Models and Representations in Science

Models are essential for the acquisition and organization of scientific knowledge, for they are one of the most important instruments of scientists. Models are commonly used to better understand the entities they stand for. But for this to be possible models must represent those entities. Thus, philosophically, models raise questions in semantics (what is the representational function that models perform?), ontology (what’s the representational reality of models?), epistemology(how do we know that models represent the actual world?), and, of course, in philosophy of science (how do models relate to theory and to the natural world?).

 
The purpose of the workshop is to address these kinds of questions or similar ones that arise in the context of assessing the role of scientific models. Thus, scholars interested in this topic and covering different aspects of the issue (formal, logical, mathematical, semantical, ontological, epistemological…) are kindly invited to participate to this workshop.
 
Working language: English
Organizers: Sorin COSTREIE & Richard DAVID-RUS
 
CURRENT SUBMISSIONS (31st of august)

Descriptive imaginary and the theory of models in natural science

Chiriac Horia Costin
Romanian Academy, Iasi Branch

	 ACKNOWLEDGEMENT: This paper was made within The Knowledge Based Society Project supported by the Sectoral Operational Programme Human Resources Development (SOP HRD), financed from the European Social Fund and by the Romanian Government under the contract number POSDRU 89/1.5/S/56815.  

The dynamics of scientific representations in natural science is a quite complex phenomenon which in the last decade became the main subject for the theory of scientific models. In order to develop a coherent theory about the way in which models represent parts of the physical real, philosophers and historians of science extended the notion of model, superposing it partially over those of theory or representation. Today this theory of scientific models gained remarkable conceptual refinement and variety, becoming a central theme in philosophy of science literature. However, the evolution of scientific representations involves social, psychological and cultural aspects that call for further conceptual innovation by philosophers of science. The present paper aims to underline the possible link between the theory of scientific models and the concept of descriptive imaginary, focusing on the major stylistic changes in designing scientific models on one hand and in conceiving their relation with experimental data on another hand. We believe that the use of descriptive imaginary as an investigation tool will contribute to the better understanding of the historical evolution of scientific models in natural science, the so-called scientific revolutions becoming real turning points in the dynamics of descriptive imaginary.

Models of the Mind

Valentin Sorin Costreie
Romanian Academy, Iasi Branch

ACKNOWLEDGEMENT: This paper was made within The Knowledge Based Society Project supported by the Sectoral Operational Programme Human Resources Development (SOP HRD), financed from the European Social Fund and by the Romanian Government under the contract number POSDRU 89/1.5/S/56815. 

My presentation tackles the notion of “mental representation”, which is a fundamental concept of the Computational Theory of Mind, according to which cognition fundamentally involves representations. My aim is to see what kind of ‘representations’ do cognitive scientists have in mind in this case, and so to critically assess these various notions of ‘mental representation’.

Understanding through modeling – some considerations

Richard David-Rus
Romanian Academy, Iasi Branch

	 ACKNOWLEDGEMENT: This paper was made within The Knowledge Based Society Project supported by the Sectoral Operational Programme Human Resources Development (SOP HRD), financed from the European Social Fund and by the Romanian Government under the contract number POSDRU 89/1.5/S/56815.  

The topic of scientific understanding attracted the attention of the philosophers of science in recent years. Though in its infancy, there are already positions claiming or denying the independence and specificity of the topic from the subject of scientific explanation. In my presentation I will discuss some recent accounts regarding the above debate and will contextualize it to the subject of scientific modeling. I will argue for the pertinence and importance of an approach on non-explanatory understanding in a modeling context.

Semantics in terms of "model transformers"

Alexandru Dragomir
University of Bucharest

My presentation will concern the dynamic "turn" in modal and epistemic logic. The semantics of dynamic logics is offered in terms of "model transformers" (or "epistemic state transformers") that guide the changing of one Kripke model that describes an epistemic situation into another. Such a design makes the logics capable of representing the dynamics of knowledge and the information flow in a group of agents.

 

Possible and Impossible Models

Craiţa Florescu
University “Alexandru Ioan Cuza”, Iasi / POSDRU 63663 “CommScie”

	 This research has been funded by the project POSDRU/89/1.5/S/63663 "CommScie" 

This presentation will be focused on model-theoretical issues. I am working on a proper understanding of the distinction between possible and impossible worlds, and also on the construction of a mechanism that should be able to generate both these two types of worlds, and the principle distinguishing them from each other. According to my hypothesis, this principle should also generate two accessibility relations, one for the normal, or possible worlds, and one for the impossible worlds. It would be shown, then, that notions like our usual concept of “necessity” do not depend semantically on the non-normal accessibility relationship, which is another way to say that the non-normal worlds are not connected to the actual world through the established semantic means, but through other means, which do not affect the scope of the typical concept of “necessity”. Thus, the laws considered as „necessary” in the actual world could be proven to be necessary regardless of our ability to imagine impossible worlds where non-normal things would happen - for instance, one and one would add to something other than two.

