Monkey business?

« previous post | next post »

Carolyn Y. Johnson, "Author on leave after Harvard inquiry", Boston Globe 8/10'/2010:

Harvard University psychologist Marc Hauser — a well-known scientist and author of the book “Moral Minds’’ — is taking a year-long leave after a lengthy internal investigation found evidence of scientific misconduct in his laboratory.

The findings have resulted in the retraction of an influential study that he led. “MH accepts responsibility for the error,’’ says the retraction of the study on whether monkeys learn rules, which was published in 2002 in the journal Cognition.

Two other journals say they have been notified of concerns in papers on which Hauser is listed as one of the main authors.

It is unusual for a scientist as prominent as Hauser — a popular professor and eloquent communicator of science whose work has often been featured on television and in newspapers — to be named in an investigation of scientific misconduct. His research focuses on the evolutionary roots of the human mind.

Heidi Ledford ("Harvard morality researcher investigated for scientific misconduct", Nature 8/10/2010) gives links to the three papers that have apparently been under scrutiny. All involve experiments on non-human primates, but the list of papers under investigation does not include the monkey paper from Hauser's lab that we've discussed most extensively here: W. Tecumseh Fitch and Marc D. Hauser, "Computational Constraints on Syntactic Processing in a Nonhuman Primate", Science 303(5656):377-380, 16 January 2004.

That paper tested cotton-top tamarins' habituation to very simple stimulus patterns (ABAB and ABABAB vs. AABB and AAABBB),  and on that basis made general claims about their ability to process infinite classes of grammars generating indefinitely long strings of symbols. At the time, this seemed to me to be one of the most enthusiastically over-interpreted results ever to appear in a major scientific journal.

In addition, I felt that the experiments didn't really engage the mechanisms for composing and decomposing linguistic messages at all, but rather explored the mechanisms of auditory texture discrimination, whose connection to linguistic processing is at best unproved. And even in regards to texture perception, I thought that several simple and plausible explanations were being disregarded in favor of more complex (and headline-worthy) stories.

For details, see

"Hi Lo Hi Lo, it's off to formal language theory we go", 1/17/2004
"Humans context-free, monkeys finite-state? Apparently not", 8/31/2004
"Rhyme schemes, texture discrimination and monkey syntax", 02/09/2006
"Learnable and unlearnable patterns — of what?", 02/25/2006
"Starlings", 4/27/2006
"The texture of time: Even educated fleas do it", 11/24/2009

For a nice summary, see Geoff Pullum's post "Recognizing grammar (or door chime changes, or anything)", 6/22/2009.

And for a broader survey of some of the headline-worthy evolutionary issues that the tamarin experiments were meant to engage, see "JP versus FHC+CHF versus PJ versus HCF", 8/25/2005.

Like many other linguists, Geoff and I have felt from the beginning that the results of Hauser's monkey experiments were of dubious relevance to the evolution of speech and language. Now we're forced to question whether there were any reliable results at all.

The Globe article says,

Much remains unclear, including why the investigation took so long, the specifics of the misconduct, and whether Hauser’s leave is a punishment for his actions.

But it doesn't look good. As Neuroskeptic points out ("Hauser Of Cards", 8/10/2010), "the only author who appears on all of the papers known to be under scrutiny is Marc Hauser himself".

One more general comment — it's long past time for scientists to start publishing their raw data along with their conclusions, and for publishers to take steps to make this not only possible but easy. Reproducible research: it's the right thing to do.

[Update 8/12/2010 — survey of reactions here; more from Carolyn Johnson at the Boston Globe here; Nicholas Wade, "Expert on Morality Is on Leave After Research Inquiry", NYT 8/11/2010; and "Inquiry on Harvard Lab Threatens Ripple Effect", 8/12/2010.]

[8/13/2010 — more commentary on the lack of information in Carolyn Johnson, "Harvard to rectify journal works", Boston Globe 8/13/2010.]

