Learning in networks and in communities of practice

In the context of an article on MOOCs I co-author, one of the reviewers raised the question of how networks and communities, in particular networks for learning and communities of practice, are different from each other. Without tracing the complex history of learning in social networks, one can safely say that all of it is rooted in the pioneering work on social learning done by such people as Albert  Bandura and, later on, Jean Lave and Etienne Wenger. Was Bandura interested in all kinds of learning, primarily formal, school-based learning, Lave and Wenger thank their fame to their development of the concept of a community of practice. The participants of such communities learn by peripheral participation, they claim, that is through their presence in the many professional discussions that take place in the context of the community. Thus learning is social, as with Bandura, but also almost accidental.

Lave and Wenger, also in their later work, stress the importance of strong links between the people that participate in a community of practice. Lave and Wenger's case studies are about people who see each other frequently in the course of their jobs or occupations and exchange 'war' stories. They discuss the problems and challenges they have had to face, the solutions they came up with. These stories become part of the group lore and are exchanged, often at informal occasions (coffee breaks rather than team meetings). John Seely Brown and Paul Duguid in their book The Social Life of Information tell a similar tale about people who repair xerox machines. They illustrate how detrimental it is to the quality of the repairmen's work if this social learning dimension is forgotten, for instance by making it next to impossible for them to meet informally, in an effort to increase efficiency. So people do learn in communities of practice, if only to improve and fine-tune their existing knowledge. They learn in virtue of the tight social network that the community constitutes, with (almost) everybody connected to everybody else through regular and topic-bound interactions.

Networked learning does not focus on the strong links but emphasises the importance of weak and latent links. Chris Jones was one of the first to establish this as a principle. Of course, weakly let alone non-linked people do not learn from each other, precisely for want of the social interactions that learning requires. However, the importance of weak and latent links for networked learning is their potential to develop, if needed, into the kinds of strong links that sustain communities (of practice). So a learning network is thought to consist of weak, strong and latent (not-yet-existent) links. Strongly linked individuals learn from each other, not or weakly linked individuals learn from each other once they get connected. This requires i) that there is a need to learn from each other through a shared interest, problem, challenge; ii) that the learners are somehow brought together if the need arises. Note that the strength of a learning network relative to a community of practice is that many more people may be mobilised to become a member of one's community of practice, should the need arise.

It is well-known that our mental ability to maintain strong relationships is limited (Roger Dunbar believes it is in the order of 150 people that we can keep track of; see also Sutcliffe et al.), so technology is needed ensure that the right kinds of people, the people that have the potentially to help solve the problem at hand, can indeed be mobilised. In our own work, we have worked on such mobilising technologies. I discuss some of them in a forthcoming book chapter (Sloep). We have also coined the term 'ad-hoc transient group (community)' to denote the emergence - in the course of applying mobilising technologies -  of small groups that are dedicated to the particular problem at hand. Although such groups may disappear once the problem is solved, they may last and become or become integrated in genuine communities of practice.

This then leads to the following delineation of learning networks and communities of practice. Where a community of practice capitalises on the existing social ties between the community members for social learning to occur, a learning networks mobilises weak and latent links for that purpose. Doing so, leads to a conception of learning networks in which any particular instance of a learning networks consists of a non-exhaustive ensemble of partially overlapping, waxing and waning communities of practice; non-exhaustive because there may be learners who are in the network but in no community of practice (the ultimate lurker if you like); overlapping as learners may be a member of several communities of practice; waxing and waning to emphasise the dynamics of the structures. These dynamics are captured by the notion of ad-hoc transient groups, communities of practice in statu nascendi, which may grow and last, or disappear. So, the concepts of a community of practice and a learning network are notably different, but interdependent, indeed complimentary.

For the sake of completeness and not unjustly to claim originality, others have stumbled upon this idea of a network as an ensemble of communities of practice. Jon Dron and Terry Anderson wrote a paper talking about collectives, networks and groups, more recently Carolin Haythorthwaite discussed networks, crowds and communities. No doubt, others have expressed similar ideas.

Note added October 23d. This blog post has been reproduced as a Tally Fox Insight.


References
Bandura, A. (1977). Social Learning Theory. Oxford, England: Prentice Hall.
Brown, J. S., & Duguid, P. (2000). The social life of information. Boston Mass: Harvard Business School University Press.
Dron, J., & Anderson, T. (2009). How the Crowd Can Teach. In J Dron & T. Anderson (Eds.), How the Crowd Can Teach. Handbook of Research on Social Software and Developing Community Ontologies. Hershey, New York: IGI Global.
Haythornthwaite, C. (2011). Learning Networks, Crowds and Communities. In Proceedings of the 1st International Conference on Learning Analytics and Knowledge, LAK’11 (pp. 18–22). New York, NY, USA: ACM.
Dunbar, R. I. M. (1993). Co-Evolution Of Neocortex Size, Group Size And Language In Humans. Behavioral and Brain Sciences, 16, 681–735. Retrieved from http://www.lifewithalacrity.com/2004/03/the_dunbar_numb.html
Jones, C. (2008). Networked learning: weak links and boundaries. Journal of Computer Assisted Learning, 24(2), 87–89.
Lave, J., & Wenger, E. (1991). Situated learning: legitimate peripheral participation. Cambridge: Cambridge University Press.
Sloep, P. B. (2013). Networked professional learning. In A. Littlejohn & A. Margaryan (Eds.), Technology-enhanced Professional Learning: Processes, Practices and Tools (p. 97-108). London: Routledge.
Sutcliffe, A., Wang, D., & Dunbar, R. I. M. (2012). Social Relationships and the Emergence of Social Networks. Journal of Artificial Societies and Social Simulation, 15(4), 3. Retrieved from http://jasss.soc.surrey.ac.uk/15/4/3.html