science communication

Conflict of Pinterest? Social media and conflicts of interest in scholarly peer-review

Posted on Updated on

peer review

In 1665, the concept of scholarly peer-review was introduced by Henry Oldenburg, the founding editor of the scientific journal Philosophical Transactions of the Royal Society (National Academy of Sciences, 2009) – a highly regarded journal that continues to publish high-quality science today. In general, the process of peer-review involves submitting one’s work to scrutiny by topical experts in a given field of research. Such a process is meant to serve a gatekeeping function, preventing the propagation of sub-par or incorrect scholarly information among research disciplines, thus ensuring that only accurate and truthful information populate scientific knowledge. Although it was introduced more than 350 years ago, scholarly peer-review remains a cornerstone of contemporary science and an important component of ensuring the validity and rigor of the scientific process.

While many factors play a role in ensuring ethical and responsible peer-review, and the process itself is far from perfect, avoiding ‘conflicts of interest’ is an important aspect of peer-review—that is, ensuring that reviewers provide honest and unbiased critiques on a piece of scholarly work. With respect to peer-review, conflicts of interest can present themselves in any situation where a reviewing researcher of a piece of scholarly work may be financially, professionally, or personally involved with the authoring researcher(s). Such intimate connections between reviewers and authors, including personal relationships (romantic or otherwise), can potentially result in a lack of critical judgement during peer-review and result in a breakdown of scientific integrity. Although scientists are typically diligent in avoiding conflicts of interest, novel and unfamiliar ways of making personal connections have the potential to create situations of conflict that scientists may misjudge and deem non-conflicting.

One relatively novel way of establishing contact is through online social media. While social media primarily serve a social purpose, many platforms have been utilized by scientists for outreach and communication purposes (among others; Van Noorden, 2014Collins et al., 2016). One platform that has been highly popular among the scientific community is Twitter. With Twitter, users can tailor their following to include other users with shared interests. In a world of online communication, however, people with common interests can often feel emotionally and personally connected with online contacts prior to meeting and engaging with them physically. For example, online dating has become a primary way of meeting potential partners, and contacts often feel an emotional connection prior to meeting. Furthermore, 64% of people maintain that common interests are the most important factor in driving online connection. As such, while the maintenance of close personal relationships requires physical contact (Dunbar, 2016), feelings of personal connection can be established from online communication with individuals sharing common interests. Thus, because scientists often share similar views and have become active on social media in recent years, it seems likely that virtual relationships (albeit non-romantic ones) through social media have the potential to create conflicts of interest, despite the many benefits of social media for researchers. This is particularly applicable to platforms where scientists can tailor their following to have a high proportion of followers with common research interests, such as Twitter.

This downside to social media and science became apparent when I was recently invited to review my 16th manuscript of 2017. I graciously declined to review the manusript for two reasons: 1. I didn’t really have the time to review the manuscript and felt I’ve put in my time so far this year, and 2. I felt that I may not have been able to give a fair and unbiased review of the manuscript because of a close connection with the primary author via Twitter, despite never meeting the author in person. Since declining the review, I’ve spent quite some time thinking about this idea and have honed in on a few key questions: how often does #ScienceTwitter get asked to review the work of close Twitter followers, how do potential reviewers respond in such situations, and does this have the potential to compromise the integrity of impartial peer-review? Preliminary results of an admittedly unscientific survey of #ScienceTwitter (with a low sample size, for now) can shed light on the former two questions, and suggests that it is not uncommon for scientists on social media to be asked to peer-review the work of close Twitter followers. Furthermore, when asked to review the work of close followers, #ScienceTwitter users overwhelmingly accept to review the work—indeed I have reviewed the work of Twitter followers in the past.

This, to me, seems inherently problematic yet not widely recognized, and further underscores the benefits and need for a fully-blinded peer-review system (or maybe a fully-open peer-review system). The question remains, however, as to whether this truly constitutes a conflict of interest and if it can indeed introduce bias into the peer-review process. I have some ideas for addressing the latter question (but will need the help of some progressive and enthusiastic journal editors) and plan to pursue the Twitter poll in a more formal fashion.

So, I want to know what you think – do you envision social media a source of conflict of interest in scholarly peer-review and how big of a problem do you think it might be? Let me know what you think in the comments section below!

