The Journal of Neuroscience has received notification of an investigation by the Perelman School of Medicine at the University of Pennsylvania, which supports the journal's findings of data misrepresentation in the article “Intraneuronal APP, Not Free Aβ Peptides in 3xTg-AD Mice: Implications for Tau Versus Aβ-Mediated Alzheimer Neurodegeneration” by Matthew J. Winton, Edward B. Lee, Eveline Sun, Margaret M. Wong, Susan Leight, Bin Zhang, John Q. Trojanowski, and Virginia M.-Y. Lee, which appeared on pages 7691–7699 of the May 25, 2011 issue. Because the results cannot be considered reliable, the editors of The Journal are retracting the paper.
From RetractionWatch we learn that the Journal has also issued a submission ban to three of the authors:
According to author John Trojanowski ... he and Lee have been barred from publishing in Journal for Neuroscience for several years. Senior author Edward Lee is out for a year.
This is the first time I have ever heard of a Journal issuing a ban on authors submitting papers to them. This is an interesting policy.
If this were a case of a conviction for academic fraud, the issues might be a little clearer. But as it turns out, it is a very muddy case indeed.
A quote from the last author:
In a nut shell, Dean Glen Gaulton asserted that the findings in the paper were correct despite mistakes in the figures. I suggested to J. Neuroscience that we publish a corrigendum to clarify these mistakes for the readership of J Neuroscience
The old "mistaken figures" excuse. Who, might we ask is at fault?
RetractionWatch quotes the second-senior author Trojanowski:
Last April, we got an email about an inquiry into figures that I would call erroneously used. An error was made by [first author] Matt Winton, who was leaving science and in transition between Penn and his new job. He was assembling the paper to submit it, there were several iterations of the paper. One set of figures was completely correct – I still don’t know what happened, but he got the files mixed up, and used erroneous figures
Winton has apparently landed a job as a market analyst*, providing advice to investors on therapeutics for Alzheimer's Disease. Maybe the comment from Trojanowski is true and he was in a rush to get the paper off his desk as he started the new job**. Maybe. Maybe there is all kinds of blame to go around and the other authors should have caught the problem.
Or maybe this was one of those deliberate frauds in which someone took shortcuts and represented immunohistochemical images or immunoblots as something they were not. The finding from the University's own investigation appears to confirm, however, that a legitimate mistake was made.
...so let us assume it was all an accident. Should the paper be retracted? or corrected?
I think there are two issues here that support the Journal's right to retract the paper.
We cannot ignore that publication of a finding first has tremendous currency in the world of academic publishing. So does the cachet of publishing in one Journal over another. If a set of authors are sloppy about their manuscript preparation, provide erroneous data figures and they are permitted to "correct" the figures, they gain essentially all the credit. Potentially taking credit for priority or a given Journal level away from another group that works more carefully.
Since we would like authors to take all the care they possibly can in submitting correct data in the first place, it makes some sense to take steps to discourage sloppiness. Retraction is certainly one such discouragement. A ban on future submissions does seem, on the face of it, a bit harsh for a single isolated error. I might not opt for that if it were my decision. But I can certainly see where another scientist might legitimately want to bring down the ban hammer and I would be open to argument that it is necessary.
The second issue I can think of is related. It has to do with whether the paper acceptance was unfairly won by the "mistake". This is tricky. I have seen many cases in which even to the relatively uninformed viewer, the replacement/correct figure looks a lot crappier/dirtier/equivocal than the original mistaken image. Whether right or wrong that so-called "pretty" data change the correctness of the interpretation and strength of the support, it is often interpreted this way. This raises the question of whether the paper would have gained acceptance with the real data instead of the supposedly mistaken data. We obviously can't rewind history, but this theoretical concern should be easy to appreciate. Maybe the Journal of Neuroscience review board went through all of the review materials for this paper and decided that the faked figure sealed the acceptance? For this concern it really makes no difference to the Journal whether the mistake was unintentional or not, there is a strong argument that the integrity of its process requires retraction whenever there is significant doubt the paper would have been accepted without the mistaken image(s).
Given these two issues, I see no reason that the Journal is obligated to "abide by the Penn committee’s investigation" as Trojanowski appears to think they should be. The Journal could accept that it was all just a mistake and still have good reason to retract the paper. But again, a ban on further submissions from the authors seems a bit harsh.
Now, I will point out one thing in this scenario that chaps my hide. It is a frequent excuse of the convicted data faker that they were right, so all is well. RetractionWatch further quotes the senior author, Lee:
...the findings of this paper are extremely important for the Alzheimer’s disease field because it provided convincing evidence pointing out that a previous report claiming accumulation of intracellular Abeta peptide in a mouse model (3XFAD) is wrong (Oddo et al., Neuron 2003), as evidenced by the fact that this paper has been cited by others for 62 times since publication. Subsequent to our 2011 J. Neuroscience paper, others also have found no evidence of intracellular Abeta in the 3XFAD mice (e.g. Lauritzen et al., J. Neurosci, 2012).
I disagree that whether the figures are correct and/or repeatable is an issue that affects the decision here. You either have the correct data or you do not. You either submitted the correct data for review with the manuscript or you did not. Whether you are able to obtain the right data later, whether other labs obtain the right data or whether you had the right data in a mislabeled file all along is absolutely immaterial to whether the paper should be retracted.
The system itself is what needs to be defended. Because if you don't protect the integrity of the peer review system - where authors are presumed to be honest - then it encourages more sloppiness and more outright fraud.
*An interesting alt-career folks. One of my old grad school peeps has been in this industry for years and appears to really love it.
**I will admit, my eyebrows go up when the person being thrown under the bus for a mistake or a data fraud is someone who is no longer in the academic science publishing game and has very little to lose compared with the other authors.