nach oben

2009 | Buch

Kapitel lesen Erstes Kapitel lesen

Epistemological Aspects of Computer Simulation in the Social Sciences

Second International Workshop, EPOS 2006, Brescia, Italy, October 5-6, 2006, Revised Selected and Invited Papers

herausgegeben von: Flaminio Squazzoni

Verlag: Springer Berlin Heidelberg

Buchreihe : Lecture Notes in Computer Science

Enthalten in: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik" , Springer Professional "Wirtschaft"

Einloggen, um Zugang zu erhalten

Über dieses Buch

This volume collects the revised versions of the invited and selected papers that were presented at the Second EPOS––Epistemological Perspectives on Simulation––Workshop, held in Brescia, Italy, in October 2006. EPOS is a bi-annual cross-disciplinary workshop on simulation originally established by Ulrich Frank and Klaus G. Troitzsch, with a first e- tion held in Koblenz in July 2004. EPOS aims to provide a forum for scholars from various disciplines, such as the social sciences, computer sciences, engineering and natural sciences, who are interested in discussing epistemological aspects of computer simulation across disciplinary boundaries. The common belief behind the workshop is the recognition that the time has come to seriously reflect on epistemological and methodological preconditions, processes and consequences of simulation as a research tool. During the fist edition in Koblenz 2004, a number of interesting topics were ca- fully addressed: the link between theory and simulation models, the empirical vali- tion of agent-based models in the natural and the social sciences, the relation between models and truth, as well as the role of stylized facts in evidence-based models. A good cross-disciplinary atmosphere permeated the workshop, making possible the exchange of knowledge and ideas beyond any disciplinary boundary. The first EPOS proceedings were edited by Ulrich Frank and Klaus G. Troitzsch and published in the Journal of Artificial Societies and Social Simulation, Vol. 8, No. 4, 2005.

Inhaltsverzeichnis

Frontmatter

EPOS-Epistemological Perspectives on Simulation: An Introduction

Abstract

There is strong evidence that computer simulation is increasingly recognized as an important analytical tool in many social sciences disciplines and fields. During the last ten years, a number of new journals, which are devoted to this field, have been founded and others have increased their influence (i.e., JASSS, CMOT, Social Science Computer Review, Autonomous Agent and Multi-Agent Systems, Journal of Economic Dynamics and Control, Computational Economics, Computational Management Science). Special issues and extensive reviews of the literature have been published in influential and standard journals. At the same time, new international associations and societies were born, with an increasing number of members (i.e., ESSA in Europe, NAACSOS in North America, The Society for Computational Economics), many research centers and institutes have been successfully launched, many workshops, conferences and congresses are organized every year (with the first world congress on social simulation held in 2006 in Tokyo and the second one in Washington in 2008), and an open market of tools and simulation platforms (i.e., Swarm, Repast, Ascape, NetLogo), based on a vast community of developers and users, is steadily growing.

Ulrich Frank, Flaminio Squazzoni, Klaus G. Troitzsch

Invited Papers

The Epistemologies of Social Simulation Research

Abstract

What is the best method for doing simulation research? This has been the basis of a continuing debate within the social science research community. Resolving it is important if the community is to demonstrate clearly that simulation is an effective method for research in the social sciences. In this paper, we tackle the question from a historical and philosophical perspective. We argue that the debate within social simulation has many connections with the debates that have echoed down the years within the wider social science community about the character of social science knowledge and the appropriate epistemological and methodological assumptions on which social science research should rest.

Nigel Gilbert, Petra Ahrweiler

From Simulation to Theory (and Backward)

Abstract

Of late, due to the perceived advantages of the generative paradigm (Epstein 2002, 2005; Arthur 2004), a generative variant of agent based modelling and simulation, i.e. Agent Based Generative Social Simulation (ABGSS), has received a great impulse. In this paper, weak ABGSS, i.e. the thesis that growing a social effect is necessary but insufficient to explain it, is supported. Casting a critical eye on the debate about generative simulation, it will be argued that ABGSS needs to be fed by, but at the same time provides feedback to, two theoretical complxements which must be formulated prior and independent of simulation itself: (a) bottom-up theory of general local rules, and of the process from them to macroscopic effects; (b) theory of downward causation, showing how local rules are modified by the effects they contribute to achieve. This twofold thesis will be carried out while discussing three main examples of social phenomena: the witness effect, Schelling’s segregation model and the ethnic homogeneity of violence, and the minority game.

