A Network Approach to Social Learning: An Application to Decision-Making in a Socio-Environmental System

• In Event: 099 - Section on Mathematical Sociology Paper Session. Open Topic on Mathematical Sociology

Abstract

This paper examines networks of mental models to understand deliberation and ‘social learning’ in a small group charged with collaboratively managing natural resources. Participants included ranchers, conservationists, and rangeland managers. They were asked to develop a management plan for the cattle present on a piece of California rangeland. We surveyed 4 groups of respondents (a homogeneous group from each category of ranchers, conservationists, and government rangeland managers as well as one mixed group) about the management decisions they would make. In particular, we were interested in how these respondents linked the methods they would use to the goals they set. This perspective results in a bipartite network of goals and methods. They then discussed their recommendations in small groups. After the group discussion, we then asked each individual to make any changes they felt were appropriate to their network of responses about goals and methods that they had given before discussion. Our dataset consists of before and after networks of each individual’s perception of the problem. We then apply a Bayesian construction of the Cultural Consensus model (Butts 2003) to examine similarities in the group and the network structure of what individuals learned from their group discussion. We use baseline models to examine what kinds of change occurred in each individual’s network. We find that the most change and ‘learning’ occurred not in adding new goals and methods, but adding new relationships between the goals and methods already understood.

Author

• Lorien Jasny, University of Maryland