The subject of crowd dynamics has been increasingly topical this year, with the London riots being particularly notable.  The Economist have a nice article in their Christmas edition on a range of crowd dynamic issues, from how shoppers move during rush hour to pilgrims on their way to Mecca.  Interesting stuff.

The riots are what interest me however as there have been direct questions asked about the role social media has played in the riots that struck London this summer.  New research from Oxford University has attempted to shed light on the exact role social media played in the riots.

The research found that, as with other large groups, rioters contain a small group of influencers at the centre of the crowd (not literally) that have considerable influence in directing the masses.  The researchers termed these people spreaders and suggested that they play a crucial role in ensuring messages reach large numbers.

Early adopters were another important group identified by the researchers.  They lead the movement and help to recruit the spreaders.  That they tend to be spread throughout the network however limits their influence after those early moments.

The research team analysed Twitter activity during riots in Spain this summer.  During this time over 87,000 users and 580,000 tweets were dissected by the team, with some interesting results.  They found two distinct processes in the spread of information via social media.  Firstly the early adopters recruited others to their cause.  The spreaders then disseminated information far and wide.  The spreaders were key to the whole thing because they themselves were connected to people with strong personal networks, so their effective reach was very high.

The key to things building momentum was the amplification effect of many users sending out messages at once.  It was found that a few people was not enough to build the critical momentum the 'movement' required.  The social proof of others engaging actively in a topic helped recruit yet more people, and a movement was formed.

Lead author Dr Sandra Gonzalez-Bailon, Research Fellow at the OII, said: 'Digital media has played an important role both in the recent wave of mobilisations in the Arab world and in protests across Western countries, such as the Occupy movement across cities worldwide. This is the first empirical study analysing the mechanisms behind protest recruitment by means of online networks. It shows that mass mobilisations depend not on the influence of central users, who are nonetheless crucial for their growth, but on the actions of many users in local networks that will ultimately reach the influential core.

'Why people back a cause is based on many factors and relates to what is happening in the offline world. By examining the collective behaviour of online users, we have established that most people are influenced by what those around them do. If they are exposed to many messages calling for action within a short time frame, they are more likely to respond to this apparent urgency and join in. This creates recruitment bursts that can translate into a global cascade with truly dramatic effects, as the massive demonstrations and the wave of occupations that followed shows.'

Of course no Twitter user is an island, so it needs to be remembered that people will have gained information from lots of other sources as well, and indeed they would have been influenced by things far from Twitter's reach.  It does however seem to chime with wider research into crowd dynamics.

As a light hearted way to end, here is a famous video of how a crowd forms at the Sasquatch music festival in 2009.