academic publishing Archives

24/11/201425/11/2014

The Allure of Couzins: Self-organising collective groups

Every now and then you stumble on a paper that changes everything for you. Typically something of a personal zeitgeist moment, it opens your eyes to a whole new world of potential and can spin your own research out in new directions, or encourage a complete re-orientation of your goals. In this new series, we are going to profile some of our favourite papers and maybe share the inspiration a little wider.

I don’t get out from behind my computer much, but when I do, my favourite engagement with real animals is to watch swirling flocks of starlings or trails of ants in search of food.

During my PhD I was working on the evolution of behaviours in social groups when I stumbled on a paper as I was searching through my supervisor Prof Graeme Ruxton’s publications. This was one of the first collective behaviour papers I read, and it was a huge eye-opener for me, pandering to both my love of biology and computers. Collective memory and spatial sorting in animal groups by Couzin, Krause, James & Ruxton brought home for the link between individual behaviour, self organisiation, emergent behaviour, complex biological pattern formation and how evolution could exploit this system and shortcut adaptive strategies without the need for invoking complex cognitive processes.

The concept itself wasn’t new: Craig Reynolds in 1986 demonstrated that interacting individual computer animals, which he termed boids, following three basic rules of separation, alignment and cohesion could generate a variety of complex group level patterns akin to biological swarming, shoaling and flocking. This simple computer simulation showed how the interactions themselves could create coordinated group-level behaviour without a need for centralized control, or for agents to possess any knowledge of their surroundings beyond their nearest neighbours. Instead, the patterns are an emergent property of the system of interacting agents that arise through a process of self-organisation. Craig Reynolds went further, and showed that information could be transmitted through the group so that obstacles (or predators) could be avoided by individuals responding to their neigbours avoidance, without having to actual see the obstacle or threat for themselves. Such characteristics have clear selection benefits in an evolutionary sense whereby there are cheap, effective ways to gain benefits of living in large collective groups.

What Couzin et al did was to show in even more detail the ecological and evolutionary relevance of these systems. They described in detail how subtle changes to individuals’ behaviours, manifested in adjustments to the radii that define whether they avoid, align or cohere with their neighbours, could arise in abrupt changes to the group-level pattern. A system dominated by attraction and avoidance tends to produce swarming behaviour around a relatively stationary point whereas one dominated by alignment produces shoals that move in a rather rigid, elongated diamond-like formation. In between there exists an intermediate state in which the group spontaneously starts to rotate in a torus (ring-doughnut) structure. The key point here is that no-one in these groups knows what a torus is, never mind how to achieve it, and nor is there a leader telling them to swim in a circle – rather, it is an inevitable consequence of the aggregation of the interactions between individuals with a mid-range tendency for alignment. An extra quirk to the system is that though the individuals are completely bereft of any brain or memory, the system shows evidence of memory – or hysteresis as physicists would refer to it. Individuals starting in a swarming pattern and increasing alignment will move through the rotating torus and on to the rigid diamond-like structure with individuals locked into a particular place in the group. However, if you start with the rigid structure, and reduce individuals’ tendency for alignment, they skip the torus structure and revert straight to swarming. In this manner, the group has memory of what it was doing in previous states, even though the individuals have no such memory. This is surely a rather cheap way for interacting ants, fish, birds, or maybe even the neurons of our brain to encode a sense of memory or history without having to explicitly encode and record the details of past states.

Perhaps most relevant from a behavioural ecology perspective, Couzin et al went further and explored how changes to an individuals behaviour relative to the group could alter its location. Increasing speed, decreasing turning rate or increasing one’s tendency for alignment would see an individual end up towards the front of the group. Reducing ones tendency for avoidance would see an individual move to the centre of the group. This is beautifully simple. Without any knowledge of the group structure, or where one is in the group, simple behavioural rules linked to internal state can now allow an individual to navigate the group. For example, if hungry, simply speed up and you will be at the front with first access to food. Once sated, you can seek out the relative safety of the middle by reducing your how much ‘personal space’ you require.

Prof Iain Couzin has gone on to show myriad intricacies of collective groups in terms of how they can make decisions as a group, and how locust swarms are driven by similar properties. During my PhD I had the pleasure to meet with Iain Couzin, study under the tutelage of Graeme Ruxton, share code with Richard James, and collaborate with Jens Krause (I didn’t get to meet the last author Nigel Franks, but now that I’m back on the conference tour there is still hope!)

Author

Andrew Jackson

Photo credit

wikimedia commons

16/02/201414/02/2014

3D dinosaurs are bringing down the Ivory Tower

Adam raised the point of science communication in his last blog post of how science should be communicated to a mainly interested and receptive public. The main question when thinking about science communication is “how should we do it?” However a second question, arising from this one would be “who should do it?”

I believe having an interest in popular science is the first step for starting a career as a scientist. Personally, I was influenced by Jurassic Park, which gave me the idea that in a distant future I would (could) become a paleontologist. I believe this influence is shared by a vast majority of young paleontologists of this Bakker dinosaur’s revolution generation. So, to me, even if Jurassic Park bent some scientific rules (the most obvious one being bringing stones to life), the film created a renewed interest in its related field of palaeontology.