Rosaria Conte

Selected Papers

Talking about ABSS: Functional Descriptions of Models

Abstract

Social simulation research lacks a common framework within which to integrate empirical and abstract models. This lack reflects an epistemological divide within the field. In an attempt to span that divide and in hope that it will lead to subsequent work on integrating abstract and empirical agent based social modelling research, I suggest here that a possibly suitable framework would derive from the mathematical notion of a function as a mapping between a well specified domain and a well specified range. The use of the function as an informal framework for the discussion of epistemological issues such as prediction, validation and verification is demonstrated as well as its use for structuring controversy about modelling techniques and applications. An example is drawn from the literature on opinion dynamics to explore the latter use.

Scott Moss

What Does Emergence in Computer Simulations? Simulation between Epistemological and Ontological Emergence

Abstract

Emergence is generally considered a fundamental property of complex systems. Being a central but notoriously ill defined notion, concepts of emergence fundamentally oscillate between epistemological and ontological interpretations. The paper relates these philosophical perspectives of emergence to the interpretation of emergence in computer simulation. It concludes that most arguments point to the fact that computer simulation deals with epistemological emergence only. However, there is no conclusive argument that computer simulation in principle is unable to model ontological emergence. Finally, the paper argues for mathematics being a restricted description what concerns all possible emergent levels not yet realised.

Alex Schmid

Emergence as an Explanatory Principle in Artificial Societies. Reflection on the Bottom-Up Approach to Social Theory

Abstract

This article investigates the notion of emergence in Artificial Societies. Roughly, two competing approaches to the foundations of social science exist: A micro foundation of social theory on the one hand and a notion of an emergent holistic social theory on the other. This dichotomy re-appears also in Artificial Societies. It will be argued that philosophical decisions made on the methodological level of how to interpret the concept of emergence will result in different sociological theories. This will be demonstrated by re-examining considerations on emergence undertaken by Joshua Epstein, who argues for a micro foundation of social theory. These considerations are then settled in the context of the long-lasting debates about emergence in sociology and philosophy of science. Considerations from the complexity theory and Philosophy of Science will be utilised to develop a concept of emergence which leads to the notion of an autonomous social sphere. It is demonstrated by two examples that this concept can be applied to Artificial Societies.

Martin Neumann

Reconstruction Failures: Questioning Level Design

Abstract

In front of unsuccessful models and simulations, we suggest that reductionist and emergentist attitudes may make it harder to detect ill-conceived modeling ontology and subsequent epistemological dead-ends. We argue that some high-level phenomena just cannot be explained and reconstructed from unsufficiently informative lower levels. This eventually requires a fundamental viewpoint change in not only low-level dynamics but also in the design of low-level objects themselves, considering distinct levels of description as just distinct observations on a single process.

Camille Roth

Narrative Scenarios, Mediating Formalisms, and the Agent-Based Simulation of Land Use Change

Abstract

The kinds of system studied using agent-based simulation are intuitively, and to a considerable extent scientifically, understood through natural language narrative scenarios, and that finding systematic and well-founded ways to relate such scenarios to simulation models, and in particular to their outputs, is important in both scientific and policy-related applications of agent-based simulation. The paper outlines a projected approach to the constellation of problems this raises – which derive from the gulf between the semantics of natural and programming languages. It centers on the use of mediating formalisms: ontologies and specialised formalisms for qualitative representation and reasoning. Examples are derived primarily from ongoing work on the simulation of land use change.

