Saturday, March 18, 2017

Over-citation: my papers that should be cited less often

Ever write a great paper – an important one – and publish it to great expectations? “Surely everyone will love this paper,” you think. It is going to be a barn-burner. It is going to bust Web of Science – maybe even Google Scholar – with citations. Then, as the weeks and months and years go by, pretty much nothing happens. The paper gets a few citations (mostly from your own group), a few people seem to have read it, but not much else. And you think, “How did this happen”? “That was one of my best papers ever – it should be more widely cited.” Perhaps you start to think, “Maybe folks just missed it. If I could only get it in front of them again, people would recognize its greatness and it would go viral.” So you write a blog about “hidden gems” or you emphasize the paper on your website or you send out a few tweets or all of the above. And …. nothing happens. So you carry a (mild) resentment to your retirement, where you give your “exit seminar” and talk about your great work that just didn’t get the attention it deserved. (Yes, I have seen this happen.) Well, this post is about the exact opposite situation – papers that get way more attention than they deserve.

When one applies for a research grant, one usually has to talk about how wonderful one is – at least partly in relation to publications and citations. This need usually takes one to Web Of Science or Google Scholar to find out numbers of citations and H-indices and so on. Whenever I do this (such as yesterday while preparing a grant application), I see my top cited papers. I look at some of them and think, “Well, yeah, that paper was indeed useful and influential” but, about the same amount of time, I think “What the hell, why does THAT paper have so many citations?” So, I thought I would here take the opposite tack to the usual “papers of mine that should be cited more” and write about “papers of mine that should be cited less.” In doing so, I first need to point out that there isn’t anything wrong with these papers, they simply seem to have received more attention (or at least citations) than their content might deserve – or that we, as authors, expected.

One choice for an over-cited paper might be a short note we published in Conservation Biology about how species distribution models that predict massive extinction under climate change generally ignore evolution and are therefore probably often wrong. Models of this sort look at the abiotic conditions where a species is currently found, ask how the geographical distribution of those conditions is expected to change into the future, and then – if the conditions currently occupied by a given species in a given area shrink excessively – make a prediction of likely extinction. The problem, of course, is that species might evolve to occupy the changing abiotic conditions as selection forces them to do so – which is the only point we made in this paper. This point is certainly correct and many papers have now shown that such modelling is likely to be wrong much of the time, partly because of evolution. Yet it just seems so obvious as to not warrant a citation and – really – all our note did was point out that evolution could be rapid and that it could cause a mismatch between predicted and realized future species distributions. Does this rather obvious insight in a very small note really deserve 200+ citations in 7 years?

And the third most cited paper on Eco-Evolutionary Dynamics is ....
(coauthors redacted to protect the innocent)
Another choice for an over-cited paper might be the introduction we wrote to a Philosophical Transactions of the Royal Society special issue on Eco-Evolutionary Dynamics. The introduction simply pointed out that evolution could be rapid and that evolution could influence ecological process, before then summarized the papers in the special issue. Again, nothing wrong with the paper, but a summary of papers in a special issue is hardly cause for (soon) 300+ citations, nor is that typical of such a summary. I here assume that people are citing this paper mainly for the first two general points we make as listed above. This is fine, but excellent papers that treat eco-evolutionary dynamics as a formal research subject, rather than a talking point, are out there and should be cited more. Indeed, several papers in that special issue are precisely on that point, and yet our introduction is cited more. Similar to this example of over-citation, I could also nominate the introduction to another special issue (in Functional Ecology) – which is my fourth most cited paper (437 citations).

Why are these “OK, but not that amazing” papers so highly cited? My guess is that two main factors come into play. The first is that these papers had very good “fill in the box” titles. For instance, our PTRSB paper is the only one in the literature with Eco-Evolutionary Dynamics being the sole words in the title. Thus, any paper writing about eco-evolutionary dynamics can use this citation to “fill in the citation box” after their first sentence on the topic. You know the one, that sentence where you first write “Eco-evolutionary dynamics is a (hot or important or exciting or developing) research topic (REF HERE)” The Functional Ecology introduction has much the same pithy “fill in the box” title (Evolution on Ecological Time Scales) and, now that I look again, so too does the Conservation Biology paper (Evolutionary Response to Climate Change.) The second inflation factor is likely that citations beget citations. When “filling in the box”, authors tend to cite papers that other authors used to fill in the same box – perhaps partly because they feel safe in doing so, even if they haven’t read the paper. (In fact, I will bet that few people who cite the above papers have actually read them.) One might say these are “lazy citations” – where you don’t have to read anything but can still show you know the field by citing the common-cited papers.

