Navigation – Plan du site

AccueilTransversalitésDossiers200112ème Colloque Européen de Géogra...Facing urban complexity: towards ...

2001
12ème Colloque Européen de Géographie Théorique et Quantitative, St-Valéry-en-Caux, France, 7-11 septembre 2001
209

Facing urban complexity: towards cognitive modelling. Part 1. Modelling as a cognitive mediator

Saisir la complexité urbaine : vers une modélisation cognitive. Partie 1. La modélisation comme médiateur cognitif
Sylvie Occelli

Résumés

Depuis vingt ans les questions de la complexité sont un sujet de réflexion important dans la modélisation des systèmes urbains. Bien qu’elles aient contribué à une révision critique des approches couramment appliquées, leurs conséquences sur l’efficacité (et le sens) de l’activité de modélisation ont été rarement discutées. La prise en considération de la complexité dans la modélisation des systèmes urbains, néanmoins, a influencé aussi bien les sujets de recherche que l’activité de modélisation en leur donnant de nouvelles potentialités. Quoique reconnues, celles ci sont encore insuffisamment explorées. Nous suggérons que ces potentialités tiennent à un changement général de l'activité de modélisation, d'un rôle "structural", à un nouveau rôle, que nous appelons "cognitif" (le structural/cognitive shift). Cet article présente les premiers résultats d’une réflexion visant à illustrer les raisons de ce changement et ses conséquences sur la modélisation. Une conséquence de ce déplacement relève spécifiquement du nouveau rôle de la modélisation en tant que médiateur cognitif entre les processus d’abstraction propres à une activité de modélisation et l’environnement dans lequel elle se place. Ceci ouvre de nouvelles perspectives à la modélisation des phénomènes spatiaux.

Haut de page

Texte intégral

Introduction

1Over the last twenty years, complexity issues deeply affected (also) modelling approaches in geography, i.e. they pointed out counterintuitive and unexpected system behaviours, they exposed the question of the relationships between the ontological and epistemological dimensions ; they disclosed new horizons for modelling and fostered a quest for an exploration of the unexpected and emergent aspects of urban life (Batty and Xie, 1996, Casti, 1997, Bertuglia and Staricco, 2000).

2Whereas opening new ways of reasoning and favouring the development of new methods of analysis, the reactions of the modelling field to those issues were rarely questioned (Batty, 1994). In very few instances the effectiveness (and sense) of modelling activity was addressed (Wegener, 1994).

3Acknowledgment of complexity has been a fruitful spur for developing new and more sophisticated methods in order to improve understanding and advance geographical sciences. However its contribution to tackle urban problems in everyday life has been rather poor and mainly limited to rhetorical claims about the potentialities of the new approach (Clark, Perz-Treio and Allen, 1995).

4Recognition of the complexity issues in urban systems impacted on modelling in several ways (Occelli, 2001a) :

  • First, on a substantial ground it helped to recognize certain system behaviours, which looked counterintuitive or difficult to be clearly stated even at a conceptual level (Casti and Kalqvist, 1986).

  • Second, on a methodological ground it exposed the question of the relationships between the ontological and epistemological dimensions underlying the analysis of urban problems (i.e. the double faced statement : complexity is an intrinsic feature of phenomena and complexity depends on the knowledge of an observer).

  • Third it contributed to shake the once compact urban modellers’ community. The preoccupation about the different emphasis to be given to the continuum between information and knowledge (and within the knowledge domain to the procedural and declarative type of knowledge) was reflected in a flourishing of different modelling approaches and styles (i.e. theoretical urban modelling, GIS, SDSS, NN, geocomputation) (Longley et al., 1998).

  • Finally, the need to explore the different ways of deployment of interacting behaviours, disclosed new horizons for modelling, fostering a quest for an exploration of the unexpected and emergent, often at the frontier between the known and the unknown (i.e. the researches in artificial life and artificial society, see Conte, Hegselmann and Terna, 1997, Eptsein, 1996, Gilbert and Troitzsch, 1999).