Nicholas M. Gotts, J. Gary Polhill

Validation and Verification in Social Simulation: Patterns and Clarification of Terminology

Abstract

The terms ‘verification’ and ‘validation’ are widely used in science, both in the natural and the social sciences. They are extensively used in simulation, often associated with the need to evaluate models in different stages of the simulation development process. Frequently, terminological ambiguities arise when researchers conflate, along the simulation development process, the technical meanings of both terms with other meanings found in the philosophy of science and the social sciences. This article considers the problem of verification and validation in social science simulation along five perspectives: The reasons to address terminological issues in simulation; the meaning of the terms in the philosophical sense of the problem of “truth”; the observation that some debates about these terms in simulation are inadvertently more terminological than epistemological; the meaning of the terms in the technical context of the simulation development process; and finally, a comprehensive outline of the relation between terminology used in simulation, different types of models used in the development process and different epistemological perspectives.

Nuno David

Validation and Verification of Agent-Based Models in the Social Sciences

Abstract

This paper considers some of the difficulties in establishing verificaction and validation of agent based models. The fact that most ABMs are solved by simulation rather than analytically blurs the distinction between validation and verification. We suggest that a clear description of the phenomena to be explained by the model and testing for the simplest possible realistic agent rules of behaviour are key to the successful validation of ABMs and will provide the strongest base for enabling model comparison and acceptance. In particular, the empirical evidence that in general agents act intuitively rather than rationally is now strong. This implies that models which assign high levels of cognition to their agents require particularly strong justification if they are to be considered valid.

Paul Ormerod, Bridget Rosewell

Abductive Fallacies with Agent-Based Modeling and System Dynamics

Abstract

Increasing usage of computer simulation as a method of pursuing science makes methodological reflection immanently important. After discussing relevant philosophical positions Winsberg’s view of simulation modeling is adapted to conceptualize simulation modeling as an abductive way of doing science. It is proposed that two main presuppositions determine the outcome of a simulation: theory and methodology. The main focus of the paper is on the analysis of the role of simulation methodologies in simulation modeling. The fallacy of applying an inadequate simulation methodology to a given simulation task is dubbed ‘abductive fallacy’. In order to facilitate a superior choice of simulation methodology three respects are proposed to compare System Dynamics and Agent-based Modeling: structure, behavior and emergence. These respects are analyzed on the level of the methodology itself and verified in case studies of the WORLD3-model and the Sugarscape model.

Tobias Lorenz

Algorithmic Analysis of Production Systems Used as Agent-Based Social Simulation Models

Abstract

Algorithmic analysis of models is a standard tool in general, but is rarely attempted in the context of computer models for agent-based social simulation. We explore the algorithmic analysis of simulation models that take the form of production systems as defined in computer science. Several implemented analysis algorithms for a particular type of production system are described, including algorithms for model abstraction and for agent discovery. Examples of the use of these algorithms are given and their significance and potential considered. In particular, it is explained how an algorithm for model abstraction, developed in the context of production system models, has been successfully applied to the Iruba model of a guerrilla war, a complex multi-agent model programmed in C, a general purpose programming language.

Jim Doran

The Nature of Noise

Abstract

The idea of noise is now widespread in many fields of study. However to a large extent the use of this term is unexamined. It has become part of the practice of science without entering to a significant extent as part of its explicit theory. Here I try to produce a clearer and more coherent account of the term. I start with a picture of noise from electrical engineering. I then generalise this to the widest conception: that of noise as what is unwanted. A closely related conception is noise as what is unexplained. A particular case of this later usage is where a source of randomness can be used to stand-in for this residual. I argue that noise and randomness are not the same. I explore the possible relation between noise and context, and propose a new conception of noise: namely that noise is what can result from an extra-contextual signal. I finish with an application of the analysis of noise to the relation of determinism and randomness.

Bruce Edmonds

Backmatter

Titel: Epistemological Aspects of Computer Simulation in the Social Sciences
herausgegeben von: Flaminio Squazzoni
Verlag: Springer Berlin Heidelberg
Electronic ISBN: 978-3-642-01109-2
Print ISBN: 978-3-642-01108-5
DOI: https://doi.org/10.1007/978-3-642-01109-2