Of course, I too sometimes take the lazy citation strategy. Sometimes when I am busting out an introduction and initially write “This [topic here] is a (hot or important or exciting or developing) research area (REF HERE)”, I simply fall back to my usual set of citations that I haven’t looked at for years and years. Doing so is a quick, easy, and safe way to simply move on to the more interesting stuff that really requires reading papers. Or, if I don’t know what to cite, but I know I am stating a well-known fact, I will simply search for the topic on Google Scholar to see what is most cited and then check the title and abstract to make sure citing it is safe. Perhaps this is a bad scholarship – or perhaps it is clever efficiency in the sense that these citations don’t really matter. They are generally known phenomena that have been discussed before and for which detailed additional reading would simply be a waste of time – so I am not exactly condemning “lazy citations” here.

My final closing point is that numbers of citations to a paper don’t always reflect the originality, importance, and quality of the paper. Sometimes papers are dramatically under-cited given their quality and potential importance. Sometimes papers are dramatically over-cited given their quality and importance. Of course, this point isn’t a new one but perhaps I am making it in a slightly novel way.


1.       Patrick Nosil first pointed out to me the “fill in the box” citation-inflation phenomenon.
2.       While writing this post, I noticed that the Google Scholar link for the Conservation Biology paper doesn’t even list me as an author – irony!
3.       No disrespect to my co-authors on the papers discussed above. In fact, my favorite part of all of the above papers was the collaborative writing efforts they involved. Clearly, we did a great job in the writing!

4.       Of course, I have my own papers that I think are way under-cited, particularly several awesome ones published in PLoS ONE (an analysis here). Check it how Humans are less morphologically variable (within populations) than are other animals and Bear predation drives the evolution of salmon senescence in unexpected ways. (And, no, I didn’t write this post simply to plug these under-cited papers.)

Friday, March 3, 2017

Maladaptation to chemical exposure – what may be happening and where do we go from here?

Guest post written by Mary Rogalski

Industrial, residential, commercial, and agricultural development greatly benefit human populations, but with the unintended and widespread consequence of increasing the release and availability of chemical pollution.  Surface runoff and atmospheric deposition introduce a complex mixture of heavy metals, pesticides, pharmaceuticals, and other contaminants into our water bodies. While some pollutants are regulated in an effort to protect human and environmental health, we have very little knowledge of how pollution exposure affects organisms over the long-term. To effectively manage pollution risk, we need to have a better understanding of these consequences.

Exposure to chemical pollution can have a host of negative effects on organisms, including reduced reproductive output, and at high enough doses, death. Individuals in the wild have been found to vary in their sensitivity to pollution exposure, both within and among populations. Based on these negative effects on individuals and the variation in sensitivity, most evolutionary biologists would likely predict that pollution exposure should select for more tolerant individuals. In other words, populations should be able to adapt to exposure.

This was my hypothesis when I set out to investigate the evolutionary consequences of long-term exposure to heavy metals in Daphnia populations in New England lakes (Rogalski 2017). Daphnia (aka “water fleas”) are tiny crustacean zooplankton that are extremely efficient at grazing algae in lakes but are also pretty sensitive to contaminant exposure. To my surprise, I found the opposite of my predicted trend. Daphnia had evolved to become more sensitive to metal exposure over decades of increasing contamination.

 Image of a Daphnia ambigua mother (approx. 1 mm in length). Photo: Eric Lazo-Wasem
 When I first started to see this result, I thought there must be some mistake. However, the pattern was repeated. I saw the same increase in sensitivity to copper following increasing historic exposure in two different populations, and also in response to cadmium in a third population.  In one lake, thirty years after peak copper levels, the sensitivity remained.

Grey points show copper or cadmium contamination through time. Black points show copper or cadmium sensitivity of individual Daphnia clones hatched from different time periods. LC50 is a measure of acute sensitivity, with lower LC50s indicating greater sensitivity. Further details on the study.

I tried to think of what could explain my unexpected result. Surely adaptation must really be happening here, right? But my study is clearly showing the opposite trend. Perhaps there’s an evolutionary trade off at play? Or some other reason why the populations have not only failed to adapt to metal exposure but also became more sensitive?

Alexander Lake, the study site where Daphnia have become more sensitive to rising cadmium concentrations

While the evolutionary pattern is striking and repeated, at this point I just don’t have enough information to understand the mechanism underlying the pattern. I certainly wouldn’t rule out the possibility that these Daphnia populations are adapting to their changing environmental conditions but just happen to be getting more sensitive to acute copper and cadmium exposure. In particular, I am curious to know if the acute and chronic toxicity responses might be inversely correlated. My assays measured acute toxicity – is it possible that being good at chronic chemical exposure makes a Daphnia worse at dealing with acute exposure? At least one study looked at this question in Daphnia with mixed results, finding no evidence of such a trade-off in response to cadmium, and no obvious pattern in response to copper (Barata et al. 2000).