Ok, so that’s a good reason for popularizing science then? It’s not that easy. A fairly large amount of scientists will always look at such attempts to popularize science as a “bastard” child of the scientific adventure, a kind of side effect that makes scientists, in the best case, smile ironically. In the worst case though, they think it propagates massively erroneous ideas: like the one I pointed out above or -another that “makes me smile ironically” – that paleontologists are Indiana Jones-style adventurers that wander around the world and from time to time brush the sand and discover how velociraptors (read Deinonychus) were talking to each other.

Richard Whitley in his book, Expository Science: Forms and Functions of Popularisation, summarises the idea by writing that: “popularization is not viewed as part of the knowledge production and validation process but as something external to top research which can be left to non-scientists, failed scientists or ex-scientists as part of the general public relations effort of the research enterprise.”

This vision of popularizing science is linked to the idea that an increased output of popular science leads to a fatal decrease of legitimate and accurate scientific progress. This is known as the ‘Sagan Effect’ after Carl Sagan and the erroneous claim that his legitimate scientific output decreased after the initiation of his popular science TV series: Cosmos.

Can one popularize science and actively do science?

When one has a look at the actual numbers (scientific articles published vs. popular articles published), one expects that if popularizing science is reserved for “failed scientists”, people publishing more popular articles will publish fewer science articles. Michael Shermer, in his paper published in Social Studies of Science in 2002, brings down this popular belief by calculating the actual numbers of scientific papers published by these popular science writers.

Let’s take a look at the numbers.

As a base line, while the four PIs publishing on this blog all take part in extra curricular activities, they are primarily research scientists. Last year, they published all together 22 papers which gives a respectable average of nearly one paper per PI published every two months! In contrast, if you look at the great popular science writers of the end of the last century (say Carl Sagan or Steven Jay Gould), during their massive and influential popularization of their fields, they produced an average of one paper per month for Sagan and nearly two papers a month for Gould (and that for 35 consecutive years!). So they seem to have it all: “legitimate“ science output and successful careers as science communicators. Just looking at these figures, the myth of the trade-off between popular science writing and actual scientific production is just, as Shermer describes it, “a Chimera”!

Can one popularize science and actively do science?

Yes. And to go back to my first point, with Jurassic Park (which, in comparison to Gould or Sagans’ popular work can easily be classified as a “bad” way to do it), I believe that popularizing science is not a problem: in an altruistic world, scientists should focus on contributing to the advancement of their field by any means possible, not exclusively on their own amount of publications. Thankfully, most scientists manage to combine both popular communication and research output efficiently. I think that one important and still altruistic way to contribute to the advancement of a field is definitely science popularization. Richard Whitley says that “by successfully combining claims to universal validity and social utility through popularization, [Sagan and Gould] laid the foundation for the present domination and expansion of the sciences.” These contributions have a real long-term benefit to the advancement of the field. Firstly, the ideas of the authors of popular science become fixed in the public consciousness and secondly (most importantly to me), they inspire new generations of scientists that will, later on, personally contribute to the advancement of their field.

Is this bad for “legitimate” scientific production? Not at all! By popularizing evolution in a realistic (or perhaps not so realistic) way, Gould and Jurassic Park spark the interest and imaginations of the next generation of potential scientist who, through their research, will make direct contributions to scientific research!

It’s a positive feedback cycle: excellent scientists generate good science communicators; good scientific communication generates scientific interest and curiosity; scientific interest produces scientists and some of these scientists will go on to become good scientific communicators…

Author: Thomas Guillerme, guillert[at]tcd.ie, @TGuillerme

Image Source: wikimedia commons

31/01/201421/01/2014

NERD club transferrable skills: reviewers, rejections and responses

Academic publishing: the currency of any research career. It’s all very straightforward; take your most recent ground-breaking results, wrap them up into a neat paper, choose the perfect journal, allow said paper to persuade an editor and reviewers of your brilliance and bask in the reflective glow of getting your research out into the world. Whether you see this rosy scenario as a target or delusional and unattainable aspirations, things rarely work out so smoothly. Instead, every researcher must learn to deal with the topic of one of our recent NERD club discussions; reviewers, rejections and responses. As a collective of staff, postdocs and postgraduate students, here are our thoughts on the dos and don’ts of dealing with the three r’s of academia.

1) Reviewers

Some journals invite authors to suggest reviewers or editors for their papers. If this happens, pick who you think is the “best” person whether that’s because the person is an expert in your field (although see our final point below), likely to give a fair review or because they are familiar with your work. Only suggest people to be reviewers if they have published themselves i.e. aim at the level of senior graduate student or from post-doc upwards. It’s also a good idea to choose someone that you’ve cited a lot in your manuscript (no harm to get on their good side). Equally, if you gave a conference talk recently, remember that person who seemed so interested in and enthusiastic about your work in the pub afterwards – chances are that they might be a fair and favourable reviewer. We also thought that, if you feel it’s necessary and you have a good reason, it might be a good idea to make an editor aware of people who you would prefer not to review your paper. However, be warned of the rumours that some editors may prefer to ignore such preferences and deliberately choose people from that “exclusion” list as reviewers.

