Despite enabling almost instantaneous global communication, email appears not to have made the world a more close-knit community.
Duncan Watts and colleagues at Columbia University in New York conducted a massive email experiment to test the theory of 鈥渟ix degrees of separation鈥�, i.e. that everyone in the world can be linked through just six social ties.
More than 60,000 people from 166 different countries took part in the experiment. Participants were assigned one of 18 target people. They were asked to contact that person by sending email to people they already knew and considered potentially 鈥渃loser鈥� to the target. The targets were chosen at random and included a professor from America, an Australian policeman and a veterinarian from Norway.
Advertisement
The researchers found that it in most cases it took between five and seven emails to contact the target. Watts says this shows that email has not fundamentally changed the way social ties are created.
鈥淚n this experiment, the internet is simply the tool we use to transmit messages,鈥� Watts told 麻豆传媒, in an email. 鈥淐ompared with offline interactions like work, school, family, and community, I don鈥檛 see email as being a particularly compelling medium for generating social ties.鈥�
The concept of six degrees of separation emerged from a similar postal experiment conducted by the social psychologist Stanley Milgram in 1967. Milgram asked volunteers to send a package by mail to one of a hundred people chosen at random.
Virtual friends
In the email study, participants were also asked to send a message to the researchers to explain who they chose to forward their message, and why. This revealed which types of social bond were most likely to help a message reach its target.
The researchers did see some internet-only relationships in the experiment, but these accounted for only six per cent overall. By far the most successful bonds were found to be work-related ones. And messages were also more likely to reach their target if they were forwarded to someone of the same sex.
The researchers were surprised to discover that message chains did not rely on a few highly connected individuals, so-called 鈥渉ubs鈥�. Previous research by Watts and fellow Cornell mathematician Steven Strogatz had suggested that such 鈥渉ubs鈥� were important to all social chains.
Understanding the nature of social networks can have important implications. For example, it can help scientists model the rapid spread of infectious diseases.
Journal reference: Science (vol 301, p 827)