An issue of data ownership
March 23, 2016
An interesting retraction of an Editorial expression of concern hit the Twitts:
https://twitter.com/schneiderleonid/status/712278127519645696
The Editors and publisher have withdrawn an Expression of Concern previously contributed by noted neuroscientist David Amaral, with his agreement.
The original version of this Comment ‘Expression of Concern’ published by D. Amaral has been withdrawn by the Publisher in relation to the paper: ‘Organization of connections of the basal and accessory basal nuclei in the monkey amygdala’ by Eva Bonda, published in Volume 12, pp. 1971-1992 (doi: 10.1046/j.1460-9568.2000.00082.x). The review carried out at the University of California at Davis in December 2001 (brought to the publisher’s attention in February 2016) concluded that the allegation against Eva Bonda described in the commentary ‘Expression of Concern’ by D. Amaral did not meet The Office of Research Integrity’s definition of research misconduct, and was not pursued further.
That November 2000 Expression of Concern read, in part:
It has recently come to my attention that Eva Bonda has published a paper in the European Journal of Neuroscience entitled, ‘Organization of connections of the basal and accessory basal nuclei in the monkey amygdala’ ( Bonda, 2000). The data described in this paper were produced by my students and me at the University of California, Davis. Support for carrying out the experiments that produced these data was provided by the National Institute of Mental Health, through grant MH 41479 for which I am the Principal Investigator.
..The publication of this single-authored paper was totally unauthorized. Eva Bonda was a postdoctoral fellow in our laboratory.
Ok, so PI asserts ownership of data collected in his lab. Fine, fine… Typical story of postdoc who thinks that she owns and controls her data? And the PI was blocking publication for reasons unknown. We all have been down the various roads of he said/she said often enough to imagine a variety of scenarios where we might alternately side with the trainee or the postdoc.
Intriguing!
She had access to the preparations that were described in the paper. However, she did not carry out any of the experimental procedures involved in making the tracer injections reported in this paper. These injections were made by other students in the laboratory and by me. Moreover, other than processing the tissue from a small minority of the reported cases, it was the technical staff of our laboratory rather than Eva Bonda that carried out the histological processing of the reported experiments.
Ah. Well that sounds bad. This suggests it is a little more like theft of credit from more people than just the PI. I happen to disagree with the not-infrequent pose of postdocs on the internet that they own and control “their” data that they generated in the laboratory of a given PI. But that is much more of an arguable position than is taking data generated by many people other than one’s self and asserting control/ownership from a position that is not the PI.
Amaral finishes by making the charge of academic misconduct against Bonda very explicit:
In my view, the appropriation and publication of these data is a serious breach of scientific ethics. I have asked the Editor of the European Journal of Neuroscience to take appropriate action including publication of this Expression of Concern. Upon consultation with the Office of Research Integrity, Public Health Service, US Department of Health and Human Services, the agency responsible for protecting the integrity of NIH funded research programs, the UC Davis campus has agreed to initiate a review of the allegations of research misconduct. Based on the outcome of this review, further actions, including request for full retraction, may be taken concerning this.
Of course, the recent retraction of the Expression of Concern indicates that Bonda, the postdoc, was exonerated of misconduct charges in 2001!
Wow. Why did it take Amaral 15 years to retract his accusations? This seems spectacularly dickish to me.
And given the fact that the postdoc was not found guilty of misconduct by the University, it really questions the factual basis of his assertions in the original Expression of Concern. If I were the postdoc in question, I might have launched a counter accusation of professional misconduct. Depending, of course, on the details of the inquiry and what each party did and did not do. The exoneration of the postdoc may simply have been a lack of proof of intent, rather than any disagreement over the facts.
I notice, however, an interesting poll put up by an individual who both was RTing the tweets that alerted me to this situation and apparently co-published with Amaral.
https://twitter.com/mrhunsaker/status/712639543531319296
Gee, I wonder what the nature of the dispute was between Amaral and Bonda?
The subject of this poll is the juxtaposition of “good data” with “high quality standards” of the PI. Given what Amaral does, I’m going out on a limb and assuming we are talking about how pretty the immunohistochemical images are or are not (the Bonda paper is nearly all immuno-staining pictures).
DrugMonkey 
March 23, 2016 at 3:05 pm
I say this to people all the time: Authorship is both credit AND responsibility. If you can’t take responsibility for *something* in a paper, then you don’t get authorship. And if the authors aren’t willing to take responsibility for *everything* in the paper, then the paper shouldn’t get published.
