Lies, Mistakes & More: These Scientific Papers Got Nixed in 2017

By Christopher Wanjek

Lies, exaggerations, criminal acts, unbridled irony, alternative facts, fake news … No, we're not talking about 2017 politics. This is the 2017 world of science.

This past year, hundreds of scientific papers were retracted from professional journals. In the majority of cases involving these retractions, the reason was an innocent, yet sloppy, error in the methodology of the experiment that the authors themselves caught. But for quite a few papers, the retractions reflected scientific misconduct and a not-so-innocent attempt to tweak the data — or make it up entirely. What follows are five notable retractions from 2017, culled from the Retraction Watch blog.

Runners-up: May the farce be with you

So many retractions, so little time. There were many more retracted papers that almost made this 2017 "top five" list, such as several that attempted to "prove" a connection between vaccines and autism. One, titled "Systematic Assessment of Research on Autism Spectrum Disorder and Mercury Reveals Conflicts of Interest and the Need for Transparency in Autism Research," wins for irony: The authors didn't reveal the fact that they were associated with organizations involved in demonstrating a vaccine-autism connection. 

Elsewhere, to demonstrate that some journals will publish anything, the blogger Neuroskeptic managed to get four journals to accept a clearly fictitious study, authored by Lucas McGeorge and Annette Kin about "midi-chlorians," the intelligent entities that give Jedi their powers in "Star Wars." And then there was "The art of writing a scientific article," which was published in the Journal of Science Communications and cited almost 400 times. The citations are real; the paper and the journal (with an "s" on Communications) don't exist.

5. I purr, therefore I am

It took 35 years, but Bruce Le Catt was finally called out for the feline he was. Le Catt, being a cat, wrote a rather catty critique of an article written by David Lewis and published in the Australasian Journal of Philosophy. Lewis, who died in 2001, was an American-born philosopher best known for his concept of modal realism, a view that all possible worlds are as real as the actual world. Perhaps there are worlds, for example, in which cats can write … that is, write intelligibly … OK, write intelligibly in words that people other than the cat's owner can understand. Maybe such a world existed in Lewis' mind because, it seems, he was Le Catt, writing a critique of himself. (Philosophers are a fun bunch.)

The 35-year-old ruse — that would be 100 plus in cat years — was an inside joke that was known to a few philosophers of Lewis' generation, including Michael Dougherty of Ohio Dominican University in Columbus, Ohio. Dougherty, who is currently writing a book about scientific integrity, asked the journal to let people know that Le Catt was a pseudonym for Lewis, so that — if nothing else — the younger generation of philosophers would know that Lewis was critiquing himself.

4. Faked to the third degree

If a paper with fake authors and fake funders is published through a fake peer-review process, would it still be fake, or would all of the fakes cancel out? Seems like a philosophical question best handled by Bruce Le Catt (see above). Here are the facts as best they are known: In 2015, a group of Chinese scientists published an article in the Journal of Molecular Neuroscience titled "Nucleolin Promotes TGF-β Signaling Initiation via TGF-β Receptor I in Glioblastoma." (Don't worry so much about what the title means because, as mentioned, there's not much truth associated with this study.)

In June 2017, the journal retracted the article because the funding source stated in the paper wasn't the funding source; one of the co-authors confirmed he wasn't involved in the research or writing of the paper and knew nothing of the study; the senior writer confirmed he was not involved in the submission process and did not support its publication; and, as the editors wrote in their retraction, there is "strong reason to believe that the peer review process was compromised."

Regarding that last point: This paper is one of more than 100 articles retracted in 2017 by Springer, the German-based publishing company that publishes Molecular Neuroscience and nearly 3,000 other scientific journals. Springer has been investigating fraudulent peer review, where the authors themselves or paid consultants provide the glowing review. Since 2012, more than 500 papers have been retracted because of a faked peer review, the vast majority of which has been from China, according to Retraction Watch.

3. If only the data were as solid as bone

Japanese researcher Yoshihiro Sato, who died in January 2017, was a respected scientist who published his work in such prestigious journals as Neurology, Bone and JAMA. But now, it seems, editors everywhere have a bone to pick with him. As of December 2017, 23 of Sato's papers have been retracted because of falsified data, questions about authorship or plagiarism.

Sato investigated therapies to reduce hip fractures, and his studies seemed to indicate that vitamin D and various genetic drugs worked wonders in frail, older patients who had had a stroke or who had Parkinson's disease or dementia. But the findings were a little too good to be true. A 2016 statistical analysis of Sato's studies, led by Mark Bolland of the University of Auckland in New Zealand, raised doubts about the validity of the results. Sato admitted to cooking the data; he also confessed that — as an honorary gesture — he had added co-authors who hadn't participated in those studies. Since then, JAMA and other journals have issued warnings to readers, asking that they not be swayed by Sato's body of research, which dates back to the 1990s. Many more retractions will likely come in 2018.

2. Fishy retraction a blow to environmentalists

Many wanted it to be true.

In June 2016, two researchers from Uppsala University in Sweden published an alarming study in the prestigious journal Science, stating that European perch larvae prefer to eat tiny beads of polystyrene rather than natural food. Ingesting these plastic beads, which are barely visible to the human eye, slows a fish's growth and makes it more likely that it will be eaten by predators, who then have the plastic inside them, the researchers said. The news media ingested the artificial tidbit, too, as the study was reported widely. Many environmentalists quickly latched on to the study as proof of the harm plastic pollution is causing.

But many scientists just as quickly challenged the study, with some wondering if the study actually had been conducted at all. By December 2016, Science stated that the study was under investigation. The researchers couldn't produce the full data; they claimed the data were lost when their laptop was stolen soon after the paper was published. After a deep dive, Sweden's Central Ethical Review Board (CEPN) determined that the researchers had been scientifically dishonest and couldn't have conducted a study that was thorough enough to produce the data they claimed they had. Science retracted the paper in May. That Science even accepted the paper is "remarkable," CEPN stated in its review.

1. Mindless eating or mindless science?

Any way you slice it, 2017 was a bad year for Brian Wansink, director of the Food & Brand Lab at Cornell University and author of the popular book "Mindless Eating: Why We Eat More Than We Think." Wansink has published influential studies, now questioned, that purported that children will choose healthy food, such as an apple, over a cookie if the apple has an Elmo sticker on it. But Wansink's troubles started in November 2016 when, in a blog post, he offered one of his graduate students some odd advice. He told her that, when faced with null results (meaning that the data doesn't support the hypothesis), why not salvage the data and use it for a different study. 

The student ultimately ended up publishing five papers, all of which were about people eating pizza at an all-you-can-eat Italian restaurant buffet.

The blog post, now deleted, raised concerns among many scientists about the quality and integrity of Wansink's own research. And so they investigated, and found a multitude of problems in Wansink's methodology and statistical analysis that went back for years. Cornell University investigated Wansink's research, as well, and found what it called "mistakes," but not misconduct. More than 50 of Wansink's papers are facing close scrutiny, and in the past year, Wansink has corrected and republished at least eight and has retracted four articles, including the one in JAMA Pediatrics about Elmo and apples. That's the way the cookie crumbles.