NEWS

‘Avalanche’ of spider-paper retractions shakes behavioural-ecology community

Allegations of fabricated data have prompted a university investigation and some soul-searching.

Search for this author in:

Three large white spiders sitting of strands of web strung horizontally across the tops of some leaves.

Studies of social spiders such as Stegodyphus dumicola (pictured) have come under close scrutiny.Credit: Bernard Dupont (CC BY-SA 2.0)

A complex web is unravelling in the field of spider research. On 5 February, McMaster University in Hamilton, Canada, confirmed that it was investigating allegations that behavioural ecologist Jonathan Pruitt fabricated data in at least 17 papers on which he was a co-author.

Since concerns about his work became public in late January, scientists have rushed to uncover the extent of questionable data in Pruitt’s studies. Publishers are now trying to keep up with requests for retractions and investigations. According to a publicly available spreadsheet maintained by Daniel Bolnick, an evolutionary biologist at the University of Connecticut in Storrs, seven papers have been retracted or are in the process of being retracted; five further retractions have been requested by Pruitt’s co-authors; and researchers have flagged at least five more studies as containing possible data anomalies.

Pruitt, who is reportedly doing field research in Australia and the South Pacific, told Science last week that he had not fabricated or manipulated data in any way. He did not respond to multiple requests from Nature for comment on the mounting list of retractions, or the accusation that he fabricated data.

His research looks at how different personalities form within communities of social spider species that live in groups, and it has implications for emerging ideas on how animal behaviours evolve in the context of their environment.

Many who are close to Pruitt are stunned. Noa Pinter-Wollman, a behavioural ecologist at the University of California, Los Angeles, who has co-written 20 papers with Pruitt, says that the process has been “emotionally devastating”. Pruitt collected the data for five of their co-authored papers. “It was hard to lose trust in someone you’ve worked so closely with and trusted for so long,” she says.

A tangled web

The retractions started in mid-January, when authors of a paper in The American Naturalist1 pulled it, citing “irregularities in the raw data”. These were data that Pruitt had provided, showing how long it takes social spiders to resume typical behaviours after a disturbance, such as a simulated attack from a predator.

After a second retraction2, in Proceedings of the Royal Society B, Kate Laskowski, a behavioural ecologist at the University of California, Davis, who had co-authored both studies with Pruitt, wrote a blogpost about her role in the retractions. She had found multiple stretches of data that had been copied and pasted to represent findings for multiple spiders. When Pruitt’s explanations failed to account for the anomalies, she requested that the journals retract the papers, reportedly with Pruitt’s consent.

“Then, hell broke loose,” says Niels Dingemanse, a behavioural ecologist at Ludwig Maximilian University in Munich, Germany, who has helped to uncover the data issues.

More than 20 scientists — co-authors, peers and other interested observers in the field — mobilized to pore through the data in almost 150 papers on which Pruitt is a co-author, looking for evidence of manipulated or fabricated numbers. They found similar signs of copy-and-paste duplications. In at least one instance, researchers identified formulae inserted into a published excel file, designed to add or subtract from a pasted value and create new data points.

Several have stated publicly and privately that they believe this to be clear evidence of fraud. Dingemanse says that his mind was made up by the “avalanche of retractions” in progress, as well as the mounting piles of irregular data. “It is hard to believe these data are not fabricated,” he says.

Rethinking behaviour

The full extent of any data fraud and its impact on the field is hard to quantify. Pruitt had written “a lot of really impressive papers” and was regarded by many as a “rising star”, says María Rebolleda-Gómez, a microbial ecologist at Yale University in New Haven, Connecticut. In 2018, Pruitt moved to McMaster University from the University of California, Santa Barbara, as part of the prestigious Canada 150 Research Chair programme.

He has published several high-profile studies, such as one in Proceedings of the National Academy of Sciences (PNAS)3 and one in Nature4 that addresses a contentious debate about group selection in evolution. Researchers have questioned the data in both. A retraction has been requested by Pruitt’s co-authors on the PNAS paper, with Pruitt’s approval; a representative of Nature says that the journal has been notified about the questionable data and is looking into it. (Nature’s news team is editorially independent of its journal team.)

The 17 papers that include questionable data from Pruitt have been cited more than 900 times, and it will take scientists a while to sort out which ideas have been supported elsewhere in the literature and which will need to be retested. “My guess is the impact will probably be pretty big,” Laskowski says.

A few scientists online have latched on to the scandal as proof that the entire field of animal-personality research is suspect. But Bolnick pushes back against this idea, saying that the actions of one person “shouldn’t be used to generalize about a whole discipline”.

A field rallies

A spokesperson for McMaster University confirmed that the institution was investigating, but would provide no further comment on issues of research integrity. The University of California, Santa Barbara, where Pruitt did most of the work in question, declined to comment on the specific case but said that it “would cooperate with any other institution conducting an investigation”.

After initially agreeing to answer Nature’s questions by e-mail, Pruitt has not responded to multiple requests for comment. Last week, he told Science that the issues being raised were attributable to mistakes in data management.

Many of the researchers who have spent their time digging into his data find this explanation inadequate, however. Given the difficulties that the situation has caused for his students and collaborators, Bolnick says, it was a “notably insufficient response”. As editor-in-chief of The American Naturalist, the first journal to issue a retraction, Bolnick has become the de facto point person for tracking problems in Pruitt’s papers. He estimates that the majority of his time over the past week has been spent fielding e-mails and calls about the matter.

Laskowski and Pinter-Wollman say that although the wave of retractions deals a blow to behavioural ecology, they are heartened by how quickly the community has acted to set the scientific record straight. They acknowledge that researchers have lessons to learn about making data publicly available — by one estimate, more than 60% of Pruitt’s data-containing papers are in journals with no data-sharing requirements — and about carefully checking data that they receive from colleagues. But they are both optimistic that these lessons will ultimately strengthen the field.

“Despite all I’ve gone through, I’m going to continue to trust collaborators,” says Pinter-Wollman, emphasizing that science improves through the sharing of expertise and experience. “It’s hard to grow in a vacuum.”

Nature 578, 199-200 (2020)

doi: 10.1038/d41586-020-00287-y

Updates & Corrections

  • Clarification 13 February 2020: Comments from Noa Pinter-Wollman have been given added context to better reflect her sentiments

References

  1. 1.

    Laskowski, K. L., Montiglio, P.-O. & Pruitt, J. N. Am. Nat. 187, 776–785 (2016).

  2. 2.

    Laskowski, K. L. & Pruitt J. N. Proc. R. Soc. B 281, 20133166 (2014).

  3. 3.

    Wright, C. M., Holbrook, C. T. & Pruitt, J. N. Proc. Natl Acad. Sci. USA 111, 9533–9537 (2014).

  4. 4.

    Pruitt, J. N. & Goodnight, C. J. Nature 514, 359–362 (2014).

Download references

Nature Briefing

An essential round-up of science news, opinion and analysis, delivered to your inbox every weekday.