Misconduct found in Harvard Animal Cognition Lab

[Alyrium Denryle]

http://www.newscientist.com/article/dn1 ... s-lab.html

Misconduct found in Harvard animal morality prof's lab

* Updated 14:39 11 August 2010 by Peter Aldhous

Do animals think? For many years, each time the media wanted insights into this most intriguing of questions, Marc Hauser was the scientist of choice. His Harvard University lab produced a string of intriguing papers suggesting that monkeys have uncannily human-like mental capacities, and Hauser was always ready with an eloquent explanation.

Now that work lies under a cloud, after the Boston Globe broke the news that a three-year Harvard investigation has found "evidence of scientific misconduct" in his laboratory.

At least one scientific paper is being retracted, questions are swirling around the rest of Hauser's work, and Harvard is refusing to comment. All of which leaves other researchers annoyed that the university has not disclosed the details of the scandal, even though it has left the field of animal cognition in confusion.

"It's important that Harvard release the data that they questioned, because otherwise you don't know what to believe," says Herbert Terrace, a specialist in primate cognition at Columbia University in New York.

While Hauser has enjoyed a high public profile – his work has featured regularly in New Scientist, among other publications – some of his scientific colleagues contacted by New Scientist say they had already questioned his conclusions. Primatologists who spend their career working with just a few species have "a high level of intuition" about what their favourite animals are capable of, explains Frans de Waal of Emory University in Atlanta, Georgia. "Marc would come out with findings that we found hard to follow."
Roots of language

Hauser made his name through studies suggesting that cotton-top tamarins and rhesus macaques were capable of mental feats previously seen as human specialisms.

In the paper that is now being retracted, for instance, published in Cognition in 2002 (vol 85, p B15), Hauser played recordings of series of syllables to tamarins and found that the animals seemed to recognise sequences that followed a similar pattern, even when syllables themselves were different. This capacity is thought to be crucial to children's ability to learn languages – but Hauser's work suggested it emerged earlier in primate evolutionary history for some other purpose.

The current Harvard investigation isn't the first time that Hauser's findings have come under question, however. After he published a paper in 1995 suggesting that cotton-top tamarins can recognise themselves in a mirror (Proceedings of the National Academy of Sciences, vol 92, p 10811), the researcher who first demonstrated this ability in chimps, Gordon Gallup of the State University of New York at Albany, asked to see videos of the experiments. "I was shocked when I sat down and reviewed the tapes," says Gallup. "There wasn't a shred of evidence."

Gallup says that he was worried about Hauser's over-enthusiastic interpretation of his results. But according to other scientists in the field, Harvard decided to investigate his lab only after students who had worked there came forward with allegations of data falsification – a much more serious charge.
Paper trail

It is not clear how much of Hauser's work may be based on flawed data. Some journals have been told of problems with papers they have published. Science, for instance, was informed on 27 June that an investigation had examined record-keeping in a 2007 paper exploring whether various primate species are able to understand human gestures that indicate the presence of hidden food (Science, vol 317, p 1402) – but the journal has so far received no further details.

Meanwhile, related experiments showing the responses of wild rhesus macaques to human gestures, which were described in a 2007 paper in Proceedings of the Royal Society B (vol 274, p 1913), have been repeated. A correction states that the experiments were repeated after "the discovery of incomplete video records and field notes".

Given that Hauser's lab has worked on many aspects of cognition, including the evolution of morality and monkeys' abilities to count and perform simple arithmetic, scientists in the field are frustrated by Harvard's insistence that it will provide no further details at present. "As a general policy, reviews of faculty conduct are considered confidential," university spokesman Jeff Neal told New Scientist.

Hauser is now on leave for a year, and a message to his email was met with an auto-reply: "I am on leave, working furiously on a book, and thus will only be checking email irregularly."
Right to silence?

"Marc Hauser's rights are being protected. Harvard's rights are being protected," complains Michael Tomasello, a specialist in chimpanzee cognition at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany. "Who's looking out for the scientific community? We're in a quandary."

Even some of Hauser's Harvard colleagues are calling for the university to end the confusion by revealing the details of its investigation. Howard Gardner, who works on human cognition and education, told New Scientist: "Keeping proceedings secret simply produces rumours, and, in the era of blogging, these can be wild."

While the recriminations grow, some researchers hope that the affair will encourage those studying animals' mental capacities to be more careful before making extraordinary claims from behavioural observations, because such observations require careful checks to ensure reliable recording and correct interpretation.

Gallup, for instance, argues that several findings about mirror self-recognition in animals are based on over-interpretation of flimsy data. "Marc's not alone in that respect," he says.

"One benefit of this will be that people will be more careful about going out on a limb," predicts Terrace.

Well, his career is over...

[Surlethe]

I wouldn't say that his career is necessarily over; it depends on what the investigation finds. The big problem right now is uncertainty, not falsehoods: nobody is sure what to believe. If we know which data are good and which are false, we're good to go.

By the way, I split the big three page transhumanism tangent.

[Anguirus]

My god, was there only ONE on-topic post?

Although part of the problem is that there's not much to say yet. Everyone is pissy because Harvard won't disclose which studies might be considered fraudulent!

[Ziggy Stardust]

My on-topic post got drowned in the transhumanism crap, so to repost:

It definitely doesn't look good for him. The question is whether the Harvard investigation will actually accuse of him of outright falsification or not. It was certainly sloppy science on his part, but when it comes to animal behavior, especially of those animals that we consider to be "smarter" than others, it is definitely easy to read too much into what you are seeing, and ascribe meaning that isn't there. I mean, either way his career is dead, except possibly as a lecturer somewhere. He certainly isn't getting any big research grants ever again ...