Journal of Neuroscience retraction, typically opaque, from author with history of errors

jneuroscienceThe Journal of Neuroscience has retracted a 2011 paper by an international group of scientists, including the prominent Maryland researcher Ronald Dubner, but readers won’t know why.

As the notice “explains“:

At the request of the authors, the following manuscript has been retracted: “Spinal 5-HT3 Receptor Activation Induces Behavioral Hypersensitivity via a Neuronal-Glial-Neuronal Signaling Cascade” by Ming Gu, Kan Miyoshi, Ronald Dubner, Wei Guo, Shiping Zou, Ke Ren, Koichi Noguchi, and Feng Wei, which appeared on pages 12823–12836 of the September 7, 2011 issue.

The paper has been cited five times, according to Thomson Scientific’s Web of Knowledge.

Absent more information — and neither Dubner nor the journal’s editor, John Maunsell, have replied to our requests for comment — we’re left to speculate. But using the data points we now have from previous notices, it looks like Koichi Noguchi may be the common thread here.

Noguchi was a co-author on a 2008 paper in the Journal of Neurochemistry that was first corrected — in a mega kind of way — then retracted. As the retraction notice indicated:

The quantification graphs of RT-PCR in Figs.1, 6, 10 cannot be reproduced from the existing data, and the first 2 bands in Fig.10 (Iba1 and IL-1beta) were mistakenly created from an identical band.

Noguchi also was a co-author of a 2006 paper in the Journal of Neuroscience that in 2009 picked up this  erratum notice:

In the article “Suppression of the p75 Neurotrophin Receptor in Uninjured Sensory Neurons Reduces Neuropathic Pain after Nerve Injury” by Koichi Obata, Hirokazu Katsura, Jun Sakurai, Kimiko Kobayashi, Hiroki Yamanaka, Yi Dai, Tetsuo Fukuoka, and Koichi Noguchi, which appeared on pages 11974–11986 of the November 15, 2006 issue, there was an error in the legends for Figs. 1E, 2D, and 4, C and D. These legends described “n=4.” However, the quantification of RT-PCR and Western blots was carried out from three, not four, samples in each experiment. All statistically significant values in all figures were obtained from 3 samples. Therefore, all data were scientifically correct, only the number of samples was incorrect.

And the Journal of Clinical Investigation has issued corrections for at least two articles on which Noguchi’s name appears, one from 2007 and one from 2005. In both cases, problematic figures were to blame.

In other words, the Journal of Neuroscience stands as a monument to editorial muteness in what appears to be a case of pretty widespread sloppy, at best — and therefore unreliable  — science. Its silence in the matter is anything but Solomonic, although it is codified in the journal’s retraction policy, which states:

The Journal will retract an article at the authors’ request at any time without requiring explanation. At the authors’ option, the retraction notice may simply state that the article has been retracted at the authors’ request. Alternatively, the authors may provide a brief explanation of the error(s) prompting the retraction. However, statements of retraction may not assign blame to specific authors or laboratories.

Readers of this blog might recall that we have disagreed on this score with Maunsell before. As he told us last year:

Retraction Watch, may be doing science a disservice by failing to emphasize that retractions fall into two very different categories: those initiated by authors to remove articles containing errors, and those initiated based on an investigation that finds fabrication, falsification, plagiarism or other unethical behavior.  In searching your web site I could find no material that lays out this distinction for visitors.  I worry that scientists visiting your site will be led to believe that all retractions arise from misconduct, which is far from the truth.  By failing to make clear that retraction is an important mechanism for correcting honest mistakes, your site may imply to scientists that they must avoid retractions at all costs.  You might greatly enhance and extend your site by providing your readers with general information about ethical behavior in scientific publication (there are many suitable links) and especially by emphasizing that there are different types of retractions.