[8/14/2010 — Nicholas Wade, "In Harvard Lab Inquiry, a Raid and a 3-Year Wait", NYT 8/13/2010:

Marc Hauser’s academic career was soaring when suddenly, three years ago, Harvard authorities raided his laboratory and confiscated computers and records. […]

In January this year, a faculty committee at last completed its report, said to contain eight charges against Dr. Hauser. But the report was kept secret and nothing changed until this month when someone showed The Boston Globe a letter about the investigation from Dr. Hauser to his faculty colleagues. […]

The captive animals, a colony of some 40 cotton-topped tamarins, may have contributed to the difficulties in Dr. Hauser’s laboratory. It is difficult to get the tamarins to pay attention, especially after the monkeys get used to experimenters.

“With some of these methods it was never clear to me how one could obtain meaningful results,” said a person with experience in Dr. Hauser’s lab, who requested anonymity for fear of retaliation. “The monkeys were often either jumping around, or not moving at all, and you rarely got the sense of an unambiguous response.”

Other experimental problems have come to light with three articles investigated by the Harvard committee. In two, the supporting data did not exist. Dr. Hauser and a colleague repeated the experiments, and say they got the same results as published. In a third case, Dr. Hauser retracted an article published in the journal Cognition in 2002 but gave the editor no explanation of his reason for doing so.

Whatever the problems in Dr. Hauser’s lab, they eventually led to an insurrection among his staff, said Michael Tomasello, a psychologist who is co-director of the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany, and shares Dr. Hauser’s interest in cognition and language.

“Three years ago,” Dr. Tomasello said, “when Marc was in Australia, the university came in and seized his hard drives and videos because some students in his lab said, ‘Enough is enough.’ They said this was a pattern and they had specific evidence.”

Wow. ]


  1. Dan Everett said,

    August 11, 2010 @ 9:04 am

    Publishing the raw data is very important. For phonology articles, the sound files should be downloadable. And so on. Having said that, it is extremely time-consuming to process the data in a way that either colleagues or the general public can follow. That is no excuse for not doing it, though.

    [(myl) With the right data-collection procedures in place from the beginning, and with decent technical support for relevant kinds of data publishing (which does not now exist), the burden would be much lighter. And since the velocity of research interactions would be materially increased — because of lowered barriers to entry for others, and because authors themselves would be able to extend prior work much more easily — the overall effect would be to speed research up, not slow it down. In my opinion, this increase in research productivity is the real reason to upgrade to a paradigm of reproducible research. Avoiding/catching fraud is a minor side-benefit.]

    Regarding Hauser's woes, these are now our woes too. Creationists and proponents of Intelligent Design, opponents of Climate Change, and so on, will be delighted by this news. "You don't need to pay attention to science 'cause those guys just make the data fit their theories anyway." A very sad day.

    [(myl) I sincerely hope that this turns out to be nothing more than carelessness with experimental records, for the sake of the field as well as for Marc's sake.]

  2. GF said,

    August 11, 2010 @ 9:52 am

    What is the evidence for the sentence "It is unusual for a scientist as prominent as Hauser […] to be named in an investigation of scientific misconduct."? I personally find that a ridiculous thing to say.

    [(myl) You should direct your question to Carolyn Johnson at the Boston Globe, who wrote the article from which I quoted that sentence. But taken literally, it's clearly true — which scientists "as prominent as Hauser" have been "named in an investigation of scientific misconduct" within, say, the past decade? I can't think of any, offhand, but there surely have not been very many.

    Meanwhile, you've given us no reason to care about what you do or don't find ridiculous. Do you yourself have anything to offer the discussion beyond your unexplained emotional reactions?]

  3. Bill Benzon said,

    August 11, 2010 @ 11:01 am

    I don't know what I think of this situation. I agree, Mark, about the flashy over interpretation in the 2004 paper. But over interpretation seems to be the coin of the realm, as in the case of mirror neurons and Theory of Mind. For all I know it's taught in basic methods courses.

