Nature inspired adaptive strategies for information sharing: individual vs. social information

Social learning is an effective way to reduce uncertainty about the environment, helping individuals to adopt adaptive behaviour cheaply. Although this is evident for learning about temporally stable targets, such as acquisition of an avoidance of toxic foods, the utility of social learning in a temporally unstable environment is less clear, since knowledge acquired by social learning may be outdated. An individual can either depend entirely on its own foraging information (individual forager) or that provided by the environment or shared by other agents. We are interested in scenarios where individual foraging might be a useful and effective strategy and how the topology and distribution of resources in the network/environment might affect this.

The work has advanced through examination of the behavioural, evolutionary and game-theoretic underpinnings of well-known biological social systems, and used to understand distributed applications through the creation of a simulation environment, in which the similarities between existing and proposed adaptive information dissemination protocols and the foraging behaviour of ants, bees and similar creatures is modelled to investigate the dynamics of social interaction in the context of resource discovery and information dissemination.

The simulation model discussed should hopefully help us understand some of the contexts in which cooperation emerges, is beneficial or not, and to what extent. Also, we should understand how the various costs can be minimised/optimised, by implementing dynamic strategies.

Ongoing implementations include scenarios like modelling trust in the system, altruism, and misinformation or malicious agents.