5We argue that although complexity has put the classical modelling activity in a kind of distress, it is also disclosing new potentialities, which so far are still largely unnoticed. These are primarily related to what the authors have called the structural-cognitive shift, and involve both the contents and general role of modelling activity (Occelli and Rabino, 1999a).

6This paper is a first part of a work aimed to illustrate the main features of this shift and discuss its main consequences on the modelling activity. In the paper the most noticeable aspects of novelty resulting from this shift are discussed. These are related to the new role of modelling which acts as a cognitive mediator, between the components of an abstraction processes as they act in the classical role of modelling (the so called internal loop of a modelling activity) and the external environment to which modelling activity belongs (the so called external loop).

7An effort is made to emphasize the new potentials of the cognitive mediation underlying a modelling activity in tackling the complexity issues.

Modelling as a cognitive mediator

8The major trends of change now occurring in the modelling field can be epitomized in what has been recently called the structural-cognitive shift (Occelli, 2001b). This basically refers to a broader view of the very concept of modelling (Occelli and Rabino, 1999b) :

  • from a notion according to which modelling is an activity through which an understanding of the organizational structure of an urban system is obtained (the structural perspective) ;

  • to a notion in which modelling is an activity for testing, exploring, creating and communicating knowledge about certain urban phenomena (the cognitive perspective).

9This has several implications, relating to the socio-cultural, epistemological and technological aspects involved in a modelling activity (Occelli, 2000, 2001a, 2001b, Occelli and Rabino, 2000), which result in changes in scope, role and functions of modelling.

10These changes affect the two fundamental loops of a modelling activity (Occelli, 2000, 2001b), i.e. the internal loop (modelling process) concerning the conventional steps underlying a process of abstraction and the external loop (modelling domain) representing the general context of modelling.

  • 1 In the Figure, the lower loop, called internal, refers to the conventional steps underlying a proce (...)

11Figure 1 describes the two loops showing how they are centred on the observable, the phenomena in the geographical world, which are subject to investigation1.

12A major consequence of the structural/cognitive shift is that modelling does not refer only to the internal loop but it involves to a greater extent the components of the external loop.

Figure 1 : A scheme of the modelling activity : the internal and external loops

  • 2 A few aspects of this discussion as far as the spatial analysis is concerned are dealt with in Case (...)

13A number of arguments can be identified to support this claim2. Here we would suggest that there are at least three main sources, which concern issues of cognition, knowledge epistemology and consciousness.

14We will briefly hint at them in the following.

151. The advancements in the field of cognitive sciences, and in particular those dealing with the relationships between the contrasting approaches of naturalistic explanation of mind and hermeneutic understanding of meaning (see, Havelange, 1998). In this respect, some major research questions likely to provide significant contributions to modelling are related to :

  • the formulation of an objective approach to the question of mind, able to overcome the axiomatic features of mathematical formalization. This was based on a number of steps, i.e. the working of mind was related to that of a formal language, the attribution of meanings was solved by establishing correspondences between the formal system and the real world and the introduction of a spatialisation of mind by means of computing machine ;

  • the recent developments in the hermeneutic approach, in which a number of themes were increasingly concerned with cognitive issues, i.e. the importance of human action in the sociological theory, the reflexivity as a rational control for actors’ behaviour, the temporal and contextual location of action, the language as a medium of practical activity (ordinary language cannot be replaced by a technical language which would solve its imprecision and ambiguities), the tacit (or considered as consolidated) comprehension ;

  • an enlargement of the idea of the centrality of the technical mediation in that techniques offer a novel point of view to approach the phenomena of mind as they allow us to tackle the material inscription of action and social structuring (Vink, 2000).