    Where I'm at a loss is whether or not there is any worthwhile generalization here. I've come to believe that these days the human sciences are just drifting along, going nowhere in particular, because that's what the institutional imperatives are. Every ambitious alpha-style investigator gets his or her own lab and is allowed his or her theory. And, I suppose, one of the simiplest routes to that is through over-interpretation. I've been listening to calls for unity and synthesis for over thirty years and have come to regard them as but another invitationi to specialization. Did Hauser just go too far in marking out his own turf? Is this merely an individual failing, or is there an institutional aspect — by which I don't mean Harvard so much as the whole system of which Harvard is a part.

    The quest for individual and institutional glory has always been there. What I'm wondering is whether or not that's pretty much all we've got these days.

    [(myl) Good questions, and hard ones. My own beliefs, which I won't try to justify here, are generally optimistic ones:

    * over the past few decades, there's been a lot of sound scholarship, science, and technology development in the general areas of speech, language, and communication;
    * "curiosity-driven" individual-investigator research remains a crucial and productive part of the process;
    * larger-scale and more goal-directed efforts are also important, and have been successful recently not only in the engineering end of the field, but in some scholarly and scientific enterprises as well.

    That's not to say that there are no problems. I continue to think that the turn away from linguistic analysis in the social sciences and the humanities is a problem for all concerned. There's a worrisome divergence among physiologists, psychophysicists, and engineers in the area of audition, compared to relative harmony and coherence among the analogous people in vision. And so on.

    As for the institutional aspect, I can't speak to the atmosphere at Harvard, but I think that the ecology of Science and Nature is worth a look. For a non-biological article to be published there, it has to seem understandable and revolutionary to a general scientific audience, which promotes over-interpretation. And reviewers are in a bind ("am I going to torpedo the only psychology article to appear in Nature this year?") as well as being chosen for eminence rather than article-specific knowledge, which promotes under-reviewing. (And see here for some quantitative evidence that over-interpretation is associated with higher impact factors even in biology.)]

  4. Chris said,

    August 11, 2010 @ 11:28 am

    I think Dan Everett's concern is legitimate. Just look at what happened with "climategate." We already know the general public does not analyze the specifics of science in the news, they just read headlines.

  5. Neuroskeptic said,

    August 11, 2010 @ 11:29 am

    "Mr. Everett: It's sad because scientists have lied."

    No one or more scientists might have lied. We don't know what's going on here.

  6. MattF said,

    August 11, 2010 @ 11:56 am

    No one's mentioned the freakishly ironic subtext here, with respect to Hauser's recent work. I can't help thinking that we're in "compulsion-to-confess" territory.

    And, frankly, I wouldn't be too saddened if the various ev-dev theories get smacked around a bit. It's possible that we (that is to say, we scientists) have been somewhat remiss in not calling out bullshit sooner.

  7. Dan Everett said,

    August 11, 2010 @ 12:34 pm

    Louis Leakey was the center of a scandal at the beginning of his career that discredited him for a long period of time. He made a claim. Someone came to check the data. The data couldn't be found. It was careless and, perhaps, overinterpretation. But Leakey learned from that, eventually recovered from this blow to his reputation and produced lots of wonderful science after that. We all wish for such a happy ending to this story.

    Moreover, whenever your research somehow catches a reporter's eye, you are going to come under greater scrutiny. And there will be people who are very happy when you fail. The last exchange I had with Marc, a few months ago, was about schadenfreude in the Amazon. There is a temptation to that when we disagree with someone's theory in a situation like this. But we should be aware that one misplaced bit of data, if your work has attracted a lot of attention, could bring about a similar caca-storm for any of us. We don't know what happened in the Hauser case. Harvard should make that clear. But I do believe that being too quick to cast stones or to underestimate the damage to all of science from incidents like this is bad for all involved.

    If, however, lying and outright fraud took place, then this is even worse. But we don't know.