Yet after having spent a lot of time reflecting on my results, I no longer find the trend so unexpected.
First of all, while maladaptation has received relatively little attention by evolutionary biologists, a metaanalysis by Hereford (2009) suggests that maladaptation happens fairly frequently. Of all reciprocal transplant studies examined in this metaanalysis, Hereford found that local maladaptation (defined as foreign population advantage) happened in 29% of cases. If we see evidence of maladaptation when we expect to see local adaptation nearly a third of the time, my result of increasing sensitivity to metals seems much less unexpected.

My study is not the first evidence of maladaptation to pollution conditions in wild animal populations. Researchers found that barnacles (Balanus amphitrite) in polluted estuarine environments were more sensitive to exposure to an antifouling biocide, copper pyrithione, compared with animals from less polluted conditions (Romano et al. 2010).  Rolshausen et al. (2015) found that Trinidadian guppy (Poecilia reticulate) populations have failed to adapt to crude oil pollution, despite devastating effects of exposure to oil. Also, a PhD student in the lab where I did my dissertation work, Steve Brady, found that wood frog (Rana sylvatica) populations in Connecticut ponds were more sensitive to road side environments in general, and road salt in particular compared with salamanders from forested ponds (Brady 2013). Steve found that there were also some overall fitness consequences of this increasing sensitivity to roadside environments. Interestingly, he found the opposite trend of adaptation to roads and road salt in another amphibian species, spotted salamanders (Ambystoma maculatum), inhabiting the exact same ponds (Brady 2012).

Results from Brady’s 2013 study of wood frogs.
In addition, just because pollution can have fitness consequences, I don’t think we should expect chemical exposure to act like other forces of selection such as predation, parasitism, or changing temperatures. Pollution exposure can also lead to increasing rates of developmental malformations, cause changes in sex ratios, and cause cancer. Some pollutants can bioaccumulate in tissues, including those of offspring. When the cadmium chloride that I had ordered for my toxicity trials arrived in the mail, the hazards listed on the safety sheet sounded pretty scary. In particular, cadmium exposure “may cause heritable genetic damage”. The other metal I studied, copper, has been linked with increasing mutation rates in exposed Daphnia populations. It’s not hard to imagine how some of these toxicological impacts could have accumulating consequences over the course of many generations of exposure.

One thing that is valuable about my study is that it tracks evolution through time. While local adaptation studies have provided valuable insight into how populations have evolved in response to contaminant exposure, we are missing three critical pieces of information. 1) We don’t know how pollution conditions have changed in the past in a given habitat. We can only compare organisms in polluted and unpolluted conditions; 2) we don’t know what the historical evolutionary trajectories of these populations look like; and, 3) in most cases the phenotypic responses that we observe may include genetic, plastic, epigenetic, and/or maternal effects.

In my study, I was able to address these issues. I used lake sediment archives to track both environmental and evolutionary trajectories over time. I measured metal contaminants in dated sediments to put together the history of exposure experienced by these populations over the past century. I hatched Daphnia from resting egg banks buried in sediments from high and low metal time periods. I then tested these Daphnia for sensitivity to copper or cadmium to see if the populations had evolved in their tolerance for these stressors. Since we can raise Daphnia clonally in the lab I was able to minimize any maternal effects that might have been present.

Extracting metals from lake sediments using hot block acid digestion. Photo: Sara Demars
In closing, I’ve come away with two key points from this study. First, maladaptation appears to be fairly common but our theoretical understanding of why it happens is pretty limited. This leaves us trying to explain seemingly counterintuitive results with a bit of hand waving and throwing around terms like “genetic drift”, “trade offs”, and “dispersal rates”. As evolutionary biologists we need to do more to understand what drivers may lead to maladaptation to improve our ability to both explain and predict evolutionary trends. Second, if species as different as Daphnia, barnacles, and wood frogs are becoming maladapted to pollution, we should think critically about the risk associated with multi-generational pollution exposure. How common is maladaptation to pollution exposure, and how does this affect the ability of organisms to adapt to other stressors? In what contexts might we expect to see adaptation vs. maladaptation to a contaminant? Could pollution exposure be having long-term damaging impacts on human populations? Those who oppose pollution regulation focus on the financial costs today, but the cost of inadequate pollution control for humans and other species could be much greater over the long-term.

Daphnia resting egg cases from one of the study lakes. Photo: Eric Lazo-Wasem.