We think Maunsell is being disingenuous here, and the journal is, too, with the following caveat to its retraction policy:

In retraction statements, The Journal clearly distinguishes between author-initiated retractions and those initiated by the editors because of violations of the Society for Neuroscience’s guidelines for Responsible Conduct Regarding Scientific Communications.

The journal’s guidelines piously intone that

SfN believes that progress in understanding the nervous system benefits human welfare. Such progress depends on the honest pursuit of scientific research and the truthful representation of findings. While recognizing that both error and differences among individuals in the interpretation of data are natural parts of the creative process, the Society for Neuroscience affirms that the success of the entire scientific endeavor is jeopardized by misconduct, in the form of plagiarism, fabrication, or falsification of data. By entering the profession, neuroscientists assume an obligation to maintain the highest level of integrity in all scientific activities.

One of the specific points in the guidelines makes clear that tinkering with data is a no-no:

1.2.1. Intentional, knowing, or reckless fabrication or falsification is misconduct and will lead to action by the Society (see Procedures for Dealing with Allegations of Unethical Scientific Conduct). No data may be put in a scientific communication that have not actually been collected or observed (fabrication), nor may data be altered in any way (falsification) other than by mathematical transformations that are commonly accepted or clearly explained in the manuscript. This includes numerical data as well as images.

Which brings us to the point: Either the journal knows what went wrong with the retracted paper and chooses not to tell its readers, or it turned a blind eye to the problems and allowed the authors to skulk off without an honest accounting of the matter. In neither case should readers feel confident that their interests  — nor the interests of their society and their specialty — are being well-served.

Hat tip: @hysell

0 thoughts on “Journal of Neuroscience retraction, typically opaque, from author with history of errors”

  1. “Retraction Watch, may be doing science a disservice by failing to emphasize that retractions fall into two very different categories: those initiated by authors to remove articles containing errors, and those initiated based on an investigation that finds fabrication, falsification, plagiarism or other unethical behavior.”
    If only life were that simple… Most of retractions do not fall into either category. Typically, people retract fraudulent papers preemptively because the stench surrounding them becomes unbearable. Alternatively, editors are gracious enough, or find it beneficial for themselves, to allow people to retract bogus papers.
    “the first 2 bands in Fig.10 (Iba1 and IL-1beta) were mistakenly created from an identical band” – mistakenly?! How on Earth can this happen? Any mention of “creating” bands screams fraud.

  2. Unless I’m very much mistaken (viewing this on a cellphone browser), Fig 5c and 6d share a beta actin loading gel. Same deal for 2c and 6e. If this is findable on a tiny screen, there is likely more to be seen upon closer scrutiny.

  3. On the screen, we found a lot more dishonorable mistakes. Look at Fig 6D(IL-18R) and 7F(IL-1beta). Same blot was shared. There is likely more, if you’re looking for.

  4. This paper and its copy/paste Western blot figures is yet another example demonstrating why journals need to insist that authors present full blot and gel images in supplemental materials, with indications showing which portions of which blots/gels were used in each figure in the main paper.

    1. I have heard that Journal of Neuroscience does not allow supplemental material
      (If true, one more questionable practice from
      this laughable journal)

  5. Journals often use the “errata” mechanism to correct unintentional errors. This seems to be a good way to correct honest mistakes such as n=3 instead of n=4 where the overall scientific conclusion of the paper remains valid. An actual retraction of a paper seems extreme for correcting honest mistakes where the conclusion remains valid. It doesn’t seem to me that an author would request a retraction unless there’s more than meets the eye here – afterall, most of the scientists I know are ambitious, career-oriented creatures who would not be inclined to risk the potential career and reputation consequences of a full retraction. Are any of the retracted papers ever corrected and re-submitted (i.e. because they were honest but needed some re-work)? My doubts about that lead me to believe that the two types of retractions (honest vs. dishonest) is in large part mythical. If there is an ‘honest’ retraction, what would it hurt for the journal to give a reason citing the mistakes and then simply stating that the corrections invalidated the scientific conclusion?

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.