  8. Lane Greene said,

    August 11, 2010 @ 11:36 am

    It is bad to see someone behaving badly for its own sake, of course. But Dan Everett is right that it is *also* sad, and potentially of far greater consequence, when it fuels doubt about the work of other, honest and ethical, researchers. This goes doubly when that doubt is motivated by anti-scientific prejudice, religious mania or greed.

    As a former climate and energy correspondent who covered the "Climategate" ruckus, I found myself disappointed with the behaviour of some of the scientists involved, but it did almost nothing to my belief that the overwhelming balance of evidence favors man-made climate change. Sadly, when I talked to non-specialists who were also sceptics or denialists, they simply pointed to "Climategate" as an excuse to stopper their ears. They really thought that a few scientists misbehaving disproved the whole enterprise. If climate change truly takes off, the consequences will be far greater than any concern I have for Phil Jones's or Mike Mann's reputations.

    This is actually a smaller deal in the global balance of thing – so far, denial of evolution does not threaten the planet's survival. But Dan is right. This is depressing in a way that could go beyond Marc Hauser.

  9. Josef Fruehwald said,

    August 11, 2010 @ 11:53 am

    One scientist is under investigation for misconduct. The nature of that misconduct isn't known yet, but sure, it might include lying. However, to conclude that scientists have lied is to make a rather large generalization from a very small sample. There are, in fact, very many people who we could call "scientists" who have not lied. The proportion of liars over all must be exceedingly small. And, even if it was a little bigger than I suppose, most people who aren't scientists wouldn't care. Marc Hauser is rather unique in researching and writing about big, public interest questions, like "Where does morality come from." Most day to day work of scientists is trying to establish facts which are incredibly mundane.

    I actually think the points made by Bill Benzon are much more damning, but I think speak more to stagnation of progress. Under his view, if Hauser was fudging the results, it was not to really stick it to his creationist, ID "enemies". If anything, it would have been to impress is competitor fellow scientists.

  10. J. W. Brewer said,

    August 11, 2010 @ 1:41 pm

    Contra MattF, I don't see why the subtext should be ironic at all. Anyone who enjoys the hobby of making up evolutionary biology Just-So Stories but who can't immediately come up with one about why lying (if that's what it turns out to be) could have been (and therefore must have been) an adaptive trait selected for among our hominid ancestors back on the savannah ought to be drummed out of the profession. And if it wasn't actual lying but sloppily overinterpreting sketchy data in a way that enhanced the social position of the hominid that was doing it . . . that's a pretty easy Just-So Story too.

  11. Jonathan Badger said,

    August 11, 2010 @ 3:49 pm

    @GF I think Mark based that comment on affairs lik the infamous "Baltimore Case" in the 1980s. Rather than target David Baltimore (a Nobel Laureate), the blame for unreproducible data was put on one his collaborators who was an assistant professor at the time. Of course in that case the whole thing was a tempest in a teapot as no fraud was ever proven; bad data is not fraudulent in and of itself.

    [(myl) The statement that GF quoted came from the 8./10/2010 article by Carolyn Johnson in the Boston Globe, whose opening paragraphs I quoted verbatim. Johnson offered no evidence for this assertion.

    If you take it to mean that high-profile public-intellectual scientists like Marc Hauser are rarely named in an investigation of scientific misconduct (which is what the article literally says), then it's clearly true. I can't think of any other examples in the past decade, for example, depending on how you define "a scientist as prominent as Hauser".

    If you take it to mean (as I guess you do) that such things are commoner among grad students, postdocs, junior faculty, and unknown senior researchers, there's a sense in which this is true — there are many episodes every year — but also a sense in which it's false — there's no evidence (known to me) that the proportions are any different.]

