Display Options

Text Size

Layout

NOTE: To use the advanced features of this site you need JavaScript turned on.

Home Research Repeated Interaction
Repeated Interaction

Investigator: Edmund Chattoe-Brown

Where many theories exist across the social sciences, how can simulation be used to integrate and compare them so that social science as a whole can progress?

Social actors may reason about what it is best for them to do but sometimes individual rationality, as a result of collective action problems, does not lead to socially preferred outcomes (Dresher 1961, Hardin 1968). One recognised solution to the problem of social order which interests anthropology, sociology, economics, psychology, management and political science is the effect that repeated interaction has on people's willingness to cooperate when compared to the "one shot" case (Axelrod and Hamilton 1981, Axelrod and Dion 1988).

In repeated interaction, actors can have their reputations tarnished, can be ostracised, punished, imitated and so on. This insight has led to a huge and diverse literature on repeated interaction in games and especially the Prisoner's Dilemma (Poundstone 1992). However, the methodological issue this raises for simulation (which has not been addressed by any other theory building tool either) is how do we synthesise competing theories effectively to develop a progressive research agenda?

Each theory about how group cooperation occurs through networks (Tomochi 2004), spatial regularity (Lindgren and Nordahl 1994), choice and refusal of partners (Stanley et al. 1993), reputation (Yao and Darwen 1999), learning (Macy 1999) and many other mechanisms has prima facie plausibility but what we want to know is how, in real social contexts, do these different processes operate together to generate distinctive patterns of co-operation and competition?

This strand of research will involve developing a synthetic simulation of repeated strategic interaction based on the existing literature. This will serve three purposes:

  • To identify lacunae in the framework for studying strategic interaction based on the current existence of multiple partial theories
  • To calibrate and validate the model as far as possible using existing data while identifying gaps in that data with a view to framing appropriate data collection tasks and research questions for future research
  • To develop a methodology (based on a substantive case study) for using simulation to synthesise social science theories and thus facilitate progressive interdisciplinary research in other areas

The objective of this strand of research (like the others) is simultaneously to meet a substantive challenge in social science (contributing to knowledge of repeated strategic interaction) and a methodological challenge to simulation (synthesis of existing theories).