In contrast, when selecting potential reviewers or editors, don’t choose someone who you have thanked in the acknowledgements of your manuscript. These are usually people who helped out or offered advice at some stage during the research so they would have a conflict of interest when it comes to reviewing the manuscript. Of course, you can also use this guideline to your advantage by seeking advice from and therefore acknowledging people who you definitely want to avoid as potential reviewers… Also from a conflict of interest point of view, don’t suggest your main collaborators, close friends or people from your own institution as potential reviewers. Similarly, for obvious reasons, don’t choose someone as a potential reviewer if you know that they dislike you or your work!

Once you have considered all these points, our final piece of advice about choosing reviewers or editors is don’t get hung up on it! Your preferred reviewers may not accept the manuscript, particularly if they are the senior, time-limited experts in your field. Finding reviewers for a manuscript is ultimately a lottery so, having thought about a few of our suggested guidelines, there’s no point in agonising over the process too much.

2) Rejection

It’s never pleasant to think about but your chances of rejection in all aspects of academia are high. Rejection of a manuscript can be particularly disheartening as it represents a dismissal of months if not years of your hard work. Our main piece of advice is not to do anything hasty. However unfair, pedantic or ridiculous the reasons which “justify” the rejection may initially seem, their bitter sting usually mellows if you take the time to sleep on it (after getting cross and having some comfort food/drink/other activities…). Similarly, don’t ruin your Friday evening/weekend/holiday by obsessively checking your emails for an editor’s response which will more likely than not be negative.

After hopefully returning to a somewhat more rational state, try to take the reviewer’s comments on board. The constructive ones will help you to make a better paper and parts which reviewers didn’t understand are often because you could have clarified your points better. Ultimately, it’s important to be realistic; the chances of rejection of a manuscript are high so be prepared to resubmit elsewhere (think back to our previous advice about working down the list of journals for which you feel your paper is suited). You could even have another version of the paper drafted and formatted for the next journal on your list even before you receive a response from your initial submission. It may take an initial investment in time and energy but at least the next version of the paper is then ready to go if you’re unlucky enough to receive a rejection from your first journal of choice. The most important advice to emerge from our discussion was to be thick skinned and not to take rejection personally. You can’t publish without experiencing rejection so it’s important to share your experiences (both good and bad) and to listen to the advice and woes others. You’re not alone!

Our final cautionary point was that you don’t necessarily have to accept rejection. If your paper was rejected on the basis of reviews which you think were poor, biased, unfair or just completely off the mark, it might be worth arguing your case with the editor and/or reviewers. HOWEVER, only take up this tactic in exceptional circumstances where you have a VERY strong case to back up your points. If you get a reputation for being petulant, argumentative, obstinate or just downright rude it can only serve to seriously aggravate an editor and damage your future publishing prospects if not your wider research reputation.

3) Responding to comments

Hooray! You’ve got through the reviewing gauntlet, dodged the cold blow of outright rejection and now there are just a few reviewers’ comments standing between you and potential publication glory. We came up with lots of dos and don’ts for how to deal with reviewers’ comments.

Most importantly, be polite and positive. No one was obliged to review your work so thank the reviewers for their comments and suggestions and consider adding them into the paper’s acknowledgements. Never be aggressive or rude and only write responses which you would be happy to say to a reviewer or editor face to face. Respond to all comments, no matter how trivial they may seem and show that you’re willing to make more changes if necessary. Don’t just ignore the bits that you don’t agree with. Show the editor that you have dealt with every comment by cross referencing the changes you made to the manuscript. To do this, either refer to the revised line numbers or, more preferably, cut and paste the sections that you have modified into your responses so that the editor doesn’t have to keep moving among different pages to check what has been changed.

Being polite and positive doesn’t mean that you have to be a push-over. If you receive a comment which is impractical, beyond the scope of your paper or just downright wrong, argue your point (with legitimate back up) but always retain a courteous tone throughout. If you have to deal with a whole slew of comments which are particularly off the wall or aggravating, ask someone to read your response before sending them to the editor – it’s always easier for someone else to pick up a passive aggressive tone which, despite your best efforts, may have crept into your writing.

If reviewers disagree on a particular point, either justify the changes you have made (don’t just ignore one reviewer’s comment) or your reasons for not making any changes. Don’t feel obliged to do everything that a reviewer suggests. Don’t ruin the flow of your text with awkward sentences which were clearly just inserted to please a particular reviewer. If you have good reasons (not just stubbornness or obstinacy) for sticking to your original ideas then make them clear. Remember that you can always get in touch with the editor if you get unrealistic or conflicting instructions or if you’re unclear about what you are expected to change.

So there’s our collective field guide to the trials and tribulations of academia. They’re by no means exhaustive but they’re definitely a good starting point. However, as academic publishing seems to require good fortune and timing as much as scientific rigour, research merit and an eye for a good story, there’s no magic formula for how to succeed, no matter how carefully you follow NERD club’s collective wisdom…

Happy publishing!

Author: Sive Finlay, sfinlay[at]tcd.ie, @SiveFinlay

Image Source: justinholman.com