  12. Jesse Hochstadt said,

    August 11, 2010 @ 8:22 pm

    @MattF: A little tangential, but what does ev-dev have to do with anything? AFAIK, the term refers to the same thing as evo-devo, which Wikipedia ( defines as follows: "Evolutionary developmental biology (… informally, evo-devo) is a field of biology that compares the developmental processes of different organisms to determine the ancestral relationship between them, and to discover how developmental processes evolved." It is concerned with embryonic development, the underlying genetics, and how they are modified over phylogeny.

    Do you perhaps mean the far more speculative field of evolutionary psychology?

  13. newsouthzach said,

    August 12, 2010 @ 12:52 am

    Hendrik Schon, formerly of Bell Labs, was reasonably prominent in the condensed-matter physics community until the 2003-ish discovery of a massive fraud involving outright falsification of data for publications in Science, Nature, and other high-profile journals. You might not have heard of him, though — wildly different field.

    [(myl) I do know about Schön's case, and it's certainly relevant to a discussion of breaches of scientific trust (as are many other well-documented episodes). But his case is not relevant to GF's query about the facts behind the assertion that "It is unusual for a scientist as prominent as Hauser […] to be named in an investigation of scientific misconduct." Schön was a postdoc at Bell Labs; Hauser is a full professor at Harvard, and a major public intellectual who "makes frequent appearances on various NPR shows, as well as television and international radio."

    Within the past year, the NYT featured Marc Hauser's work in two major stories by Nicholas Wade, one David Brooks column, a Freakonomics blog post, and three other blog posts. Within the year before the news of Jan Hendrik Schön's problems came out, the NYT didn't mention him even once.]

  14. Jason Eisner said,

    August 12, 2010 @ 6:49 am

    This 2009 book by Eugenie Samuel Reich is a full journalistic account of the Schön case aimed at lay readers:

    Here's a long review with some discussion:

    (Full disclosure: I slightly know both the author (a distant relative) and, coincidentally, the reviewer.)

  15. noahpoah said,

    August 12, 2010 @ 7:42 am

    With respect to sharing data, what counts as the raw data for the monkey studies in question? The videotapes? The transcriptions of the videotapes? Both?

    [(myl) Digital video files with time-marked annotations would be a good (and easy) place to start.]

    Also, as a side (snarky) note, how many phonology articles actually have sound files associated with them?

    [(myl) An increasing amount of speech research is done using evidence from published speech corpora. A few recent papers of this kind that I've been associated with are here, here, here. Some other examples are here and here.

    For several decades, at least a plurality of published papers on child language acquisition have been based on published (raw) data, or have been accompanied by publication of the associated recordings, usually via CHILDES.

    And in many areas of speech and language engineering, claims of results that aren't based on publicly-available raw material will receive a very skeptical response, to the point that such results may be hard to publish unless there are special circumstances.]

  16. Jongseong Park said,

    August 12, 2010 @ 10:36 am

    As long as we are talking about prominent scientists being named in an investigation of scientific misconduct, you can't find too many cases in recent memory that are more high-profile than that of disgraced stem cell researcher Hwang Woo-Suk. He became a household name in Korea with a series of breakthroughs including the creation of human embryonic stem cells by cloning, only to create a national scandal when much of the data was subsequently revealed to have been fabricated.

    [(myl) Yes, I agree that Hwang was arguably "a scientist as prominent as Hauser", even in the U.S. I thought of him in connection with this discussion, and falsely remembered that his work had not gotten much international coverage before the scandal broke. But a quick check of newspaper archives shows that I was wrong.]

  17. Greg Laden said,

    August 12, 2010 @ 5:32 pm

    Dan: Louis Leakey was the center of a scandal at the beginning of his career that discredited him for a long period of time.

    That's a little like saying "Some kid had a birthday party at Chucky Cheese" … Are you referring to the bet over the hand axes?

  18. the other Mark P said,

    August 13, 2010 @ 5:50 am

    Just look at what happened with "climategate."

    What happened with climategate was that high profile scientists refused to supply evidence of their claims to open scrutiny. Indeed they actively colluded to avoid it, in defiance of official information laws.