This was a massive fuck up by everyone involved, in my opinion. It’s symptomatic of how scientific publications have turned from essential pieces of communication into professional prizes to be wrestled over. And that has happened because (I know I am a broken record on this, but it’s true) we’re all too lazy to actually read and think about each other’s work.)
And I don’t see this changing any time soon. There are too many researchers, and people will be lazy forever. The next big advance will be an AI search engine and text reader that finds papers and summarizes key points, with cross correlation so you can easily identify consensus opinion, knowledge gaps, and opposing bits of evidence. NIH will help develop this to reduce reviewer and administrative workload. Your proposal will be triaged unless it addresses an AI-identified knowledge gap and your approach is consistent with AI-determined consensus opinion and acknowledges the opposing evidence. Welcome to the future.
LikeLike
March 23, 2016 at 3:29 pm
And if the authors aren’t willing to take responsibility for *everything* in the paper, then the paper shouldn’t get published.
Don’t you think there are situations where this puts too much power in the hands of the PI to exploit the postdoc?
This was a massive fuck up by everyone involved, in my opinion. It’s symptomatic of how scientific publications have turned from essential pieces of communication into professional prizes to be wrestled over.
But, as you note, this is the reality. Peer review was surmounted in this case and the University failed to sustain any ethics charge against the postdoc. The accusation from the PI appears to be over credit/ownership rather than over faked or incorrect data.
Sorry, but this smells like a PI holding up a legit paper. Reasons? who knows. Not sure the specific reasons even matter. This incident speaks to the imbalance of power and the need for trainees to be productive on a reasonable timeline.
LikeLike
March 23, 2016 at 3:31 pm
It wasn’t long ago when a graduate student or post-doc could publish a single-author paper without advisor being a co-author. It was common enough in Physics and Chemistry that I personally know few cases (and I am not that old). Sometime advisor was not a co-author because he wouldn’t want to put his name if he did not agree with the main findings of the paper or sometime out of gentlemanship because he thought that he didn’t contribute intellectually enough to get an authorship. A simple acknowledgement was enough. Similarly acknowledgements were given to everyone else who helped in the study but didn’t contribute intellectually. Whatever happened to that time.
LikeLike
March 23, 2016 at 4:43 pm
15 years. Wow.
LikeLike
March 23, 2016 at 5:19 pm
DM: Oh, yes, I absolutely 100% agree that this totally smells like a dickish PI not letting a paper get out the door. The PI is clearly a shitty mentor, based on the fact that he totally failed to train the postdoc with regard to publishing ethics and procedures, and totally failed to properly weight the postdoc’s need for a publication with his need for a high profile pub. I don’t know why we all agree that research misconduct should ruin your career and bar you from future funding, but training misconduct is OK. As we’ve seen, you basically have to rape a trainee to get people to take action. And even that is spotty.
But all that said, the postdoc totally should not have published that paper either.
Shitty PI doing shitty training such that a postdoc did a shitty thing.
Interestingly, I didn’t see any accusation that the data or conclusions were incorrectly or misleadingly represented by the postdoc. The PI’s only problem seemed to have to do with permission and authorship. Again, this is a sign that speedy and accurate scientific communication has taken a back seat to wrestling over professional prizes. Science (the profession) has corrupted science (the process).
LikeLike
March 23, 2016 at 6:02 pm
@RS: that was before we started counting papers. Back then, it was all a very old-boys network based on reputation. Everyone in the field knew who the student’s advisor was and gave that advisor props for raising such a great student. So the advisor could be famous even if he wasn’t on those specific papers.
Also, my understanding is that in the fields where (and when) this was practiced, grant-getting was a smaller part of the operation. My understanding is that a lot of such practices came from fields where faculty were salaried from universities and didn’t need to fight for grants (which meant more stability and less grubbing for grain).
In the thunderdome, every paper matters.
LikeLike
March 23, 2016 at 6:12 pm
So TOD, what would have been the non-shitty thing for the postdoc to do that didn’t simultaneously hinder their own career?
It’s easy to acknowledge that the PI was being dickish, but that doesn’t make the PI actually publish data that was apparently publication worthy. So what recourse does the postdoc have in this scenario? Everything seems to ultimately fall into one of two categories: don’t publish (this would include everything from unsuccessfully lobbying to publish the paper to abandoning the lab as well as the time and effort put into producing the data that will never be published) or publish on their own (which is apparently shitty).
