By Jan Willem de Gee, Anne Urai & Tobias Donner.
This blog post expresses the opinion of the authors and has been agreed upon by the members of the lab. The post does not represent the official position of the University Medical Center Hamburg-Eppendorf (UKE), or of the Department of Neurophysiology & Pathophysiology, UKE.
Scientific authorship is a decisive factor for a career in academia. Yet, it seems to us that the criteria for authorship are not discussed in the scientific community as much as one might expect for a matter of such importance. Also, the practice seems to vary vastly between labs within our field (neuroscience) as well as between different scientific disciplines within the life sciences. The rules governing decisions about authorship often seem implicit to many involved. This especially holds for the more junior researchers (research assistants, Master students, and PhD students) before their first publication of a scientific paper (especially as lead author). More senior researchers obviously have more experience with the publication process, but this previous experience will also depend on the idiosyncrasies of their previous training. For instance, the level of engagement of the PI in the project, and the time it takes for the PI to engage in a manuscript draft, vary considerably from lab to lab.
The consequence is that people collaborate under implicit, sometimes misaligned assumptions. This can be counter-productive and frustrating, and can culminate in conflict between colleagues. What’s more: it is unnecessary. Recommendations for such criteria are available, and defining a code of conduct for a lab is straightforward.
So, we recently allotted a lab meeting to develop such a code for our lab (we should have done it a lot earlier, but better late than never…). We took the International Committee of Medical Journal Editors’ (ICMJE) recommendations as a starting point (see below). The Society for Neuroscience subscribes to these guidelines as well. We discussed these recommendations, collected input for further criteria, and then discussed some difficult example cases, followed by further discussion among the three authors of this blog post. The initial draft of the post was then shared among all lab members for comments, and we fixed our policy in the following lab meeting. The result of this process is what you are reading now.
The ICMJE guidelines
Here is a quote of the key part of this text:
“The ICMJE recommends that authorship be based on the following four criteria:
- Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND
- Drafting the work or revising it critically for important intellectual content; AND
- Final approval of the version to be published; AND
- Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
In addition to being accountable for the parts of the work he or she has done, an author should be able to identify which co-authors are responsible for specific other parts of the work. In addition, authors should have confidence in the integrity of the contributions of their co-authors.”
In addition, the ICMJE also lists examples of activities that alone do not qualify for authorship:
“Contributors who meet fewer than all 4 of the above criteria for authorship should not be listed as authors, but they should be acknowledged. Examples of activities that alone (without other contributions) do not qualify a contributor for authorship are acquisition of funding; general supervision of a research group or general administrative support; and writing assistance, technical editing, language editing, and proofreading.”
Some of the activities in the above list rarely occur in isolation. For example, it is rare (though not impossible) that a PI acquires a grant for the project leading to a paper, without having made substantial contributions to the conception or design of the work. Nonetheless, the likelihood of this occurring will increase with the volume of the grant and the freedom to deviate from the original work plan that is associated with that grant. In any case, we feel that PIs should not confine their role to the acquisition of funding but continue to keep in touch with the project during all phases, and should play a major part in the interpretation and write-up of the results of each project.
Our policy
We concluded to subscribe to the ICMJE guidelines as a gold standard for our lab, with the following amendments:
- Regarding the first criterion in the ICMJE list, the term “the work” is ambiguous. We make the additional distinction: A substantial contribution to the data analysis satisfies this criterion, as long as agreed upon beforehand with all members of the project team. This is irrespective of whether or not the analysis performed is reported in the final version of the paper. This may become relevant, for example, if the project team develops an improved version of an analysis down the line, which is then conducted by another lab member. All analyses conducted during the project are important for shaping the work reported in the final paper.
- Regarding the second criterion in the ICMJE list, we add that intellectual contributions to the paper do not always have to take the form of work on the manuscript text directly. Intellectual contributions can be just as valuable in the form of critical, in-depth discussion of the manuscript text.
- Engaging in project-related communication in a timely This includes responding to questions and issues arising during the conduct of the project, but also editing or commenting on circulated drafts of the manuscript.
The time it will take a co-author to respond to project-related queries will of course depend on the number of projects an individual is involved in, and other responsibilities that an individual may have at the time. But each author should feel obliged to respond to these queries (typically from the lead author) as soon as possible and to re-adjust their priorities. At the same time, it is the lead author’s responsibility to keep all authors informed about the status of the project and send a heads-up early in advance if the author expects to reach a major project milestone (i.e., first analysis, first manuscript draft) in the near future. That will enable the co-authors to allot time for this in their schedule.