    What Mark says about the need for openness in publishing data and findings goes double when financial and policy issues are at stake.

    When data is not published, and the given reason is "you will just try to show that we are wrong" then I believe that the science being touted by the scientists in question lacks credibility. I think the public was right to see that lack of openness as evidence that the findings were doubtful.

    (Of course Mann and Jones might be right despite their poor behaviour.)

    I believe that if the state pays for research, then it should be published in full. If researchers do not wish to share their material, then they should find private sources of funding.

  19. Bill Benzon said,

    August 13, 2010 @ 6:41 am

    The New York Times has now published an article on this case. They don't seem to have more concrete information than the Globe did. Their piece ends in a paragraph of murky implication:

    “The people who really know what’s happened are students, current and former,” said a scientist who asked to remain anonymous because of Dr. Hauser’s continuing power in the field. “They are very unhappy about how Harvard has handled this, and they feel things are being swept under the rug.”

    I haven't got the foggiest idea what's going on. But the ambiguity and uncertainty is not good. If Harvard's engaged in a cover-up, that's certainly not good. If they're keeping silent because they've not yet gotten to the bottom of things, that's not good either, but in a different way.

  20. Dan Everett said,

    August 13, 2010 @ 7:21 am

    To Greg: I am referring to the Louis Leakey controversy over Reck's dating and tools, yes. Of course he had already lost funding and his morals questioned and the scientific community was against him because of an impending divorce and on-going affair.

    C.S. Peirce's entire career was destroyed because, beyond his weird personality, he was seen entering a hotel far from home with a woman he didn't happen to be married to.

    Ideas of ethics may have changed. But sometimes people can enjoy rehabilitation from even the most serious charges. Leakey did. Peirce did not.

    But Mark is right, I think. For Harvard to name Hauser is very unusual. A similar controversy in the humanities at Emory has been remembered recently in the Chronicle of Higher Education ( In the latter case colleagues have not been forgiving at all.

  21. john riemann soong said,

    August 13, 2010 @ 3:08 pm

    Sometimes the raw data is over 40 GB+ worth of material.

    When I was writing up my results, the material presented became about 1+ MB.

    The problem of access to the raw data seems a little difficult.

    [(myl) With respect, this is already wrong, rapidly sliding off into the land of patent nonsense. Given that terabyte drives are now O($100), the cost to archive your 40GB of data is negligible — O($4). If you follow appropriate collection and archiving practices, the cost of indexed internet access would also be low and falling rapidly. And for distribution of the whole thing — if anyone is interested — your whole dataset will fit on one dual-layer blu-ray disk, with 10GB to spare, for a media cost of O($2).

    I'm leaving out the fixed costs of providing this sort of infrastructure — space, people, power, and so on. But in terms of the physical substrate required, it's probably much more expensive to make a xerox copy of your research report than it is to archive or distribute data on the scale that you describe.

    A couple of hundred hours of video is a bigger deal — but, with respect, so is Marc Hauser. And in a few years, a few hundred hours of video will be as negligible (in terms of storage and distribution) as a few hundred hours of audio are now.]

  22. Sili said,

    August 14, 2010 @ 1:06 pm

    the other Mark P,

    This is not the place to discuss climate science and I wholly respect Liberman's right to delete my comment. But.

    The data that were repeatedly requested are in the public domain. The last commission to investigate the brouhaha could collect the data and reproduce the necessary computer programmes from the published articles in something like three days. The repeated bothering of CRU was wholly uncalled for. A fishing expedition designed largely to harass.

    That said, conspiring to subvert FoI legislation is indeed ethically unacceptable and the scientists have rightly been chastised for this. What I personally don't understand is why the Hadley centre did not have a dedicated officer to deal with this deluge of PoI requests instead of leaving it to the scientists. I can only assume that it's a question of someone in management not wanting to spend money on such a salary despite it being a far better use of resources than destroying the working environment of the people trying to do research.

