The STAP story: Is widespread distrust the way forward?

As we wait for the final verdict on the STAP case, one thing is certain: when it comes to life outside the lab, the principles of scientific thinking become hazy. The very foundation of science, to collect evidence and build upon sound reason is blurred in the current environment of publication retractions and media frenzy with the very mention of an unsuccessful attempt in replicating a study. While cases of misconduct such as that of Hwang Woo-suk are responsible for creating this hype, a knee-jerk reaction of an accusatory tone in response to any non-reproducible study seems unwise as the way forward. Alas, according to ACS Nano, "the number of blogs, twitter messages, etc. in which individuals accuse others of academic fraud is steadily rising in science.”

We understand the importance of concepts like the classic fallacy of composition: one cannot infer that something is true/false of the whole from the fact that it is true/false of some part of the whole. Yet, some forget its practical application. In the STAP case, the image is claimed to be an error. Vacanti reasons that one unintended error does not nullify the sound science on which this study is based. He states that the image error does “not affect the overall content of the published reports, the scientific data or the conclusions." It would be a big mistake to dismiss any study underpinned by solid science. Skepticism is an integral feature of scientific thinking, but it must be guided by strong logic and of course, a bit of patience. Brash conclusions lead to more bad than good. Even if accusations turn out to be unfounded, the accused has to live with his/her name forever attached to the incident. 

The Knoepfler Lab Stem Cell Blog is running STAP polls on their website, much in the manner of primetime television. This does nothing but stoke popular sentiment, which may distract away from the path to discovering the real issues. In the media, rather, discussions about the nature of replication need to be at the forefront. We know its merits, but we don’t seem to be on the same page in terms of the challenges of reproducing a study and when to treat a claim as fabrication versus a problem with replication. For example, Riken’s scientists in their technical tips say, “Despite its seeming simplicity, this procedure requires special care in cell handling and culture conditions, as well as in the choice of the starting cell population” and that it is “absolutely reproducible.” Concerning results that can't be replicated, Dr. Bruce Alberts, editor of Science, explains that the increasing involvedness of experiments is a factor. He says, "It has to do with the complexity of biology and the fact that methods [used in labs] are getting more sophisticated."

The complex nature of today's scientific experiments require a more comprehensive speculation and discussion concerning reproducibility. Such an approach may need to be interjected into the mainstream community as much as the implications of misconduct are, for a balanced view. A balanced view may shed light on underlying issues. Indeed, some issues are driven by competition and may lead to intentional misconduct in some cases such as the lack of double-blind experiments, which are rarely enforced in preclinical settings. Without these double-blind conditions, there is a risk that researchers may selectively publish positive results driven by a desire to get high-impact publications. In other cases, raw data is unavailable to rival scientists and even when it is, the supplied data is not detailed enough.

However, aside from these intended acts, other possible factors also need to be weighed with each case. Researchers replicating the study need to thoroughly understand of the nuances of the methods, and their responsibility in meticulously conducting such a study and proactively working to succeed with the attempt. We also have to consider that differences in labs, equipment, and materials may lead to different results. Lastly, there is the possibility that the more variables in an experiment, the more likely small errors will pile up and sway conclusions. Paying attention equally to such unintended factors affecting reproducibility will not only prevent wrongly defaming someone’s character but also allow for a fuller understanding of the dynamics of reproducibility. It is this insight that will advance research behavior.

Overall, if we are going to advance as a community during a time of dramatic shifts in scientific publishing, we are going to need a more robust interpretation regarding the concept of replication. Most importantly, for the interim, the community must keep their faith in the self-correcting nature of science while remaining grounded in solid scientific thinking – where nobility and sound judgment lead the way to truth.