INVESTIGATING the social influences that lead people to carry out copycat suicides can be difficult when your subjects are, by definition, dead.
Sociologists have long been aware of the phenomenon and have identified the existence of 鈥減oint clusters鈥� in which individuals copy suicides within their social groups, and 鈥渕ass clusters鈥� that happen after reports of celebrity suicides appear in the media.
Alex Mesoudi of Queen Mary, University of London, developed a computer model community of 1000 people, divided into 100 social sub-groups. Individual 鈥減eople鈥� were programmed to react as realistically as possible to the influence of friends, celebrities and the media, and assigned suicide risks based on sex, age and race.
Advertisement
Running the model through 100 generations supported the point cluster idea 鈥� people were more likely to commit suicide either because they had learned this trait from their friends, or because suicidal people are more likely to associate together (PLoS ONE, ).
Mass clusters within the simulation seemed to need the publicity provided by the frenzied media reporting that follows celebrity deaths. For example, there was a 12 per cent jump in suicides in the US in the month following Marilyn Monroe鈥檚 death from a sleeping pill overdose in 1962.