Friday, January 24, 2014

Explaining things with flow

This week our group’s first full paper on modelling socio-economic systems came out in PLoS One. The idea for this article came up when I saw Hans Rosling’s presentations of Gapminder. When you see how Rosling's bubbles representing countries float up from poverty and low life expectancy towards a better world you can’t help be inspired by the power of data.

Watching Rolsing’s videos, I also thought about fish movements! We have been using dynamical systems and Bayesian model selection to identify the models which best fit the rules of motion of fish. If we can think of countries as fish then we could use the same methods to find the rules of motion of countries in a socio-economic space, rather than a fish tank.


There turned out to be quite a lot of differences, as well as similarities, between fish and countries. But keeping this initial analogy in mind, Shyam Ranganathan and Viktoria Spaiser have developed a toolbox in R for finding the best set of differential equation models that describe socio-economic change. In the paper, we apply these methods to democracy and economic growth. You can see more about this in the video on the right. Not quite up to Rosling enthusiastic standard of simplifying things, but I hope it illustrates the idea behind the paper.

We are now applying this method to look at the demographic transition, the changes in cultural values across the world, as well as segregation in schools. Most of these projects are based at the Futures Institute in Stockholm. The idea in the paper is to present the tools, rather than make predictions. But we hope that future work will lead to useful insights in to how the world is changing and what we should do about it. 

Please contact us if you have data that you think can be analysed using this method. 



Monday, January 20, 2014

Anti-social extinction

During a recent trip to New Zealand, I visited a couple of bird conservation projects. One was for the rarest of thefour species of Kiwi birds, the other for yellow-eyed penguins. Both these conservation projects are doing an amazing job keeping the animals living in their natural environment, while allowing tourists like me to get a look atthem and develop some insight in to conservation issues.

A yellow-eyed penguin couple and baby penguin.
My wife commented that it was ironic that, given I prefer group-living animals, we were visiting two anti-social species. Kiwis pair for life and jointly care for one egg at a time. But when the egg hatches they first ignore the newly hatched bird, and when it they gets bigger they just chase it away. The yellow-eyed penguins also form long-standing couples (although homosexuality, divorce and season long infidelities do occur) and invest in joint parental care. But once their offspring is big enough to fend for itself, it has to find its own territory.

A group of blue-eyed penguins
This anti-social behaviour makes conservation difficult, simply because more space is required for a sustainable population. I was struck most by this when we saw a colony of blue-eyed penguins on a nearby cliff. These birds also form breeding pairs, but are happy to live in close proximity to other blue-eyes and a vast array of seagulls. Blue-eye penguins are more numerous and wide spread in New Zealand and, although we shouldn’t become blasé, they are not endangered.

What I hadn’t thought of before is that the difficulty of conserving solitary-living species could help explain the evolution of sociality. Animals that are difficult to conserve are prone to extinction when their environment changes. Penguins that tolerate each other have to share local nesting resources, but if space suddenly becomes limited they don’t waste time and energy fighting with each other. Territorial penguins might be able to chase off more tolerant penguins and dominate the best nests, but if their territory comes under attack by a predator, for example, they are not welcome at their neighbours place.

Developing an argument about evolution of territoriality requires a model defined in terms of selfish individuals, not species success. My evolutionary penguins are playing some form of a ‘tragedy of the commons’ game, where in a stable environment selective pressure acts to make them more territorial. In an unstable environment, where territories can suddenly become unusable, selection may favour greater tolerance. When danger arises penguins that tolerate each other can congregate in the few available sites. Provided that penguins can only migrate locally, then local extinctions of territorial penguins who start fighting over the few available resources can lead to the evolution of greater tolerance for neighbors. 

The penguin model I sketch above is related to the idea of r and K selection popular in the 1970s and 80s, and now replaced by more general life history models. In environments where everything is stable (limited by the carrying capacity K) it is better to defend your territory. When territories are shifting (growth rate r is more important) then defense of these territories is less important and it is more important to do well in the space that can be found. I have to admit I don't know much about the literature that links together evolution of co-operation with life history strategies. I found one review of co-operation (a bigger step than just tolerating each other). Maybe people working on the evolution of co-operation should stop bickering over nothing and get a bit more biological detail in their models.

Disclaimer: I don't think my penguin model is necessarily a good description of what actually happened in real penguins. It turns out that modern yellow-eyed penguins may well have arrived with humans in New Zealand.

Friday, January 17, 2014

A picture of two penguins


I have been having a bit of an extended holiday over Christmas till now. Since New Year I have been in Australia and New Zealand. From Monday, I will be in Sydney on sabbatical until the start of April. Visiting Ashley Ward and his growing research group.

And I will be back to blogging soon. We have a paper coming out on Monday on our dynamical systems methods for studying socio-economic systems, I am going to write something inspired by my holiday about why solitary living animals might be more at risk of extinction than those living in groups. I have had a couple of students finish interesting projects, one on evolutionary game theory of poker and one on studying the pop charts, which I want to write about. And I might even do an analysis of my Twitter following (still quite low, but growing).

Photo: Elise Sumpter
In the meantime, here is a picture of a couple of yellow-eyed penguins. Sadly, two of just 4000 left on the planet.