Scientific models – between empirical adequacy and contribution to scientific knowledge

Corina Grigoriu
University of Bucharest

The complexity of the phenomena sometimes makes it necessary for scientists to develop idealized theoretical models, in order to better understand, systematize, explain, and predict these phenomena. The models are constructed by means of disregarding certain factors which are believed to be irrelevant to the question in hand. But models which take into account only a few parameters are faced with the charge of ceasing to represent the phenomenon they were meant to represent. And we are thus left with the dilemma of how a model that does not really represent a reality can tell us something relevant and increase our understanding of it. On the other hand, complex models which may be said to be empirically adequate seem to be unable to perform another important function of a theoretical model: that of shedding light on the underlying mechanisms of the phenomenon in question. What I wish to discuss is precisely this relationship between a model’s empirical adequacy and its contribution to the acquisition of scientific knowledge. 

Scientific Models and Mathematical Understanding

Gabriel Tarziu
Romanian Academy, Iasi Branch

	ACKNOWLEDGEMENT: This paper was made within The Knowledge Based Society Project supported by the Sectoral Operational Programme Human Resources Development (SOP HRD), financed from the European Social Fund and by the Romanian Government under the contract number POSDRU 89/1.5/S/56815.  

What I am aiming at in this paper is to offer an elucidation of the role of mathematics in scientific understanding by analyzing the way mathematics is used in scientific models. Does the mathematics used in science convey understanding in any way? Can we take it as explanatory? A superficial look at this issue reveals three potential answers. The first and maybe the most persuasive one is that mathematics doesn’t convey understanding in a scientific context. This answer is most attractive for someone who adopts a causal model for scientific explanation. From this perspective “underlying causal mechanisms hold the key to our understanding of the world” (Salmon 1984: 260). Another answer takes mathematics as explanatory and the type of understanding mathematics is seen as producing in a scientific context to be identical with what we find in a purely mathematical context. At first sight, the details of this answer can be specified in terms of Kitcher’s unificationist account of explanation. On this account, what increases our understanding is unification of regularities. From this perspective, understanding in mathematics is not so different from understanding in science. Another answer takes mathematics as explanatory but sees the understanding conveyed by it as being of a special kind. A large part of my paper will consist in spelling out the details of this kind of understanding. As I take it, mathematics provides structural understanding of physical phenomena. In order to show this, I will argue, we have to analyze the role of mathematics not in scientific theories but in scientific models, especially in those cases where models help in applying theories that are otherwise inapplicable.

 

Scientific Understanding Through Idealization Control

Iulian Toader
University of Bucharest

I consider the claim that scientific idealization indicates not that some fictional parameters of a physical system are necessary for explanation, but rather that some real parameters are unnecessary. For example, friction is unnecessary for explaining certain mechanical phenomena, and neglecting it does not prevent one from answering relevant questions about what causes these phenomena, but allegedly helps one do so. I contend, however, that there are concrete cases in foundational physics that seem to show not that real parameters (e.g., finite volume) are explanatorily unnecessary, but that fictional parameters (e.g., infinite volume) are explanatorily necessary. Such are indicated, for example, by some limit-taking procedures in the thermodynamics of equilibrium processes. I further argue that even if one resists such counterexamples, it still does not follow that answering relevant causal questions by neglecting some real parameters is the best approach to scientific understanding. For fictional parameters, even if they are not explanatorily necessary, may provide the same epistemic quality of knowledge, but may do so more efficiently than real parameters, even if the relevant causal questions remain unanswered. Finally, I propose and defend an account of scientific understanding based on what I call idealization control, which obtains if one can establish that the deviation from actual observation of scientific results obtained through idealization is smaller than what may be due to experimental error. 

Dynamic mental representations and the causal role challenge

Cristinel Ungureanu
Independent Researcher

Embodied mind theory (EMT) represents a significant change for the cognitive sciences: from the textual paradigm of the classic cognition to the processual one of the dynamic systems theory. According to the new paradigm, the relation between mind and world could not be a linear one. Mental states are not static symbols standing in semantic relation with external objects, but topological entities, evolutions in space-time that constitute their objects. Some radical upholders of EMT argue that, given the dynamic relation of the cognitive agent with the world, there is no place for the concept of “mental representation” in the scientific explanations (Chemero 2009). In a recent paper we have advanced the thesis that the dynamic agent-world interaction can be interpreted as being representational. On a certain threshold of complexity, the dynamic system develops stable attractors and attractor landscapes which could be interpreted as standing for something outside the system. In the dynamic paradigm, the mental states are not to be identified with the bare neural states, but with the higher-level dynamic states emerged from the brain-body-world coupling. In this paper we mean to deepen this thesis by answering to the following possible philosophical objection: if representations are not reducible to bare neural states, how is their causal relevance possible? Given the specificity of causation in a dynamic system, we will show that mental representations manifest their causal role downwardly – as order parameter (by setting limit conditions for the neural activity) and upwardly – as control parameters (by determining the emergence of the meta-representational level). Only the meta-representational level certifies the existence of mental representations.

TFE
-->