Part of the collage posted on the Biology Letters website that a correction note says provides evidence of the number of lionfish used in experiments in an Australian lab.

O. Lönnstedt et al., Biology Letters 10, 10.1098 (CC-BY 4.0)

Can you spot the duplicates? Critics say these photos of lionfish point to fraud

How many fish really appear in the photo collage above? The answer bears on whether a study about lionfish social behavior, published in Biology Letters in 2014, was fabricated—and whether Oona Lönnstedt, a marine biologist formerly at Uppsala University (UU) in Sweden who made up data in a 2016 Science paper, committed an earlier fraud. The case also raises fresh questions about whether senior scientists working with Lönnstedt, who was then a Ph.D. student, properly oversaw and took responsibility for her work.

Last year, Lönnstedt and her co-authors posted the collage on the Biology Letters website in what appeared to be an attempt to end questions about whether the scientists really caught enough fish to carry out their behavioral experiments. But critics say the colorful ensemble appears to include many photos of the same fish, and in some cases doctored duplicates of the same photo—which would undermine the authors’ defense.

The lionfish study was done in 2012, when Lönnstedt was a student at James Cook University (JCU) in Townsville, Australia. But the suspicions about it resemble those that discredited the 2016 Science study of the effects of microplastics on fish larvae. There, too, researchers questioned whether Lönnstedt had collected the claimed number of fish and wondered how she could have recorded reams of behavioral data without videotaping the experiments. In 2017, both UU and a national Swedish ethics panel confirmed the doubts about the Science paper, co-authored with UU biologist Peter Eklöv. It was retracted and Lönnstedt, who maintained her innocence, lost her job.

Lönnstedt, who has left research, did not respond to requests for comment sent by email and to her residence. Her co-authors on the Biology Letters paper, Maud Ferrari and Douglas Chivers, both of the University of Saskatchewan in Saskatoon, Canada, say they don’t know whether Lönnstedt used as many lionfish as the paper claims. But they dispute that the collage, which they say Lönnstedt produced, was meant to dispel doubts about the number. (Chivers was also an external co-supervisor of Lönnstedt’s Ph.D. research.)

The paper described a series of lab experiments showing that zebra lionfish (Dendrochirus zebra) flare their pectoral fins in a striking display to recruit other fish for a collective hunt. The authors speculated that fin movements could even convey information about the prey or its location, similar to the honey bee’s “waggle dance.” “These are highly complex animals with advanced social behaviors, and they are ridiculously good at catching prey,” Lönnstedt told National Geographic at the time.

A whistleblower who asked not to be identified because he’s worried it might hurt his career tells Science he first sent questions about the data to Biology Letters in May 2016, before the Science paper had come out. Once the Science paper was retracted, he voiced concerns that the Biology Letters paper, too, might be made up.

The experiments described by Lönnstedt would have required catching 86 zebra lionfish and 16 spotfin lionfish (Pterois antennata). At the Lizard Island Research Station on Australia’s Great Barrier Reef, where all three co-authors worked in 2012, researchers are required to record their catch. When the whistleblower checked an online record of fish collections, he found that Lönnstedt caught only 12 zebra lionfish and three spotfin lionfish during her 2012 trip.

Zebra lionfish and spotfin lionfish aren’t common around Lizard Island, says Anne Hoggett, one of the station’s two directors, who has worked there for decades and says she’s “very familiar” with the marine fauna. “If I was advising someone proposing a project that required such numbers, I would suggest that it would be extremely difficult to obtain them in a reasonable period of time.”

In February 2018, Biology Letters published an expression of concern about the paper and said it was carrying out an investigation. In November 2018, the three authors published a correction explaining that, in contrast to their earlier statement, they had reused some fish in the study, lowering the numbers required to 40 zebra lionfish and nine spotfin lionfish. To provide “evidence of the number of lionfish,” the correction said, the authors published the collage of 50 fish images in the supplementary material. (A PDF with the full collage is here.)

Two images from the collage seem to show two different fish (top). But when one is mirrored and enlarged, they are identical (bottom).

