Commentary / Japan

Internet raises misconduct risk

by Takamitsu Sawa

The ever growing use of computers and the Internet has drastically changed the ways in which academic and scientific research projects are conducted.

The Internet has made it much easier to collect information related to a given research theme. In preparing theses, researchers can rely on computers to process images and sentences. So, they now spend most of their time with their eyes glued to the computer screens.

Although computers have contributed greatly to making research work far more efficient than in the past, there is no denying that they have saddled researchers with a potential hotbed for research misconduct. Suspicions about misconduct swirl about online.

On April 1, an investigation committee of Riken, a government-backed comprehensive research institute, decided that manipulation and fabrication of images had been detected in one of two papers on a new procedure for producing stem cells known as STAP (stimulus-triggered acquisition of pluripotency) cells, published in the scientific journal Nature, and that the chief author of the two papers had committed “research misconduct.”

The same committee also stated that the first 20 pages of her 120-page doctoral dissertation were copied from the website of the U.S. National Institutes of Health (NIH) and that some of the cited references in the paper were copied from other papers. All of these acts of misconduct were presumably due to the use of computers in the course of writing papers and the use of the Internet in searching related literature.

There must be a large number of “copy and paste” cases in writing papers regardless of the discipline. Most researchers must have been tempted, at least a few times, to copy sentences or images from other papers that appear on the Internet while they were writing their papers. But very few researchers actually resort to copy-and-paste presentation of data because of ethical standards that prohibit such research misconduct and because of all the developed software on the market capable of unmasking plagiarism.

It isn’t difficult for authors to change images to suit their purposes by using computers. One professor told me that he has been plagued by copy-and-paste instances involving students writing their master’s theses. Resorting to this tactic for the most crucial part of a paper cannot be tolerated.

As for the doctoral dissertation by the chief author of the two STAP papers, some allowances may be in order as she copied certain statements from the NIH public website simply to explain stem cells and to compensate for her inability to express her thoughts in clear English.

In other words, what she did was “borrow,” but not necessarily “steal,” some statements from the website. Nevertheless, there is no denying that her action constituted plagiarism and, therefore, misconduct.

According to the aforementioned professor, copy-and-paste tricks are used most extensively by students studying in Japan from other countries. If those students were given the benefit of the doubt, one could say they are supplementing their insufficient command of the Japanese language.

A large majority of academic publications use a “referee” system of rigorously selected experts who are asked to review each contributed paper to verify trustworthiness, novelty and originality.

On Jan. 29, many newspapers carried articles saying that when a paper on STAP cells was first submitted to the publishers of Nature in February 2013, it was rejected after one referee scathingly criticized it for “ridiculing a hundred years of history of cell biology.”

Subsequently the paper was revised with the addition of two world-famous stem cell experts from Riken as coauthors, and the revised versions were approved for publication.

While referees are kept anonymous, there are two methods for evaluating papers submitted for publication: the “single blind system” where authors’ names are made known to referees and the “double blind system” in which referees don’t know who the authors are.

The journal Nature does not follow the latter system at present, and referees spend only seven to 10 days to review each paper, according to a lecture given Dec. 7, 2012, by Dr. Magdalena Skipper, senior editor of Nature, at the Okinawa Institute of Science and Technology. (The details can be found the institute’s website.)

From my experiences, which are limited to research in the field of economics, it appears highly likely that referees preconceive their opinions by simply looking at the authors’ names.

While I have no way of knowing who coauthored the STAP cell paper submitted in February 2013, those who wrote the revised versions included two well-known researchers of regenerative medicine affiliated with Riken.

It is said that the trustworthiness of the STAP cell theses is not to be verified, even by stem cell experts, until the duplicability of the research is confirmed. Since the referees are only human and not infallible, their preconception based on the list of authors might well have played a crucial role in determining whether the two revised papers on STAP cell discovery were worthy of publication.

It is very possible that the referee who denounced the original STAP cell paper for ridiculing cell biology history also served as one of the referees for the revised versions. The addition of the two authorities on stem cell research to the list of the authors must have led that referee to change his or her evaluation of the papers to one of approval.

If it is shown that either STAP cells did not exist or that the research is not duplicable, we could presume that the referee who denounced the February 2013 paper was intuitively correct in calling the paper’s claims “impossible.”

In just about every field of research, it has been found that an expert reviewer’s conclusion that what is said in a paper is “impossible” has an 80 to 90 percent chance of being correct. This means that an expert’s intuition is worthy of trust.

Yet, a scientist tends to be bound by traditional paradigms, making it highly likely that a paper that could actuate a paradigm shift will be rejected.

Even Albert Einstein rebuffed the uncertainty principle of quantum mechanics — saying he could not imagine God rolling dice — when he wrote a letter in 1926 to Max Born, one of the early figures in quantum mechanics.

This shows that, like many other scientists, Einstein was unable, at that point, to free himself from the deterministic paradigm of classical mechanics.

Generally speaking, those who serve as referees must keep in mind that nothing is impossible. Because a traditional paradigm has a tendency to bind the minds of scientists, it is the responsibility of editors of academic and scientific publications to do all they can to prevent consigning to oblivion those papers that could bring about a paradigm shift.

Therefore, editors must possess an intuitive power that surpasses all others.

Takamitsu Sawa is president of Shiga University.