Skip to content

Harvard Researcher May Have Fabricated Data

Harvard authorities have made available information suggesting that Marc Hauser, a star researcher who was put on leave this month, may have fabricated data in a 2002 paper.

“Given the published design of the experiment, my conclusion is that the control condition was fabricated,” said Gerry Altmann, the editor of the journal Cognition, in which the experiment was published.

Dr. Hauser said he expected to have a statement about the Cognition paper available soon. He issued a statement last week saying he was “deeply sorry” and acknowledged having made “significant mistakes” but did not admit to any scientific misconduct.

Dr. Hauser is a leading expert in comparing animal and human mental processes and recently wrote a well-received book, “Moral Minds,” in which he explored the evolutionary basis of morality. An inquiry into his Harvard lab was opened in 2007 after students felt they were being pushed to reach a particular conclusion that they thought was incorrect. Though the inquiry was completed in January this year, Harvard announced only last week that Dr. Hauser had been required to retract the Cognition article, and it supplied no details about the episode.

On Friday, Dr. Altmann said Michael D. Smith, dean of the Faculty of Arts and Sciences, had given him a summary of the part of the confidential faculty inquiry related to the 2002 experiment, a test of whether monkeys could distinguish algebraic rules.

The summary included a description of a videotape recording the monkeys’ reaction to a test stimulus. Standard practice is to alternate a stimulus with a control condition, but no tests of the control condition are present on the videotape. Dr. Altmann, a psychologist at the University of York in England, said it seemed that the control experiments reported in the article were not performed.

Some forms of scientific error, like poor record keeping or even mistaken results, are forgivable, but fabrication of data, if such a charge were to be proved against Dr. Hauser, is usually followed by expulsion from the scientific community.

“There is a difference between breaking the rules and breaking the most sacred of all rules,” said Jonathan Haidt, a moral psychologist at the University of Virginia. The failure to have performed a reported control experiment would be “a very serious and perhaps unforgivable offense,” Dr. Haidt said.

Dr. Hauser’s case is unusual, however, because of his substantial contributions to the fields of animal cognition and the basis of morality. Dr. Altmann held out the possibility of redemption. “If he were to give a full and frank account of the errors he made, then the process can start of repatriating him into the community in some form,” he said.

Dr. Hauser’s fall from grace, if it occurs, could cast a shadow over several fields of research until Harvard makes clear the exact nature of the problems found in his lab. Last week, Dr. Smith, the Harvard dean, wrote in a letter to the faculty that he had found Dr. Hauser responsible for eight counts of scientific misconduct. He described these in general terms but did not specify fabrication. An oblique sentence in his letter said that the Cognition paper had been retracted because “the data produced in the published experiments did not support the published findings.”

Scientists trying to assess Dr. Hauser’s oeuvre are likely to take into account another issue besides the eight counts of misconduct. In 1995, Dr. Hauser published that cotton-top tamarins, the monkey species he worked with, could recognize themselves in a mirror. The finding was challenged by the psychologist Gordon Gallup, who asked for the videotapes and has said that he could see no evidence in the monkey’s reactions for what Dr. Hauser had reported. Dr. Hauser later wrote in another paper that he could not repeat the finding.

Full story