Doesn’t the dickishness of the PI sitting on publication quality data (that is apparently below their own threshold for publication) that was funded by public money and will never see the light of day outweigh the shittiness of a postdoc publishing on their own? In fact isn’t a postdoc publishing on their own pushing science towards speedy and accurate scientific communication? That doesn’t seem very shitty to me, particularly given the context.
LikeLike
March 23, 2016 at 6:18 pm
Whatever grievances the student had – and I’m sure there were some – you cant go around throwing papers in without your PIs approval. That’s not good cricket.
LikeLike
March 23, 2016 at 6:47 pm
Green Fluorescent Postdoc: You’re blaming the PI for (maybe) withholding a publication that the postdoc needed? What about the postdoc that denied authorship to other people (including the PI) that share credit/responsibility for the work?
The only victims of this train wreck, as far as I can tell, are 1) the lab mates who might have also worked on the project and 2) the committees that had to sort through all this bad behavior.
LikeLike
March 23, 2016 at 11:09 pm
These published data seem rather unimportant.
LikeLike
March 24, 2016 at 2:09 am
“The PI is clearly a shitty mentor, based on the fact that he totally failed to train the postdoc with regard to publishing ethics and procedures”
TOD, this is stupid.
LikeLike
March 24, 2016 at 10:58 am
So, it lo0ks like most people agree that it’s really a dick move to sit on one’s trainees’ papers. But those of you saying that publishing solo was also not OK – what should this postdoc have done? What if their support system (committees, other faculty, other department heads, etc.) have been equally ineffective in helping them get the work out? Should he/she note everyone’s effort in the acknowledgements but not make them authors? Or just give up?
Despite the shitty funding climate, this kind of shenanigans still happen. The RO1s still get funded because of the reputation of the PI for good work (and the preliminary data is multiple draft manuscripts), the thesis committees get tired of holding students up because their boss is being a jerk, and a steady stream of foreign postdocs hoping to get faculty jobs in their home country (where $$$ for this topic is more plentiful) keep rolling through. Permadocs and technicians stick around because they have unusually stable positions given the current state of funding. Trainees with drafts of papers get comments like “this is really great, really nicely done. Let’s submit to that really great journal ASAP, as soon as I read it”, and it goes on like this for years.
So again – what should this postdoc have done? Personally, I’ve moved on and am working my ass off in postdoc-land to make up for that blank space on my CV (despite having no intentions/delusions of getting a faculty job), but I suspect that as soon as I find a job outside of the academic world, I’m throwing in the towel on my PhD work. It’s SO hard to do that because multiple outside readers have also convinced me that the work is good and a valuable contribution to the field, but unfortunately in the absence of papers that reference is still helpful and even pre-pub avenues will be considered treason by the PI. It’s easier to settle than to fight, but I feel that I personally am doing the field a huge disservice by not finding a way to communicate these results.
LikeLike
March 24, 2016 at 11:19 am
This is the rub. I see no structural solutions that do not depend, ultimately, on the PI to do the right thing.
LikeLike
March 24, 2016 at 1:12 pm
@rs Publishing without the PIs name, or advisor’s name, still occurs and is common in some areas of math and computer science. When you get crossover people at the postdoc level, it can lead to conflicts.
If you get a crossover person at the PI level, you get some surprising results. Check on Erik Winfree at Caltech. His lab’s website lists publications of his lab members. He is a co-author on some, but not all, of those publications.
The BSD’s need to be careful about swinging the ethics ax, because students/postdocs could turn it around easily and claim that the PI does not deserve authorship. Most statements about authorship include a line saying something like “solely providing funding is not sufficient”.
LikeLike
March 24, 2016 at 1:25 pm
The solution is for all of us to quit counting publications and comparing impact factors, and actually get to know and think about the work of colleagues. Then there will be less motivation for PIs to sit on results, for postdocs to fret about a lack of publications, and for people to get all bent out of shape about authorship.
As for what this particular postdoc could have done… She could have learned earlier that the PI is a dick and moved to a lab that was better suited for launching her career. I’m sorry but bad PIs will exist as long as trainees continue to make and then settle for poor choices.
And I also like the fact that this normally very PC group is happily describing the PI as a ‘dick’ and his actions ‘dickish’, without noting that this is fairly sexist. God forbid we’d refer to the postdoc as a ‘c*nt’.