O. Lönnstedt et al., Biology Letters 10, 10.1098 (CC-BY 4.0), ADAPTED BY E.PETERSEN/SCIENCE

But now the supposedly exonerating data have come under suspicion as well. In a report he sent to Chivers earlier this year and that Science has seen, former JCU marine scientist Peter Ridd wrote that in two instances, the same photo had been used twice in the collage. In each case, one of the photos was mirrored or its contrast or colors appeared to be altered.

On top of that, Ridd wrote, many photos that weren’t identical appeared to be of the same fish. From metadata in the PDF file of the collage, he reconstructed a partial chronology of the original images. In several instances, photos taken consecutively—but placed far apart in the collage—showed very similar-looking fish in almost the same position. “It would be remarkable if one could get a second [different] fish into this precise position and take a photo in the very next shot from the camera,” Ridd wrote.

(JCU dismissed Ridd in 2018 for violating the university’s code of conduct after he repeatedly criticized research quality control at the Australian Research Council Centre of Excellence for Coral Reef Studies, which is based at JCU. In April, a judge ruled that his termination was unlawful, and earlier this month JCU was ordered to pay Ridd AU$1.2 million in damages. JCU has appealed.)

Science asked Elisabeth Bik, an independent image expert based in San Francisco, California, to study the collage. She, too, found two pairs of duplicate images. The fact that one-half of each pair was cropped or mirrored “might suggest that these are not honest error, but a potential intention to mislead,” she writes. (Later Bik confirmed a third duplication.) Bik also found a set of 12 zebra lionfish photos that, although not identical, “look so similar that one could suspect they were taken from the same animal,” and also three very similar looking spotfin lionfish.

Mike Rossner, a former managing editor of the Journal of Cell Biology who now runs Image Data Integrity, a San Francisco-based company, concurs that three pairs of images in the collage “do indeed appear to be duplicates derived from the same source images.”

The whistleblower says he brought these problems up with Biology Letters in January. A spokesperson for the Royal Society in London, the journal’s publisher, says the matter is still under investigation. “A summary of the findings has been passed to the authors for response and we cannot comment further at this stage,” he wrote in an email to Science.

Chivers and Ferrari don’t dispute that the same fish appear multiple times in the collage. But in an email to Science, they say Lönnstedt had produced the collage years earlier for a presentation, and it wasn’t meant to show the number of fish used in the study. The published correction note mistakenly presents the collage as evidence, they say, because of “an unfortunate error” during the editing process at Biology Letters, which they say the journal has acknowledged. They want the correction corrected.

In their email, the two scientists say they were at Lizard Island at the same time in 2012 as Lönnstedt and helped her design the study and the fish tanks she used, but did not see her actually conduct the experiments. “We had no reason to think anything was suspicious,” they write. Because lionfish are nocturnal, they believed the experiments took place “in the middle of the night,” when they were asleep. “We have no firsthand knowledge of the difficulty of catching fish” at Lizard Island, Chivers and Ferrari add. “We snorkel, but do not dive.”

Josefin Sundin, a former colleague of Lönnstedt at UU who blew the whistle on the Science study, says those answers remind her of the reaction of Eklöv, who said he had almost no involvement in the actual work for the Science paper or knowledge about the way it was carried out. “Chivers was Lönnstedt’s supervisor,” Sundin says. “He should have known what she was doing.” But Chivers says there is “nothing unusual about an advanced Ph.D. student, postdoctoral fellow, or collaborator working on their own without me or another senior researcher watching over their shoulder.”

After the Science scandal, JCU promised to do an investigation of the research Lönnstedt published while at the university—a total of 20 papers, 10 of them co-authored with Chivers, Ferrari, or both. But that investigation has yet to start, a JCU spokesperson says.

The whistleblower says the Biology Letters paper should have been retracted by now. He says the failure of journals and universities to take misconduct allegations seriously is “unethical.” For whistleblowers, he says, “It’s exhausting. And depressing.”