CBI graphic: Publishing success and proximity, color-coded on Longwood Avenue
This week’s New Yorker features a story entitled “Groupthink: The science of the team effort.” The piece jumps from questions about benefits of brainstorming to the intellectual virtues of proximity.
It ends up at MIT’s legendary Building 20 — the 1942 on Vassar Street lab that stood near the Stata Center site. But first, the story stops briefly at another cluster of Boston area brain power – Longwood Avenue. Home to three top hospitals and Harvard Medical School, the neighborhood offers a quantifiable example of the benefits of “co-location.”
In late 2010, a team from the Harvard Med School Center for Biomedical Informatics published a paper in PLoS One on the correlation between the physical distance between collaborators and the scientific impact of their research. The researchers, including Kyungjoon Lee and Isaac Kohane, looked at articles published by Harvard biomedical investigators from 1993 to 2003.
Click here for an interactive map.
“Each collaboration was geocoded to the precise three-dimensional location of its authors. Physical distances between any two coauthors were calculated and associated with corresponding citations,” they wrote. The results “provide striking evidence for the role of physical proximity as a predictor of the impact of collaborations”
Or as Kohane put it in the New Yorker article: “If you want people to work together effectively, these findings reinforce the need to create architectures that support frequent, physical, spontaneous interaction.”
This previously unpublished 2010 interview offers a little background:
Q. This study looked at research up to 2003. Since then, advances in digital technology and the Internet have made it much easier to share images and data. Do you think your findinsg would stand if you looked at 2005-2010?
Dr. Kohane: On the one hand the web has been around since the early 90’s and email has been used by academics even earlier……I would wager a nice bottle of wine that the findings will hold. Serendipity at the water cooler is likely to continue to be an important driver of impactful and original science.
Dr. Lee: It would not make much sense at the moment (to look at recent data.) It takes several years until each articles gets to its maximum citation. Our data shows that it takes about 5 to 10 years to reach the maximum. Therefore, articles from 2005 to yesterday are too young to consider their potential in the future. They’re just not exposed long enough to be cited. It takes quit a long time to be cited in another article after the publication.
Q. Outside of the NIH campus, it might be hard to find so many researchers working as closely together as they do in Longwood. Do you think these findings will apply to other settings?
Dr. Kohane : If we were to look at other large biomedical nexi (San Francisco, Cambridge UK, Houston Texas) I would venture that we would find similar findings. However, you are quite right, where there is insufficient critical mass, locality will likely matter less.
Q. How far is your office from Dr. Lee’s?
Dr. Kohane: Now his is about 20 meters. Before the study it was about 100 meters and an extra floor.
Q. Is there a trade-off between traffic, parking and publishing success?
Dr. Kohane: If traffic and parking were less problematic then I am sure the frequency of meetings across campuses would be greater. But for all we know, pre-ordained meetings might actually reduce innovation and impactful publication. Perhaps your readers know better.