A group of neuroscientists in Germany and Hungary is calling for the retraction of two of their recent papers after discovering a fatal error in the research.
Myriam Sander, a memory researcher at the Max Planck Institute for Human Development in Berlin, took to social media on Wednesday to alert her followers to the decision. In what Sander called the “most difficult tweet ever,” she wrote:
Sander said the two articles are “Age-related declines in neural selectivity manifest differentially during encoding and recognition,” which appears in the April 2022 issue of the Neurobiology of Aging, and “Tracking Age Differences in Neural Distinctiveness across Representational Levels,” published in April 2021 in the Journal of Neuroscience.
She wrote:
The revelations were received positively on Twitter, where Sander’s followers praised the transparency with which her group handled the issue. Among the comments were this one from Eiko Fried, a mental health researcher at the Leiden University in The Netherlands, who wrote:
Darren Dahly, a statistician, wrote:
And psychologist Scott Furtwengler posted:
This is not the first time a researcher has taken to Twitter to announce a retraction of their work. Some other cases:
- ‘I’m starting the year off with something I didn’t expect to ever do: I’m retracting a paper.’
- “Study linking police violence and black infants’ health is retracted”
- Nobel winner retracts paper from Science
Update, 1400 UTC: Sander tells us:
I am still overwhelmed by all these positive reactions regarding the tweet. I was not at all expecting that.
We (my co-authors and I) simply thought that, since we usually announce new papers on twitter, we should also say when we retract one. Quiet self-retraction doesn’t make any sense, if you want to correct the scientific record and prevent that other researchers try to build on your wrong findings. So, yes, we have been very much gratified by the response to the tweet thread. I even got emails from colleagues offering help for the reanalysis.
One of the editors reacted immediately to my mail and put us in contact with the office to draft the formal retraction note, but I did not yet hear back from them.
I haven’t yet heard back from the other one.My predoc Claire Pauley, who is the first-author of one of the papers, discovered the error while working on further analyses for another paper. She got some results that seemed weird, and therefore checked again all code. She discovered that trial onset were not correctly defined. She then contacted our colleague Malte Kobelt who had originally drafted these scripts and worked with the data before (he is the first author of the other paper that we need to retract) to check whether this was really the case. Both then informed me and we discussed all further steps together – we did some more checks and first reanalyses and informed all co-authors as well as our Institute’s Director. After some preliminary analyses we are now pretty sure that this error affects all results, we decided to retract the papers. I then wrote the emails to the editors and the tweet. Following a suggestion of our Director, I also plan to directly inform colleagues from the aging neuroscience community who cited our work or from which we know that they are working on similar topics. We are also still working on removing the reference or adding a warning on all the internet sites where these papers are listed.
In sum, we realize that a retraction is not so easy – if you / retraction watch has further suggestions how to handle such a situation, we happily welcome them.
Let me end by saying that for some of us co-authors, this is of course a very difficult situation, but the consequences for my predoctoral student Claire are even more severe, since she needs the publications for her dissertation and is now running out of time and financial support… I am not yet sure how we can handle this, but I am very proud of her integrity and hope that we find a good solution. All the public support will certainly help…
Like Retraction Watch? You can make a one-time tax-deductible contribution by PayPal or by Square, or a monthly tax-deductible donation by Paypal to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].
Like this retracted neuroscience article, the article by Joscha Legewie on policing and infant health was retracted right after it was published in late 2019 and for similar reasons (data processing errors). But it was still cited in 2020 and 20201 by others. Are there data on the time lag to retraction and the citation inertia of a retracted article?
I appreciate the authors doing the right thing.
Is it just me, or does describing this as the “most difficult tweet ever” seem over the top? Given all the bad things happening around the world every day, owning up to a scientific mistake is pretty far down the list of bad things that could happen to someone or their loved ones.
Read as “the most difficult tweet [regarding our research] ever,” no, it is not over the top. Context counts.
Kudos to the authors for being so quick with the alert and so transparent about the whole thing. The preprocessing pipeline is one of the big black holes in science. That’s where a lot of mistakes can happen — and will never be spotted.
Points for Claire Pauley for doing the right thing and checking and re-checking code that produced weird results. She didn’t shrug her shoulders and look the other way; she went all the way to the bottom of this. That already qualifies her for the PhD she’ll eventually get: being a fearless, hard-nosed, data-drilling scientist. In other words, the best kind.
“This error AFFECTS ALL RESULTS, we decided to retract the papers”
Well done, because the results are not valid.