Monday, April 19, 2021

Is biological adaptation and publishing papers the same thing?

 

Guest post from Pim Edelaar:

Publish or perish ... You have probably heard this “advice” before. And while there are many alternative ways to be a valuable and contributing academic, if you do research then publishing your science is indeed an important responsibility. And at the same time a daunting task. There are many books and postings about how to write a good paper (e.g. Andrew´s entry “How to write/present science: BABY-WEREWOLF-SILVER BULLET”: https://ecoevoevoeco.blogspot.com/2014/10/how-to-writepresent-science-baby.html), but here I wanted to explore whether there are underutilised strategies to get your paper published where you want to get it published. I will intersperse this with some parallel observations on organisms that face a similar ordeal but then in the eco-evolutionary arena, and to what extent we, as researchers, may have some blind spots with respect to how organisms solve this.



Two years ago Dan Bolnick and I published a paper in Trends in Ecology and Evolution (Edelaar & Bolnick 2019: https://pimedelaar.org/publications/) where we presented a classification of the processes that can increase the match between an organism and its environment. Because the better that match, the greater its ecological performance and the higher its expected fitness. And in order to explain how to apply this classification to real life, in talks I sometimes use human examples, and ask the audience for strategies that can result in the desired match. Like the publishing of a paper in a good journal. This typically results in a long list of possibilities, and it turns out that these can always be nicely categorised and understood using our classification. Interestingly enough, the audience often does not mention some additional strategies that are also viable. So we proposed that the classification of our paper can help us to focus our attention on such under-recognised strategies, not only in publishing but in biology in general.

Our classification is composed of four processes (the paper goes into the derivation and justification of these, but I will not repeat that here). The first one is “good old” natural selection. In terms of publishing, that would perhaps be equivalent to sending a randomly generated paper to a random journal, and hope for the best. If you use this strategy, then keep reading because I think you can do better. (Although some predatory publish-for-pay journals or conferences might accept virtually anything, even near-random papers: https://pdos.csail.mit.edu/archive/scigen/). For the other three processes the organism (so you as writer!) takes an active role, and either changes its phenotype (your paper) or the environment (the journal). This second option may seem a bit odd at first sight, but if what is important is the match between a phenotype and its environment to perform well (the match between a paper and the journal to be published), then it really does not matter which of the two interacting components is changed in order to improve this match. So changing the environment is a perfectly valid, and sometimes under-recognised, strategy to achieve one’s goals. 

Still, let’s first look at the thing we tend to spend most time on, the development of our manuscript. This is what we have been taught to do as researchers, and what we could perhaps compare to the development of an organism, resulting in a phenotype. What strategies do we have to make the best paper? (Important note here: I am talking about the best paper in the eyes of the journal, since we are only focussing here on getting it published. Whether it in fact is a good paper depends on what readers think. And these views don´t always align, so we may need to take this into consideration and reach some compromise. Unfortunately this is even more relevant for research proposals: you need to convince the funders, virtually the only people who will ever read your proposal.) 

So, how do we nicely “develop” a good paper? We can look for an interesting or unresolved topic, design clever or ambitious experiments, use cutting-edge methods or materials, collect or analyse valuable data, make appealing figures, write a useful or provocative discussion, write clearly and engaging (see Andrew’s really useful entry again, if you didn´t do this before: https://ecoevoevoeco.blogspot.com/2014/10/how-to-writepresent-science-baby.html). To do this better, read and think, practice and edit, get feedback from other people, collaborate with other people, etcetera. 



 My technician Fatima working with our optogenetic fruit flies – cutting-edge materials, but not that you can see that ...

Those are the usual strategies. But if our only goal is to get a paper published, then alternative strategies do come to mind, some questionable and some outright rejectable. For example, adding names of famous people to your list of co-authors, hoping for better reviews or editorial preferences. Double blind reviewing (http://robertfeldt.net/advice/double_blind_reviewing/) hopefully puts the scope for this dubious practice to rest. Or selective citation, where you misrepresent what has been written before, or you ignore citations that counter your message (one of the hardest things to spot as a reviewer). Or only presenting the statistical analyses that resulted in significant results, while ignoring equally valid approaches or models or outlying data (https://en.wikipedia.org/wiki/Researcher_degrees_of_freedom). Or even fabricating data. Of course I am not trying to give you ideas of how to get your paper published using strategies that are not accepted by the scientific community, I simply want to show that conceptually they align with the strategies for producing a good paper – it is all about producing the product that matches best with what the journal is looking for. If we think about the development of a biological organism, then indeed there is no ethics and fairness involved and the most unpleasant appearances, weapons, toxins and behaviours might appear, as long as they get the job done. 



