Corie blogged a few days ago about the brewing neuroscience kerfuffle at MIT. Briefly, a promising young professor may, or may not, have decided against MIT because she was going to be denied a collaboration with a key researcher. Now today the Globe leads with the headline “MIT vows to press its scientists to collaborate” – an interesting mix of words.

I am far from qualified to judge the situation. Though I’m across the street from most of these folks, the Computer Science & Artificial Intelligence Lab is a world away. I don’t know any of them.

But I am utterly fascinated by the interplay of policy and personality, competition and collaboration. It speaks to the issues we’re working on at Science Commons in a unique way: we got into this research thinking it was about the law and using the law to help people collaborate and innovate – like Creative Commons did for culture. But in science, it’s so much more complex than that. Prestige and competition appear to play a greater role in the collaboration, or non-collaboration, between laboratories, than do for example…patents…copyrights…contracts. It’s one of the reasons we’re also focused on the integration of technology, law, and policy to make scientists see the benefits of an open approach by giving them questions they wouldn’t otherwise be able to ask. Apropos of this discussion, we’re doing a lot of work in neuroscience on this front.

“Pressing scientists to collaborate”. Makes me think of collaboration as a constantly shifting piece of game theory. Is it worth collaborating with you? A few roleplaying scenarios:

1. No, if you can’t help me, or if you can hurt me
2. Yes, if you can help me, or are more famous than I am
3. Yes, if by not collaborating with you, you choose to work with a different lab with the same tools and get a paper out ahead of me

This doesn’t even count the two graduate students or postdocs who are surreptitiously trading antibodies and animal models, taking the T with an eppendorf tube and some shaved ice between MIT and the Farber. In that case, the collaboration is deniable unless the two involved get enough data to make a publicly acknowledged collaboration worthwhile.

So what is the impact of adding to this complex calculus, which scientists seem to learn through guild and rites in graduate school, an institutional policy of collaboration? What would that policy look like? How would you track its success and impact? I’m curious and eager to watch and see what happens here.