Tuesday, August 26, 2008
Dunbars number, virtual communities and trust
Mass collaboration has become a hot topic, in the wake of Open source actionists and books like "Wickonomics", and "We-think". According to Tapscott and Williams, authors of the first mentioned, mass collaborating the Linux-way represent a revolution in how people collaborate on complex matters. Much of the same ideas is reflected in James Surowieckis book "The wisdom of crowds". It seems like when millions of people are capable og collaborating the sultions and ideas they generate outdistance the performance of smaller groups. However is big size always size beneficial? And isnt there certain limits?
Other pieces of research has argued that there exist an absolute limit for collaborative groups. Most notably perhaps, Robin Dunbar have argued that for there is a correlation between size of cortex and the size of groups for a specie can manage to coordinate and sustain over time. Based on cultural and historical case studies of human groups , he argue that there is a limit for humans on appr 150 persons (i.e the Dunbar number). See here for his paper Coevolution of neocortical size, group size and language in humans.... His argument is that for a group not to disolve, a significant amonunt of time (appr 42%!) must be used to social grooming. If not, the group wil loose its cohesion and distrust evolve. The most controversial part of Dunbars argument comes when he relates this to a more general theory of language development. He argued that: "My suggestion, then, is that language evolved as a "cheap" form of social grooming, so enabling the ancestral humans to maintain the cohesion of the unusually large groups demanded by the particular conditions they faced at the time."Thus humans have reduced the amount of time used on grooming by the use of "language technology".
If we accept Dunbars ideas, it raises some interesting questions tregarding virtual communities and mass collaboration. Does new technologies like email, social software and mobile phones speed up the grooming? Or does it, on the contrary, make grooming more cumbersome due to lack of proximity? Can we now extend the Dunbars numbers to over 200 or are 150 relations more than we can handle in a virtual world? The mass collaboration ideas seems to suggest that there are now almost no limits as to how many who may collaborate on a given project... Yet empirical studies suggest that in online communities, larger than 150 members is rare. Using social network analysis, Chris Allen has hypothesized that different group sizes impact a group's behavior and their choice of processes and tools. Based on empirical data from MMOG and online communities, he has suggested that for non-survival groups, the equivalent Dunbar number falls somewhere between 60-90. Thus 150, may bee too large for these communities.
I believe that it is true that it is difficult to develop active and cohesive online communities or virtual teams larger than this size. Yet we should probably note that group cohesion is not an absolute measure. It depends on what kind of obligations and tasks a group is doing, and it can (like trust) be founded on other sources other that interpersonal grooming. High risk tasks with high interdependencies needs high trust and cohesion, while low risk collaboration require lower levels. Most cases of mass collaborations - liw Wikipedia- is really more cases of mass coontribution, where a number of people contribute on limited areas with low risks, encouraged by the prospect of a positive reputation, economic rewards.Further it should be noted that cohesion can be founded on other sources than grooming. As often is tha case for political groups and cultural movements, strong identities of ideologies act as important basis for large scale collaboration across distances. Thus only looking at interpersonal grooming may bee too simple.
For many real life groups there are also a question of interpersonal history. Not all groups are 100% virtual, and if some of the members have former relations based on FTF relationships this, may create stronger ties within the group. Such ties obviously needs less social grooming than others. So, even if Allen may be right that online communities hardly grow abow 60-90 people, this is often possible for communities with strong common identities and/or with simpler task and objectives.