162. The current debate about the epistemological issues concerning the distinction/articulation between the ontological and epistemological dimensions in an analytic enquiry. This relates to unsolved questions underlying the contrasting approaches conventionally held in relating human experience and the external world, i.e. the subjectivist’s in which what matters is the individual subject and its experience and the objectivist’s in which a world reality is postulated which is observer-independent. The need to reconcile these contrasting views has been at the basis of a number of approaches, such as constructivism or epistemological subjectivism. A major consequence in terms of the structural-cognitive shift, is that modelling (and primarily the encoding and decoding processes) requires to make explicit its main dimensions of analytic enquiry, i.e. (Occelli and Rabino, 2000) :

  • The degree of explanation we expect to achieve by means of a modelling activity i.e. the level of insights or knowledge level we (as analysts or urban system experts) should be satisfied with. Both the holism/reductionism and predictable/unpredictable dualisms are involved in this dimension :

  • The kind of system structure we refer to in our investigation, that is the kind of degree of control on the actors’ decisions we build into the model, i.e. the control mechanisms underlying market and state regimes, the interactions occurring between micro and macro levels.

  • 3 Searle (2000) summarizes a few salient traits of the current debate about consciousness : ’Consciou (...)

173. A last but not least relevant topic, which is also related to the previous ones, is the increasing importance of the questions concerning consciousness3. In spite of its major acknowledgement in the domain of cognitive sciences (Pachoud and Zalla, 2000), the debate on consciousness raises a number of questions for modelling urban phenomena as well as for the modelling activity itself. On a general ground, it requires to revisit the "mistaken conception that science could only deal with the objective third person phenomena, but not with the inner qualitative subjective experiences that make up our conscious life" (Searle, 2000, p. 97). To deal with consciousness, therefore, brings to the fore a number of questions relating not only to the relationships between the subjective and objective, but also to values, beliefs, ethical judgements, social and moral norms. As these are recognized to be essential drives of human action, both their ontological and epistemological dimensions are likely to play an increasingly important role in orienting modelling activity.

18All the above arguments show that the connections between the modelling process (internal loop) and modelling domain (external loop) do not refer only to the observable but involve several components of both loops, although in a different ways. Modelling as a cognitive mediator therefore becomes an activity, which primarily enables us to establish, reinforce and articulate the various connections likely to exist between the internal and external loops.

19From a sharper perspective, Marney and Tarbert (2000) give further evidence to this point. They emphasise that the purpose of research (and hence scientific progress) is to map the construct-domain hypothetical statements (i.e. the mental constructs used for the purpose of providing organising principles to interpret real world) onto protocol-domain real-world experience. They also claim that this mapping is not just a chain of causes and effects, but can be multiple and manifold.

20To some extent, this advocacy about the potentials of cognitive mediation underlying a modelling activity, parallels the emphasis given in the analysis of cognitive strategies on processes called conceptual blending, i.e. a bridge from abstract models and more tangible material embodiments of cultural models (Hutchins, 1998).

21To reflect these new potentials of modelling activity the scheme of fig.1, thus transforms into the one shown in fig.2.

Figure 2 : Modelling activity as a cognitive mediator

22What are, therefore, the essential features of modelling as a cognitive mediator ?

23We believe that an exhaustive answer to this question would require a deeper research endeavor than that undertaken here. We would suggest, however, that the modelling task and technological interface are major dimensions, which substantially contribute to shape the features of the modelling activity as a cognitive mediator. In particular, as far as the former dimension is concerned, there are two features that are certainly relevant.

  • 4 In this respect, because of the intrinsic irreducibility of complex system, simulation models as a (...)
  • 5 The recent development of simulation in the geographical field and more generally in the social sci (...)

24A first feature, which we will call action oriented, stems directly from the issues in the mainstream of the complexity studies. As Cowan and Pines (1994) put it : "Achieving comprehensive "stereoscopic" views of most complex systems usually demands that they be described in more than three dimensions…"(p. 710). To adopt a cognitive mediation perspective, therefore, entails to consider a multiplicity of system views in a coordinated way. Such an approach involves what Zeleny (1996) has indicated as "relating descriptions of objects into coherent complexes. The relationships among objects are not simply out there to be captured, but are being continually constructed and reconstructed and re-established by the knower" (p.212). When considering the internal loop of the modelling process (see, Fig.1), this action oriented feature, for example, gives us the possibility to reverse the conventional direction of the abstraction process, i.e. moving anti-clockwise, from the observable to system models, mental models then theories. As a result, one major focus of modelling becomes the shared representations of the observable. In this respect, experimentation made possible by models plays a crucial role in connecting system models, individual mental representations and theories4. Theories lose their unique role of "depository" of scientific truths, guiding the formation of individuals’ mental models. They acquire a more pragmatic role, as means for validating models. In mental models, in addition, greater attention is given to awareness and consciousness. As they become more orientated at "cognition", models (of the observable) are necessary means for exploring alternative courses of action, adjusting them and assessing their viability (0ccelli, 2000). This endeavour also implies a general change in the function of the modelling activity, from analytical / procedural versus explorative/ cognitive approach5. Some characteristics of the action-oriented features as far as the classical dimensions of modelling are concerned, i.e. scope, role and function, are summarised in Tab.1.

