PhD Defense by Jonas L. Juul

Title: Synchronisation, shocks and contagion in a connected world

Abstract: Biological, engineered and social systems often consist of many interacting parts. Sometimes the interactions between these smaller components give rise to macroscopic behavior. From stem cells joining their efforts to create a fetus capable of living, to social interactions allowing diseases to spread, interacting populations on many different scales exhibit dynamics that are important to study and understand. In this dissertation, numerical, analytical, and statistical methods are used to study the dynamics of systems consisting of interacting components. The thesis contains three parts.

In Part I, a system of coupled oscillating stem cells in mouse embryos is studied. The stem cells create the complex spatial pattern of vertebrae precursors - somites - in mice. We develop a theoretical framework for a recently proposed mechanism for the size and timing of somite creation. We use the theory to establish previously unknown relations between key variables in the biological system, and show that the experimental values of these variables are consistent with the proposed mechanism. We also suggest experiments capable of falsifying the proposed mechanism. In another study, we show how to control experimentally observed phase waves. This method could be used as a means of testing whether these waves are critical to the timing of somites, which is a widespread hypothesis in the field.

In Part II, we study the impact of contagion in networked populations. Firstly, we consider a model of complex contagion with synergy. Combining analytical and numerical analyses, we find that the synergistic effects determine which nodes can be infected and which cannot. Secondly, in another study, we examine the impact of anti-establishment nodes in a model of complex contagion describing the spread of two competing products. The model is inspired by recent anti-establishment outcomes in elections and referendums. We find that a very small number of anti-establishment nodes can cause an otherwise insignificant product to become the most adopted at equilibrium. We also argue why a few nodes can have such a considerable influence. Finally, we study what impact mutant contagion, such as the 1918 Spanish flu, has in structured populations. We analytically derive a scaling law for this impact in infinite-dimensional networks. We find that numerical analyses and existing empirical results for meme-spreading on Facebook support this law. 

In Part III of this dissertation, telecommunication in human populations is studied. In one paper, we examine correlations between narcissistic personality traits and social behaviour. We find correlations between the number of social contacts nodes have and their narcissistic scores. We also find evidence of homophily in so-called narcissistic admiration. The observations are in agreement with some existing hypotheses regarding the psychology of grandiose narcissists. In the final paper, we examine telecommunication patterns following terror attacks in several Western European cities. We find that, although both females and males change behaviour following terror attacks, there are significant gender differences between the changes of behaviour.

Participating via Zoom by following link: https://deic.zoom.us/j/8177252756