate students both in the United Kingdom and the United States, as well as on her personal experience in scientific collaborations. The idea to write this article came from attending a session led by Simon Inger on creating and managing effective multidisciplinary collaborations.

Simon Inger spent 13 years as a researcher in geology, a subject that has long occupied the intersection of chemistry, physics, biology, engineering, and other disciplines. Many of the ideas in this article crystallised when, in his current role as a training and development specialist, he planned a conference on how to do interdisciplinary research. The conference drew on the expertise of research leaders who had crossed boundaries and joined up disciplines, funding bodies that are trying to find ways to support innovative collaborations, and the day-to-day experience of researchers as they try to build and maintain relationships, get funds, and get published.

Comments, suggestions? Please send your feedback to our editor.

DOI: 10.1126/science.caredit.a0700006

10.1126/science.caredit.a0700006