 

Current modelling activity

Modelling activity as a cognitive mediator

Scope

Problem solving

Problem making. Shared representations for conscious action

Role

Meaningful description. Simplification is advocated

Different descriptive hypotheses. Coordinated redundancy is retained

Function

Analytic/procedural. The most pertinent descriptions are looked for

Explorative/ cognitive. Different descriptive hypotheses are investigated

Table 1 : Action oriented characteristics of modelling as a cognitive mediator

  • 6 Underlying this need are several facts, which are mainly relevant in the external loop of the model (...)
  • 7 Harrison and Livingston (1982) highlighted the nexus between metaphor, myth and models and emphasiz (...)
  • 8 Although we do not have any definitive evidence to support this thesis, the experience gained in re (...)

25A second trait, which we will call meaning oriented, derives from a need of critical revising those contrasting views conventionally held when relating human experience and the external worlds, i.e. subjective and objective approach, tacit and explicit knowledge, local and global knowledge, hard and soft sciences, common sense and scientific knowledge6. Modelling as a cognitive mediator thus implies a sharper focus on the relationships between what in Fig.1 has been indicated system knowledge (in the external loop) and knowledge levels of the observable (in the internal loop). This means that in order to achieve shared representations of the observable shared form-meaning pairs (i.e. a language) are also required. Language is an essential mechanism to both distinguish between relevant objects in an environment and "coordinate our action in a social domain" (Zeleny, 1996, p. 212). Its role in relation to the modelling activity has been rather neglected although some authors (Harrison and Livingston, 1982) addressed the relevance of the nature of language in geographical understanding7. Here we contend that, because its connecting (and interfacing) ability, modelling, as a cognitive mediator can be very effective in making it easier to achieve shared meanings which ultimately can widen the action possibilities for individuals (Hutchins and Hazlehurst, 19958). A few characteristics of the meaning oriented features of modelling as cognitive mediator are summarized in Tab.2.

 

Current modelling activity

Modelling activity as a cognitive mediator

Actors involved

Few particular actors : experts, decision-makers

Many actors : experts, stakeholders, general public

Language

Many languages exist. Only the natural language is universal

A shared lexicon can be dynamically construed by means of the interaction among actors

Table 2 : Meaning oriented characteristics of modelling as a cognitive mediator

Some final remarks. What about urban complexity and model complexity ?

26The main thesis of this paper is that "notwithstanding their manifold impacts on the modelling field, complexity issues are opening new potentials for modelling which so far are still overlooked". These are related to a general structural-cognitive shift, which affects modelling activity as a whole.

27We sustained that a few relevant aspects of novelty resulting from this shift pertain to a new role of modelling which acts as a cognitive mediator, between a so called internal loop, i.e. the conventional steps underlying a process of abstraction, and a so-called external loop, i.e. the general context of a modelling activity.

28As the observable is not the only link between the two loops, modelling as a cognitive mediator, allows us to establish and investigate the various relationships likely to be established between them A few arguments supporting this claim were mentioned, resulting from several issues concerning cognition, knowledge epistemology and consciousness.

29An effort was also made to identify the salient features of modelling as a cognitive mediator. In this respect, the modelling task and technological interface were mentioned as major dimensions in shaping these features. As far as the former dimension is considered, in particular, both the action oriented and meaning oriented features were considered as primarily relevant.

