A team of physicists has retracted a paper from Science after they discovered mistakes in their data and statistical analysis when following up on their work.
The paper “A room-temperature single-photon source based on strongly interacting Rydberg atoms” published in 2018, garnered 117 citations, according to Clarivate’s Web of Science. Of these, 97 citations came after the authors corrected the paper in 2020 to adjust for updated calibrations, which they said did not affect the conclusions of the article.
This correction was unrelated to the reasons for retraction, said corresponding author Tilman Pfau, a professor of physics at the University of Stuttgart in Germany.
The now-retracted paper described a way to generate single photons on demand at room temperature rather than in extremely cold conditions. The approach could have “various applications in quantum information processing and communication,” the authors wrote.
According to the retraction notice, after a recent internal reevaluation of the data, the authors “detected mistakes in data filtering and statistical calculations in one of the study’s figures.” The authors reevaluation of the data does not support the study’s conclusions, hence the retraction, the notice stated.
The authors found the errors in their previous data analysis while working on an update to the experiment, but with more advanced equipment, Pfau said.
“While analyzing this new data we realized that we did subtle mistakes in the analysis and the filtering of the ‘old’ data,” Pfau told Retraction Watch. He added that the experiment’s low coincidence rate, an important factor in quantum analysis, was “not enough to draw conclusions with sufficient confidence in retrospect.” All authors agree with the retraction, he said.
A spokesperson for Science said it’s encouraging “to see examples like this where authors uncover an error and work quickly to correct the record.” Holden Thorp, the editor in chief of the Science family of journals, has previously expressed his desire to change the retraction process, such as implementing a two-step process to speed up retractions.
Pfau said his team handled the case as transparently as possible: “In order to avoid such mistakes in the future we will intensify our training in the statistical analysis of data and will also publish alongside with future papers, the original data and the analyzing scripts.”
Like Retraction Watch? You can make a tax-deductible contribution 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].
Seems like an innocent mistake
You know, this is an argument for a version of ‘retraction’ that equals “We discovered we were wrong”, as opposed to “Someone else discovered we were wrong”. One should be a retraction; the other needs a different word. Rejection seems already taken; however “retroaction” appears to be a valid word. There’s also “retaliation” and “repercussion” to consider. Thoughts?
Or “honorable retraction” v “dishonorable retraction”
How about “withdrawal” for the voluntary kind? “The authors withdrew their paper” vs “The journal retracted its publication.”
The term “withdrawal” is reserved for manuscripts that haven’t been officially published.
Work quickly? After 6 years and 117 citations?
Yes! 117 citations and probably more in the mill need to be reviewed. It would be more useful to define the error exactly in detail as soon as it is discovered. Get the error notice out ASAP and then deal with its materiality, retract or no, honorable or no, etc.