References

Collins K, Shiffman D, Rock J (2016) How are scientists using social media in the workplace? PLoS One 11: e0162680. doi:10.1371/journal.pone.0162680

Dunbar RIM (2016) Do online social media cut through the constraints that limit the size of offline social networks? Royal Society Open Science 3: 150292. doi:10.1098/rsos.150292

National Academy of Sciences, Committee on Science, Engineering, and Public Policy (2009) On being a scientist: a guide to responsible conduct in research, third edition. The National Academies Press, Washington. 63 pp.

Van Noorden, R (2014) Online collaboration: scientists and the social network. Nature 512: 126-129. doi:10.1038/512126a

Advertisements

Common names suck; stop using them

Posted on Updated on

This afternoon I engaged in a Twitter conversation with some colleagues regarding the use of the term dreissenid in the context of “dreissenid mussels”. Colleague A wanted to know if dreissenid should be italicized. I assured her that it indeed does not, because Dreissenidae is a family of mussels containing 3 genera and is not a single genus (to which she obliged). Colleague B then questioned this and asked what to do if using the term when only referring to the genus Dreissena, whereby I suggested using a more specific term (i.e., Dreissena spp.). Colleague A then responded that she originally wanted to use the term to describe only the genus Dreissena, and that this was common practice. Then I got annoyed (again) at common names in general…

So which is it – does dreissenid refer to the family Dreissenidae or the genus Dreissena?

Figure 1. Total number of dreissenid mussel species when “dreissenid” refers to the family Dreissenidae (16) versus the genus Dreissena (7). Data obtained from MUSSELp (http://mussel-project.uwsp.edu/index.html).
Figure 1. Total number of dreissenid mussel species when “dreissenid” refers to the family Dreissenidae (16) versus the genus Dreissena (7). Data obtained from MUSSELp (http://mussel-project.uwsp.edu/index.html).

The answer is that it’s commonly used for both. Although many scientists may not care about or acknowledge this, the interchangeability of common names across different taxonomic resolutions can be problematic for a number of reasons.

Let’s first look at a relatively simple example. Say I published a paper on “dreissenid mussels” in the Journal of Crappy Nomenclature, and in the introduction made the claim that there are 16 species of dreissenid mussels. Without context, the reader has no idea as to whether there are 16 species within the family Dreissenidae or 16 species within the genus Dreissena unless they search this information themselves (there are 16 species in the family Dreissenidae; Figure 1).

Likewise, let’s say that in the same paper I was to claim that dreissenid mussels reside in supraterranean (above ground) freshwater systems. While that is true for the genus Dreissena, there exists a subterraneous genus of Dreissenidae (Congeria; resides solely in cave river systems). Again, without context, the reader would be left searching such information. Unfortunately, many readers would not recognize the need to search for this information and would likely apply the information obtained from the two statements outlined above in the context of how they interpret the term “dreissenid mussels”, which may be correct or incorrect depending on my definition of “dreissenid mussels”. Thus, in subsequent publications obtaining information from my hypothetical paper on dreissenid mussels, information may be incorrect, but nonetheless become “common knowledge”.

Figure 2. Extant Dreissenidae species of the genus A) Congeria (Congeria kusceri), and B) Dreissena (Dreissena polymorpha).
Figure 2. Extant Dreissenidae species of the genus A) Congeria (Congeria kusceri), and B) Dreissena (Dreissena polymorpha).

While the above examples may appear extreme, particularly for those who study these mussels, the points still stand – and for many more taxa than the example herein. Researchers conducting work on species new to them must learn as much about their new study species and related taxa as possible. In this way, using common names interchangeably across levels of taxonomic resolution can easily create problems for these researchers and the propagation of incorrect biological information may result. Furthermore, other problems with common names arise when even more generic terminology is used, like “cushion stars”.

Ultimately, there are two ways to solve the problems outlined above: either define the range of taxa (up front) that a common name being used encompasses, or stop using common names all together. If we are to follow the biological writing rules of Dr. Pechenik (i.e., more concise = better), scientific works would benefit from the elimination of common names (for example, “Dreissena spp.” consumes less space than “dreissenid mussels”, and the former would not require a formal definition). Not only does the use of precise taxonomic nomenclature reduce verbiage, but it would remove the potential for misinterpretation with respect to the breadth of biological processes across various levels of taxonomic resolution. That, and we would negate complex Twitter conversations regarding how to use common nomenclature and have more time to spend on writing our actual papers…

So, in conclusion, just stop using common names. They suck.