30To deepen our discussion would require a closer look at how these features are deployed in both the modelling domain and modelling process. As a corollary, an analysis of how and by which means the cognitive mediation of the modelling activity is established between the two loops would also be necessary. We contend that the form of this cognitive mediation is not unique but could be instantiated in several ways. To some extent, this reflects the intrinsic creativity of the modelling activity.

  • 9 This ultimately reflects the more basic and encompassing need to organize the different ways for re (...)

31Nonetheless, we realize that if modelling should effectively cope with the complexity of today cities an effort to give a frame of reference to this multiplicity of modelling instances might be worthwhile9. One possible approach to organize this framework might be to focus on the kind of questions which modelling, as a cognitive mediator should address, as in the example shown in Table 3.

 

Abstractions

Mental models

Knowledge levels of the observable

Model of the observable

System knowledge

What are the prevailing conventional approaches and/or cultural models ?

How do we think about our problems ?

What are the relevant information and how they are retrieved ?

What are the meaningful representations ? How do they connect ? How can be shared ?

Geographically social relevant phenomena

What kind of issues is legitimised ? Which ones are still hidden ?

What are the differences in the perceptions of issues ?

To which extent does conventional information affect people awareness in the definition of certain issues ?

What are the most appropriate communication media to share the representations of certain issues ?

Action domain

What is the performance of the currently adopted practices ?

What is the experience (success and failure) of the actions undertaken ?

What did we learn from undertaking certain actions ?

What do we expect from our actions ?

Table 3 : Some questions to be addressed by the modelling activity as a cognitive mediator 

Haut de page

Bibliographie

Ballot G. and Weisbuch G. (eds.) (2000) Applications of Simulations to Social Sciences, Hermes, Paris.

Batty M. (1994) A Chronicle of Scientific Planning. The Anglo-American Modeling Experience, Journal of the American Planning Association, 60, 7-16.

Batty M., Xie Y. (1996) Possible Urban Automata, in Besussi E. and Cecchini A. eds. Artificial Words and Urban Studies, Daest, Convegni n.1, Venezia, 191-220.

Bertuglia C.S, Staricco L. (2000) Complessità, autoorganizzazione, città, Franco Angeli, Milano.

Casti J.L. (1997) Would-be worlds, Wiley, New York.

Casti J.L., Karlqvist A. (eds.) (1986) Complexity, Language, and Life : Mathematical Approaches, Springer, Berlin.

Clark N., Perz-Treio F. and Allen P. (1995) Evolutionary Dynamics and Sustainable Development. A system Approach, Edward Elgar, Aldershot.

Casetti E. (1999) The evolution of Scientific Disciplines, Mathematical Modeling and Human Geography, Geographical Analysis, 31(4), 332-339.

Conte, R., Hegselmann, R. and Terna, P. eds. (1997) Simulating Social Phenomena. Lecture Notes in Mathematical Systems 456, Springer, Berlin.

Couclelis H. (1998) Geocomputation in Context, in P. Longley, S.M. Brooks, R. McDonnell R. and B. MacMillan eds., Geocomputation : A Primer, Wiley, New York, 17-29.

Cowan G.A., Pines D. (1994) From Metaphors to Reality ? in Cowan G.A., Pines D. and Meltzer D. eds., Complexity. Metaphors, Models and Reality, Proceedings Volume in the Santa Fe Institute Studies in the Science of Complexity, Addison Wesley, Reading MA.

Gilbert N. and Troitzsch K.G. (1999) Simulation for the Social Scientist, Open University Press Philadelphia.

Harrison R.T. and Livingston D.N. (1982) Understanding in Geography : Structuring the Subjective, in Herbert D.T. and Johnston R.J. Eds. Geography and the Urban Environment., Wiley, New York, 1-40.

Havelange V. (1998) Le social en débat : cognition ou interprétation, Intellectica, 1-2,9-55.

Hutchins E. (1998) Mental Models as an Instrument for Bounded Rationality, www : cogsci.ucsd.edu/hutchins.

