Computational models and social simulations

In a world where data are becoming everyday more easily available, such business and social sciences as management, economics, sociology and psychology are struggling to better identify and understand how human behaviors, everyday interactions and decision-making processes are related together: indeed, the main actor of social systems – the human being – still remains today the most important and unpredictable factor.

However, the standard methods of investigation commonly applied by the aforementioned disciplines, even considering a perfect knowledge of individual decision-making rules and behaviors, does not guarantee the possibility of predicting the macroscopic structure (Epstein, 2007) and the emergent properties of the complex systems that we see in reality (e.g. economic markets, crowds, online communities).

To fill this gap, new methods of research, in line with the recent computational power resources, have arisen during the last decade during the last decade: in fact, Computational Social Science, an ambitious interdisciplinary field, is able today to provide advanced computing techniques capable of “growing up” phenomena from the bottom-up (e.g. from individual behaviors to groups formation, from isolated businesses to innovation diffusion). These methods, grounded on the paradigm of the Generative Science (Epstein, 2007), have proven to bring to counterintuitive insights about social and organizational phenomena.

Among these methods, Social Simulations (also known as Agent-Based Models or ABMs) are “a type of computer simulation that is used to study how micro-level processes affect the macro-level outcomes” (Hughes et al., 2012). Undeniably, their ultimate value lies on the ability to investigate social processes and emergent phenomena, which would be too complex (or even, unethical) to investigate in the reality, by recreating in a virtual world individual and social behaviors through dedicated programming languages and computer softwares (e.g. NetLogo, SeSam. Repast).

In this way, APRESO encourages the development of computational models, such as Social Simulations, which specifically aim to address social issues or to recreate and investigate organizational social processes. APRESO intends to support and foster the scientific knowledge through innovative methodologies on themes of interested for the social and organizational research and consultancy.

As an example, the main themes investigated so far by means of Social Simulations include:

  • Teamwork and team performance
  • Conflict and negotiation
  • Environmental and ecological-economic modeling
  • Innovation diffusion
  • Prosocial and selfish behavior
  • Cultural change in society and in organizations
  • Group formation
  • Opinion dynamics
  • Models of trust and reputation
  • Dynamics of organizational populations
  • Policy modelling
  • Computational models of economic processes
  • Models of social segregation and inclusion
  • Social network dynamics
  • Emergence of social norms,
  • Developing of cognitive models for plausible agents

To those who might start working on computational simulations we would suggest reading the guide provided by Axelrod and Tesfatsion (2005), “A Guide for Newcomers to Agent-Based Modeling in the Social Science”. Furthermore, the following resources constitute a valuable starting point to discover the scientific proposal of the Generative Science and the contribution bring to the research on social and organizational systems born thanks to the Computational Social Science:

  • Buchanan, M. (2009). Economics: Meltdown modelling. Nature, 460(7256), 680–682.
  • Cioffi-Revilla, C. (2014). Introduction to Computational Social Science. Principles and Applications. London, UK: Springer-Verlag.
  • Epstein, J. M. (2007). Generative Social Science: Studies in Agent-Based Computational Modeling. Princeton: Princeton University Press.
  • Epstein, J. M. (2008). Why Model?. Journal of Artificial Societies and Social Simulation, 11(4).
  • Gilbert, N. (2008). Agent-Based Models. SAGE Publications, 153(153), 98.
  • Gilbert, N., & Troitzsch, K. G. (2005). Simulation for the social scientist (2nd ed.). Maidenhead, United Kingdom: Open University Press.