  23. Hauser Apologist said,

    August 17, 2010 @ 5:47 am

    @myl re: "Digital video files with time-marked annotations would be a good (and easy) place to start."

    Having worked on similar experiments before, I think you've got that exactly wrong: Digital video files with time-marked annotations would be the difficult-to-accomplish ideal endpoint!

    Most commonly you record on a fuzzy VHS. Then, if your project requires it (like if you have poor undergrad intro psych students who need course credit helping to code your study), you begrudgingly spend hours and hours transferring those videos to an iMac.

    And then if you need time coded annotations on those videos, well, first you spend dozens of hours starting at a fuzzy video of a tamarin. You black out every once in a while, and the rest of the time you bang your head against the wall, wondering if you'll ever finish your dissertation. You find yourself wondering if monkey meat tastes best raw or slightly seared, like a good slab of Ahi. And then you just make up some results and claim the files were lost or corrupted when they begin an investigation into your unreplicable results :)

    But seriously, the technology I've most often seen in animal psych labs doesn't come close to making digital video files with time-marked annotation a good and easy place to start. You'd think with the big grants that a lot of these places seem to get it wouldn't be a problem, but maybe just caring for the animals eats into most of that (especially with CTT, which I think are endangered)? Sure, let's aim high and all, but you're shooting for the moon!

    [(myl) If this is true, then the people who run those labs are bigger fools than I think they are. HD digital video cameras are now available for a few hundred dollars. And many programs for interactive annotation of digital video are easily available, and some of the good ones are free — one among literally dozens is here. So let's review the required investment. Digital video recorders: a few hundred dollars each, or maybe up to $1,900 if you want to be unnecessarily professional about it ; software for interaction video annotation: zero; hardware for digital video annotation: zero (use existing computers). Total cost: $1,000-2,000.

    These prices have come down a lot in recent years, so the calculation would have been different 10 or 15 years ago — though it would still have been feasible, and worth doing.

    If the videos are "fuzzy" then the people making them need to read a couple of manuals or take a course. And as for the dozens of hours spent staring at them — why not record the fruits of that labor in a permanent record that you and others can build on? With a little bit of design work, it's lots easier than writing stuff down with pencil and paper, or making notes in Microsoft Word, or whatever you're used to doing — and extraction of the data for statistical analysis then takes a single menu selection, not hours of painful and error-prone scribal transfer.

    Failure to run a lab along these lines would be nothing short of foolish.]

  24. Nancy Hall said,

    August 29, 2010 @ 5:02 pm

    Although it's not an issue with monkeys, a big problem with sharing raw data in areas like phonology would be getting IRB permission. Our IRB has to be coaxed to even let the researcher keep the data more than 3 years (they would like all audio recordings to be destroyed at that point); I think they'd hit the roof at a request to make participants' recordings freely downloadable, even with appropriate permission from the participants.

    [(myl) This is basically a matter of educating your IRB (and your researchers) about appropriate standards for various areas of research. My experience is that some IRBs are dominated by rules of thumb, derived from biomedical practice, which have no standing whatever in the underlying laws, policies, or principles about protection of human subjects. And some researchers in the social and behavioral sciences accept these exaggerated precautions, and even participate in exaggerating them further. At Penn, we've published recordings of interviews, meetings, and telephone conversations involved literally tens of thousands of speakers from all over the world, collected using protocols with full IRB approval. We've been doing this since 1992, without a single complaint.

    Publication of data, however, need not be equivalent to "making participants' recordings freely downloadable" — where necessary or appropriate, there can and should be users' agreements to protect the rights of various parties involved, including experimental subjects.

    If you'd like to change your IRB's attitude towards such things, get in touch with me and I can share some experience, advice, copies of protocols, etc. You should also look for allies elsewhere at your institution, for example among the people in your institution's oral history program, who presumably get a different attitude from the IRB than what you're describing.]

RSS feed for comments on this post