Pascal Roggero

Using Soclab for a Rigorous Assessment of the Social Feasibility of Agricultural Policies

This paper presents a theoretical and methodological framework to take into consideration the social dimension in a sustainable development project. To do this, the authors have developed the SocLab software environment, which implements a formalization of a well-established sociological theory, and enables the modeling of social organizations, to analyze their properties and to simulate social actors’ behaviors. SocLab was used to assess the social acceptability of new agricultural practices more in line with the preservation of water resources and natural environments, in a well defined context. The paper shows how it was used and presents the main results.

Compte rendu d’une recherche interdisciplinaire entre sociologues et informaticiens. De la sociologie de l’action organisée au logiciel SocLab

We report a collaboration between sociologists and computer scientists about the formalization of a well-experienced theory, the sociology of the organized action. After a brief introduction to this theory, the paper presents our formalization of this theory and a sample example of its application. The following sections address the validity of the results obtained in this way, the scientific outcomes and the requirements of an interdisciplinary collaboration between sociologists and computer scientists.

La modelización y simulación computacional como metodología de investigación social

El presente trabajo se propone brindar una introduccion critica y sistematica a la metodologia de modelado y simulacion computacional en ciencias sociales. El uso de metodos computacionales para simular procesos sociales se ha expandido con vigor en los ultimos veinte anos en las distintas disciplinas de las ciencias sociales; sin embargo, continua siendo una metodologia poco conocida y empleada por los investigadores sociales en America Latina. El objetivo de este trabajo es contribuir a la reflexion epistemologica, teorica y metodologica de la simulacion computacional en ciencias sociales, atendiendo a la dimension politica que atraviesa toda praxis cientifica.

Cooperation is not always so simple to learn

In this paper, we propose to study the influence of different learning mechanisms of social behaviours on a given multi-agent model (Sibertin-Blanc et al. 2005). The studied model has been elaborated from a formalization of the organized action theory (Crozier and Friedberg 1977) and is based on the modelling of control and dependency relationships between resources and actors. The proposed learning mechanisms cover different possible implementations of the classifier systems on this model. In order to compare our results with existing ones in a classical framework, we restrain here the study to cases corresponding to the prisoner’s dilemma framework. The obtained results exhibit variability about convergence times as well as emergent social behaviours depending on the implementation choices for the learning classifier systems (LCS) and on the LCS parameters. We conclude by analysing the sources of this variability and giving perspectives about the use of such a model in broader cases.

Modelos basados en agentes: aportes epistemológicos y teóricos para la investigación social

Agent-based models (ABM) constitute a new generation of computer based methods that allow the modeling of the structure of a complex system and simulation of its dynamic evolution over time. The use of ABM constitutes a methodological tendency expanding in contemporary social sciences; however, these models remain quite unknown and not frequently taught in the field of sociology. They therefore represent a minor methodological alternative among social researchers. The purpose of this work is to introduce social scientists to the main ideas of the agent-based models as they are articulated with certain theoretical and methodological problems of the social sciences. The first section puts into question the relationship between agent-based models and complex systems in a critical epistemological perspective. Subsequently, the contributions of ABM to social research are analyzed, and in the third section, their application in the framework of a specific discipline, political science, is assessed. Finally, a practical example of a simulation based on agents from the classic work of Thomas Schelling about racial segregation is developed.

Sur le lien entre pensée et système complexes

La these defendue est celle de la pertinence de l’articulation entre la pensee complexe et l’approche des systemes complexes. Alors que la premiere pourrait trouver dans la seconde la dimension methodologique semblant lui manquer, la seconde pourrait emprunter a la premiere un riche cadre epistemologique qui apparait lui faire defaut. Deux projets en cours temoignent de tentatives pour concretiser cette articulation.