A case of dip in standards

January 26, 2007

Neuroscientist found guilty of 'deception' but not of 'fabrication' in paper on sheep memory. Phil Baty reports on a case that raises questions about integrity and peer review

An internationally renowned UK academic has been found guilty of "deception" in the reporting of research results in relation to a paper published in one of the world's leading science journals.

Keith Kendrick, head of the Laboratory of Cognitive and Behavioural Neuroscience at the Babraham Institute in Cambridge, is set to be disciplined - but not sacked - over misconduct relating to a 2001 paper he wrote for the journal Nature , called "Sheep don't forget a face".

After a disciplinary investigation prompted by inquiries from The Times Higher in August last year, Professor Kendrick was also found guilty this week of the "negligent deviation from accepted practice in carrying out research", relating to how he ran his research group at the time.

A correction of the work will be agreed with Nature , but Professor Kendrick has been cleared of fabricating data, as a whistleblower had originally suggested.

Michael Wakelam, director of the Babraham, said: "I will shortly personally review procedures and compliance with them to ensure that exemplary standards are upheld without exception."

But the verdict will raise more questions about the general integrity of UK research and the ability of the top journals and their peer reviewers to identify scientific malpractice.

The Babraham began a six-month investigation into the validity of data behind the Nature paper after a whistleblower raised concerns via The Times Higher .

The whistleblower claimed that although the paper had recorded that a group of sheep were found to remember the faces of other sheep in a series of experiments carried out up to 800 days after the first memory tests, the sheep in question had actually been killed before the reported final phase of experiments.

The disciplinary panel reported this week that Professor Kendrick explained that there had been another, entirely separate, set of sheep from which the latter set of data had been obtained. These data were based on experiments conducted earlier, in 1996-97.

The panel said it was "unfortunate and regrettable" there were no primary data from these earlier tests but Professor Kendrick's "possible explanations" for the lack of data were "not unreasonable, given the flooding of his laboratory in 2001".

It found that there was "substantial and convincing evidence" that the earlier group of sheep had existed and the relevant experiments were carried out on them but said that the "acceptance of Dr Kendrick's explanation inevitably raises further questions".

The panel found that the published paper "omits significant details of experimental design that should have been made available to the reviewers and subsequently to the readers of the published paper". It said: "This falls under the rubric of 'deception in reporting results of research'."

The panel also criticised the "standards for scientific management" of Professor Kendrick's group at that time. The panel did not take full evidence about how the group was managed now. "This falls under the rubric of 'negligent deviation from accepted practice'," the panel said.

Professor Kendrick also revealed during the investigation that there may be "incorrect statistical analysis" of some data.

Now, the panel said, it was "imperative for all stakeholders that the published record be fully corrected". It recommended that the data be reanalysed with an independent expert and that a "corrigendum" be sent to Nature to clarify the involvement of two different sets of sheep.

The panel said it had not recommended sacking because the most serious allegation, fabricating results, had been dismissed and that offences under the "deception in reporting results" were of "omission rather than commission" and that offences under "negligent deviation" "while serious were not reckless".

The Babraham would not confirm what disciplinary sanction had been imposed.

Professor Kendrick said he was "very satisfied" the panel had cleared him of the allegation of fabrication.

He said: "I accept that the allegation has arisen mainly due to confusions caused by my providing insufficient methodological details in this brief (one page) format of paper. This was not intentional, and I have always made it clear during the resultant investigation that I was willing to set the record straight."

phil.baty@thes.co.uk

HOW THE STORY UNFOLDED

NOVEMBER 2001

"Sheep don't forget a face" was published in Nature

August 2006

The Times Higher used the Freedom of Information Act to request the data underpinning the research in the Nature paper and records on the sheep deaths. The Times Higher agreed to waive its Fo... request to allow a full investigation by the Babraham. A preliminary investigation committee (PIC) was set up under Barry Keverne, professor of behavioural neuroscience at Cambridge University.

The PIC concluded that there was a need for a formal investigation.

A new investigation panel was set up under Len Stephens, head of Babraham's Inositide Laboratory.

NOVEMBER

Dr Stephens's panel concluded with a recommendation to move to afull disciplinary hearing.

A disciplinary panel was set up, chaired by Chris Lamb, director of the John Innes Institute.

DECEMBER 4

Kendrick's disciplinary hearing. Not completed, so moved to a second day.

DECEMBER 18

Second phase of Kendrick's disciplinary hearing.

JANUARY 2, 2007

Deadline for disciplinary panel's report.

JANUARY 23

Babraham issued a full statement.

  • Whistleblower investigates CALL NOW ON 020 7782 3298


Who is Kendrick and what was the research about?

THE MAN

Keith Kendrick, one of the UK's top neuroscientists, is head of the Laboratory of Cognitive and Behavioural Neuroscience at Cambridge University's Babraham Institute.

His work on sheep and face recognition has offered insight into the workings of the human brain.

But he is also renowned for his research into how neural systems are organised to control key functions in animals and humans. His work has covered stress and depression, love and sex, diet, gambling and music.

Professor Kendrick works to increase the public understanding of science, notably as the Gresham Professor of Physic (and the other biological sciences) at Gresham College in London, which has offered free lectures for the public for 400 years.

His Babraham group has also contributed to public exhibitions and shows including the Cheltenham Science Festival, the Royal Show, the BA festival and the Science Museum.

Professor Kendrick is listed on the independent Science Media Centre's list of 'technical experts', available to provide comment or opinion on issues of public interest.

THE RESEARCH

Keith Kendrick's 2001 Nature paper, "Sheep don't forget a face", provided headline writers with a field day.

"A baa-d memory? Not us say sheep", said the Daily Express . "I remember ewe," said the Scotsman (and several others). "Nothing woolly about sheep's faculties," offered The Guardian .

But the implications and possible future applications of the work were serious. The paper was listed as one of the "key publications" produced by Professor Kendrick's Babraham Laboratory.

It appeared to establish remarkable similarities between the brains of sheep and humans. The researchers discovered "hierarchical" structures in the sheep's brains, similar to humans, in which separate sections of the brain were dedicated to different tasks in relation to registering faces.

Professor Kendrick said in a statement this week that there was no question that sheep can recognise faces. "Sheep face recognition abilities have been established in many other publications from as far back as 1987 and replicated independently by a scientific group in France and in independent PhD, undergraduate and even high-school projects in the UK and US," he said.

"In short, the discovery that sheep can recognise faces is an extensively corroborated scientific fact and was never in question."

You've reached your article limit.

Register to continue

Registration is free and only takes a moment. Once registered you can read a total of 3 articles each month, plus:

  • Sign up for the editor's highlights
  • Receive World University Rankings news first
  • Get job alerts, shortlist jobs and save job searches
  • Participate in reader discussions and post comments
Register

Have your say

Log in or register to post comments