Up until yesterday, I thought I knew the fundamental drivers as to why social networks have been successful. Our human needs e.g. family, success in our careers and other self-centered drivers were being released. The Dunbar’s number perspective (that social networks enable us to manage more relationships) is another side of the same coin, in that we use relationships to satisfy these needs and the more we have, the happier we are.

However I attended a lecture yesterday at my old university Imperial College, that challenges this or at the very least muddies the waters.

Self-organisation is a concept I’ve accepted innately since studying, but I’ve never really thought about it in detail and certainly haven’t mapped it into things I encounter every day. It’s a phenomenon (re)introduced to me in the inaugural lecture of Pierre Degond, Professor of Applied Mathematics at Imperial College yesterday evening.

Professor Degond did an excellent job of explaining in simple-to-understand terms how his team have got to the point of being able to categorise and model self-organising or ’emergent’ systems. If you’d like to listen to the details of his presentation Imperial have made a recording of his lecture available here:

The most interesting aspect of his presentation for me was his introduction of the notion of ‘exotic’ particles. These tend to be groups of simple things e.g. molecules, birds, and they exhibit the following traits:

  • Coordination behaviour e.g. it has been observed that sperm swims in a circle around a circuit, even though it’s not that smart.
  • Self-organisation e.g. pedestrians self-organise into lanes to improve efficiency.
  • Pattern formation e.g. network formation in living tissue and birth of cities.

Professor Degond performed a simple ‘bristle bot’ experiment, 23 minutes into his lecture embedded above, to demonstrate how these simple systems organise themselves. To the surprise of the audience, instead of the simple robots organising themselves into a line due to volume deprivation, they clustered together in a stationary group.

Pierre Degond uses 'Hexbots' or 'Bristlebots' to demonstrate emergent behaviour.

Pierre Degond uses ‘Hexbots’ or ‘Bristlebots’ to demonstrate emergent behaviour.

These traits constitute an emergent behaviour. It’s not a behaviour encoded in the individual particles, but is still exhibited by groups of simple ‘exotic’ particles like a flock of birds. Humans also are exotic in this sense as per the pedestrian example above, so my question is: in addition to satisfying our personal needs, does this natural state of organisation contribute to why we use social media?

It’s not yet clear why this emergent behaviour exists – which, for me, makes it rather exciting.

A further interesting point Professor Degond made is that if there’s too much noise a system does not ‘self-organise’. Perhaps this is a limiting factor now experienced by more mature social networks which are underpinning the company strategies of Facebook (diversifying to Instant Messaging), LinkedIn (acquired) and Twitter (struggling to retain users).


Leave a Reply