LikeLike
March 24, 2016 at 1:30 pm
Yes, TOD, these epithets have exactly the same valence for USian audiences. #massiveeyeroll Stop being a dick.
LikeLike
March 24, 2016 at 1:31 pm
Most statements about authorship include a line saying something like “solely providing funding is not sufficient”.
Those “statements” have not been agreed to by most biomedical scientists, are being promulgated by asshole humanities majors who have no call to stick their nose so ignorantly in the business of science and are rightfully ignored for the bullshit that they are.
and in case anyone thinks my position on this is kneejerkedly defending PIs such as myself for their exploitative ways, these self-same statements also say that merely collecting the data is not justification for authorship (screwing over the techs) either whereas editing a few words of text in the manuscript (BSD buddy of the PI) may be sufficient. I also reject this.
LikeLike
March 25, 2016 at 10:17 am
“As for what this particular postdoc could have done… She could have learned earlier that the PI is a dick and moved to a lab that was better suited for launching her career. I’m sorry but bad PIs will exist as long as trainees continue to make and then settle for poor choices.”
It’s the postdoc’s fault!!11!!1!!
LikeLike
May 16, 2016 at 3:17 pm
Your view does not seem to reflect the actual facts about the handling of the case, which appears to have been mishandled under the policies.
Your view stated: “. . . the recent retraction of the 2000 Expression of Concern indicates that . . . the postdoc, was exonerated of misconduct charges in 2001! Wow. Why did it take . . . [the PI] 15 years to retract his accusations? . . . . Given the fact that the postdoc was not found guilty of misconduct by the University, it really questions the factual basis of his assertions in the original Expression of Concern.”
– – – –
The online information you cited indicates that the PI alleged an ethical violation of misappropriation by the postdoc for the 2000 paper of data that were created by others in the laboratory in which the postdoc had no significant role [“. . . had access to the preparations . . . did not carry out any of the experimental procedures involved in making the tracer injections reported . . .These injections were made by other students in the laboratory and by me. Moreover, other than processing the tissue from a small minority of the reported cases, it was the technical staff . . . that carried out the histological processing of the reported experiments. The 2000 Expression of Concern by the PI stated: “I have asked the Editor of the European Journal of Neuroscience to take appropriate action including publication of this Expression of Concern. Upon consultation with the Office of Research Integrity, Public Health Service, US Department of Health and Human Services, the agency responsible for protecting the integrity of NIH funded research programs, the UC Davis campus has agreed to initiate a review of the allegations of research misconduct. Based on the outcome of this review, further actions, including request for full retraction, may be taken concerning this.”
There apparently was only a “review” (NOT a misconduct inquiry or investigation) by the University, with no further action described. The Withdrawal in 2016 done by the Publisher [not by the PI] of that EOC states: “The original version of this Comment . . . has been withdrawn by the Publisher . . . . The review carried out at the University of California at Davis in December 2001 (brought to the publisher’s attention in February 2016) concluded that the allegation . . . did not meet The Office of Research Integrity’s definition of research misconduct, and was not pursued further.”
– – – –
Of course ORI said it did not meet the ORI definition of misconduct — ORI policy since 1992 has excluded authorship disputes between collaborators or former collaborators from ORI’s definition of plagiarism — see at http://ori.hhs.gov/ori-policy-plagiarism.
– – –
However, this declination of ORI to consider the allegation of misappropriation under ORI’s policy is NOT at all relevant to whether the university or the publisher would consider that the postdoc did not qualify as an author or as sole author of the 2000 paper containing data on research that was conducted by others in the PI’s laboratory. The 2016 Withdrawal of the 2000 EOC by the publisher fails to explain why this ethical issue was not addressed, as it appears, under the journal’s own Author Guidelines:
“EJN follows the Code of Conduct of the Committee on Publication Ethics (COPE) and handles cases of research and publication misconduct accordingly (http://publicationethics.org/about). As per recommendation of ICMJE http://www.icmje.org/recommendations/browse/roles-and-responsibilities/defining-the-role-of-authors-and-contributors.html), authorship at EJN is based on the following four criteria: “i) Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; ii) Drafting the work or revising it critically for important intellectual content; iii) Final approval of the version to be published and iv) Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved […] All those designated as authors should meet all 4 criteria. And all who meeting all 4 criteria should be identified as authors. Those who do not meet all four criteria should be acknowledged […]
http://onlinelibrary.wiley.com/journal/10.1111/(ISSN)1460-9568/homepage/ForAuthors.html
LikeLike