A prominent German neuroscientist committed scientific misconduct in research in which he claimed to have developed a brain-monitoring technique able to read certain thoughts of paralysed people, Germany’s main research agency has found.
The DFG’s investigation into Niels Birbaumer’s high-profile work found that data in two papers were incomplete and that the scientific analysis was flawed — although it did not comment on whether the approach was valid. In a 19 September statement, the agency, which funded some of the work, said it was imposing some of its most severe sanctions to Birbaumer, who has positions at the University of Tübingen in Germany and the Wyss Center for Bio and Neuroengineering in Geneva, Switzerland. The DFG has banned Birbaumer from applying for its grants and from serving as a DFG evaluator for five years. The agency has also recommended the retraction of the two papers1,2 published in PLoS Biology, and says that it will ask him to return the grant money that he used to generate the data underpinning the papers.
“The DFG has found scientific misconduct on my part and has imposed sanctions. I must therefore accept that I was unable to refute the allegations made against me,” Birbaumer said in a statement e-mailed to Nature in response to the DFG’s findings. In a subsequent phone conversation with Nature, Birbaumer added that he could not comment further on the findings because the DFG has not yet provided him with specific details on the reasoning behind the decisions.
Birbaumer says he stands by his studies, which he says, “show that it is possible to communicate with patients who are completely paralysed, through computer-based analysis of blood flow and brain currents”.
The DFG also found that Ujwal Chaudhary, first author of both of the PLoS Biology papers and a member of Birbaumer’s team at the University of Tübingen and the Wyss Center, had committed scientific misconduct. The agency banned Chaudhary from applying for its grants and from serving as a DFG evaluator for three years. Chaudhary did not respond to a request for comment from Nature.
The misconduct findings against Birbaumer and Chaudhary relate to research conducted in 2013–14, in which they worked with four people with the neurodegenerative condition motor-neuron disease, also known as amyotrophic lateral sclerosis, who were being cared for at home by relatives. The scientists recorded the patients’ brain activity using sensors on their scalps. In a 2017 paper1, Birbaumer and his colleagues reported that their analysis of the recordings allowed them to determine whether the patients were silently answering ‘yes’ or ‘no’ to simple questions. The paper attracted extensive media attention.
In the summary of its investigation committee’s findings, the DFG says that the scientists did not film patient examinations in full, did not appropriately show details of their analyses in the papers and made false statements.
The DFG stressed that the scientists had a “special responsibility” towards seriously ill people participating in innovative research. It said that they had not met this responsibility, “in particular by failing to document exactly the entire research procedure”.
Birbaumer says that filming often had to be interrupted to meet the participants’ immediate care needs, such as the need to suction saliva from their mouths. “For this reason, we did not upload data that we collected but had to declare as not analyzable in the publication. In addition, we did not describe every single step of the complex data evaluation and did not fully document it with accompanying video recordings,” he said in his e-mailed statement.
The DFG and the University of Tübingen opened separate investigations into the work in earlier this year, after a whistle-blower raised concerns about the research. Martin Spüler, who was then a postdoc in informatics at Tübingen, said he was unable to reproduce the findings when he reanalysed the published data. An independent expert commissioned by the DFG confirmed Spüler’s findings, as did two additional whistle-blowers, according to the agency’s statement. The DFG commission found that the researchers had not analysed their data correctly. Four other co-authors of the studies were not investigated.
Birbaumer and his team published a rebuttal to Spüler’s criticisms2 in April this year — the second paper whose retraction the DFG recommended. PLoS Biology added expressions of concerns to both studies shortly after the DFG announced the findings of its investigation.
The University of Tübingen’s investigation into Birbaumer and Chaudhary’s work concluded in June that the two had committed guilty of scientific misconduct.