Hutchins E. and Hazlehurst B. (1995) How to invent a lexicon : the development of shared symbols in interaction, in Gilbert N. and Conte R. (eds.) Artificial Societies. The Computer Simulation of Social Life, UCL Press, London, 157-189.

Longley S.M., Brooks. R., McDonnell R. and MacMillan B. eds. (1998) Geocomputation : A Prime, Wiley, New York.

Minski M. (2000) Common sense-Based Interfaces, Communications, 43, 8, 66-74.

Occelli S. (2000) Revisiting the Concept of Accessibility : Some Comments and Research Questions, in Janelle D.G., Hodge D.C., (eds.), Information, Place and Cyberspace, Springer, Berlin, 279-302.

Occelli S. (2001a) Cognition in urban modelling : a preliminary overview, Paper to be presented at the 2001 ARCo meeting, Lyon, 15-16/12/2001.

Occelli S. (2001b) Why modelling : the Cognitive Drive, Paper presented at the INPUT Meeting, Tremiti, 26-29/6/2001.

Occelli S. and Rabino G.A (1999a) The PF.US model for simulating a Post-Fordist urban system. Proceedings of the 6th Conference CUPUM, Venice, September 8-11, 1999.

Occelli S. and Rabino G.A (1999b) Rationality and Creativity in Urban Modelling, Urbanistica, 113, 2000, 25-27.

Pattee H. (1986) Universal principles of measurement and language functions in evolving systems, in Casti J.L., Karlqvist A. (eds.) Complexity, Language, and Life : Mathematical Approaches, Springer, Berlin, 268-281.

Ryan T. (2000) The Role of Simulation Gaming in Policy-Making, Systems Research and Behavioural Science, 17, 359-364.

Searle J.R. (2000) Consciousness, Intellectica, 2, 85-110.

Pachoud B. and Zalla T. (2000) Enjeux et significations du "retour de la coscience" en science cognitives, Intellectica, 2, 9-34.

Vinck D. (2000) Objets intermédiaires comme inscription de la cognition collective, Médiation Technique et Cognition. Cognition situé, individuelle et collective, Septième Ecole d’Eté, ARCo, Bonas, 10-21 Juillet.

Wegener M. (1994) Operational Urban Models. Journal of the American Planning Association, 60, 17-29.

Zeleny M. (1996) Knowledge as a coordination of action, Human System Management, 155, 211-213.

Haut de page

Notes

1 In the Figure, the lower loop, called internal, refers to the conventional steps underlying a process of abstraction and forming the main steps in a classical modelling activity of urban phenomena. This has its roots in the positivistic assumptions held in the mainstream of social sciences.

The following elements are involved :

1. The abstractions : theories (or metaphors) steering our understanding of the observable.

2. The mental models of the observed phenomena, i.e. the representations built by an individual which guide his/her day-to-day action in a spatial environment.

3. The knowledge levels of the observed phenomena, i.e. the windows of observation we adopt in building an explicit model of the observable. This presupposes an interrelated choice concerning both the theoretical and methodological ingredients and operational aspects of the model.

4. The model of the observable, i.e. the shared descriptions of the phenomena that can be understood by communicating individuals.

The circular clockwise links between them show how the encoding /decoding steps of the modelling process typically develop. It proceeds from observation to the formulation of concepts and more formal models of it (the so called encoding process), and terminates referring back these latter to the observed reality (through the so called decoding process). Encoding and decoding, together, define the ‘cognitive contents’ of a modelling activity.

The upper (external) loop refers to the modelling domain, i.e., the historical and socio-cultural domain in which the process of abstraction occurs. It reminds us that modelling activity has to confront itself with the urban issues it aims to address, as well as with the socio-cultural and technological context in which it is developed. The following elements are emphasised in fig.1 :

1. The system knowledge. This represents the overall information level, i.e. that information embodied in both the theoretical body of a discipline and folk knowledge, which characterise a certain system in a given historical period.
2. The socially relevant urban phenomena. This reflects those societal issues taking to the fore in a certain historical period, i.e. equity, pollution, sustainability, which affect the collective awareness of processes and phenomena in geographical space.
3. The action domain. This relates to the set of actions (i.e. urban behaviours and policy measures) agents (individuals, firms and organizations) decide to implement to purposely modify their behaviours thus affecting their social, economic, spatial and natural environment, and ultimately the observable.

