NB. This book was sent to me as a review copy.


The idea of this book is relatively simple, but the consequences are huge, and in fact some of the ideas are far more subtle and complex than they may first appear.

Essentially this book is based on a series of experiments which Damon Centola has run, which are all related to changes in behaviour which can be tracked, and made to occur, through a social network (in the broadest sense of the word). This is the study diffusion in a network.

The fundamentals of the research lie on two distinctions: One in the complexity of a contagion/behaviour, meaning how many connections with others who have the contagion/behaviour do you need until you adopt it, and the other in the topology of the social network, meaning loosely, how much like a street where each person only talks to their neighbours, versus a small world-network where there are a lot of disparate connections does the network look like.

If one lives on a street where you only talk to your neighbours, and someone gets a disease which can be spread by a single contact, then the disease spreads slowly. If you start to include long jumps from different areas of the street where people move around over longer distances, the disease can spread much faster.

One might look at this and conclude that everything spreads much faster when a network is of this ‘small-world’ form with lots of distant connections. However, one quickly realises that actually, such a topology can slow the diffusion of a complex behaviour, like a fashion, or a health-related activity, or a technology. In those cases it is important for us to see a few people around us take up this behaviour before we start to do it.

This has enormous implications for trying to spread practices like the taking of medication, taking on a healthy diet, or the spreading of political ideologies. This research means that someone who wants to spread a behaviour should try and engineer the network in which they are working such that its structure is efficient at making the contagion spread, and stick.

Computer simulations as well as real world experiments are used in this research and the hypotheses about what type of behaviour spreads well on what types of network are tested.

It’s clearly an area of research with a lot of implications, particularly in the world today where misinformation can be quickly spread on a social network and political tendencies can be swayed by one’s connections. Understanding this may allow us to build social networks which are more robust to the spread of certain information, or that easily allow for certain behaviours to be engendered within a society.

Overall the book is well written and engaging, with plenty of discussion about the experiments that go into the conclusions, and on reading it, it is clear that there is a lot more to be done so that we can better implement lasting health strategies, and political engagement amongst many, many other complex behaviours.

How clear is this post?