Implementing the policy
Our hope is that making these guidelines explicit in the lab, and making them available to each of our external collaborators, will ensure appropriate credit assignment whenever a project culminates in a publication. Guideline dissemination should happen as early as possible, so as to give each prospective author the opportunity to fulfil each of the requirements. It will also give the lead author or PI the opportunity to ask for adjustments where necessary.
Specifically, we converged on the following procedure for our own lab:
- All lab members are expected to be aware of these guidelines; new lab members should be pointed to these guidelines by the PI upon their start.
- For each new project, prospective co-authors (i.e., collaborators) will be pointed to this blog post. This should happen at the start of the project, or (if they join the project later) by the time that they get on board. The responsibility for doing so lies with the lead author and PI. If the above was missed or ineffective, all collaborators are encouraged to alert the lead author and PI to this issue.
- Research assistants who help substantially with any of the items from point 1 in the ICMJEl list for a project led by someone else will be offered the option to become co-authors when they start to work on the project. If they accept, they are expected to contribute, to the best of their abilities, to all the remaining criteria from the list.
- Whenever individuals are uncertain about what is expected from them in the collaboration, or what they can expect from the other collaborators, they should talk about it to the lead author and the PI; there is no such thing as a stupid question in these matters.
- Whether or not a lab member is a paid employee or an unpaid M.Sc. student has no implications for decisions about authorship. All that matters is the actual scientific contribution.
- Whenever allowed by the journal, we will largely adhere to the CRediT Taxonomy to describe each author’s contributions in a detailed and specific way (e.g. “person X analysed the behavioural data and fitted the drift diffusion model, person Y analysed the MEG data, etc.”).
- We value collaboration with other labs – and labs are different, just as people are different. While harmonizing authorship practice will certainly benefit the scientific community, we do respect that different groups may use different policies. We feel strongly that the code we have here set up for our lab is the way to go, but we will not impose our rules on external collaborators. The lead author and/or PI should point external collaborators to our internal code early in the project.
- Practical implementation of fixed rules pertaining to authorship can be challenging. We think it is important to use the above guidelines exactly as such, not as absolute laws. While we will strive to adhere to them in each project running in our lab, we will allow ourselves the freedom to deviate from them in special cases (see below for concrete examples). When this is required, the deviation should be agreed upon with each co-author.
Some difficult cases
In what follows, we present a non-exhaustive list of realistic scenarios in which we can imagine some deviation from these guidelines to be necessary.
Author is added in a late stage of the project.
Consider a situation in which a control experiment or new data turn out to be important in order to address concerns emerging during the final stage of the analyses or during the revision. Suppose a lab member, who is not an author on the initial submission, has already acquired data that can be used for exactly this purpose. This situation occurs frequently in our lab, where different projects often build on, or are cross-linked with, one another. In this situation, that lab member may be asked to contribute the data and analyses, and consequently join the author list. Our top priority is making the science we report as compelling and insightful as possible.
Lead author leaves science or moves to another lab after a substantial body of the work has been completed, but before the final version of the paper is submitted.
Consider a situation in which a PhD student or postdoc leaves the lab soon after submitting a manuscript for publication to a journal. Even if that person would like to invest more time into pushing this through all the way to publication, this might not always be feasible in practice. In such a case, we may invite another lab member to carry the project through until publication, and join the author list. This needs to be coordinated among all authors early on, and it is the lead author’s and PI’s responsibility to do so.
Combination of multiple advanced techniques and approaches in a complex story for a single paper.
Consider a manuscript that features several different approaches, each of which are technically challenging, or one that includes data from different humans and animals. The different parts of such complex stories will often have been contributed by different lab members, perhaps even by team members from different labs. In this case we can imagine that not every author can take full responsibility for each aspect of the work. But again, the expectation would be that each author does so as much as possible.
Little correspondence with middle authors between initiation and write-up of a project.
In some projects where both the questions and technical issues are clearly defined, it can be most efficient if the lead author and the PI carry the project all the way through, instead of discussing each and every step along the way with all collaborators. In these cases, it is OK if middle authors make smaller contributions during the course of the analyses leading to the paper, and instead focus on their contribution during the interpretation and write-up phase.
Further reading:
- ICMJE’s authorships guidelines: http://www.icmje.org/recommendations/browse/roles-and-responsibilities/defining-the-role-of-authors-and-contributors.html
- CRediT Taxonomy:
http://openscholar.mit.edu/sites/default/files/dept/files/lpub28-2_151-155.pdf
Twitter authorship discussion (tweets selected by Anne Urai): https://storify.com/AnneEUrai/authorship-discussion