Aarhus University Seal

How management dealt with the Nature article case

A scientific article published in Nature in 2016 has been the subject of criticism from peers for a number of years, and it has given rise to an important case for the university's Research Practice Committee. The matter has been mentioned several times in the newspaper Berlingske and most recently in other media. The media coverage could give the impression that the management at the faculty has passively looked on and not dealt with the matter. This is certainly not correct and there is still an ongoing proces. The following is an outline of how the department management team and the dean have dealt with the criticism that followed in the wake of the article in Nature entitled: Exploring the quantum speed limit with computer games.

[Translate to English:] Foto: Jesper Rais/AU Foto
[Translate to English:] Foto: Jesper Rais/AU Foto

April 2016
Two academic staff at the Department of Physics and Astronomy (IFA) at AU are the corresponding author and co-author, respectively of an article in Nature. The article is hereinafter referred to as Nature 2016.

July-August 2017
A researcher from Harvard University contacts the corresponding author and asks for a game code that the Harvard researcher considers is important for the results in Nature 2016.

April 2018
The Harvard researcher publishes an article in Physical Review criticising Nature 2016.

October 2018
The article in Physical Review leads to growing internal criticism at IFA during the summer and early autumn 2018. It develops from academic criticism to questions being raised about breaches of research integrity. As a consequence of this, in October, Head of Department Ulrik Uggerhøj (UIU) contacts the faculty's advisor for responsible conduct of research. The adviser investigates the matter and answers UIU that he is convinced that there is no problem with Nature 2016, although that there is a need to explain the Nature article in light of the article in Physical Review.

November 2018
The co-author sends an account to UIU and the deputy head of department, Lars Bojer Madsen (LBM), about his view of the Harvard researcher's statement in the April 2018 article.

April 2019
A researcher at the Department of Computer Science at AU publishes an article on? arXiv. In this article, he supports the Harvard researcher's views on the conclusions in Nature 2016.

May 2019
At meeting is held at IFA to find an understanding of the views of the local parties involved. The meeting is attended by the management of IFA, the corresponding author and co-author of Nature 2016, as well as the employees from the department primarily behind the criticism of Nature 2016.

8 July 2019
UIU receives a communication from the secretariat of the Research Practice Committee in response to a request for advice on the matter. The secretariat is of the opinion that the information forwarded sounds like an academic disagreement and advises UIU to contact the adviser for responsible conduct of research at the faculty again. On the same day, UIU asks the adviser to look into the case again, which he agrees to do.

12 July 2019
The researcher at the Department of Computer Science at AU asks the corresponding author to forward the codes for the first time.

8 August 2019
The faculty's adviser for responsible conduct of research informs UIU verbally about the results of his further investigations. The adviser has spoken to two of the parties involved and has assessed that no significant changes have occurred. The adviser therefore believes that there is no reason to do any more.

August-October 2019
UIU repeatedly asks the corresponding author to follow up on the public response to the criticism that he, at his own initiative, had assured the researcher from the Department of Computer Science at AU he would do in the early summer. This response is issued on 31.10.and corrected on 16.12, when the code error is discovered.

November, 2019
UIU is first informed verbally, then in writing (7.11.2019), that the researcher at Harvard has not received a game code after his inquiry in 2017, although the researcher had not  although the researcher had not since requested it.

15 November 2019
UIU orders the corresponding author to forward the computer codes with deadline 1.12.2019. Before the order, the co-author had notified UIU that they were looking for the code. However, UIU considers that the pressure to get the codes sent does not allow for a longer waiting time. UIU rejects a request to postpone the deadline, and the codes are forwarded as ordered on 1.12.19.

13 December 2019
The head of department at the Department of Computer Science, Kaj Grønbæk (KG), informs UIU that the researcher from the Department of Computer Science has found a sign error in the code in question. The researcher has written a memo related to this sign of error, which is published later (12.03.2020).

16 December 2019
The corresponding author and the co-author inform UIU about their own discovery of a serious sign error in the computer code. Briefings regarding the error and its significance take place over the following month.

3 February 2020
UIU informs academic staff at IFA about the errors found in the calculations behind Nature 2016, and that parts of the article's wording and figures are incorrect.

4 February 2020
The corresponding author and co-author send a report to Nature about the consequences of the error. With the knowledge and approval of the corresponding author and co-author, UIU forwards this report to KG and the researcher from the Department of Computer Science.

4-5 February 2020
UIU informs key foundations by telephone about the discovery of the error.

17 February 2020
Meeting between UIU and the then dean of Nat, Lars H. Andersen (LHA), where possible initiatives are discussed. Extraordinary meeting of the department management team for IFA, where it is decided that the matter is to be reported to the Research Practice Committee.

21 February 2020
Meeting between Rector Brian Bech Nielsen (BBN), LHA and UIU, where BBN and LHA agree to report the matter to the Research Practice Committee. The matter is submitted to the Research Practice Committee on 4.3.2020.

5 May 2020
Nature publishes notice of retraction of Nature 2016.

7 May 2020
Current Dean Kristian Pedersen (KP) receives the statement from the Research Practice Committee from BBN. The Research Practice Committee says that the delay in forwarding the code from the corresponding author and co-author (Nature2016) was questionable research practice. BBN and KP both approve the statement. The secretariat of the Research Practice Committee has a discussion with the Danish Committee on Scientific Dishonesty (NVU) about the inquiry, and in an email dated 31 March 2020, the NVU secretariat, in consultation with the chair of NVU declares the case outside of NVU's remit.

On this basis, the Research Practice Committee at Aarhus University assesses the matter with a view to assessing whether it is "questionable research practice" and so covered by the Research Practice Committee‘s remit.

11-19 May 2020
UIU receives a statement from the Research Practice Committee from KP. The corresponding author and co-author are informed orally about the statement from the Research Practice Committee at meetings on 12.5, and the other parties involved from IFA are notified at a meeting on 19.5, where the researchers from IFA present the additional point of criticism that a code without the crucial sign error exists dated May 2017. This information led to initiation of a clarifying process.

14 June 2020
KP and BBN discuss how KP can help clarify any outstanding points of criticism and look ahead.

15 June 2020
UIU asks the co-author for a meeting to explain an unresolved problem. In UIU’s opinion, the co-author submits a satisfactory explanation at the meeting.

10 August 2020
The corresponding author informs UIU that Nature 2016 was retracted on 22.7.

9 September 2020
Meeting between UIU and the corresponding author to clarify unanswered questions about details of the matter regarding the timing of acknowledgement of the code error. In UIU’s opinion, the corresponding author submits a satisfactory explanation at the meeting.

23 October 2020, 17 November 2020 and later dates
KP holds separate meetings with the parties in order to listen to and understand points of criticism, and to find a way to clarify these. Work is still ongoing regarding the onward process.