 

Assuming you stick with accepted practices (highly recommended, for the sake of good science and yourself), then one common and useful advice here is that it helps to write the paper with a specific journal or at least a specific audience in mind, since this will increase the probability that your paper will be a good match with your target journal or a similar one. So you do all of that ... and then comes that dreaded rejection. Now what? After we have dealt with this mentally (see Andrew’s entry on this: https://ecoevoevoeco.blogspot.com/2016/02/dealing-with-rejection.html), we look for another journal, and resubmit our manuscript. But before we do that, it of course would be wise to adapt our manuscript in order to improve the match. Whereas the previous section was more about the innate development of the first submission of the paper, this response of producing a changed submission is comparable to adaptive phenotypic plasticity, where the environment (the journals) rule, and the organism (the manuscript) responds. 

What strategies do we have to adapt our paper best to the next journal? Apart from adhering to journal requirements (oh, the joy of reformatting citations ...), it is often a good thing to take reviewer comments seriously, and improve your paper where you can. Never resubmit the same manuscript unchanged to another journal! The chances are that (1) you really could have made a better product (previously rejected papers are more cited: https://www.sciencemag.org/news/2012/10/scientists-may-feel-rejected-rejection-rare-publication-study-finds), and (2) one of the reviewers for your new submission is the same as for your old submission, setting yourself up for a nice rejection again. 

Now maybe somebody noticed, above I already mentioned a strategy that is actually not about the phenotype, but about the environment: “write the paper with a specific journal in mind” and “we look for another journal”. And of course you would be wise to carefully select your journal to improve your chances of acceptance! Again, if the challenge is to obtain the best match between your paper and the journal, you should not ignore this other side of the interaction! So check which journals might be interested in publishing your paper. Going through your references might give you some ideas, and if you have read many papers or browsed many journal issue indices, you might have a feel for what different journals tend to publish (both in terms of topics and in terms of quality). Advisors and colleagues can also help with identifying a good target journal. Journal editors and potential reviewers may hate me for this comment, but if you have time to spare, you may aim a bit high in terms of journal quality in the beginning, if you think your paper has a realistic chance of getting accepted there. But be warned that “going down the ladder” can take a lot of time, energy and perseverance (or what some call stubbornness, a waste of people’s time, or stupidity). I think my personal record is with Edelaar et al. 2019 (https://pimedelaar.org/publications/), which was submitted 9 times (Nature, Ecology Letters, Science, Nature Ecology & Evolution, Nature Communications, eLife, Nature Ecology & Evolution again, Evolution, and finally Proceedings of the Royal Society B), because I was (am) convinced it is a damn good paper that deserved (and got) a good place. But that took over 3 years, and a lot of reviewer effort (so let’s hope it gets read and cited). If you follow this strategy a lot (and I think most people do to some extent?), it would be nice if you also reviewed your fair share of the manuscripts seeking their spot in the light (read: you really should ... See “The Golden Rule of Reviewing”: https://www.journals.uchicago.edu/doi/full/10.1086/598847). (You can show your contributions to science as a reviewer on Publons, e.g. see my page: https://publons.com/researcher/1175333/pim-edelaar/metrics/)

Not ignoring the environment as a target is also true for organisms: the phenotype-environment match can be improved by tweaking both components, simultaneously or independently. Indeed, there is overwhelming evidence that organisms select aspects of their environment in order to increase their expected fitness: they select habitats, prey types, sexual partners, social partners, and so on. While selecting what to eat and with whom to mate might not appear to be comparable (and should not be confused – https://www.youtube.com/watch?v=ycuzVQmVj44), it all has to do with what we called in our TREE paper’s classification “selection of the environment”, in the sense of selecting the environment with which the phenotype then has to interact. And the outcome of this choice might depend on how the phenotype is, or in our case, how our paper is. So choose wisely ...



