Computational simulations are now widely employed to study the behaviour of social systems, examples being market behaviours, and social media population behaviours. These methods rely heavily on game theoretical modelling, usually employing populations of software agents to emulate the behaviour of human populations. Researchers construct models, usually based on known games, and empirical data, and use these to explore how the population reacts to changes. Many behaviours that are not well understood in social systems can be accurately captured and understood using these techniques.
An example is the Groupthink problem, that is commonly defined as a psychological phenomenon arising within a group of people, where the pursuit of conformity in the group results in an irrational decision-making outcome. It has been well studied in the management and psychology literature. The best course of action to maintain group cohesion leads to a very poor outcome for the group. A project exploring Groupthink would develop new computational methods for modelling Groupthink that can account for real-world complexity in agents’ behaviour, and build realistic enough models that can fit past and present empirical instances.
Related Reading: Avoiding the Pitfalls of the Dunning-Kruger Effect and Groupthink
Required knowledge
This project is suited for a applicants with good skills in mathematics and computing, and an interest in game theoretic modelling methods. PhD applicants with honours or research masters degree in an applicable computing discipline should contact me by email at Carlo.Kopp@Monash.edu. Curriculum vitae, references and an academic transcript of results are to be presented on request. For PhD applicants for whom English is not their first language, documentary evidence of English proficiency is required (IELTS 6.5).