2 A few aspects of this discussion as far as the spatial analysis is concerned are dealt with in Casetti (1999).

3 Searle (2000) summarizes a few salient traits of the current debate about consciousness : ’Consciousness is a biological phenomenon like any other. It consists of inner qualitative subjective states of perceiving, feeling and thinking. Its essential feature is unified qualitative subjectivity…. Consciousness differs from other biological phenomena in that it has first person ontology. But ontological subjectivity does not prevent us from having epistemic objectivity’ (p.98).

4 In this respect, because of the intrinsic irreducibility of complex system, simulation models as a general analytic approach underlying modelling, enables us to go through the different knowledge levels associated with complex systems, thus empowering our capacity of understanding (Ocelli, 2001b)

5 The recent development of simulation in the geographical field and more generally in the social sciences can be primarily associated to this change (see Conte, Hegselmann and Terna, 1997, Gilbert and Troitzsch, 1999, Ballot and Weisbuch eds. 2000). The potentials of simulations are being acknowledged also the in policy making context, where, because of the increasing complexity, it gives the possibility to develop ‘an ability to appreciate wholes and gain system awareness of the issues surrounding policies’ (Ryan, 2000, p.359)

6 Underlying this need are several facts, which are mainly relevant in the external loop of the modelling activity (Occelli and Rabino, 1999a, i.e. the increasing complexity of the policy making context, the changing role of the observer (i.e. the expert or modeller), whose role as a ‘maven’ no longer holds and a greater awareness of the importance of experience and learning in increasing the capacity of the individual to steer his/her own course of life

7 Harrison and Livingston (1982) highlighted the nexus between metaphor, myth and models and emphasized how this nexus intertwines with both the psychology-sociology-logic and insight-contextualization-explanation. The discussion advocated the centrality of metaphor as the fundamental human means in this connection, i.e. a means that unites reason (i.e. categorization, entailment and inference) and imagination (Harrison and Livingston, 1982, p.30) by means of language.

Pattee (1986) theorized the importance of language for living systems in the concept of semantic closure, i.e. the complementary primitiveness of measurement and linguistic meaning.

8 Although we do not have any definitive evidence to support this thesis, the experience gained in recent model applications (Occelli, 2001b), suggests that modelling plays a role in :

Enabling us to navigate among different conceptual constructs, i.e. geographical metaphors, making it possible to establish a link between certain of them.

Allowing us to connect conventional knowledge, i.e. that incorporated in the mainstream geographical metaphors, with ordinary knowledge, i.e. that of common sense as experienced in real-life situations. This points to a kind of isomorphism, which is perhaps unusual in conventional approaches, but which however, can provide a common platform for ‘objective’ understanding and ‘subjective’ comprehension capable of significantly improving the communication between experts and general public.

9 This ultimately reflects the more basic and encompassing need to organize the different ways for representing human knowledge (Minski, 2000).

Haut de page

Pour citer cet article

Référence électronique

Sylvie Occelli, « Facing urban complexity: towards cognitive modelling. Part 1. Modelling as a cognitive mediator  », Cybergeo: European Journal of Geography [En ligne], Dossiers, document 209, mis en ligne le 19 mars 2002, consulté le 28 mars 2024. URL : http://journals.openedition.org/cybergeo/4179 ; DOI : https://doi.org/10.4000/cybergeo.4179

Haut de page

Auteur

Sylvie Occelli

IRES - Istituto di Ricerche Economico Sociali del Piemonte, Via Nizza 18, 10125 Turin, Italy Tel : ++39/11/6666462
occelli@ires.piemonte.it

Articles du même auteur

Haut de page

Droits d’auteur

CC-BY-4.0

Le texte seul est utilisable sous licence CC BY 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search