Workshop on Models - Abstracts

Gabriele Contessa|
LSE
Scientific Models, Representation and Similarity

In this paper, I develop and defend an account of how models represent real systems in which the notion of similarity plays a central role. A similarity account of scientific representation has been proposed by Ronald Giere. In the paper, I attempt to apply Giere's account to a concrete case in which we want to use a model to predict a specific aspect of the behaviour of a specific system. I argue that Giere's account fails to account for this and similar cases. I then offer an alternative account in which the notion of similarity plays a different role.

Erika Mattila|
University of Helsinki
From properties of models to a simulation platform: Lifespan of an infectious disease model

How to build an individual-based simulation model on Hib-diseases and programme a computer interface related to that model? This question lies behind a micro-scale analysis, which focuses on how the properties from a set of Hib-related models were integrated into a single model on Hib (Haemophilus influenzae type b). Along the integration process, the analysis also reconstructs how a computer-interface, a simulation platform, based on this model, was build for predictive uses. The thrust for a detailed analysis of the properties of models and how they "travel" into a single integrated model is mainly motivated by Boumans's (1999) prominent view in his analysis of the ingredients of a small business cycle model. I will pay a special attention to the integration practices, like optimisation and calibration, and discuss, how the properties of the previously built models function in the integration process.

Damien Fennell|
LSE
Mathematical Equations and their Causal Interpretations: Herbert Simon's Concept of Causal Order

This paper begins by presenting a problem in which mathematically acceptable manipulations of a set of linear deterministic equations changes the intuitive causal reading intended for those equations. In order to resolve this problem, it develops Herbert Simon's work on causal order to present an interpretive schema for sets of deterministic linear mathematical equations according to which their causal content, as models, can be set out explicitly. With this in place, it then identifies and critically discusses certain properties of the key causal concepts that are assumed when interpreting mathematical equations using Simon's methods.

Tarja Knuuttila|
Helsinki School of Economics
Parsers as Epistemic Artefacts

Parsers are language-technological models that assign morphological and syntactic structure (but not semantic interpretation) to written input texts of any length and complexity. A parser provides an example of a model the scientific status of which is neither predominantly tied to its representative function nor based on any explicit theory. As such it poses the question of how its epistemic value should be assessed. I shall argue that a promising way to approach models like parsers is to treat them as epistemic artefacts, that is, as constructed things that can give us knowledge in various ways and which also, in themselves, provide us new objects of knowledge.

Share:Facebook|Twitter|LinkedIn|