Choose wisely ...

There is one last process that organisms might employ to improve the environment they have to interact with, and we called that “adjustment of the environment”, since it does not involve the choice out of a set of available alternatives (as for selection of the environment), but the changing of the characteristics of the single option they work with. Together with selection of the environment, it is often not recognised as a distinct process of adaptation, and commonly treated as “just a behaviour”. Narrow-sense niche construction falls into this domain (making physical constructions like nests and burrows for example), but also somehow changing the characteristics of the organisms an individual might interact with: changing the perception and behaviour of a competitor or a (potential) sexual or social partner, changing the behaviour of an infected host, those kinds of things. Humans are masters in this strategy, and if you look around and compare what you see with what the undisturbed landscape and biological community would look like if there were no humans, then I hope it is clear what I mean. We adjust our environments to how we are and what we need to perform well (or at least, we think so).

Paradoxically enough, I think this is one of the more under-recognised “strategies” in publishing, where the aim is to change the journal such that its fits our paper. It is hardly ever suggested when I ask people for publication strategies in talks, so not only in biology but even with respect to our own lives it appears to be a bit of a blind spot. But the communication with the editor and reviewers falls under this umbrella: writing an appealing and convincing cover letter, responding well to editorial and reviewer comments (that does not mean doing everything they ask for, but at least politely explain why some changes have not been implemented), and perhaps even talking to an editor at a conference or via a pre-submission inquiry. One of the comments editors keep making in public, is that they want to work with authors to get the best paper possible, but I think this interaction also involves (perhaps unconsciously) movement (understanding) from the editor’s part on what the paper is about. Alternatively, the writer takes up the role of the journal: editing a special issue usually means that you can publish one or more of your own papers in that issue. I don´t think this is the main motivation for people editing special issues, but I do think people see that as a nice side effect for their hard work as editors, and it may swing the balance towards undertaking this endeavour. More questionable practices also fall under adjustment of the environment: proposing favourable colleagues as reviewers (how many people regularly suggest their worst critics?), excessive citing of papers written by the handling editor and reviewers, or even blunter types of bribing (for which I don´t know examples, but I am pretty sure it has been done or attempted) or threatening and bullying of editors or reviewers. (Please don´t do these things.) And again, for biological organisms nothing is out of bounds, so the most intricate ways to manipulate their abiotic and biotic environments have evolved. Like a parasite that manipulates an infected ant to swell up and change colour and to perch motionless at the top of a plant, such that it might get eaten by a bird that mistakes it for a juicy ripe berry – all of that so the parasite gets into the bird, its vector for dispersal (https://www.nationalgeographic.com/animals/article/parasite-transforms-ants-into-berries).



A normal and an infected ant

Where does this exercise of revising our goal of publishing our paper in a desired journal using the classification framework leave us? Perhaps you picked up an extra suggestion on how you can increase the chances of getting your paper published – in an honest way. Perhaps you know of a strategy that I missed, and it would be great if you could write these in the comments. My prediction is that it can always be classified as one of these three processes: somehow changing the paper to match with the journal, some sort of selective process regarding the journal to match with the paper, or some sort of adjustment of the journal to match with the paper. (What do you think, is this still true in a new world with Open Access, public repositories, Peer Community In (PCI) journals, self-publishing, etc.?). The same is true if we consider the phenotype-environment match in biology: I think that anything adaptive that individuals do can be classified by these three processes (changing the phenotype, selecting the environment, adjusting the environment). Likewise, I think that this classification can help us to recognise specific behaviours and strategies for what they are, and how they compare with other seemingly unrelated strategies (e.g. choosing what to have for breakfast, and with whom to have it, involve similarities in choice and constraint). And perhaps most of all (or at least I hope so), it can help us recognise that adaptation is not only about the organism adapting to its unmoveable and ever-demanding environment: organisms have tons of ways to adapt their environment to themselves. Even though we know this, it is not always something we take into account when designing research projects or interpreting results, so doing that more often in your papers should be interesting and valuable. And that happen to be two traits that help getting your papers published ...

 

Guidelines for archiving data AND code

The following is a cross-post from the Editor's blog of The American Naturalist, developed with input from various volunteers (credited ...