Treat your published papers as those of your competitors

Feb 18 2016 Published by under Careerism, Conduct of Science

No scientist should *ever* be afraid to publish a finding that contradicts their prior publications.

9 responses so far

  • whizbang says:

    Of course, this works best if you can explain why there's a contradiction...

  • baltogirl says:

    A grad student in a famous collaborator's lab (to use your terminology, a bsd) reported certain results of a mouse backcross he very likely did not ever do in our joint CNS paper (this was a minor finding in that particular paper; but still-! that's called fraud!). A few years later we actually did that cross in my lab and got quite different results - with deep implications for the biology of the system.
    While it was embarrassing, I had no problem contradicting the earlier paper in a later paper.
    More important than anything else is making sure the truth is out there.

  • drugmonkey says:

    Are you saying he didn't do it simply because your results were different?

  • baltogirl says:

    I suspect he didn't actually do it because it takes time to cross mice, and he added this sentence late in the game (ie after review- and review was quick). But I don't know for sure he didn't do it- he could have started it earlier. Frankly his advisor should have made sure that there was a written record of this cross. BTW JAX also could not reproduce this result.

  • Laffer says:

    My PhD supervisor said it's totally fine for the lab to publish a result that contradicts a previous one from the lab if it's based on new information/assays/etc. Better us than someone else. Plus, two papers for the price of one!

  • blatnoi says:

    I suspect what he really means, and can't get across in a short tweet, is that to get a paper in CNS people often rush data and do sloppy work. Then when they come back to it and look at it in more detail, they get results that contradict the work and instead of retracting earlier findings, which were done in a hurried and bad way, they say that this is the 'self-correcting nature of science' and get yet another publication in contradicting those earlier glamour mag findings.

    As for me, that kind of attitude, caused me to lose a year and a half of my postdoctoral work because my results contradicted that mechanism that the lab was famous for. I would have only lost one year, but the boss couldn't bring himself to tell me outright that the stuff was not publishable for another half a year and kept saying "this is not the right time", or "we have to wait a bit and maybe get a patent" and I was too stupid to understand the hints and kept working on it because to me it seemed like at least two very good publications there, until he outright said it. The publication gap probably cost me a few interviews when looking for positions.

    Still, can't complain since the boss gave me a good letter and I did get publications in other stuff, and eventually a job. But still a bit bitter...

  • blatnoi says:

    And of course, I would never refuse to publish a result that contradicts my own earlier findings in my own lab. Especially double down on that due to my experience.

  • Dusanbe says:

    Blatnoi's take is right- this tweet refers to the PIs who use CNS as a pre-print server to communicate first, ask questions later.

  • Grumble says:

    What about the club of people whose podunk society-level-or-below (would that be 7th tier?) journal article contradicts the CNS paper of a famous soon-to-be Nobel laureate? That would describe one of my first ever papers.

Leave a Reply